subject:"\[Wikimedia\-l\] Paid translation"

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬

2018-02-28 1:03 GMT+02:00 Tim Landscheidt :

> Then of course there is the more fundamental problem: If
> those 100,000 monolingual speakers do not speak other lan-
> guages, have no access to encyclopedias, etc., how do they
> interact with a computer now, which web sites do they visit,
> etc.?
>

Quite possibly, they don't visit any websites.

Can Wikipedia be a first website in a given language? Of course.

Who if not Wikipedia? In a lot of languages, the first, and sometimes the
only written work is a translation of the Bible or of the UDHR. (Reminder:
The Bible was the first work that was published in a lot of European
languages, too.) These are usually made by some kind of a funded initiative
that comes from religious or human rights organizations. Why shouldn't it
be a translation of 10,000 Wikipedia articles? Why shouldn't it be an
initiative from Wikimedia or another educational organization?


> I just have a very hard time to imagine a community of
> 100,000 people under those circumstances who are only held
> back by not having access to a Wikipedia.  On the contrary,
> this reminds me very much of traditional development prac-
> tices where third world countries always seem to urgently
> need to buy what first world countries have to sell.  IMHO,
> there is a considerable risk that this creates unhealthy de-
> pendencies.
>

Hey, if people don't want it, they don't have to read it, but I suspect
that if you *let* people read useful information about geography, medicine,
public policy, economics, etc., they will use it.

But in very simplified terms, I see it as a competition between UN, JW,
Facebook, and Wikimedia, and Wikimedia is hardly even participating. UN is
a fine organization, but not very useful in people's daily life. Religious
materials' contribution to development of publishing and literacy
throughout history can't be denied, but the usefulness of their content can
be questioned. Facebook is useful to a lot of people, and it can be
localized easily, but it would be kind of depressing if that's the only
thing that people do in their language. And Facebook is very actively
trying to reach to the farthest corners of the world and get people
connected.

And this leaves Wikimedia, which is hardly doing anything proactive to get
its materials *actually* written in more languages. We are making
*technologies* for translation—Wikidata, Content Translation, and more—and
they are used by thousands of translators to write in dozens of languages,
but we are not doing anything proactive to expand the coverage of languages
beyond the usual suspects: the 70 or so languages that John Erling
mentioned in the email that started this thread. The ~70 big languages take
care of themselves. We've been saying that the rest of the languages can
take care of themselves, but that is naïve.
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-28 1:25 GMT+02:00 James Salsman :

> > I was not trying to say that everybody
> > should learn English. The point I was
> > trying to make there is that knowing
> > English is a privilege and that it is easy
> > to not notice it.
>
> I agree with that, too. How is teaching language different relative to
> the Foundation Mission than teaching subjects of encyclopedia
> articles?
>
>
We are not *teaching* encyclopedia articles. We are making it possible to
write them and to read them. It is not the same thing as teaching subjects.

Should we do teaching? Maybe, but since it's different from making it
possible to write and read, I'm afraid it would be losing focus.

Is there anything bad about teaching languages? Of course not. It's great.
I'm just not sure that it's the right thing for Wikimedia to do, when
Wikimedia should be busy getting even better at its main thing: wiki
articles.

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Gnangarra

A person language is a key part of their culture, their knowledge, and
their identity to truly understand a concept its best shared in its
original language.  Since our goal is to freely share the sum the all
knowledge we should be endeavouring to encourage every culture to use its
own language.   Indigenous languages have and continue to be suppressed by
the colonial languages making any decision to deny a language project, or
to translate an article based on one written in another language is
political decision that has greater impact.   Wikipedias have by the very
nature of what we have created become to be seen as part of a
languages(cultures) recognition and online identity.

Tim its not hard to imagine a community of 100,000 who are held back
because there is no Wikipedia in their language when you look at how much
Wikimedia projects are now at the centre of knowledge systems on the web.
 For the last four years I have been working with the Noongar community to
establish a Wikipedia in Noongar. The noongar language is widely used
within English language here in Western Australia such is impact that we
have over 2 million people who use noongar in their daily lives yet it gets
called Australian english.  The influence of noongar goes beyond the words
and permeates through the Western Australian culture to understand that
impact one needs to be able to access that knowledge.

So how do those people who are monolingual interact with computers at the
moment, its really quite simple they dont computer literacy in Indigenous
communities is well behind that of the colonial based language communities
in the same country.  In the process of reaching out for that knowledge we
need to ensure we do more than just take.

On 28 February 2018 at 07:25, James Salsman  wrote:

> > I was not trying to say that everybody
> > should learn English. The point I was
> > trying to make there is that knowing
> > English is a privilege and that it is easy
> > to not notice it.
>
> I agree with that, too. How is teaching language different relative to
> the Foundation Mission than teaching subjects of encyclopedia
> articles?
>
>
> On Tue, Feb 27, 2018 at 1:11 PM, Amir E. Aharoni
>  wrote:
> > 2018-02-27 21:23 GMT+02:00 James Salsman :
> >
> >> Languages are taught by authoritative dictionaries (after people, and
> >> ahead of almost all other similar reference books.)
> >>
> >
> > ... Yeah, and building an authoritative dictionary is considerably harder
> > than building a (de facto) authoritative encyclopedia. Despite, I have
> > enormous respect for Wiktionary, and great (great!) hopes about Lexical
> > Wikidata.
> >
> >
> >> Wiktionary has multiple teaching functions whether we want it to or
> >> not: https://curve.coventry.ac.uk/open/items/efe362e1-fe80-4c90-
> >> bc1e-4ab2d9bbae20/1/
> >>
> >
> > Why not :)
> >
> >
> >> Amir, you know it would not be losing focus because of what you said
> >> in your talk at https://www.youtube.com/watch?v=P_xJaqQV71s
> >>
> >
> > Um... thanks for the publicity :)
> >
> > But no, that's not what I said. I was not trying to say that everybody
> > should learn English. The point I was trying to make there is that
> knowing
> > English is a privilege and that it is easy to not notice it. Of course,
> if
> > that point didn't come through, it's my fault.
> >
> > --
> > Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
> > http://aharoni.wordpress.com
> > ‪“We're living in pieces,
> > I want to live in peace.” – T. Moore‬
> > ___
> > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
> > New messages to: Wikimedia-l@lists.wikimedia.org
> > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
>
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
>

-- 
GN.
Noongarpedia: https://incubator.wikimedia.org/wiki/Wp/nys/Main_Page
WMAU: http://www.wikimedia.org.au/wiki/User:Gnangarra
Photo Gallery: http://gnangarra.redbubble.com
Out now: A.Gaynor, P. Newman and P. Jennings (eds.), *Never Again:
Reflections on Environmental Responsibility after Roe 8*, UWAP, 2017.  Order
here

.
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread James Salsman

> I was not trying to say that everybody
> should learn English. The point I was
> trying to make there is that knowing
> English is a privilege and that it is easy
> to not notice it.

I agree with that, too. How is teaching language different relative to
the Foundation Mission than teaching subjects of encyclopedia
articles?


On Tue, Feb 27, 2018 at 1:11 PM, Amir E. Aharoni
 wrote:
> 2018-02-27 21:23 GMT+02:00 James Salsman :
>
>> Languages are taught by authoritative dictionaries (after people, and
>> ahead of almost all other similar reference books.)
>>
>
> ... Yeah, and building an authoritative dictionary is considerably harder
> than building a (de facto) authoritative encyclopedia. Despite, I have
> enormous respect for Wiktionary, and great (great!) hopes about Lexical
> Wikidata.
>
>
>> Wiktionary has multiple teaching functions whether we want it to or
>> not: https://curve.coventry.ac.uk/open/items/efe362e1-fe80-4c90-
>> bc1e-4ab2d9bbae20/1/
>>
>
> Why not :)
>
>
>> Amir, you know it would not be losing focus because of what you said
>> in your talk at https://www.youtube.com/watch?v=P_xJaqQV71s
>>
>
> Um... thanks for the publicity :)
>
> But no, that's not what I said. I was not trying to say that everybody
> should learn English. The point I was trying to make there is that knowing
> English is a privilege and that it is easy to not notice it. Of course, if
> that point didn't come through, it's my fault.
>
> --
> Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
> http://aharoni.wordpress.com
> ‪“We're living in pieces,
> I want to live in peace.” – T. Moore‬
> ___
> Wikimedia-l mailing list, guidelines at: 
> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
> https://meta.wikimedia.org/wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 
> 

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Tim Landscheidt

"Amir E. Aharoni"  wrote:

> […]

> On a more practical and less ideological note, I should note that even
> though I didn't run the numbers, I strongly suspect that translating 10,000
> articles to 100 languages is considerably cheaper than teaching 7 billion
> people English.

Definitely, but your argument was:

| […]

| If there is no substantial Wikipedia in such a language, these people can't
| read Wikipedia in *any language* because they are monolingual. Most likely
| they cannot read any any encyclopedia in any language. They need a
| Wikipedia not in order to preserve the language, but to have access to
| *any* encyclopedic knowledge.

| […]

A large part of humanity *has* access to a reasonably main-
tained Wikipedia in a language they understand, not to speak
of traditional encyclopedias in schools and libraries.

Then of course there is the more fundamental problem: If
those 100,000 monolingual speakers do not speak other lan-
guages, have no access to encyclopedias, etc., how do they
interact with a computer now, which web sites do they visit,
etc.?

I just have a very hard time to imagine a community of
100,000 people under those circumstances who are only held
back by not having access to a Wikipedia.  On the contrary,
this reminds me very much of traditional development prac-
tices where third world countries always seem to urgently
need to buy what first world countries have to sell.  IMHO,
there is a considerable risk that this creates unhealthy de-
pendencies.

Tim

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 21:23 GMT+02:00 James Salsman :

> Languages are taught by authoritative dictionaries (after people, and
> ahead of almost all other similar reference books.)
>

... Yeah, and building an authoritative dictionary is considerably harder
than building a (de facto) authoritative encyclopedia. Despite, I have
enormous respect for Wiktionary, and great (great!) hopes about Lexical
Wikidata.


> Wiktionary has multiple teaching functions whether we want it to or
> not: https://curve.coventry.ac.uk/open/items/efe362e1-fe80-4c90-
> bc1e-4ab2d9bbae20/1/
>

Why not :)


> Amir, you know it would not be losing focus because of what you said
> in your talk at https://www.youtube.com/watch?v=P_xJaqQV71s
>

Um... thanks for the publicity :)

But no, that's not what I said. I was not trying to say that everybody
should learn English. The point I was trying to make there is that knowing
English is a privilege and that it is easy to not notice it. Of course, if
that point didn't come through, it's my fault.

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Jean-Philippe Béland

I don't think it would be losing focus as it fits directly in the mission
of the movement to share the sum of human knowledge, since languages are
knowledge in themselves.

Yes I agree that Wikiversity could be used for that, but this project
really needs support to get to current standards of "online courses", and I
don't see much push in that direction.

JP


On Tue, Feb 27, 2018 at 2:23 PM James Salsman  wrote:

> Languages are taught by authoritative dictionaries (after people, and
> ahead of almost all other similar reference books.)
>
> Wiktionary has multiple teaching functions whether we want it to or
> not:
> https://curve.coventry.ac.uk/open/items/efe362e1-fe80-4c90-bc1e-4ab2d9bbae20/1/
>
> Have you seen how much Wiktionary has been growing in Brazil?
> https://blog.searchmetrics.com/us/2018/02/14/seo-world-rankings-2018/
>
> Amir, you know it would not be losing focus because of what you said
> in your talk at https://www.youtube.com/watch?v=P_xJaqQV71s
>
> Best regards,
> Jim
>
>
> On Tue, Feb 27, 2018 at 11:14 AM, Amir E. Aharoni
>  wrote:
> > Well... Not that teaching languages—big or small—is bad, but wouldn't we
> be
> > losing focus if we got into it?
> >
> > Wikibooks and Wikiversity can theoretically be places for teaching. Are
> > they good at it? Probably not. Should they be made better? Maybe.
> >
> > בתאריך 27 בפבר׳ 2018 19:52,‏ "Jean-Philippe Béland" <
> jpbel...@wikimedia.ca>
> > כתב:
> >
> > Amir,
> >
> > I agree with everything you said, especially that languages are knowledge
> > in themselves, but I must say that Wikimedia is not doing much in an
> effort
> > to teach languages to people. Why isn't there more effort at the WMF or
> as
> > a movement to try to develop a platform to teach languages?
> >
> > Jean-Philippe Béland
> > Vice President and Programs Coordinator, Wikimedia Canada
> > Coordinator, Wikimedians of North American Indigenous Languages User
> Group
> > ___
> > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> > wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> > wiki/Wikimedia-l
> > New messages to: Wikimedia-l@lists.wikimedia.org
> > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> > 
> > ___
> > Wikimedia-l mailing list, guidelines at:
> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and
> https://meta.wikimedia.org/wiki/Wikimedia-l
> > New messages to: Wikimedia-l@lists.wikimedia.org
> > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
>
> ___
> Wikimedia-l mailing list, guidelines at:
> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and
> https://meta.wikimedia.org/wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread James Salsman

Languages are taught by authoritative dictionaries (after people, and
ahead of almost all other similar reference books.)

Wiktionary has multiple teaching functions whether we want it to or
not: 
https://curve.coventry.ac.uk/open/items/efe362e1-fe80-4c90-bc1e-4ab2d9bbae20/1/

Have you seen how much Wiktionary has been growing in Brazil?
https://blog.searchmetrics.com/us/2018/02/14/seo-world-rankings-2018/

Amir, you know it would not be losing focus because of what you said
in your talk at https://www.youtube.com/watch?v=P_xJaqQV71s

Best regards,
Jim


On Tue, Feb 27, 2018 at 11:14 AM, Amir E. Aharoni
 wrote:
> Well... Not that teaching languages—big or small—is bad, but wouldn't we be
> losing focus if we got into it?
>
> Wikibooks and Wikiversity can theoretically be places for teaching. Are
> they good at it? Probably not. Should they be made better? Maybe.
>
> בתאריך 27 בפבר׳ 2018 19:52,‏ "Jean-Philippe Béland" 
> כתב:
>
> Amir,
>
> I agree with everything you said, especially that languages are knowledge
> in themselves, but I must say that Wikimedia is not doing much in an effort
> to teach languages to people. Why isn't there more effort at the WMF or as
> a movement to try to develop a platform to teach languages?
>
> Jean-Philippe Béland
> Vice President and Programs Coordinator, Wikimedia Canada
> Coordinator, Wikimedians of North American Indigenous Languages User Group
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
> ___
> Wikimedia-l mailing list, guidelines at: 
> https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
> https://meta.wikimedia.org/wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 
> 

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

Well... Not that teaching languages—big or small—is bad, but wouldn't we be
losing focus if we got into it?

Wikibooks and Wikiversity can theoretically be places for teaching. Are
they good at it? Probably not. Should they be made better? Maybe.

בתאריך 27 בפבר׳ 2018 19:52,‏ "Jean-Philippe Béland" 
כתב:

Amir,

I agree with everything you said, especially that languages are knowledge
in themselves, but I must say that Wikimedia is not doing much in an effort
to teach languages to people. Why isn't there more effort at the WMF or as
a movement to try to develop a platform to teach languages?

Jean-Philippe Béland
Vice President and Programs Coordinator, Wikimedia Canada
Coordinator, Wikimedians of North American Indigenous Languages User Group
___
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Jean-Philippe Béland

Amir,

I agree with everything you said, especially that languages are knowledge
in themselves, but I must say that Wikimedia is not doing much in an effort
to teach languages to people. Why isn't there more effort at the WMF or as
a movement to try to develop a platform to teach languages?

Jean-Philippe Béland
Vice President and Programs Coordinator, Wikimedia Canada
Coordinator, Wikimedians of North American Indigenous Languages User Group
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Vi to

(This thread is getting terribly interesting)

I generally think Wikipedia should be a strictly non interfering observer
for various aspects, language included. I fear if a wiki tries to set a
model for a language it may be a model which doesn't represent the reality
of that language: small wikis are often monopolized by a few users. That's
not a fault per se but it may introduce a significant bias in linguistic
models used.

About one of Amir's emails I think a "small" Wikipedia edition is sign of a
series of situations, one of the most common of is an endangered language.
While planning should differentiate between endangered and non endangered
language I think most of problems we have to face are related to languages
endangered at various levels.

On a more practical and less ideological note, I should note that even
though I didn't run the numbers, I strongly suspect that translating 10,000
articles to 100 languages is considerably cheaper than teaching 7 billion
people English.

I don't why but I tend to second your suspects :p


Vito

2018-02-27 16:53 GMT+01:00 Peter Southwood <peter.southw...@telkomsa.net>:

> If the people creating the basic encyclopaedic terminology and style in
> the language are native speakers, then it would not be a thing imposed from
> outside. It would be a development within the language, just like it was
> with the languages that already have encyclopaedias. The basic
> encyclopaedic terminology and style in languages that have then also had to
> be created before it existed, it just happened earlier. Living languages
> evolve to deal with the realities of the present. Those which don’t, tend
> to die out as they become less useful. Cheers, Peter
>
> -Original Message-
> From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> Behalf Of Vi to
> Sent: Tuesday, February 27, 2018 1:43 PM
> To: Wikimedia Mailing List
> Subject: Re: [Wikimedia-l] Paid translation
>
> I see Amir's points, which are pretty reasonable, but I fear this would
> suit languages with a significant presence on the web.
>
> Among them I agree with points 1, 3 and 4 while I'm not sure about #2
> "creating basic encyclopedic terminology and style in that language", if we
> want to preserve a language we shouldn't create a thing.
>
> By the way I was wondering my concerns about cultural colonization may be
> addressed -for wikis which has some contents (let's say at least 1000
> articles)- by starting expanding existing articles instead of translating
> new ones. This would solve the problem of choosing what to translate though
> would leave problems about the perspective contents are created.
>
> Vito
>
> 2018-02-27 12:31 GMT+01:00 Amir E. Aharoni <amir.ahar...@mail.huji.ac.il>:
>
> > 2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz <
> > psychosl...@culture-libre.org>:
> >
> > >
> > >
> > > Le 24/02/2018 à 18:08, Vi to a écrit :
> > >
> > >> *finally I think paid translators would hardly turn into stable
> > >> Wikipedians.
> > >>
> > >> I think this misses an important point that is, we don't need the
> > initial
> > > translator to turn into a sustaining editor, we need the article to
> > evolve
> > > with call to action incentives. And articles which don't exist at
> > > all – even as a stub – or don't meet an audience of potential
> > > contributors will never catch such an evolving cycle.
> >
> >
> > This is one of the issues with what I alluded to in my earlier email
> > in this thread: the privilege that the "big" languages have. It's the
> > privilege of already having other encyclopedias, textbooks, public
> > education, etc., in this language. A lot of languages don't have these
> > things. When you speak a language that has had these things before
> > Wikipedia came along, it's hard to perceive the world like a person
> > who speaks a language that doesn't perceives it.
> >
> > If you define the purpose of paying somebody to translate as "turning
> > the paid translator" into a sustaining editor, then this is indeed
> > likely to fail.
> >
> > But if you define the purpose differently, it may succeed. For
> > example, you may define the purpose as one or more of the following:
> > * Demonstrating that it's possible to write an encyclopedia in that
> > language
> > * Creating basic encyclopedic terminology and style in that language
> > * Creating a bunch of basic articles that would appear in
> > interlanguage links in Wikipedias from bigger languages (English,
> > French, etc.)
> > * Creating a bunch of basic artic

Re: [Wikimedia-l] Paid translation

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬

2018-02-27 18:04 GMT+02:00 Tim Landscheidt :

> "Amir E. Aharoni"  wrote:
>
> > […]
>
> > Sometimes it is, but there is something much bigger: There are many
> > languages that
> > 1. are alive in speech (and possibly in writing)
> > 2. are not in danger of extinction
> > 3. have a large number of monolingual speakers (let's say 100,000+)
>
> > If there is no substantial Wikipedia in such a language, these people
> can't
> > read Wikipedia in *any language* because they are monolingual. Most
> likely
> > they cannot read any any encyclopedia in any language. They need a
> > Wikipedia not in order to preserve the language, but to have access to
> > *any* encyclopedic knowledge.
>
> > I speak a revitalized language, and I'm very well aware of its history.
> > Language preservation and revitalization are lovely things. But it's not
> > the main point of what Wikimedia does.
>
> "Need a Wikipedia" sounds like a great idea when you are
> selling Wikipedias, but for progress, betterment of humani-
> ty, sustainable development, etc. I think teaching those
> monolingual speakers a second language (for example English)
> is far preferable as it not only enables them to access to a
> few hundred or thousand articles someone paid to have trans-
> lated, but all articles of the English Wikipedia, plus every
> English article, every English book, every English blog, ev-
> ery English video on the InterNet.
>
> It also grows them not only intellectually, but also removes
> economical barriers for trading with other groups.
>
> Tim
>
>
> ... Yeah, it's a tempting thought. Without English we wouldn't be able to
have this conversation, and do thousands of other things.

And yet, that's exactly what we as Wikimedia are not supposed to do, for
reasons that mathieu stumpf guntz suggests: not only what is written in a
language is knowledge; language itself is also knowledge.

On a more practical and less ideological note, I should note that even
though I didn't run the numbers, I strongly suspect that translating 10,000
articles to 100 languages is considerably cheaper than teaching 7 billion
people English.
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Tim Landscheidt

"Amir E. Aharoni"  wrote:

> […]

> Sometimes it is, but there is something much bigger: There are many
> languages that
> 1. are alive in speech (and possibly in writing)
> 2. are not in danger of extinction
> 3. have a large number of monolingual speakers (let's say 100,000+)

> If there is no substantial Wikipedia in such a language, these people can't
> read Wikipedia in *any language* because they are monolingual. Most likely
> they cannot read any any encyclopedia in any language. They need a
> Wikipedia not in order to preserve the language, but to have access to
> *any* encyclopedic knowledge.

> I speak a revitalized language, and I'm very well aware of its history.
> Language preservation and revitalization are lovely things. But it's not
> the main point of what Wikimedia does.

"Need a Wikipedia" sounds like a great idea when you are
selling Wikipedias, but for progress, betterment of humani-
ty, sustainable development, etc. I think teaching those
monolingual speakers a second language (for example English)
is far preferable as it not only enables them to access to a
few hundred or thousand articles someone paid to have trans-
lated, but all articles of the English Wikipedia, plus every
English article, every English book, every English blog, ev-
ery English video on the InterNet.

It also grows them not only intellectually, but also removes
economical barriers for trading with other groups.

Tim

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Peter Southwood

If the people creating the basic encyclopaedic terminology and style in the 
language are native speakers, then it would not be a thing imposed from 
outside. It would be a development within the language, just like it was with 
the languages that already have encyclopaedias. The basic encyclopaedic 
terminology and style in languages that have then also had to be created before 
it existed, it just happened earlier. Living languages evolve to deal with the 
realities of the present. Those which don’t, tend to die out as they become 
less useful. Cheers, Peter

-Original Message-
From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On Behalf Of 
Vi to
Sent: Tuesday, February 27, 2018 1:43 PM
To: Wikimedia Mailing List
Subject: Re: [Wikimedia-l] Paid translation

I see Amir's points, which are pretty reasonable, but I fear this would suit 
languages with a significant presence on the web.

Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating 
basic encyclopedic terminology and style in that language", if we want to 
preserve a language we shouldn't create a thing.

By the way I was wondering my concerns about cultural colonization may be 
addressed -for wikis which has some contents (let's say at least 1000
articles)- by starting expanding existing articles instead of translating new 
ones. This would solve the problem of choosing what to translate though would 
leave problems about the perspective contents are created.

Vito

2018-02-27 12:31 GMT+01:00 Amir E. Aharoni <amir.ahar...@mail.huji.ac.il>:

> 2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz <
> psychosl...@culture-libre.org>:
>
> >
> >
> > Le 24/02/2018 à 18:08, Vi to a écrit :
> >
> >> *finally I think paid translators would hardly turn into stable 
> >> Wikipedians.
> >>
> >> I think this misses an important point that is, we don't need the
> initial
> > translator to turn into a sustaining editor, we need the article to
> evolve
> > with call to action incentives. And articles which don't exist at 
> > all – even as a stub – or don't meet an audience of potential 
> > contributors will never catch such an evolving cycle.
>
>
> This is one of the issues with what I alluded to in my earlier email 
> in this thread: the privilege that the "big" languages have. It's the 
> privilege of already having other encyclopedias, textbooks, public 
> education, etc., in this language. A lot of languages don't have these 
> things. When you speak a language that has had these things before 
> Wikipedia came along, it's hard to perceive the world like a person 
> who speaks a language that doesn't perceives it.
>
> If you define the purpose of paying somebody to translate as "turning 
> the paid translator" into a sustaining editor, then this is indeed 
> likely to fail.
>
> But if you define the purpose differently, it may succeed. For 
> example, you may define the purpose as one or more of the following:
> * Demonstrating that it's possible to write an encyclopedia in that 
> language
> * Creating basic encyclopedic terminology and style in that language
> * Creating a bunch of basic articles that would appear in 
> interlanguage links in Wikipedias from bigger languages (English, 
> French, etc.)
> * Creating a bunch of basic articles that would appear in search 
> results from internet search engines
>
> The existence of these things may bring in people who will become 
> volunteer sustaining editors.
>
> --
> Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי 
> http://aharoni.wordpress.com ‪“We're living in pieces, I want to live 
> in peace.” – T. Moore‬ ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ 
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ 
> wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-requ...@lists.wikimedia.org?subject=unsubscribe>
>
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 
<mailto:wikimedia-l-requ...@lists.wikimedia.org?subject=unsubscribe>


___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 
<mailto:wikimedia-l-requ...@lists.wikimedia.org?subject=unsubscribe>

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread mathieu stumpf guntz

Le 27/02/2018 à 12:42, Vi to a écrit :

I see Amir's points, which are pretty reasonable, but I fear this would
suit languages with a significant presence on the web.

Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating
basic encyclopedic terminology and style in that language", if we want to
preserve a language we shouldn't create a thing.
I think that here the term "preserving" is misinterpreted. It's not
about stuff it to put it in a nothing-should-move-anymore museum. It's
about preserving actual use of diverse language as diachronic phenomena,
ie as evolving objects.

On this regard, even largest language communities are seeing their use
changing at an increasing pace, as recognize institutions like Académie
française (not quite your average neologismophilic neo-punk band).

I think it's also good to recall that there are places where there is
not yet a a high bandwith reliable internet (or internet at all), but
that computer are accessibles. For example Libraries Without Borders[2]
are providing computer boxes, which do include some Wikimedia material
if I'm not mistaken. Although I'm not enough informed on their actions,
but it would interesting to be in contact with them if it's not already
the case. Making encyclopedia shared through travelling USB key would be
surely possible for example, but that just a sketched idea.

On the other hand, should we recall that we are losing language
diversity at an increasing pace?[3] And of course when a language die,
it's whole culture which go with it like a bush medicine engraved in
aboriginal vocabulary.[4] So really it's not about bringing knowledge to
communities with less geopolitcally influence, it's about giving mankind
a chance to loose as few as possible of valuable knowledge by diffusing
it omnidirectly.

[1] Parce qu’il doit être tout à la fois le greffier de l’usage, le
témoin de l’histoire et celui du changement le Dictionnaire de
l’Académie aura donc presque doublé de volume. En consacrant ainsi un
très grand nombre de mots nouveaux, l’Académie répond aux exigences du
temps mais elle se montre fidèle aussi à sa tradition.
http://www.academie-francaise.fr/la-langue-francaise-langue-de-la-modernite-seance-publique-annuelle

[2] https://www.librarieswithoutborders.org/
[3]
http://www.ibtimes.co.uk/unesco-half-worlds-languages-will-disappear-by-2100-1498154

[4] https://en.wikipedia.org/wiki/Bush_medicine

By the way I was wondering my concerns about cultural colonization may be
addressed -for wikis which has some contents (let's say at least 1000
articles)- by starting expanding existing articles instead of translating
new ones. This would solve the problem of choosing what to translate though
would leave problems about the perspective contents are created.

Vito

2018-02-27 12:31 GMT+01:00 Amir E. Aharoni :

2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz <
psychosl...@culture-libre.org>:

Le 24/02/2018 à 18:08, Vi to a écrit :

*finally I think paid translators would hardly turn into stable
Wikipedians.

I think this misses an important point that is, we don't need the

initial

translator to turn into a sustaining editor, we need the article to

evolve

with call to action incentives. And articles which don't exist at all –
even as a stub – or don't meet an audience of potential contributors will
never catch such an evolving cycle.

This is one of the issues with what I alluded to in my earlier email in
this thread: the privilege that the "big" languages have. It's the
privilege of already having other encyclopedias, textbooks, public
education, etc., in this language. A lot of languages don't have these
things. When you speak a language that has had these things before
Wikipedia came along, it's hard to perceive the world like a person who
speaks a language that doesn't perceives it.

If you define the purpose of paying somebody to translate as "turning the
paid translator" into a sustaining editor, then this is indeed likely to
fail.

But if you define the purpose differently, it may succeed. For example, you
may define the purpose as one or more of the following:
* Demonstrating that it's possible to write an encyclopedia in that
language
* Creating basic encyclopedic terminology and style in that language
* Creating a bunch of basic articles that would appear in interlanguage
links in Wikipedias from bigger languages (English, French, etc.)
* Creating a bunch of basic articles that would appear in search results
from internet search engines

The existence of these things may bring in people who will become volunteer
sustaining editors.

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬
___
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines and

Re: [Wikimedia-l] Paid translation

2018-02-27 13:42 GMT+02:00 Vi to :

> I see Amir's points, which are pretty reasonable, but I fear this would
> suit languages with a significant presence on the web.
>
> Among them I agree with points 1, 3 and 4 while I'm not sure about #2
> "creating
> basic encyclopedic terminology and style in that language", if we want to
> preserve a language we shouldn't create a thing.
>
> By the way I was wondering my concerns about cultural colonization may be
> addressed -for wikis which has some contents (let's say at least 1000
> articles)- by starting expanding existing articles instead of translating
> new ones. This would solve the problem of choosing what to translate though
> would leave problems about the perspective contents are created.
>
> Vito
>

It's a very common mistake to think that the purpose of Wikipedias in
"small" languages is language preservation or revitalization.

Sometimes it is, but there is something much bigger: There are many
languages that
1. are alive in speech (and possibly in writing)
2. are not in danger of extinction
3. have a large number of monolingual speakers (let's say 100,000+)

If there is no substantial Wikipedia in such a language, these people can't
read Wikipedia in *any language* because they are monolingual. Most likely
they cannot read any any encyclopedia in any language. They need a
Wikipedia not in order to preserve the language, but to have access to
*any* encyclopedic knowledge.

I speak a revitalized language, and I'm very well aware of its history.
Language preservation and revitalization are lovely things. But it's not
the main point of what Wikimedia does.

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Vi to

I see Amir's points, which are pretty reasonable, but I fear this would
suit languages with a significant presence on the web.

Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating
basic encyclopedic terminology and style in that language", if we want to
preserve a language we shouldn't create a thing.

By the way I was wondering my concerns about cultural colonization may be
addressed -for wikis which has some contents (let's say at least 1000
articles)- by starting expanding existing articles instead of translating
new ones. This would solve the problem of choosing what to translate though
would leave problems about the perspective contents are created.

Vito

2018-02-27 12:31 GMT+01:00 Amir E. Aharoni :

> 2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz <
> psychosl...@culture-libre.org>:
>
> >
> >
> > Le 24/02/2018 à 18:08, Vi to a écrit :
> >
> >> *finally I think paid translators would hardly turn into stable
> >> Wikipedians.
> >>
> >> I think this misses an important point that is, we don't need the
> initial
> > translator to turn into a sustaining editor, we need the article to
> evolve
> > with call to action incentives. And articles which don't exist at all –
> > even as a stub – or don't meet an audience of potential contributors will
> > never catch such an evolving cycle.
>
>
> This is one of the issues with what I alluded to in my earlier email in
> this thread: the privilege that the "big" languages have. It's the
> privilege of already having other encyclopedias, textbooks, public
> education, etc., in this language. A lot of languages don't have these
> things. When you speak a language that has had these things before
> Wikipedia came along, it's hard to perceive the world like a person who
> speaks a language that doesn't perceives it.
>
> If you define the purpose of paying somebody to translate as "turning the
> paid translator" into a sustaining editor, then this is indeed likely to
> fail.
>
> But if you define the purpose differently, it may succeed. For example, you
> may define the purpose as one or more of the following:
> * Demonstrating that it's possible to write an encyclopedia in that
> language
> * Creating basic encyclopedic terminology and style in that language
> * Creating a bunch of basic articles that would appear in interlanguage
> links in Wikipedias from bigger languages (English, French, etc.)
> * Creating a bunch of basic articles that would appear in search results
> from internet search engines
>
> The existence of these things may bring in people who will become volunteer
> sustaining editors.
>
> --
> Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
> http://aharoni.wordpress.com
> ‪“We're living in pieces,
> I want to live in peace.” – T. Moore‬
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
>
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz <
psychosl...@culture-libre.org>:

>
>
> Le 24/02/2018 à 18:08, Vi to a écrit :
>
>> *finally I think paid translators would hardly turn into stable
>> Wikipedians.
>>
>> I think this misses an important point that is, we don't need the initial
> translator to turn into a sustaining editor, we need the article to evolve
> with call to action incentives. And articles which don't exist at all –
> even as a stub – or don't meet an audience of potential contributors will
> never catch such an evolving cycle.


This is one of the issues with what I alluded to in my earlier email in
this thread: the privilege that the "big" languages have. It's the
privilege of already having other encyclopedias, textbooks, public
education, etc., in this language. A lot of languages don't have these
things. When you speak a language that has had these things before
Wikipedia came along, it's hard to perceive the world like a person who
speaks a language that doesn't perceives it.

If you define the purpose of paying somebody to translate as "turning the
paid translator" into a sustaining editor, then this is indeed likely to
fail.

But if you define the purpose differently, it may succeed. For example, you
may define the purpose as one or more of the following:
* Demonstrating that it's possible to write an encyclopedia in that language
* Creating basic encyclopedic terminology and style in that language
* Creating a bunch of basic articles that would appear in interlanguage
links in Wikipedias from bigger languages (English, French, etc.)
* Creating a bunch of basic articles that would appear in search results
from internet search engines

The existence of these things may bring in people who will become volunteer
sustaining editors.

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread mathieu stumpf guntz




Le 24/02/2018 à 18:08, Vi to a écrit :

*finally I think paid translators would hardly turn into stable Wikipedians.

I think this misses an important point that is, we don't need the 
initial translator to turn into a sustaining editor, we need the article 
to evolve with call to action incentives. And articles which don't exist 
at all – even as a stub – or don't meet an audience of potential 
contributors will never catch such an evolving cycle.

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Gnangarra

Agree with mathieo it needs to be something driven by the receiving
language community with WMF support rather than something being pushed in
to communities from the WMF or other projects.   Such if the Swahili
community thought that having say medical articles translated was something
it felt was needed then the WMF could support by assisting with tools, and
facilities to make it happen.

On 27 February 2018 at 18:42, mathieu stumpf guntz <
psychosl...@culture-libre.org> wrote:

> I'm not against the idea of paid translation /per se/, but it shouldn't be
> managed by the WMF, should it be only to ensure that it doesn't cross too
> far the line of non-intervention regarding editorial decisions.
>
> Debate can go on to which level it stands with this line, but to my mind
> WMF always have been mainly about hosting works, not about what will be
> published by who under which (non-)remunerated conditions. I think that it
> is important that it stay so for example due to legal reasons regarding
> responsibility of what is stated in this works.
>
> From this perspective, it would be probably better to have locale
> collective initiatives which decide what seems the more important to be
> translated and means to achieve them, should it be through paid editing
> with money coming from the said collective itself. Directly financing that
> kind of initiative would blur the line of the hosting position I think. But
> giving visibility to this kind of locale fund raising initiatives could be
> a donation in kind that would be maybe less problematic, wouldn't it?
>
>
>
> Le 24/02/2018 à 13:51, John Erling Blad a écrit :
>
>> This discussion is going to be fun! =D
>>
>> A little more than seventy Wikipedia-projects has more than 65k articles,
>> the remaining two hundred or so are pretty small.
>>
>> What if a base set of articles were opened for paid translators? There are
>> several lists of such base sets. We have both the thousand articles from
>> "List of articles every Wikipedia should have"[1] and and the ten thousand
>> articles from the expanded list[2].
>>
>> Lets say verified good translators was paid about $0.01 per word (about $1
>> for a 1k-article) for translating one of those articles into another
>> language, with perhaps a higher pay for contributors in high-cost
>> countries. The pay would also have to be higher for languages that lacks
>> good translation tools.
>>
>> I believe this would be an _enabling_ activity for the communities, as
>> without a base set of articles it won't be possible to build a community
>> at
>> all. By not paying for new articles, and only translating well-referenced
>> articles, some of the disputes in the communities could be avoided.
>> Perhaps
>> we should also identify good source articles, that would be a help.
>> Translated articles should be above some minimum size, but they does not
>> have to be full translations of the source article.
>>
>> A real problem is that our existing lists of good articles other projects
>> should have is pretty much biased towards Western World, so they need a
>> lot
>> of adjustments. Perhaps such a project would identify our inherit bias?
>>
>> [1]
>> https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip
>> edia_should_have
>> [2]
>> https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip
>> edia_should_have/Expanded
>> ___
>> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik
>> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik
>> i/Wikimedia-l
>> New messages to: Wikimedia-l@lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
>> 
>>
>
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik
> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
>



-- 
GN.
Noongarpedia: https://incubator.wikimedia.org/wiki/Wp/nys/Main_Page
WMAU: http://www.wikimedia.org.au/wiki/User:Gnangarra
Photo Gallery: http://gnangarra.redbubble.com
Out now: A.Gaynor, P. Newman and P. Jennings (eds.), *Never Again:
Reflections on Environmental Responsibility after Roe 8*, UWAP, 2017.  Order
here

.
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread mathieu stumpf guntz

I'm not against the idea of paid translation /per se/, but it shouldn't 
be managed by the WMF, should it be only to ensure that it doesn't cross 
too far the line of non-intervention regarding editorial decisions.


Debate can go on to which level it stands with this line, but to my mind 
WMF always have been mainly about hosting works, not about what will be 
published by who under which (non-)remunerated conditions. I think that 
it is important that it stay so for example due to legal reasons 
regarding responsibility of what is stated in this works.


From this perspective, it would be probably better to have locale 
collective initiatives which decide what seems the more important to be 
translated and means to achieve them, should it be through paid editing 
with money coming from the said collective itself. Directly financing 
that kind of initiative would blur the line of the hosting position I 
think. But giving visibility to this kind of locale fund raising 
initiatives could be a donation in kind that would be maybe less 
problematic, wouldn't it?



Le 24/02/2018 à 13:51, John Erling Blad a écrit :

This discussion is going to be fun! =D

A little more than seventy Wikipedia-projects has more than 65k articles,
the remaining two hundred or so are pretty small.

What if a base set of articles were opened for paid translators? There are
several lists of such base sets. We have both the thousand articles from
"List of articles every Wikipedia should have"[1] and and the ten thousand
articles from the expanded list[2].

Lets say verified good translators was paid about $0.01 per word (about $1
for a 1k-article) for translating one of those articles into another
language, with perhaps a higher pay for contributors in high-cost
countries. The pay would also have to be higher for languages that lacks
good translation tools.

I believe this would be an _enabling_ activity for the communities, as
without a base set of articles it won't be possible to build a community at
all. By not paying for new articles, and only translating well-referenced
articles, some of the disputes in the communities could be avoided. Perhaps
we should also identify good source articles, that would be a help.
Translated articles should be above some minimum size, but they does not
have to be full translations of the source article.

A real problem is that our existing lists of good articles other projects
should have is pretty much biased towards Western World, so they need a lot
of adjustments. Perhaps such a project would identify our inherit bias?

[1]
https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have
[2]
https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have/Expanded
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 



___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread Gnangarra

>
>  WWII is not an universal truth. If some small country claim the Nazis
> was
> the good guys, then they are simply wrong.
>

No even thats not entirely true for some countries WWII in Europe was foot
note, for others WWII was the trigger for escaping colonial rule. Languages
related to individual cultures do have different perspectives on events
even on en.wp some FA can tend to have bias to US/UK perspective on
events. There is no one truth for history its all about perspectives,
about the significance of differing events, and the impact those events had.

Even when it comes to less disputed topics like biota there can
differences, take Kangaroo there is referred to an Aboriginal Australian
word but in reality there are over 300 different Australian Languages and
each has their own name for a kangaroo.  They each also have different
knowledge and information simply because of the different environmental
conditions.

Paid translations is not the ideal format, it even has flaws if money is to
be spent then making tools and support projects that enable translations.
Translations risk being interpreted at paternalism with a colonial language
deciding how an indigenous language should speak about a subject.


> 
>
>
On 27 February 2018 at 17:40, John Erling Blad  wrote:

> WWII is not an universal truth. If some small country claim the Nazis was
> the good guys, then they are simply wrong.
>
> Yes there are a lot of projects where information diverge, but usually that
> is because someone added material that somehow seems more appropriate for
> readers in that specific language. Although sometimes the content is really
> wrong, and that happen on all projects.
>
>
> On Mon, Feb 26, 2018 at 12:51 PM, Gerard Meijssen <
> gerard.meijs...@gmail.com
> > wrote:
>
> > Hoi,
> > I have been involved in a translation project with professional
> translators
> > translating featured articles of the English Wikipedia. The choice for
> > featured articles was done because we expected that the content would not
> > be in dispute. We found different. Several of the translated articles
> were
> > not accepted.. one of them was about World War II.
> >
> > I have also toyed with the idea of content that is not available in the
> > language of a Wikipedia (including English). Translation is one solution
> an
> > other solution is generating basic information from the data available at
> > Wikidata. The benefit is not only to our readers; they will at least be
> > informed up to a point and another benefit will be the quality of the
> > Wikipedia involved. One problem that will be fixed is the one of false
> > friends, when red links are linked to Wikidata, the information provided
> > will always be implicitly correct. Another possibility is to provide the
> > text of a sister Wikipedia.
> >
> > We can do a better job by providing the sum of all knowledge that is
> > available to us.
> > Thanks,
> >   GerardM
> >
> > On 25 February 2018 at 15:16, John Erling Blad  wrote:
> >
> > > Sorry, but this does not make sense. The core articles apply globally.
> > > There will although be articles in additions to a list of core
> articles,
> > > but I don't try to advocate any of those lists as the one and only
> list.
> > > Actually I have toyed with an idea of automatically create a list of
> core
> > > articles, and that would identify important articles no matter if they
> > are
> > > from a big western language or a minority language.
> > >
> > > The main problem is NOT that minority languages should have articles
> > about
> > > the major cities and important philosophers, *the main problem is that
> > > minor languages can't get started because they lack content*!
> > >
> > > On Sun, Feb 25, 2018 at 2:41 AM, Vi to  wrote:
> > >
> > > > Cultural appropriation is something different, by "forcing" the
> > contents
> > > in
> > > > a minority language we would actually be at risk of implementing a
> form
> > > of
> > > > "cultural colonialism" which is the opposite of a cultural
> > appropriation.
> > > >
> > > > NOTE: I refer to "the Western" in both cultural and "Wikipedian"
> > sense: I
> > > > mean cultures with a strong presence on the web plus developed and
> > > > flourishing Wikipedia communities.
> > > >
> > > > Helping minority languages with funds/workforce is not bad in my
> > opinion,
> > > > but I think a bottom-up process must be followed, with the "bottom"
> > being
> > > > as closer as possible to relevant linguistic/cultural communities. A
> > > > Wikipedia full of "what the Westerns think is important" in a
> minority
> > > > non-Western language would definitely fail project scopes.
> > > >
> > > > This kind of problem almost does not arise with minority language
> > > > associated to Western cultures since they share the same cultural
> > > > backgrounds: back to my previous example the cultural background of
> > > > Sicilian is substantially

Re: [Wikimedia-l] Paid translation

Indeed. We can all agree that it's OK for a lot of reason to have
differences in content between projects. What these differences are is a
separate discussion.

These differences often come up when discussing translation projects in
Wikipedia, and it's important to recognize them, but it's also important
not to treat them as a blocker or to let them be too much of a distraction.


--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬

2018-02-27 11:40 GMT+02:00 John Erling Blad :

> WWII is not an universal truth. If some small country claim the Nazis was
> the good guys, then they are simply wrong.
>
> Yes there are a lot of projects where information diverge, but usually that
> is because someone added material that somehow seems more appropriate for
> readers in that specific language. Although sometimes the content is really
> wrong, and that happen on all projects.
>
>
> On Mon, Feb 26, 2018 at 12:51 PM, Gerard Meijssen <
> gerard.meijs...@gmail.com
> > wrote:
>
> > Hoi,
> > I have been involved in a translation project with professional
> translators
> > translating featured articles of the English Wikipedia. The choice for
> > featured articles was done because we expected that the content would not
> > be in dispute. We found different. Several of the translated articles
> were
> > not accepted.. one of them was about World War II.
> >
> > I have also toyed with the idea of content that is not available in the
> > language of a Wikipedia (including English). Translation is one solution
> an
> > other solution is generating basic information from the data available at
> > Wikidata. The benefit is not only to our readers; they will at least be
> > informed up to a point and another benefit will be the quality of the
> > Wikipedia involved. One problem that will be fixed is the one of false
> > friends, when red links are linked to Wikidata, the information provided
> > will always be implicitly correct. Another possibility is to provide the
> > text of a sister Wikipedia.
> >
> > We can do a better job by providing the sum of all knowledge that is
> > available to us.
> > Thanks,
> >   GerardM
> >
> > On 25 February 2018 at 15:16, John Erling Blad  wrote:
> >
> > > Sorry, but this does not make sense. The core articles apply globally.
> > > There will although be articles in additions to a list of core
> articles,
> > > but I don't try to advocate any of those lists as the one and only
> list.
> > > Actually I have toyed with an idea of automatically create a list of
> core
> > > articles, and that would identify important articles no matter if they
> > are
> > > from a big western language or a minority language.
> > >
> > > The main problem is NOT that minority languages should have articles
> > about
> > > the major cities and important philosophers, *the main problem is that
> > > minor languages can't get started because they lack content*!
> > >
> > > On Sun, Feb 25, 2018 at 2:41 AM, Vi to  wrote:
> > >
> > > > Cultural appropriation is something different, by "forcing" the
> > contents
> > > in
> > > > a minority language we would actually be at risk of implementing a
> form
> > > of
> > > > "cultural colonialism" which is the opposite of a cultural
> > appropriation.
> > > >
> > > > NOTE: I refer to "the Western" in both cultural and "Wikipedian"
> > sense: I
> > > > mean cultures with a strong presence on the web plus developed and
> > > > flourishing Wikipedia communities.
> > > >
> > > > Helping minority languages with funds/workforce is not bad in my
> > opinion,
> > > > but I think a bottom-up process must be followed, with the "bottom"
> > being
> > > > as closer as possible to relevant linguistic/cultural communities. A
> > > > Wikipedia full of "what the Westerns think is important" in a
> minority
> > > > non-Western language would definitely fail project scopes.
> > > >
> > > > This kind of problem almost does not arise with minority language
> > > > associated to Western cultures since they share the same cultural
> > > > backgrounds: back to my previous example the cultural background of
> > > > Sicilian is substantially equal to Italian one. Still, as I already
> > > wrote,
> > > > wikis in minority languages should focus on a certain aspect of wiki
> > > scope:
> > > > Wiki has roughly two main scopes: 1) sharing knowledge in a certain
> > > > language 2) also preserving the cultural heritage associated with
> > > different
> > > > languages. For languages mainly spoken as first language the "sharing
> > > > knowledge" aspect is predominant, while the second should take
> > precedence
> > > > in languages whose speakers are native speakers of a "bigger"
> language.
> > > >
> > > > Vito
> > > >
> > > > 2018-02-24 22:58 GMT+01:00 John Erling Blad :
> > > >
> > > > > Seems like this is mostly about cultural

Re: [Wikimedia-l] Paid translation

2018-02-27 Thread John Erling Blad

WWII is not an universal truth. If some small country claim the Nazis was
the good guys, then they are simply wrong.

Yes there are a lot of projects where information diverge, but usually that
is because someone added material that somehow seems more appropriate for
readers in that specific language. Although sometimes the content is really
wrong, and that happen on all projects.


On Mon, Feb 26, 2018 at 12:51 PM, Gerard Meijssen  wrote:

> Hoi,
> I have been involved in a translation project with professional translators
> translating featured articles of the English Wikipedia. The choice for
> featured articles was done because we expected that the content would not
> be in dispute. We found different. Several of the translated articles were
> not accepted.. one of them was about World War II.
>
> I have also toyed with the idea of content that is not available in the
> language of a Wikipedia (including English). Translation is one solution an
> other solution is generating basic information from the data available at
> Wikidata. The benefit is not only to our readers; they will at least be
> informed up to a point and another benefit will be the quality of the
> Wikipedia involved. One problem that will be fixed is the one of false
> friends, when red links are linked to Wikidata, the information provided
> will always be implicitly correct. Another possibility is to provide the
> text of a sister Wikipedia.
>
> We can do a better job by providing the sum of all knowledge that is
> available to us.
> Thanks,
>   GerardM
>
> On 25 February 2018 at 15:16, John Erling Blad  wrote:
>
> > Sorry, but this does not make sense. The core articles apply globally.
> > There will although be articles in additions to a list of core articles,
> > but I don't try to advocate any of those lists as the one and only list.
> > Actually I have toyed with an idea of automatically create a list of core
> > articles, and that would identify important articles no matter if they
> are
> > from a big western language or a minority language.
> >
> > The main problem is NOT that minority languages should have articles
> about
> > the major cities and important philosophers, *the main problem is that
> > minor languages can't get started because they lack content*!
> >
> > On Sun, Feb 25, 2018 at 2:41 AM, Vi to  wrote:
> >
> > > Cultural appropriation is something different, by "forcing" the
> contents
> > in
> > > a minority language we would actually be at risk of implementing a form
> > of
> > > "cultural colonialism" which is the opposite of a cultural
> appropriation.
> > >
> > > NOTE: I refer to "the Western" in both cultural and "Wikipedian"
> sense: I
> > > mean cultures with a strong presence on the web plus developed and
> > > flourishing Wikipedia communities.
> > >
> > > Helping minority languages with funds/workforce is not bad in my
> opinion,
> > > but I think a bottom-up process must be followed, with the "bottom"
> being
> > > as closer as possible to relevant linguistic/cultural communities. A
> > > Wikipedia full of "what the Westerns think is important" in a minority
> > > non-Western language would definitely fail project scopes.
> > >
> > > This kind of problem almost does not arise with minority language
> > > associated to Western cultures since they share the same cultural
> > > backgrounds: back to my previous example the cultural background of
> > > Sicilian is substantially equal to Italian one. Still, as I already
> > wrote,
> > > wikis in minority languages should focus on a certain aspect of wiki
> > scope:
> > > Wiki has roughly two main scopes: 1) sharing knowledge in a certain
> > > language 2) also preserving the cultural heritage associated with
> > different
> > > languages. For languages mainly spoken as first language the "sharing
> > > knowledge" aspect is predominant, while the second should take
> precedence
> > > in languages whose speakers are native speakers of a "bigger" language.
> > >
> > > Vito
> > >
> > > 2018-02-24 22:58 GMT+01:00 John Erling Blad :
> > >
> > > > Seems like this is mostly about cultural ownership and appropriation.
> > Not
> > > > sure if it is possible to agree on this.
> > > >
> > > > On Sat, Feb 24, 2018 at 6:08 PM, Vi to 
> wrote:
> > > >
> > > > > I'll reply to the most recent email just for laziness.
> > > > >
> > > > > I'm doubtful for a series of reasons, most of were already
> expressed
> > > in a
> > > > > better way by others:
> > > > > *a remuneration in terms of quantity will weaken the quality of
> > > > > translations unless there's a strong mechanism of quality
> > verification
> > > > > requiring a quantity of resources comparable to translations
> > > themselves;
> > > > > *articles are the result of a long process which reflects cultural
> > > > identity
> > > > > of different communities, I'm not confident with transferring

Re: [Wikimedia-l] Paid translation

2018-02-26 Thread Gerard Meijssen

Hoi,
I have been involved in a translation project with professional translators
translating featured articles of the English Wikipedia. The choice for
featured articles was done because we expected that the content would not
be in dispute. We found different. Several of the translated articles were
not accepted.. one of them was about World War II.

I have also toyed with the idea of content that is not available in the
language of a Wikipedia (including English). Translation is one solution an
other solution is generating basic information from the data available at
Wikidata. The benefit is not only to our readers; they will at least be
informed up to a point and another benefit will be the quality of the
Wikipedia involved. One problem that will be fixed is the one of false
friends, when red links are linked to Wikidata, the information provided
will always be implicitly correct. Another possibility is to provide the
text of a sister Wikipedia.

We can do a better job by providing the sum of all knowledge that is
available to us.
Thanks,
  GerardM

On 25 February 2018 at 15:16, John Erling Blad  wrote:

> Sorry, but this does not make sense. The core articles apply globally.
> There will although be articles in additions to a list of core articles,
> but I don't try to advocate any of those lists as the one and only list.
> Actually I have toyed with an idea of automatically create a list of core
> articles, and that would identify important articles no matter if they are
> from a big western language or a minority language.
>
> The main problem is NOT that minority languages should have articles about
> the major cities and important philosophers, *the main problem is that
> minor languages can't get started because they lack content*!
>
> On Sun, Feb 25, 2018 at 2:41 AM, Vi to  wrote:
>
> > Cultural appropriation is something different, by "forcing" the contents
> in
> > a minority language we would actually be at risk of implementing a form
> of
> > "cultural colonialism" which is the opposite of a cultural appropriation.
> >
> > NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I
> > mean cultures with a strong presence on the web plus developed and
> > flourishing Wikipedia communities.
> >
> > Helping minority languages with funds/workforce is not bad in my opinion,
> > but I think a bottom-up process must be followed, with the "bottom" being
> > as closer as possible to relevant linguistic/cultural communities. A
> > Wikipedia full of "what the Westerns think is important" in a minority
> > non-Western language would definitely fail project scopes.
> >
> > This kind of problem almost does not arise with minority language
> > associated to Western cultures since they share the same cultural
> > backgrounds: back to my previous example the cultural background of
> > Sicilian is substantially equal to Italian one. Still, as I already
> wrote,
> > wikis in minority languages should focus on a certain aspect of wiki
> scope:
> > Wiki has roughly two main scopes: 1) sharing knowledge in a certain
> > language 2) also preserving the cultural heritage associated with
> different
> > languages. For languages mainly spoken as first language the "sharing
> > knowledge" aspect is predominant, while the second should take precedence
> > in languages whose speakers are native speakers of a "bigger" language.
> >
> > Vito
> >
> > 2018-02-24 22:58 GMT+01:00 John Erling Blad :
> >
> > > Seems like this is mostly about cultural ownership and appropriation.
> Not
> > > sure if it is possible to agree on this.
> > >
> > > On Sat, Feb 24, 2018 at 6:08 PM, Vi to  wrote:
> > >
> > > > I'll reply to the most recent email just for laziness.
> > > >
> > > > I'm doubtful for a series of reasons, most of were already expressed
> > in a
> > > > better way by others:
> > > > *a remuneration in terms of quantity will weaken the quality of
> > > > translations unless there's a strong mechanism of quality
> verification
> > > > requiring a quantity of resources comparable to translations
> > themselves;
> > > > *articles are the result of a long process which reflects cultural
> > > identity
> > > > of different communities, I'm not confident with transferring them
> to a
> > > > different "weaker" cultures. My usage of "weaker" adjective only
> > focuses
> > > > about the strength of a cultural presence on the Internet;
> > > > *articles to be translated are at high risk of reflecting the
> cultural
> > > > identity (and biases) of the Western culture;
> > > > *finally I think paid translators would hardly turn into stable
> > > > Wikipedians.
> > > >
> > > > IMHO some paid editing may be better exploited in order to digitalise
> > > texts
> > > > of unrepresented cultures (wikisource) or preserving their
> vocabularies
> > > > (wiktionary).
> > > >
> > > > Also those languages which are secondary for all their

Re: [Wikimedia-l] Paid translation

2018-02-26 Thread James Salsman

> wonder if creating dynamic articles from Wikidata is better
> than creating static articles

Not for years to decades.

https://twitter.com/AustenAllred/status/967842020151603200



On Mon, Feb 26, 2018 at 3:02 AM, John Erling Blad  wrote:
> I wonder if creating dynamic articles from Wikidata is better than creating
> static articles. Because we lack tools for this, it is easier to do this
> offline, and as a consequence we get the static bot-articles.
>
> Den søn. 25. feb. 2018, 16.26 skrev Gabriel Thullen :
>
>> I should have joined in this discussion a little earlier. I work a lot with
>> the French Wikipedia, and we do not just translate articles from English (6
>> million articles) to French (only 2 million articles). The French community
>> is large and active, and provide a unique local perspective on the
>> different articles that are written. And when I say local, I mean that
>> things are seen differently in France than in the French speaking part of
>> Switzerland or Belgium.
>>
>> I think that we are ignoring something very important here: putting it
>> simply, Wikipedia contributors do two things. They add information to the
>> encyclopedia by improving articles or writing new ones, and they curate or
>> check the existing articles. All this talk about machine translation does
>> not address the second aspect of what the volunteer contributors do.
>> This means that we could have hundreds of thousands of articles in a
>> language with  very few active contributors. Will that small community be
>> able to oversee so many articles ?
>>
>> For example, have a look at the list of Wikipedias ordered by number of
>> articles:
>> 1. English - 5,578,081 articles - 138,479 active users - 1,230 admins
>> 2. Cebuano - 5,383,108 articles - 162 active users - 5 admins
>> 3. Swedish - 3,784,331 articles - 2,929 active users - 65 admins
>> 4. German - 2,157,495 articles - 20, 085 active users - 194 admins
>>
>> When I have some time, I will look into different ratios like number or
>> articles/active users or number of articles/number of native language
>> speakers... Now I am not saying that our Swedish friends have abused
>> machine translation of articles, but I definetly that something is not
>> quite right about the Cebuano wiki...
>> Gabe
>>
>>
>> On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten <
>> m...@anderswennersten.se
>> > wrote:
>>
>> > I am very happy to follow this thread as I believe it is addressing a
>> very
>> > relevant issue.
>> >
>> > In my mind we can divide up the different language version into 5
>> > categories:
>> >
>> > 1.Enwp,
>> >
>> > 2.the next 6-7 (de,fr, es,jp,pt,ru..)
>> >
>> > 3.the next 20 or so, where the basic workprocesses are applied
>> >
>> > 4.the next 40-50 which are struggling to generate more input then what is
>> > vandalised
>> >
>> > 5.the rest which in reality is no viable online encyclopedias
>> >
>> > And for me no 1 priority is to accept that there are these categories,
>> and
>> > that what is applicable for cat 1 and 2 is not so for 4 and 5.
>> >
>> > I believe the grant model could easily make room for subsiding good
>> > initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
>> >
>> > And I think it is very presumptuous to start talking of what technique to
>> > use and things like translation. If we open up for creative brainstorming
>> > (among the ones having the need) I think very many other ways can turn
>> up.
>> > Myself I am deeply impressed what you can create using Wikidata as a base
>> > source of info, and being from a version of type 3 I see how much my
>> > homeversion improve content with wikidata created infoboxes
>> >
>> > Anders
>> >
>> >
>> >
>> >
>> > Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
>> >
>> >> This discussion is going to be fun! =D
>> >>
>> >> A little more than seventy Wikipedia-projects has more than 65k
>> articles,
>> >> the remaining two hundred or so are pretty small.
>> >>
>> >> What if a base set of articles were opened for paid translators? There
>> are
>> >> several lists of such base sets. We have both the thousand articles from
>> >> "List of articles every Wikipedia should have"[1] and and the ten
>> thousand
>> >> articles from the expanded list[2].
>> >>
>> >> Lets say verified good translators was paid about $0.01 per word (about
>> $1
>> >> for a 1k-article) for translating one of those articles into another
>> >> language, with perhaps a higher pay for contributors in high-cost
>> >> countries. The pay would also have to be higher for languages that lacks
>> >> good translation tools.
>> >>
>> >> I believe this would be an _enabling_ activity for the communities, as
>> >> without a base set of articles it won't be possible to build a community
>> >> at
>> >> all. By not paying for new articles, and only translating
>> well-referenced
>> >> articles, some of the disputes in the communities could be avoided.
>> >> Perhaps
>> >> we should also

Re: [Wikimedia-l] Paid translation

2018-02-26 Thread John Erling Blad

I wonder if creating dynamic articles from Wikidata is better than creating
static articles. Because we lack tools for this, it is easier to do this
offline, and as a consequence we get the static bot-articles.

Den søn. 25. feb. 2018, 16.26 skrev Gabriel Thullen :

> I should have joined in this discussion a little earlier. I work a lot with
> the French Wikipedia, and we do not just translate articles from English (6
> million articles) to French (only 2 million articles). The French community
> is large and active, and provide a unique local perspective on the
> different articles that are written. And when I say local, I mean that
> things are seen differently in France than in the French speaking part of
> Switzerland or Belgium.
>
> I think that we are ignoring something very important here: putting it
> simply, Wikipedia contributors do two things. They add information to the
> encyclopedia by improving articles or writing new ones, and they curate or
> check the existing articles. All this talk about machine translation does
> not address the second aspect of what the volunteer contributors do.
> This means that we could have hundreds of thousands of articles in a
> language with  very few active contributors. Will that small community be
> able to oversee so many articles ?
>
> For example, have a look at the list of Wikipedias ordered by number of
> articles:
> 1. English - 5,578,081 articles - 138,479 active users - 1,230 admins
> 2. Cebuano - 5,383,108 articles - 162 active users - 5 admins
> 3. Swedish - 3,784,331 articles - 2,929 active users - 65 admins
> 4. German - 2,157,495 articles - 20, 085 active users - 194 admins
>
> When I have some time, I will look into different ratios like number or
> articles/active users or number of articles/number of native language
> speakers... Now I am not saying that our Swedish friends have abused
> machine translation of articles, but I definetly that something is not
> quite right about the Cebuano wiki...
> Gabe
>
>
> On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten <
> m...@anderswennersten.se
> > wrote:
>
> > I am very happy to follow this thread as I believe it is addressing a
> very
> > relevant issue.
> >
> > In my mind we can divide up the different language version into 5
> > categories:
> >
> > 1.Enwp,
> >
> > 2.the next 6-7 (de,fr, es,jp,pt,ru..)
> >
> > 3.the next 20 or so, where the basic workprocesses are applied
> >
> > 4.the next 40-50 which are struggling to generate more input then what is
> > vandalised
> >
> > 5.the rest which in reality is no viable online encyclopedias
> >
> > And for me no 1 priority is to accept that there are these categories,
> and
> > that what is applicable for cat 1 and 2 is not so for 4 and 5.
> >
> > I believe the grant model could easily make room for subsiding good
> > initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
> >
> > And I think it is very presumptuous to start talking of what technique to
> > use and things like translation. If we open up for creative brainstorming
> > (among the ones having the need) I think very many other ways can turn
> up.
> > Myself I am deeply impressed what you can create using Wikidata as a base
> > source of info, and being from a version of type 3 I see how much my
> > homeversion improve content with wikidata created infoboxes
> >
> > Anders
> >
> >
> >
> >
> > Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
> >
> >> This discussion is going to be fun! =D
> >>
> >> A little more than seventy Wikipedia-projects has more than 65k
> articles,
> >> the remaining two hundred or so are pretty small.
> >>
> >> What if a base set of articles were opened for paid translators? There
> are
> >> several lists of such base sets. We have both the thousand articles from
> >> "List of articles every Wikipedia should have"[1] and and the ten
> thousand
> >> articles from the expanded list[2].
> >>
> >> Lets say verified good translators was paid about $0.01 per word (about
> $1
> >> for a 1k-article) for translating one of those articles into another
> >> language, with perhaps a higher pay for contributors in high-cost
> >> countries. The pay would also have to be higher for languages that lacks
> >> good translation tools.
> >>
> >> I believe this would be an _enabling_ activity for the communities, as
> >> without a base set of articles it won't be possible to build a community
> >> at
> >> all. By not paying for new articles, and only translating
> well-referenced
> >> articles, some of the disputes in the communities could be avoided.
> >> Perhaps
> >> we should also identify good source articles, that would be a help.
> >> Translated articles should be above some minimum size, but they does not
> >> have to be full translations of the source article.
> >>
> >> A real problem is that our existing lists of good articles other
> projects
> >> should have is pretty much biased towards Western World, so they need a
> >> lot
> >> of

Re: [Wikimedia-l] Paid translation

Some years ago I tried to figure out whether there was some kind of
mechanism that kept the community sizes at a fixed level. Taking the
population in countries that spoke a specific language, adjusting for
access to internet, and family sizes, made me realize that most stable
projects have 0.2–0.4 ‰ contributors within a normalized language group. If
you then say a stable community consists of 10-20 users, then this creates
a pretty hard limit on the language size. A community of 10-20 users would
imply a language size of 250–1000 000 people. That makes a
Wikipedia-project out of reach for a lot of languages.

(If you somehow limit this group, for example by demanding that only
trained medics should write medical articles, then you draw those medics
from the already limited set, in effect demanding an even larger language
group to make a working community.)

To create the initial interest to make a core encyclopedia is even more
difficult. To bridge that gap, and initiate building of a sustainable
community, a core set of articles are necessary. Perhaps that core set will
be short-lived, and will be replaced with articles that somehow better
reflects the user basis at the local language, but a core set that reflects
some common ground is nonetheless necessary. The capitol of Sweden doesn't
magically disappears in Bengali. The moon doesn't magically turns into
cheese unless in a fairy tale. There are some universal constants that all
languages must adhere to, even in Sicilian Wikipedia a mafioso is a mobster
[1] (someone have messed up the interlinking)

In Norwegian Bokmål we have a few users that has this kind of weird idea
that if something lacks an explicit name, either a word or phrase, then it
should not be described. In my opinion that is nonsense. Some kind of
entity can be described, in any language, no matter if it has a name. We
describe the World as we know it, using words or phrases from the language
to do so. If what we describe has a name, then we use that name. In some
languages that means describing a specific entity is difficult because the
local language has many words and phrases for the same thing. In some other
language it might be difficult because there are no word or phrases to
describe the entity. Neither of those problems arise because the entity is
non-existing in our world, it is just difficult to describe in the specific
language.

Give people knowledge! If they need to somehow clarify that knowledge to
make it more accessible to them, then let them do that! That is why
Wikipedia is editable for everyone!

[1] https://scn.wikipedia.org/wiki/Mafiusu

On Sun, Feb 25, 2018 at 2:07 PM, Amir E. Aharoni <
amir.ahar...@mail.huji.ac.il> wrote:

> I'll start by saying that I'm one of the developers of Content Translation,
> so I'm obviously biased about this topic.
>
> A lot of good points were raised here, but there's one that is not really
> mentioned. If it sounds obvious to you, it's great, but it's not obvious to
> everyone. Here it is:
>
> More successful Wikipedia projects tend to be in languages in which there
> is an established history and tradition of:
> * elementary and higher education where teachers and professors speak to
> students in that language, and in which students write papers in that
> language
> * publishing textbooks
> * publishing encyclopedias
> * publishing dictionaries
> * translating works from (any) other languages, both fiction and reference
>
> People who can read in these developed languages should remember this
> privilege that they have: English, French, Russian, Spanish, German,
> Polish, Italian, Dutch, Czech, Japanese, Norwegian, Hebrew and a few other
> well-developed Wikipedias are written in languages in which good
> encyclopedias had already existed before Wikipedia came along. A Wikipedia
> in these languages didn't make encyclopedic knowledge available in these
> languages; it made encyclopedic knowledge *more easily* available in them.
>
> There are many other things that (probably) affect the development of a
> Wikipedia, such as web connectivity; speakers' population; speakers'
> attitude to the language; work week length (and the remaining free time);
> volunteering culture (or lack thereof); support of common operating systems
> for the language; economic indicators like GDP and HDI in the countries
> where the language is spoken; etc. I'm not aware of research that checks
> the correlation between these aspects and the development of a Wikipedia
> project in a language, but I strongly suspect that it exists for at least
> some of the above. (If anybody reading this is aware of such research, I'll
> be very happy to read it.)
>
> But it's important to go back to the first point here: The existence of
> previous encyclopedias makes it easier for writers in these languages to
> simply start writing. "An encyclopedia" is not a new concept for them. The
> culture around these languages already had well-developed scientific
> terminology

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Gabriel Thullen

I should have joined in this discussion a little earlier. I work a lot with
the French Wikipedia, and we do not just translate articles from English (6
million articles) to French (only 2 million articles). The French community
is large and active, and provide a unique local perspective on the
different articles that are written. And when I say local, I mean that
things are seen differently in France than in the French speaking part of
Switzerland or Belgium.

I think that we are ignoring something very important here: putting it
simply, Wikipedia contributors do two things. They add information to the
encyclopedia by improving articles or writing new ones, and they curate or
check the existing articles. All this talk about machine translation does
not address the second aspect of what the volunteer contributors do.
This means that we could have hundreds of thousands of articles in a
language with  very few active contributors. Will that small community be
able to oversee so many articles ?

For example, have a look at the list of Wikipedias ordered by number of
articles:
1. English - 5,578,081 articles - 138,479 active users - 1,230 admins
2. Cebuano - 5,383,108 articles - 162 active users - 5 admins
3. Swedish - 3,784,331 articles - 2,929 active users - 65 admins
4. German - 2,157,495 articles - 20, 085 active users - 194 admins

When I have some time, I will look into different ratios like number or
articles/active users or number of articles/number of native language
speakers... Now I am not saying that our Swedish friends have abused
machine translation of articles, but I definetly that something is not
quite right about the Cebuano wiki...
Gabe


On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten  wrote:

> I am very happy to follow this thread as I believe it is addressing a very
> relevant issue.
>
> In my mind we can divide up the different language version into 5
> categories:
>
> 1.Enwp,
>
> 2.the next 6-7 (de,fr, es,jp,pt,ru..)
>
> 3.the next 20 or so, where the basic workprocesses are applied
>
> 4.the next 40-50 which are struggling to generate more input then what is
> vandalised
>
> 5.the rest which in reality is no viable online encyclopedias
>
> And for me no 1 priority is to accept that there are these categories, and
> that what is applicable for cat 1 and 2 is not so for 4 and 5.
>
> I believe the grant model could easily make room for subsiding good
> initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
>
> And I think it is very presumptuous to start talking of what technique to
> use and things like translation. If we open up for creative brainstorming
> (among the ones having the need) I think very many other ways can turn up.
> Myself I am deeply impressed what you can create using Wikidata as a base
> source of info, and being from a version of type 3 I see how much my
> homeversion improve content with wikidata created infoboxes
>
> Anders
>
>
>
>
> Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
>
>> This discussion is going to be fun! =D
>>
>> A little more than seventy Wikipedia-projects has more than 65k articles,
>> the remaining two hundred or so are pretty small.
>>
>> What if a base set of articles were opened for paid translators? There are
>> several lists of such base sets. We have both the thousand articles from
>> "List of articles every Wikipedia should have"[1] and and the ten thousand
>> articles from the expanded list[2].
>>
>> Lets say verified good translators was paid about $0.01 per word (about $1
>> for a 1k-article) for translating one of those articles into another
>> language, with perhaps a higher pay for contributors in high-cost
>> countries. The pay would also have to be higher for languages that lacks
>> good translation tools.
>>
>> I believe this would be an _enabling_ activity for the communities, as
>> without a base set of articles it won't be possible to build a community
>> at
>> all. By not paying for new articles, and only translating well-referenced
>> articles, some of the disputes in the communities could be avoided.
>> Perhaps
>> we should also identify good source articles, that would be a help.
>> Translated articles should be above some minimum size, but they does not
>> have to be full translations of the source article.
>>
>> A real problem is that our existing lists of good articles other projects
>> should have is pretty much biased towards Western World, so they need a
>> lot
>> of adjustments. Perhaps such a project would identify our inherit bias?
>>
>> [1]
>> https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip
>> edia_should_have
>> [2]
>> https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip
>> edia_should_have/Expanded
>> ___
>> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik
>> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik
>> i/Wikimedia-l
>> New messages to:

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Anders Wennersten

I am very happy to follow this thread as I believe it is addressing a 
very relevant issue.


In my mind we can divide up the different language version into 5 
categories:


1.Enwp,

2.the next 6-7 (de,fr, es,jp,pt,ru..)

3.the next 20 or so, where the basic workprocesses are applied

4.the next 40-50 which are struggling to generate more input then what 
is vandalised


5.the rest which in reality is no viable online encyclopedias

And for me no 1 priority is to accept that there are these categories, 
and that what is applicable for cat 1 and 2 is not so for 4 and 5.


I believe the grant model could easily make room for subsiding good 
initiatives addressing the problem for cat 4 and 5 (and perhaps 3).


And I think it is very presumptuous to start talking of what technique 
to use and things like translation. If we open up for creative 
brainstorming (among the ones having the need) I think very many other 
ways can turn up. Myself I am deeply impressed what you can create using 
Wikidata as a base source of info, and being from a version of type 3 I 
see how much my homeversion improve content with wikidata created infoboxes


Anders



Den 2018-02-24 kl. 13:51, skrev John Erling Blad:

This discussion is going to be fun! =D

A little more than seventy Wikipedia-projects has more than 65k articles,
the remaining two hundred or so are pretty small.

What if a base set of articles were opened for paid translators? There are
several lists of such base sets. We have both the thousand articles from
"List of articles every Wikipedia should have"[1] and and the ten thousand
articles from the expanded list[2].

Lets say verified good translators was paid about $0.01 per word (about $1
for a 1k-article) for translating one of those articles into another
language, with perhaps a higher pay for contributors in high-cost
countries. The pay would also have to be higher for languages that lacks
good translation tools.

I believe this would be an _enabling_ activity for the communities, as
without a base set of articles it won't be possible to build a community at
all. By not paying for new articles, and only translating well-referenced
articles, some of the disputes in the communities could be avoided. Perhaps
we should also identify good source articles, that would be a help.
Translated articles should be above some minimum size, but they does not
have to be full translations of the source article.

A real problem is that our existing lists of good articles other projects
should have is pretty much biased towards Western World, so they need a lot
of adjustments. Perhaps such a project would identify our inherit bias?

[1]
https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have
[2]
https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have/Expanded
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 




___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Isaac Olatunde

How about training language experts in academic institutions on how to
translate contents from one language Wikipedia (Eg. English wikipedia) to
another? I believe this would be more productive than paying people
directly to contribute or translate contents.

Sometimes in 2016, I discussed with a professor of Yoruba language and Head
of Department of Yoruba language on possible collaboration between the
department and the Yoruba Wikipedia community. We agreed that students
could be assigned to translating high quality articles from the English
Wikipedia to Yoruba Wikipedia and they could be doing these translations as
part of their course work in Yoruba language.

In Nigerian universities for example, Yoruba students take "Àyan Ògbùfò
(the principle of translation) " as part of a course(s) they must pass to
be awarded a degree in Yoruba language.

We could take advantage of this and approach them on possible collaboration.

Today, I had about 30 minutes discussion with  one of the contributors to
the Yoruba language version

of  The watchtower and awake! magazine.

on possible collaboration. He was excited and agreed to be fully involved.

There are institutions and individuals  that would be interested in
translating high quality contents, we just need to reach out to them and
devise a means to get them fully involved.

Regards,

Isaac


On Sat, Feb 24, 2018 at 4:04 PM, James Heilman  wrote:

> I agree with John that it is very difficult to turn a translator into a new
> editor. I also agree with Jean-Philippe that it is key to have involvement
> of the local projects and preferable if they lead the efforts. Of the
> languages we worked in only one explicitly requested not to be involved /
> have translations from TWB.
>
> James
>
> On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> wrote:
>
> > You can turn it around; give added credits for translations from small
> > language projects and into the larger ones, that is a lot more
> interesting
> > than strictly translating from the larger language projects.
> >
> > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > jpbel...@wikimedia.ca
> > > wrote:
> >
> > > I think the request for such projects should come from the concerned
> > > language projects, same for the list of articles. If not, in my simple
> > > opinion, it is a form of coloniasm again.
> > >
> > > Jean-Philippe Béland
> > > Vice President, Wikimedia Canada
> > >
> > >
> > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad 
> > wrote:
> > >
> > > > Should have added that the remaining points are somewhat less
> > interesting
> > > > in this context. Preloading a set of articles is a bad idea, the
> > > > translators should be able to chose for themselves. Articles should
> > also
> > > be
> > > > pretty broad, not very narrow technical or medical, ie vertical
> > articles,
> > > > as the number of editors that can handle those will be pretty small.
> > > >
> > > > In particular: Do not believe you can turn a teanslator into a new
> > > editor!
> > > > You can although turn an existing editor into a translator.
> > > >
> > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad 
> > > > wrote:
> > > >
> > > > > 1) You must start with high quality content and thus all articles
> are
> > > > >> extensively improved before being proposed for translation.
> > > > >
> > > > >
> > > > > Note that to much pressure on "quality" can easily kill the
> project.
> > > > >
> > > > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > more
> > > > >> efficient than handing around word documents. Would love to see
> that
> > > > tool
> > > > >> improved further such as having it support specific lists of
> > articles
> > > > that
> > > > >> are deemed ready for translation by certain groups. Would also
> love
> > > the
> > > > >> tool to have tracking metrics for these types of projects.
> > > > >
> > > > >
> > > > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > > > >
> > > > > 4) We used volunteer translators mostly associated with our partner
> > > > >> Translators Without Borders. One issue we found was that languages
> > in
> > > > >> which
> > > > >> their are lots of translators such as French, Spanish, and Italian
> > > there
> > > > >> is
> > > > >> often already at least some content on many of the topics in
> > question.
> > > > The
> > > > >> issue than becomes integration which needs an expert Wikipedia.
> And
> > > for
> > > > >> languages in which we have little content there are often few
> > > avaliable
> > > > >> volunteers.
> > > > >
> > > > >
> > > > > I used projects below 65k articles as an example, as the chance of
> > > > > competing articles are pretty low.
> > > > >
> > > > > 5) With respect to "paying per word" the

Re: [Wikimedia-l] Paid translation

I like this!
+1000!!

On Sat, Feb 24, 2018 at 11:07 PM, Info WorldUniversity <
i...@worlduniversityandschool.org> wrote:

> John and All,
>
> As a possible complement to this discussion, CC-4 MIT
> OpenCourseWare-centric World University and School seeks to matriculate
> students in all ~200 countries' official/main languages (
> https://wiki.worlduniversityandschool.org/wiki/Languages), and may
> compensate them for work in a number of ways, including translation and
> developing machine translation (and in all 7,099 living languages
> eventually).
>
> World Univ. and Sch. donated ourselves to Wikidata in 2015 for
> co-development, and got a new WUaS Miraheze Mediawiki last year in these
> regards too.
>
> Cheers, Scott
> - https://wiki.worlduniversityandschool.org/wiki/Nation_States
> (each to become a major online University for free CC-4 OCW degrees)
>
>
>
> On Sat, Feb 24, 2018 at 1:49 PM, John Erling Blad <jeb...@gmail.com>
> wrote:
>
> > It is a long time since everyone on these projects were solely
> volunteers.
> > :)
> >
> > On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen <toddmal...@gmail.com>
> wrote:
> >
> > > Yes, and then there's always the question. If he's getting paid, why
> > aren't
> > > I? Why is he getting paid per word of article translated? Why am I not
> > > getting paid per spamvertisement deleted or vandal blocked? Why am I
> not
> > > getting paid for closing discussions that it takes hours of reading
> input
> > > and considering all sides and getting rocks thrown at me no matter
> what I
> > > do? Is that not valuable to the project as well?
> > >
> > > If you want to pay anyone, you better start paying me. I'm okay with
> the
> > > idea of being a volunteer as long as everyone is a volunteer. But if
> you
> > > start paying some people and not me, we're going to have a problem.
> > >
> > > Todd
> > >
> > > On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood <
> > > peter.southw...@telkomsa.net> wrote:
> > >
> > > > Those who pay get to select what is translated.
> > > > Cheers,
> > > > Peter
> > > >
> > > > -Original Message-
> > > > From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org]
> On
> > > > Behalf Of Jean-Philippe Béland
> > > > Sent: 24 February 2018 16:55
> > > > To: Wikimedia Mailing List
> > > > Subject: Re: [Wikimedia-l] Paid translation
> > > >
> > > > I think the request for such projects should come from the concerned
> > > > language projects, same for the list of articles. If not, in my
> simple
> > > > opinion, it is a form of coloniasm again.
> > > >
> > > > Jean-Philippe Béland
> > > > Vice President, Wikimedia Canada
> > > >
> > > >
> > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com>
> > > wrote:
> > > >
> > > > > Should have added that the remaining points are somewhat less
> > > > > interesting in this context. Preloading a set of articles is a bad
> > > > > idea, the translators should be able to chose for themselves.
> > Articles
> > > > > should also be pretty broad, not very narrow technical or medical,
> ie
> > > > > vertical articles, as the number of editors that can handle those
> > will
> > > > be pretty small.
> > > > >
> > > > > In particular: Do not believe you can turn a teanslator into a new
> > > > editor!
> > > > > You can although turn an existing editor into a translator.
> > > > >
> > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
> jeb...@gmail.com>
> > > > > wrote:
> > > > >
> > > > > > 1) You must start with high quality content and thus all articles
> > > > > > are
> > > > > >> extensively improved before being proposed for translation.
> > > > > >
> > > > > >
> > > > > > Note that to much pressure on "quality" can easily kill the
> > project.
> > > > > >
> > > > > > 3) The "Content Translation" tool developed by the WMF made
> efforts
> > > > > > more
> > > > > >> efficient than handing around word documents. Would love to see
> > > > > >> that
> > > &g

Re: [Wikimedia-l] Paid translation

Not sure what you mean by common search terms, but if it is about direct
translation of search terms to get good SEO ranking it is outside what I'm
talking about. That area will vanish completely in a coupe of years.

I've replied about medical articles previously, and why this isn't an area
where it is easy to translate articles.

I agree to both of your bullet points, but note that for point 1, creating
a core set of articles are necessary to attract interest to the project.
There are some weird ideas that these kind of projects emerge from nothing,
but it is a lot of really hard work to start them. Without a base set of
articles the projects does not attract readers, and without readers no
contributors, and without contributors no articles.

The main problem isn't the "cultural colonialism" or "cultural
appropriation" BS, it is lack of articles and thus non-existing communities.

On Sun, Feb 25, 2018 at 12:10 PM, Jonathan Cardy <
werespielchequ...@gmail.com> wrote:

> There were two presentations on paid translation at Wikimania in Gdansk. I
> think that would be 2010? One by Google.org, the other by Google.com
> (charity and corporate wings).
>
> I'm afraid my memory of the event is far from perfect. But some things
> stuck in my mind.
>
> As one would expect, many of the things that could go wrong had gone wrong.
>
> Translators were not recruited from the community and did not understand
> the need to interact with the community.
>
> The aims of the two projects were very different. .org wanted to make
> basic medical info available in a number of languages that were emerging on
> the Internet; .com wanted to give responses to common search terms in those
> languages. Bangla, Tamil and I think Telegu were among them.
>
> One, I think it was Bangla had banned a group of translators, on another
> an irate attendee explained that people who spoke his language did not want
> articles on Hollywood film stars: I suspect that shows a disconnect between
> search engine results and the wishes of wikipedians, it illustrates the
> concerns others have already raised re colonialism, and the difficulty of
> mixing volunteers and paid staff in one project.
>
> No surprise that one of the two projects was much more contentious than
> the other, and not just among Wikipedians on the target project. I can
> understand the frustration of a wikipedian volunteer who realises he is
> fixing for free work that someone else has been paid to do.
>
> I don't know whether the concern about Hollywood was just an inter
> generational thing, whether the people with access tohollywood films were
> representative of the young, or representative of the tech savvy verbally
> bilingual early adopters in that society and unrepresentative of the tens
> of millions in that language who were about to come online.
>
> But I do remember the "common search term" project being much more
> contentious than the medical one.
>
> My experience from here and several other part volunteer communities is
> that there are two golden rules to follow when mixing paid and unpaid staff.
>
> 1 Only pay people to do things that the volunteers want to have happen but
> aren't volunteering to do.
> 2 As much as possible recruit your paid staff from your community of
> volunteers.
>
> Sadly almost all my examples of getting this wrong come from this movement.
>
> Regards
>
> Jonathan / WereSpielChequers
>
>
> > On 24 Feb 2018, at 19:41, wikimedia-l-requ...@lists.wikimedia.org wrote:
> >
> > Send Wikimedia-l mailing list submissions to
> >   wikimedia-l@lists.wikimedia.org
> >
> > To subscribe or unsubscribe via the World Wide Web, visit
> >   https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
> > or, via email, send a message with subject or body 'help' to
> >   wikimedia-l-requ...@lists.wikimedia.org
> >
> > You can reach the person managing the list at
> >   wikimedia-l-ow...@lists.wikimedia.org
> >
> > When replying, please edit your Subject line so it is more specific
> > than "Re: Contents of Wikimedia-l digest..."
> >
> >
> > Today's Topics:
> >
> >  1. Re: Paid translation (Gnangarra)
> >  2. Re: Paid translation (Michael Snow)
> >
> >
> > --
> >
> > Message: 1
> > Date: Sun, 25 Feb 2018 03:05:41 +0800
> > From: Gnangarra <gnanga...@gmail.com>
> > To: Wikimedia Mailing List <wikimedia-l@lists.wikimedia.org>
> > Subject: Re: [Wikimedia-l] Paid translation
> > Message-ID:
> >

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Vi to

Any "global" list reflects (and I fear it will always reflect) the
Weltanschauung of those cultures which are stronger on the web.

I'm deeply concerned about cultures being eaten up by globalization but
attempts to preserve them should take into account the risk of ending up
preserving just "our" view of these cultures.

I also agree with WereSpielChequers' comments about mixing paid and unpaid
editing. What I think it can be done is a system of prizes/contests (maybe
evaluated by paid experts) focused on attracting people on Wikisource and
Wiktionaries, Wikipedia can follow if a critical mass is eventually reached.

Vito

2018-02-25 15:16 GMT+01:00 John Erling Blad :

> Sorry, but this does not make sense. The core articles apply globally.
> There will although be articles in additions to a list of core articles,
> but I don't try to advocate any of those lists as the one and only list.
> Actually I have toyed with an idea of automatically create a list of core
> articles, and that would identify important articles no matter if they are
> from a big western language or a minority language.
>
> The main problem is NOT that minority languages should have articles about
> the major cities and important philosophers, *the main problem is that
> minor languages can't get started because they lack content*!
>
> On Sun, Feb 25, 2018 at 2:41 AM, Vi to  wrote:
>
> > Cultural appropriation is something different, by "forcing" the contents
> in
> > a minority language we would actually be at risk of implementing a form
> of
> > "cultural colonialism" which is the opposite of a cultural appropriation.
> >
> > NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I
> > mean cultures with a strong presence on the web plus developed and
> > flourishing Wikipedia communities.
> >
> > Helping minority languages with funds/workforce is not bad in my opinion,
> > but I think a bottom-up process must be followed, with the "bottom" being
> > as closer as possible to relevant linguistic/cultural communities. A
> > Wikipedia full of "what the Westerns think is important" in a minority
> > non-Western language would definitely fail project scopes.
> >
> > This kind of problem almost does not arise with minority language
> > associated to Western cultures since they share the same cultural
> > backgrounds: back to my previous example the cultural background of
> > Sicilian is substantially equal to Italian one. Still, as I already
> wrote,
> > wikis in minority languages should focus on a certain aspect of wiki
> scope:
> > Wiki has roughly two main scopes: 1) sharing knowledge in a certain
> > language 2) also preserving the cultural heritage associated with
> different
> > languages. For languages mainly spoken as first language the "sharing
> > knowledge" aspect is predominant, while the second should take precedence
> > in languages whose speakers are native speakers of a "bigger" language.
> >
> > Vito
> >
> > 2018-02-24 22:58 GMT+01:00 John Erling Blad :
> >
> > > Seems like this is mostly about cultural ownership and appropriation.
> Not
> > > sure if it is possible to agree on this.
> > >
> > > On Sat, Feb 24, 2018 at 6:08 PM, Vi to  wrote:
> > >
> > > > I'll reply to the most recent email just for laziness.
> > > >
> > > > I'm doubtful for a series of reasons, most of were already expressed
> > in a
> > > > better way by others:
> > > > *a remuneration in terms of quantity will weaken the quality of
> > > > translations unless there's a strong mechanism of quality
> verification
> > > > requiring a quantity of resources comparable to translations
> > themselves;
> > > > *articles are the result of a long process which reflects cultural
> > > identity
> > > > of different communities, I'm not confident with transferring them
> to a
> > > > different "weaker" cultures. My usage of "weaker" adjective only
> > focuses
> > > > about the strength of a cultural presence on the Internet;
> > > > *articles to be translated are at high risk of reflecting the
> cultural
> > > > identity (and biases) of the Western culture;
> > > > *finally I think paid translators would hardly turn into stable
> > > > Wikipedians.
> > > >
> > > > IMHO some paid editing may be better exploited in order to digitalise
> > > texts
> > > > of unrepresented cultures (wikisource) or preserving their
> vocabularies
> > > > (wiktionary).
> > > >
> > > > Also those languages which are secondary for all their speakers
> should
> > be
> > > > dealt with in a different fashion. I, for one, am a native speaker of
> > > > specific variant of Sicilian, Sicilian is a secondary language to any
> > of
> > > > its speakers. Honestly, I'd find pointless to read the biography of
> > > > Leonardo da Vinci in Sicilian while I can find thousands of books
> about
> > > him
> > > > in Italian. Also I find this kind of translation creates a fake
> > > "literary"
> > > >

Re: [Wikimedia-l] Paid translation

Sorry, but this does not make sense. The core articles apply globally.
There will although be articles in additions to a list of core articles,
but I don't try to advocate any of those lists as the one and only list.
Actually I have toyed with an idea of automatically create a list of core
articles, and that would identify important articles no matter if they are
from a big western language or a minority language.

The main problem is NOT that minority languages should have articles about
the major cities and important philosophers, *the main problem is that
minor languages can't get started because they lack content*!

On Sun, Feb 25, 2018 at 2:41 AM, Vi to  wrote:

> Cultural appropriation is something different, by "forcing" the contents in
> a minority language we would actually be at risk of implementing a form of
> "cultural colonialism" which is the opposite of a cultural appropriation.
>
> NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I
> mean cultures with a strong presence on the web plus developed and
> flourishing Wikipedia communities.
>
> Helping minority languages with funds/workforce is not bad in my opinion,
> but I think a bottom-up process must be followed, with the "bottom" being
> as closer as possible to relevant linguistic/cultural communities. A
> Wikipedia full of "what the Westerns think is important" in a minority
> non-Western language would definitely fail project scopes.
>
> This kind of problem almost does not arise with minority language
> associated to Western cultures since they share the same cultural
> backgrounds: back to my previous example the cultural background of
> Sicilian is substantially equal to Italian one. Still, as I already wrote,
> wikis in minority languages should focus on a certain aspect of wiki scope:
> Wiki has roughly two main scopes: 1) sharing knowledge in a certain
> language 2) also preserving the cultural heritage associated with different
> languages. For languages mainly spoken as first language the "sharing
> knowledge" aspect is predominant, while the second should take precedence
> in languages whose speakers are native speakers of a "bigger" language.
>
> Vito
>
> 2018-02-24 22:58 GMT+01:00 John Erling Blad :
>
> > Seems like this is mostly about cultural ownership and appropriation. Not
> > sure if it is possible to agree on this.
> >
> > On Sat, Feb 24, 2018 at 6:08 PM, Vi to  wrote:
> >
> > > I'll reply to the most recent email just for laziness.
> > >
> > > I'm doubtful for a series of reasons, most of were already expressed
> in a
> > > better way by others:
> > > *a remuneration in terms of quantity will weaken the quality of
> > > translations unless there's a strong mechanism of quality verification
> > > requiring a quantity of resources comparable to translations
> themselves;
> > > *articles are the result of a long process which reflects cultural
> > identity
> > > of different communities, I'm not confident with transferring them to a
> > > different "weaker" cultures. My usage of "weaker" adjective only
> focuses
> > > about the strength of a cultural presence on the Internet;
> > > *articles to be translated are at high risk of reflecting the cultural
> > > identity (and biases) of the Western culture;
> > > *finally I think paid translators would hardly turn into stable
> > > Wikipedians.
> > >
> > > IMHO some paid editing may be better exploited in order to digitalise
> > texts
> > > of unrepresented cultures (wikisource) or preserving their vocabularies
> > > (wiktionary).
> > >
> > > Also those languages which are secondary for all their speakers should
> be
> > > dealt with in a different fashion. I, for one, am a native speaker of
> > > specific variant of Sicilian, Sicilian is a secondary language to any
> of
> > > its speakers. Honestly, I'd find pointless to read the biography of
> > > Leonardo da Vinci in Sicilian while I can find thousands of books about
> > him
> > > in Italian. Also I find this kind of translation creates a fake
> > "literary"
> > > language totally detached from reality: there's no "encaustic painting"
> > in
> > > Sicilian, still a Sicilian article about Leonardo will invent one.
> > >
> > > As a general principle we should always collect, rather than create,
> > > knowledge.
> > >
> > > Vito
> > >
> > > 2018-02-24 16:30 GMT+01:00 John Erling Blad :
> > >
> > > > My reply can be read as a bit more harsh than intended, it was
> merely a
> > > > statement about my present experience about translators in general.
> > > >
> > > > The problem with lack of contributors (and translators) in a
> > specialized
> > > > area is that there is a small community, and within this community
> some
> > > > kind of selection is made. Each time a selection is repeated the
> > > remaining
> > > > group shrinks. Specialize the selection sufficiently many times and
> > there
> > > > will be no

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Amir E. Aharoni

I'll start by saying that I'm one of the developers of Content Translation,
so I'm obviously biased about this topic.

A lot of good points were raised here, but there's one that is not really
mentioned. If it sounds obvious to you, it's great, but it's not obvious to
everyone. Here it is:

More successful Wikipedia projects tend to be in languages in which there
is an established history and tradition of:
* elementary and higher education where teachers and professors speak to
students in that language, and in which students write papers in that
language
* publishing textbooks
* publishing encyclopedias
* publishing dictionaries
* translating works from (any) other languages, both fiction and reference

People who can read in these developed languages should remember this
privilege that they have: English, French, Russian, Spanish, German,
Polish, Italian, Dutch, Czech, Japanese, Norwegian, Hebrew and a few other
well-developed Wikipedias are written in languages in which good
encyclopedias had already existed before Wikipedia came along. A Wikipedia
in these languages didn't make encyclopedic knowledge available in these
languages; it made encyclopedic knowledge *more easily* available in them.

There are many other things that (probably) affect the development of a
Wikipedia, such as web connectivity; speakers' population; speakers'
attitude to the language; work week length (and the remaining free time);
volunteering culture (or lack thereof); support of common operating systems
for the language; economic indicators like GDP and HDI in the countries
where the language is spoken; etc. I'm not aware of research that checks
the correlation between these aspects and the development of a Wikipedia
project in a language, but I strongly suspect that it exists for at least
some of the above. (If anybody reading this is aware of such research, I'll
be very happy to read it.)

But it's important to go back to the first point here: The existence of
previous encyclopedias makes it easier for writers in these languages to
simply start writing. "An encyclopedia" is not a new concept for them. The
culture around these languages already had well-developed scientific
terminology and a language style.

When I speak to people who write in Wikipedia in languages of India,
Philippines, and other developing countries, they complain about different
things from people that write in European languages. For example, they very
often complain about the difficulty of writing in an encyclopedic style and
bridging the colloquial language that common people can read and the
standardized versions of the respective languages. This makes me think that
they were standardized in a way that is problematic for *actually* writing
an encyclopedia that would be useful to the general public.

A *massive* project for writing in a language, would create a critical mass
of people who would either make the general public accustomed to reading in
this standard language, or create a new de facto standard. But I guess that
none of the current Wikipedia projects in these languages have this
critical mass of writers.

A translation project, such as what Jon Erling Blad and Lane Rasberry are
suggesting in this thread *may* create such a critical mass. It also needs
bold leaders, who will take it upon themselves Languages that are developed
today went through periods of directed development in the past; Lomonosov
did it for Russian, Diderot did it for French, and so on. This can happen
today as well. (English went through this, too, although I'm not sure which
person should be tied to it: Isaac Newton? Samuel Johnson? John Harris
(Q562265)? Alfred the Great? Probably all of them to some degree.)

I'd even go further and say that I don't agree with Lane when he says that
the WMF cannot and will never pay for content. It sounds like a given thing
to some people, but it isn't. Quite the contrary; it's imaginable that a
careful and thoughtful project of this nature can be carried out by the WMF
itself. "WMF never does this" is not a rule, and it must not be a mental
blocker. I increasingly feel that the WMF is gradually, increasingly
understanding that different languages need different kinds of resources
and support, and this may include paid content creation. (Before you jump
to conclusions: I'm a WMF staff member, but please don't understand from
this that I know about some internal project to do such a thing, or that I
am suggesting to do this. Neither thing is true. I'm just writing a sincere
stream of consciousness about my opinions and feelings, and I might be
wrong about it all.)

That said, it does make more sense to me that organizations other than the
WMF should lead such work, perhaps with some WMF funding, for the sake of
thought diversity if for nothing else. But whether it's paid for by the WMF
directly, by Wikimedia chapters, by thematic interest groups, or by
somebody else is not the main issue. What is important, is that *local*
people and

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Info WorldUniversity

John and All,

As a possible complement to this discussion, CC-4 MIT
OpenCourseWare-centric World University and School seeks to matriculate
students in all ~200 countries' official/main languages (
https://wiki.worlduniversityandschool.org/wiki/Languages), and may
compensate them for work in a number of ways, including translation and
developing machine translation (and in all 7,099 living languages
eventually).

World Univ. and Sch. donated ourselves to Wikidata in 2015 for
co-development, and got a new WUaS Miraheze Mediawiki last year in these
regards too.

Cheers, Scott
- https://wiki.worlduniversityandschool.org/wiki/Nation_States
(each to become a major online University for free CC-4 OCW degrees)



On Sat, Feb 24, 2018 at 1:49 PM, John Erling Blad <jeb...@gmail.com> wrote:

> It is a long time since everyone on these projects were solely volunteers.
> :)
>
> On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen <toddmal...@gmail.com> wrote:
>
> > Yes, and then there's always the question. If he's getting paid, why
> aren't
> > I? Why is he getting paid per word of article translated? Why am I not
> > getting paid per spamvertisement deleted or vandal blocked? Why am I not
> > getting paid for closing discussions that it takes hours of reading input
> > and considering all sides and getting rocks thrown at me no matter what I
> > do? Is that not valuable to the project as well?
> >
> > If you want to pay anyone, you better start paying me. I'm okay with the
> > idea of being a volunteer as long as everyone is a volunteer. But if you
> > start paying some people and not me, we're going to have a problem.
> >
> > Todd
> >
> > On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood <
> > peter.southw...@telkomsa.net> wrote:
> >
> > > Those who pay get to select what is translated.
> > > Cheers,
> > > Peter
> > >
> > > -Original Message-
> > > From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> > > Behalf Of Jean-Philippe Béland
> > > Sent: 24 February 2018 16:55
> > > To: Wikimedia Mailing List
> > > Subject: Re: [Wikimedia-l] Paid translation
> > >
> > > I think the request for such projects should come from the concerned
> > > language projects, same for the list of articles. If not, in my simple
> > > opinion, it is a form of coloniasm again.
> > >
> > > Jean-Philippe Béland
> > > Vice President, Wikimedia Canada
> > >
> > >
> > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com>
> > wrote:
> > >
> > > > Should have added that the remaining points are somewhat less
> > > > interesting in this context. Preloading a set of articles is a bad
> > > > idea, the translators should be able to chose for themselves.
> Articles
> > > > should also be pretty broad, not very narrow technical or medical, ie
> > > > vertical articles, as the number of editors that can handle those
> will
> > > be pretty small.
> > > >
> > > > In particular: Do not believe you can turn a teanslator into a new
> > > editor!
> > > > You can although turn an existing editor into a translator.
> > > >
> > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <jeb...@gmail.com>
> > > > wrote:
> > > >
> > > > > 1) You must start with high quality content and thus all articles
> > > > > are
> > > > >> extensively improved before being proposed for translation.
> > > > >
> > > > >
> > > > > Note that to much pressure on "quality" can easily kill the
> project.
> > > > >
> > > > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > > > more
> > > > >> efficient than handing around word documents. Would love to see
> > > > >> that
> > > > tool
> > > > >> improved further such as having it support specific lists of
> > > > >> articles
> > > > that
> > > > >> are deemed ready for translation by certain groups. Would also
> love
> > > > >> the tool to have tracking metrics for these types of projects.
> > > > >
> > > > >
> > > > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > > > >
> > > > > 4) We used volunteer translators mostly associated with our partner
> > > > >> Translators Witho

Re: [Wikimedia-l] Paid translation

2018-02-25 Thread Jonathan Cardy

There were two presentations on paid translation at Wikimania in Gdansk. I 
think that would be 2010? One by Google.org, the other by Google.com (charity 
and corporate wings).

I'm afraid my memory of the event is far from perfect. But some things stuck in 
my mind.

As one would expect, many of the things that could go wrong had gone wrong. 

Translators were not recruited from the community and did not understand the 
need to interact with the community.

The aims of the two projects were very different. .org wanted to make basic 
medical info available in a number of languages that were emerging on the 
Internet; .com wanted to give responses to common search terms in those 
languages. Bangla, Tamil and I think Telegu were among them.

One, I think it was Bangla had banned a group of translators, on another an 
irate attendee explained that people who spoke his language did not want 
articles on Hollywood film stars: I suspect that shows a disconnect between 
search engine results and the wishes of wikipedians, it illustrates the 
concerns others have already raised re colonialism, and the difficulty of 
mixing volunteers and paid staff in one project. 

No surprise that one of the two projects was much more contentious than the 
other, and not just among Wikipedians on the target project. I can understand 
the frustration of a wikipedian volunteer who realises he is fixing for free 
work that someone else has been paid to do.

I don't know whether the concern about Hollywood was just an inter generational 
thing, whether the people with access tohollywood films were representative of 
the young, or representative of the tech savvy verbally bilingual early 
adopters in that society and unrepresentative of the tens of millions in that 
language who were about to come online.

But I do remember the "common search term" project being much more contentious 
than the medical one.

My experience from here and several other part volunteer communities is that 
there are two golden rules to follow when mixing paid and unpaid staff.

1 Only pay people to do things that the volunteers want to have happen but 
aren't volunteering to do.
2 As much as possible recruit your paid staff from your community of volunteers.

Sadly almost all my examples of getting this wrong come from this movement.

Regards

Jonathan / WereSpielChequers

> On 24 Feb 2018, at 19:41, wikimedia-l-requ...@lists.wikimedia.org wrote:
> 
> Send Wikimedia-l mailing list submissions to
>   wikimedia-l@lists.wikimedia.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
>   https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
> or, via email, send a message with subject or body 'help' to
>   wikimedia-l-requ...@lists.wikimedia.org
> 
> You can reach the person managing the list at
>   wikimedia-l-ow...@lists.wikimedia.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Wikimedia-l digest..."
> 
> 
> Today's Topics:
> 
>  1. Re: Paid translation (Gnangarra)
>  2. Re: Paid translation (Michael Snow)
> 
> 
> --
> 
> Message: 1
> Date: Sun, 25 Feb 2018 03:05:41 +0800
> From: Gnangarra <gnanga...@gmail.com>
> To: Wikimedia Mailing List <wikimedia-l@lists.wikimedia.org>
> Subject: Re: [Wikimedia-l] Paid translation
> Message-ID:
>

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Vi to

Cultural appropriation is something different, by "forcing" the contents in
a minority language we would actually be at risk of implementing a form of
"cultural colonialism" which is the opposite of a cultural appropriation.

NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I
mean cultures with a strong presence on the web plus developed and
flourishing Wikipedia communities.

Helping minority languages with funds/workforce is not bad in my opinion,
but I think a bottom-up process must be followed, with the "bottom" being
as closer as possible to relevant linguistic/cultural communities. A
Wikipedia full of "what the Westerns think is important" in a minority
non-Western language would definitely fail project scopes.

This kind of problem almost does not arise with minority language
associated to Western cultures since they share the same cultural
backgrounds: back to my previous example the cultural background of
Sicilian is substantially equal to Italian one. Still, as I already wrote,
wikis in minority languages should focus on a certain aspect of wiki scope:
Wiki has roughly two main scopes: 1) sharing knowledge in a certain
language 2) also preserving the cultural heritage associated with different
languages. For languages mainly spoken as first language the "sharing
knowledge" aspect is predominant, while the second should take precedence
in languages whose speakers are native speakers of a "bigger" language.

Vito

2018-02-24 22:58 GMT+01:00 John Erling Blad :

> Seems like this is mostly about cultural ownership and appropriation. Not
> sure if it is possible to agree on this.
>
> On Sat, Feb 24, 2018 at 6:08 PM, Vi to  wrote:
>
> > I'll reply to the most recent email just for laziness.
> >
> > I'm doubtful for a series of reasons, most of were already expressed in a
> > better way by others:
> > *a remuneration in terms of quantity will weaken the quality of
> > translations unless there's a strong mechanism of quality verification
> > requiring a quantity of resources comparable to translations themselves;
> > *articles are the result of a long process which reflects cultural
> identity
> > of different communities, I'm not confident with transferring them to a
> > different "weaker" cultures. My usage of "weaker" adjective only focuses
> > about the strength of a cultural presence on the Internet;
> > *articles to be translated are at high risk of reflecting the cultural
> > identity (and biases) of the Western culture;
> > *finally I think paid translators would hardly turn into stable
> > Wikipedians.
> >
> > IMHO some paid editing may be better exploited in order to digitalise
> texts
> > of unrepresented cultures (wikisource) or preserving their vocabularies
> > (wiktionary).
> >
> > Also those languages which are secondary for all their speakers should be
> > dealt with in a different fashion. I, for one, am a native speaker of
> > specific variant of Sicilian, Sicilian is a secondary language to any of
> > its speakers. Honestly, I'd find pointless to read the biography of
> > Leonardo da Vinci in Sicilian while I can find thousands of books about
> him
> > in Italian. Also I find this kind of translation creates a fake
> "literary"
> > language totally detached from reality: there's no "encaustic painting"
> in
> > Sicilian, still a Sicilian article about Leonardo will invent one.
> >
> > As a general principle we should always collect, rather than create,
> > knowledge.
> >
> > Vito
> >
> > 2018-02-24 16:30 GMT+01:00 John Erling Blad :
> >
> > > My reply can be read as a bit more harsh than intended, it was merely a
> > > statement about my present experience about translators in general.
> > >
> > > The problem with lack of contributors (and translators) in a
> specialized
> > > area is that there is a small community, and within this community some
> > > kind of selection is made. Each time a selection is repeated the
> > remaining
> > > group shrinks. Specialize the selection sufficiently many times and
> there
> > > will be no contributors (or translators) left. It is simply a game of
> > > probabilities. Thus, to make such a project work it must have a
> > > sufficiently broad scope for the articles. Articles about public health
> > > services will probably work even for a pretty small language group, but
> > > specialized medical articles might create a problem. But then you find
> > > a retired
> > > orthopedic surgeon like Subas Chandra Rout…
> > >
> > > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman 
> wrote:
> > >
> > > > I agree with John that it is very difficult to turn a translator
> into a
> > > new
> > > > editor. I also agree with Jean-Philippe that it is key to have
> > > involvement
> > > > of the local projects and preferable if they lead the efforts. Of the
> > > > languages we worked in only one explicitly requested not to be
> > involved /
> > > > have translations

Re: [Wikimedia-l] Paid translation

I agree with the last part of Vito's message. For languages where '''all'''
the speakers speak another lingua franca, I think such process does not
have real value. The speakers will always go read in the bigger language
because the article is most likely to be better. The advantages of having
their own Wikipedia is to be able to express knowledge in their own way
according to their own culture.

Jean-Philippe Béland
Vice President, Wikimedia Canada

On Sat, Feb 24, 2018 at 12:09 PM Vi to  wrote:

> I'll reply to the most recent email just for laziness.
>
> I'm doubtful for a series of reasons, most of were already expressed in a
> better way by others:
> *a remuneration in terms of quantity will weaken the quality of
> translations unless there's a strong mechanism of quality verification
> requiring a quantity of resources comparable to translations themselves;
> *articles are the result of a long process which reflects cultural identity
> of different communities, I'm not confident with transferring them to a
> different "weaker" cultures. My usage of "weaker" adjective only focuses
> about the strength of a cultural presence on the Internet;
> *articles to be translated are at high risk of reflecting the cultural
> identity (and biases) of the Western culture;
> *finally I think paid translators would hardly turn into stable
> Wikipedians.
>
> IMHO some paid editing may be better exploited in order to digitalise texts
> of unrepresented cultures (wikisource) or preserving their vocabularies
> (wiktionary).
>
> Also those languages which are secondary for all their speakers should be
> dealt with in a different fashion. I, for one, am a native speaker of
> specific variant of Sicilian, Sicilian is a secondary language to any of
> its speakers. Honestly, I'd find pointless to read the biography of
> Leonardo da Vinci in Sicilian while I can find thousands of books about him
> in Italian. Also I find this kind of translation creates a fake "literary"
> language totally detached from reality: there's no "encaustic painting" in
> Sicilian, still a Sicilian article about Leonardo will invent one.
>
> As a general principle we should always collect, rather than create,
> knowledge.
>
> Vito
>
> 2018-02-24 16:30 GMT+01:00 John Erling Blad :
>
> > My reply can be read as a bit more harsh than intended, it was merely a
> > statement about my present experience about translators in general.
> >
> > The problem with lack of contributors (and translators) in a specialized
> > area is that there is a small community, and within this community some
> > kind of selection is made. Each time a selection is repeated the
> remaining
> > group shrinks. Specialize the selection sufficiently many times and there
> > will be no contributors (or translators) left. It is simply a game of
> > probabilities. Thus, to make such a project work it must have a
> > sufficiently broad scope for the articles. Articles about public health
> > services will probably work even for a pretty small language group, but
> > specialized medical articles might create a problem. But then you find
> > a retired
> > orthopedic surgeon like Subas Chandra Rout…
> >
> > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman  wrote:
> >
> > > I agree with John that it is very difficult to turn a translator into a
> > new
> > > editor. I also agree with Jean-Philippe that it is key to have
> > involvement
> > > of the local projects and preferable if they lead the efforts. Of the
> > > languages we worked in only one explicitly requested not to be
> involved /
> > > have translations from TWB.
> > >
> > > James
> > >
> > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> > > wrote:
> > >
> > > > You can turn it around; give added credits for translations from
> small
> > > > language projects and into the larger ones, that is a lot more
> > > interesting
> > > > than strictly translating from the larger language projects.
> > > >
> > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > > > jpbel...@wikimedia.ca
> > > > > wrote:
> > > >
> > > > > I think the request for such projects should come from the
> concerned
> > > > > language projects, same for the list of articles. If not, in my
> > simple
> > > > > opinion, it is a form of coloniasm again.
> > > > >
> > > > > Jean-Philippe Béland
> > > > > Vice President, Wikimedia Canada
> > > > >
> > > > >
> > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad  >
> > > > wrote:
> > > > >
> > > > > > Should have added that the remaining points are somewhat less
> > > > interesting
> > > > > > in this context. Preloading a set of articles is a bad idea, the
> > > > > > translators should be able to chose for themselves. Articles
> should
> > > > also
> > > > > be
> > > > > > pretty broad, not very narrow technical or medical, ie vertical
> > > > articles,
> > > > > > as the number of editors that can

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Michael Snow


On 2/24/2018 1:53 PM, John Erling Blad wrote:

The source article should meet certain standards, but do not fall in the
trap where the translated articles must themselves be better than some
imagined standard. That would lead to a defunc process.
I'm not saying a translated article must be flawless. But certainly if 
we were paying for translations, it would be appropriate to have some 
level of expectations for the quality of the result. With volunteers, 
any honest effort is encouraged, although if the quality is low enough 
to be worse than the alternative, they can be encouraged to redirect 
that effort more productively.


--Michael Snow

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

Seems like this is mostly about cultural ownership and appropriation. Not
sure if it is possible to agree on this.

On Sat, Feb 24, 2018 at 6:08 PM, Vi to  wrote:

> I'll reply to the most recent email just for laziness.
>
> I'm doubtful for a series of reasons, most of were already expressed in a
> better way by others:
> *a remuneration in terms of quantity will weaken the quality of
> translations unless there's a strong mechanism of quality verification
> requiring a quantity of resources comparable to translations themselves;
> *articles are the result of a long process which reflects cultural identity
> of different communities, I'm not confident with transferring them to a
> different "weaker" cultures. My usage of "weaker" adjective only focuses
> about the strength of a cultural presence on the Internet;
> *articles to be translated are at high risk of reflecting the cultural
> identity (and biases) of the Western culture;
> *finally I think paid translators would hardly turn into stable
> Wikipedians.
>
> IMHO some paid editing may be better exploited in order to digitalise texts
> of unrepresented cultures (wikisource) or preserving their vocabularies
> (wiktionary).
>
> Also those languages which are secondary for all their speakers should be
> dealt with in a different fashion. I, for one, am a native speaker of
> specific variant of Sicilian, Sicilian is a secondary language to any of
> its speakers. Honestly, I'd find pointless to read the biography of
> Leonardo da Vinci in Sicilian while I can find thousands of books about him
> in Italian. Also I find this kind of translation creates a fake "literary"
> language totally detached from reality: there's no "encaustic painting" in
> Sicilian, still a Sicilian article about Leonardo will invent one.
>
> As a general principle we should always collect, rather than create,
> knowledge.
>
> Vito
>
> 2018-02-24 16:30 GMT+01:00 John Erling Blad :
>
> > My reply can be read as a bit more harsh than intended, it was merely a
> > statement about my present experience about translators in general.
> >
> > The problem with lack of contributors (and translators) in a specialized
> > area is that there is a small community, and within this community some
> > kind of selection is made. Each time a selection is repeated the
> remaining
> > group shrinks. Specialize the selection sufficiently many times and there
> > will be no contributors (or translators) left. It is simply a game of
> > probabilities. Thus, to make such a project work it must have a
> > sufficiently broad scope for the articles. Articles about public health
> > services will probably work even for a pretty small language group, but
> > specialized medical articles might create a problem. But then you find
> > a retired
> > orthopedic surgeon like Subas Chandra Rout…
> >
> > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman  wrote:
> >
> > > I agree with John that it is very difficult to turn a translator into a
> > new
> > > editor. I also agree with Jean-Philippe that it is key to have
> > involvement
> > > of the local projects and preferable if they lead the efforts. Of the
> > > languages we worked in only one explicitly requested not to be
> involved /
> > > have translations from TWB.
> > >
> > > James
> > >
> > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> > > wrote:
> > >
> > > > You can turn it around; give added credits for translations from
> small
> > > > language projects and into the larger ones, that is a lot more
> > > interesting
> > > > than strictly translating from the larger language projects.
> > > >
> > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > > > jpbel...@wikimedia.ca
> > > > > wrote:
> > > >
> > > > > I think the request for such projects should come from the
> concerned
> > > > > language projects, same for the list of articles. If not, in my
> > simple
> > > > > opinion, it is a form of coloniasm again.
> > > > >
> > > > > Jean-Philippe Béland
> > > > > Vice President, Wikimedia Canada
> > > > >
> > > > >
> > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad  >
> > > > wrote:
> > > > >
> > > > > > Should have added that the remaining points are somewhat less
> > > > interesting
> > > > > > in this context. Preloading a set of articles is a bad idea, the
> > > > > > translators should be able to chose for themselves. Articles
> should
> > > > also
> > > > > be
> > > > > > pretty broad, not very narrow technical or medical, ie vertical
> > > > articles,
> > > > > > as the number of editors that can handle those will be pretty
> > small.
> > > > > >
> > > > > > In particular: Do not believe you can turn a teanslator into a
> new
> > > > > editor!
> > > > > > You can although turn an existing editor into a translator.
> > > > > >
> > > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
> > jeb...@gmail.com>
> > > > > > wrote:
> >

Re: [Wikimedia-l] Paid translation

This is not the same, and is more like the present grant system.

On Sat, Feb 24, 2018 at 8:05 PM, Gnangarra  wrote:

> this would be a good practical exercise to develop for WiR / WikiEd
> programs in universities where they can engage with International Students
> and local students studying additional languages as means of learning the
> written nuances of the individual languages.  Any funding would be better
> utilised in enabling such programs where the flow on impact is more
> likely{fact} to be lasting.  Though I can see value in using a gift/reward
> system for technically disadvantaged communities like the case presented
> about Swahili .The focus would need to be on basic health, hygiene,
> biology, science topics rather than more social or political topics.
>
> On 25 February 2018 at 01:08, Vi to  wrote:
>
> > I'll reply to the most recent email just for laziness.
> >
> > I'm doubtful for a series of reasons, most of were already expressed in a
> > better way by others:
> > *a remuneration in terms of quantity will weaken the quality of
> > translations unless there's a strong mechanism of quality verification
> > requiring a quantity of resources comparable to translations themselves;
> > *articles are the result of a long process which reflects cultural
> identity
> > of different communities, I'm not confident with transferring them to a
> > different "weaker" cultures. My usage of "weaker" adjective only focuses
> > about the strength of a cultural presence on the Internet;
> > *articles to be translated are at high risk of reflecting the cultural
> > identity (and biases) of the Western culture;
> > *finally I think paid translators would hardly turn into stable
> > Wikipedians.
> >
> > IMHO some paid editing may be better exploited in order to digitalise
> texts
> > of unrepresented cultures (wikisource) or preserving their vocabularies
> > (wiktionary).
> >
> > Also those languages which are secondary for all their speakers should be
> > dealt with in a different fashion. I, for one, am a native speaker of
> > specific variant of Sicilian, Sicilian is a secondary language to any of
> > its speakers. Honestly, I'd find pointless to read the biography of
> > Leonardo da Vinci in Sicilian while I can find thousands of books about
> him
> > in Italian. Also I find this kind of translation creates a fake
> "literary"
> > language totally detached from reality: there's no "encaustic painting"
> in
> > Sicilian, still a Sicilian article about Leonardo will invent one.
> >
> > As a general principle we should always collect, rather than create,
> > knowledge.
> >
> > Vito
> >
> > 2018-02-24 16:30 GMT+01:00 John Erling Blad :
> >
> > > My reply can be read as a bit more harsh than intended, it was merely a
> > > statement about my present experience about translators in general.
> > >
> > > The problem with lack of contributors (and translators) in a
> specialized
> > > area is that there is a small community, and within this community some
> > > kind of selection is made. Each time a selection is repeated the
> > remaining
> > > group shrinks. Specialize the selection sufficiently many times and
> there
> > > will be no contributors (or translators) left. It is simply a game of
> > > probabilities. Thus, to make such a project work it must have a
> > > sufficiently broad scope for the articles. Articles about public health
> > > services will probably work even for a pretty small language group, but
> > > specialized medical articles might create a problem. But then you find
> > > a retired
> > > orthopedic surgeon like Subas Chandra Rout…
> > >
> > > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman 
> wrote:
> > >
> > > > I agree with John that it is very difficult to turn a translator
> into a
> > > new
> > > > editor. I also agree with Jean-Philippe that it is key to have
> > > involvement
> > > > of the local projects and preferable if they lead the efforts. Of the
> > > > languages we worked in only one explicitly requested not to be
> > involved /
> > > > have translations from TWB.
> > > >
> > > > James
> > > >
> > > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> > > > wrote:
> > > >
> > > > > You can turn it around; give added credits for translations from
> > small
> > > > > language projects and into the larger ones, that is a lot more
> > > > interesting
> > > > > than strictly translating from the larger language projects.
> > > > >
> > > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > > > > jpbel...@wikimedia.ca
> > > > > > wrote:
> > > > >
> > > > > > I think the request for such projects should come from the
> > concerned
> > > > > > language projects, same for the list of articles. If not, in my
> > > simple
> > > > > > opinion, it is a form of coloniasm again.
> > > > > >
> > > > > > Jean-Philippe Béland
> > > > > > Vice President, Wikimedia Canada
> > >

Re: [Wikimedia-l] Paid translation

The source article should meet certain standards, but do not fall in the
trap where the translated articles must themselves be better than some
imagined standard. That would lead to a defunc process.

On Sat, Feb 24, 2018 at 8:41 PM, Michael Snow 
wrote:

> I think the experience I've had with translating matches up well with the
> conclusions James has outlined. Even though I'm more likely to translate
> content into English rather than out of English, the principles still hold.
>
> Trying to produce a translation without quality content in the original
> article is a frustrating and pointless exercise for the translator. Unless
> the original meets certain standards, it would be better and easier to
> write the article from scratch in the "destination" language and translate
> it back to the "source" language.
>
> Assuming we have a good article in the original language, I definitely
> encourage translators to use editorial judgment in what they carry over.
> Focusing on the lead section is one possible approach. In general, because
> we are trying to translate information and not literature, we should have
> different priorities. It is more important that the translation maintain
> fidelity to the facts than to the language and structure of the article.
> Sometimes it makes sense to pass over certain details, even a
> beginning-to-end translation might come out a bit condensed. As one reason
> for this, making some details accessible to the cultural audience in the
> new language can at times require a fair amount of elaboration, more than
> may be ideal for the context under discussion. The best approach to use is
> one of adaptation as much as translation.
>
> I don't have strong feelings about whether a paid model will work, or work
> better than purely volunteer activity, but I would be open to seeing a
> trial. The essential thing is that we find translators who can understand
> and apply standards of quality in their work, much like we would expect if
> they were editors writing entirely new articles.
>
> --Michael Snow
>
>
> On 2/24/2018 5:26 AM, James Heilman wrote:
>
>> We learned a few things during the medical translation project which
>> started back in 2011:
>>
>> 1) You must start with high quality content and thus all articles are
>> extensively improved before being proposed for translation.
>>
>> 2) A lot of languages want "less" content than is present on EN WP. Thus
>> we
>> moved to just improving and suggesting for translation the leads of the
>> English articles.
>>
>> 3) The "Content Translation" tool developed by the WMF made efforts more
>> efficient than handing around word documents. Would love to see that tool
>> improved further such as having it support specific lists of articles that
>> are deemed ready for translation by certain groups. Would also love the
>> tool to have tracking metrics for these types of projects.
>>
>> 4) We used volunteer translators mostly associated with our partner
>> Translators Without Borders. One issue we found was that languages in
>> which
>> their are lots of translators such as French, Spanish, and Italian there
>> is
>> often already at least some content on many of the topics in question. The
>> issue than becomes integration which needs an expert Wikipedia. And for
>> languages in which we have little content there are often few avaliable
>> volunteers.
>>
>> 5) With respect to "paying per word" the problem is this would require
>> significant checks and balances to make sure people are taking the work
>> seriously and not simple using Google translate for the 70 or so languages
>> in which it claims to work. We often had translations undergo a second
>> review and the volunteers at TWB have to pass certain tests to be
>> accepted.
>>
>> 6) I hired a coordinator for the translation project for a couple of
>> years.
>> The translators at TWB did not want to become Wikipedians or learn how to
>> use our systems. The coordinator created account like TransSW001 (one for
>> each volunteer) and preloaded the article to be translated into Content
>> Translation. They than gave the volunteer translator the user name and
>> password to the account.
>>
>> 7) Were are we at now? There are currently just over 1,000 leads of
>> articles that have been improved and are ready for translation. This
>> includes articles on the 440 medications that are on the WHO Essential
>> List. We have worked a bit in some 100 languages. The efforts have
>> resulted
>> in more than 5 million works translated and integrated into different
>> Wikipedias. The coordinator has unfortunately moved on to his real job of
>> teaching high school students.
>>
>> 8) The project continues but at a slower pace than before. The Wikipedian
>> and retired orthopedic surgeon Subas Chandra Rout has basically single
>> handedly translated nearly all 1,000 leads into Odia a language spoken by
>> 40 million people in Eastern India. The amazing thing is that for

Re: [Wikimedia-l] Paid translation

Well, what I wrote about was translating the articles on the lists at meta.
In addition the translators themselves chose which one they want to
translate.

On Sat, Feb 24, 2018 at 8:47 PM, Peter Southwood <
peter.southw...@telkomsa.net> wrote:

> Those who pay get to select what is translated.
> Cheers,
> Peter
>
> -Original Message-
> From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> Behalf Of Jean-Philippe Béland
> Sent: 24 February 2018 16:55
> To: Wikimedia Mailing List
> Subject: Re: [Wikimedia-l] Paid translation
>
> I think the request for such projects should come from the concerned
> language projects, same for the list of articles. If not, in my simple
> opinion, it is a form of coloniasm again.
>
> Jean-Philippe Béland
> Vice President, Wikimedia Canada
>
>
> On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com> wrote:
>
> > Should have added that the remaining points are somewhat less
> > interesting in this context. Preloading a set of articles is a bad
> > idea, the translators should be able to chose for themselves. Articles
> > should also be pretty broad, not very narrow technical or medical, ie
> > vertical articles, as the number of editors that can handle those will
> be pretty small.
> >
> > In particular: Do not believe you can turn a teanslator into a new
> editor!
> > You can although turn an existing editor into a translator.
> >
> > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <jeb...@gmail.com>
> > wrote:
> >
> > > 1) You must start with high quality content and thus all articles
> > > are
> > >> extensively improved before being proposed for translation.
> > >
> > >
> > > Note that to much pressure on "quality" can easily kill the project.
> > >
> > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > more
> > >> efficient than handing around word documents. Would love to see
> > >> that
> > tool
> > >> improved further such as having it support specific lists of
> > >> articles
> > that
> > >> are deemed ready for translation by certain groups. Would also love
> > >> the tool to have tracking metrics for these types of projects.
> > >
> > >
> > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > >
> > > 4) We used volunteer translators mostly associated with our partner
> > >> Translators Without Borders. One issue we found was that languages
> > >> in which their are lots of translators such as French, Spanish, and
> > >> Italian there is often already at least some content on many of the
> > >> topics in question.
> > The
> > >> issue than becomes integration which needs an expert Wikipedia. And
> > >> for languages in which we have little content there are often few
> > >> avaliable volunteers.
> > >
> > >
> > > I used projects below 65k articles as an example, as the chance of
> > > competing articles are pretty low.
> > >
> > > 5) With respect to "paying per word" the problem is this would
> > > require
> > >> significant checks and balances to make sure people are taking the
> > >> work seriously and not simple using Google translate for the 70 or
> > >> so
> > languages
> > >> in which it claims to work. We often had translations undergo a
> > >> second review and the volunteers at TWB have to pass certain tests
> > >> to be accepted.
> > >
> > >
> > > I'n my original email I wrote "verified good translators". It is as
> > > simple as "Has the editor contributed other articles at the project?"
> > >
> > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <jmh...@gmail.com>
> wrote:
> > >
> > >> We learned a few things during the medical translation project
> > >> which started back in 2011:
> > >>
> > >> 1) You must start with high quality content and thus all articles
> > >> are extensively improved before being proposed for translation.
> > >>
> > >> 2) A lot of languages want "less" content than is present on EN WP.
> > >> Thus we moved to just improving and suggesting for translation the
> > >> leads of the English articles.
> > >>
> > >> 3) The "Content Translation" tool developed by the WMF made efforts
>

Re: [Wikimedia-l] Paid translation

It is a long time since everyone on these projects were solely volunteers.
:)

On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen <toddmal...@gmail.com> wrote:

> Yes, and then there's always the question. If he's getting paid, why aren't
> I? Why is he getting paid per word of article translated? Why am I not
> getting paid per spamvertisement deleted or vandal blocked? Why am I not
> getting paid for closing discussions that it takes hours of reading input
> and considering all sides and getting rocks thrown at me no matter what I
> do? Is that not valuable to the project as well?
>
> If you want to pay anyone, you better start paying me. I'm okay with the
> idea of being a volunteer as long as everyone is a volunteer. But if you
> start paying some people and not me, we're going to have a problem.
>
> Todd
>
> On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood <
> peter.southw...@telkomsa.net> wrote:
>
> > Those who pay get to select what is translated.
> > Cheers,
> > Peter
> >
> > -Original Message-
> > From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> > Behalf Of Jean-Philippe Béland
> > Sent: 24 February 2018 16:55
> > To: Wikimedia Mailing List
> > Subject: Re: [Wikimedia-l] Paid translation
> >
> > I think the request for such projects should come from the concerned
> > language projects, same for the list of articles. If not, in my simple
> > opinion, it is a form of coloniasm again.
> >
> > Jean-Philippe Béland
> > Vice President, Wikimedia Canada
> >
> >
> > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com>
> wrote:
> >
> > > Should have added that the remaining points are somewhat less
> > > interesting in this context. Preloading a set of articles is a bad
> > > idea, the translators should be able to chose for themselves. Articles
> > > should also be pretty broad, not very narrow technical or medical, ie
> > > vertical articles, as the number of editors that can handle those will
> > be pretty small.
> > >
> > > In particular: Do not believe you can turn a teanslator into a new
> > editor!
> > > You can although turn an existing editor into a translator.
> > >
> > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <jeb...@gmail.com>
> > > wrote:
> > >
> > > > 1) You must start with high quality content and thus all articles
> > > > are
> > > >> extensively improved before being proposed for translation.
> > > >
> > > >
> > > > Note that to much pressure on "quality" can easily kill the project.
> > > >
> > > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > > more
> > > >> efficient than handing around word documents. Would love to see
> > > >> that
> > > tool
> > > >> improved further such as having it support specific lists of
> > > >> articles
> > > that
> > > >> are deemed ready for translation by certain groups. Would also love
> > > >> the tool to have tracking metrics for these types of projects.
> > > >
> > > >
> > > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > > >
> > > > 4) We used volunteer translators mostly associated with our partner
> > > >> Translators Without Borders. One issue we found was that languages
> > > >> in which their are lots of translators such as French, Spanish, and
> > > >> Italian there is often already at least some content on many of the
> > > >> topics in question.
> > > The
> > > >> issue than becomes integration which needs an expert Wikipedia. And
> > > >> for languages in which we have little content there are often few
> > > >> avaliable volunteers.
> > > >
> > > >
> > > > I used projects below 65k articles as an example, as the chance of
> > > > competing articles are pretty low.
> > > >
> > > > 5) With respect to "paying per word" the problem is this would
> > > > require
> > > >> significant checks and balances to make sure people are taking the
> > > >> work seriously and not simple using Google translate for the 70 or
> > > >> so
> > > languages
> > > >> in which it claims to work. We often had translations undergo a
> > > >> second review and the volunt

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Yaroslav Blanter

I actually agree with Todd, and I though this is actually a reason why WMF
staff may not edit articles (at least not from WMF accounts). I am afraid
disadvantages due to the broken symmetry will be bigger than advantages due
to actual content translated.

Cheers
Yaroslav

On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen <toddmal...@gmail.com> wrote:

> Yes, and then there's always the question. If he's getting paid, why aren't
> I? Why is he getting paid per word of article translated? Why am I not
> getting paid per spamvertisement deleted or vandal blocked? Why am I not
> getting paid for closing discussions that it takes hours of reading input
> and considering all sides and getting rocks thrown at me no matter what I
> do? Is that not valuable to the project as well?
>
> If you want to pay anyone, you better start paying me. I'm okay with the
> idea of being a volunteer as long as everyone is a volunteer. But if you
> start paying some people and not me, we're going to have a problem.
>
> Todd
>
> On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood <
> peter.southw...@telkomsa.net> wrote:
>
> > Those who pay get to select what is translated.
> > Cheers,
> > Peter
> >
> > -Original Message-
> > From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> > Behalf Of Jean-Philippe Béland
> > Sent: 24 February 2018 16:55
> > To: Wikimedia Mailing List
> > Subject: Re: [Wikimedia-l] Paid translation
> >
> > I think the request for such projects should come from the concerned
> > language projects, same for the list of articles. If not, in my simple
> > opinion, it is a form of coloniasm again.
> >
> > Jean-Philippe Béland
> > Vice President, Wikimedia Canada
> >
> >
> > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com>
> wrote:
> >
> > > Should have added that the remaining points are somewhat less
> > > interesting in this context. Preloading a set of articles is a bad
> > > idea, the translators should be able to chose for themselves. Articles
> > > should also be pretty broad, not very narrow technical or medical, ie
> > > vertical articles, as the number of editors that can handle those will
> > be pretty small.
> > >
> > > In particular: Do not believe you can turn a teanslator into a new
> > editor!
> > > You can although turn an existing editor into a translator.
> > >
> > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <jeb...@gmail.com>
> > > wrote:
> > >
> > > > 1) You must start with high quality content and thus all articles
> > > > are
> > > >> extensively improved before being proposed for translation.
> > > >
> > > >
> > > > Note that to much pressure on "quality" can easily kill the project.
> > > >
> > > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > > more
> > > >> efficient than handing around word documents. Would love to see
> > > >> that
> > > tool
> > > >> improved further such as having it support specific lists of
> > > >> articles
> > > that
> > > >> are deemed ready for translation by certain groups. Would also love
> > > >> the tool to have tracking metrics for these types of projects.
> > > >
> > > >
> > > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > > >
> > > > 4) We used volunteer translators mostly associated with our partner
> > > >> Translators Without Borders. One issue we found was that languages
> > > >> in which their are lots of translators such as French, Spanish, and
> > > >> Italian there is often already at least some content on many of the
> > > >> topics in question.
> > > The
> > > >> issue than becomes integration which needs an expert Wikipedia. And
> > > >> for languages in which we have little content there are often few
> > > >> avaliable volunteers.
> > > >
> > > >
> > > > I used projects below 65k articles as an example, as the chance of
> > > > competing articles are pretty low.
> > > >
> > > > 5) With respect to "paying per word" the problem is this would
> > > > require
> > > >> significant checks and balances to make sure people are taking the
> > > >> work seriously and not simple using Google translate for the 70

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Todd Allen

Yes, and then there's always the question. If he's getting paid, why aren't
I? Why is he getting paid per word of article translated? Why am I not
getting paid per spamvertisement deleted or vandal blocked? Why am I not
getting paid for closing discussions that it takes hours of reading input
and considering all sides and getting rocks thrown at me no matter what I
do? Is that not valuable to the project as well?

If you want to pay anyone, you better start paying me. I'm okay with the
idea of being a volunteer as long as everyone is a volunteer. But if you
start paying some people and not me, we're going to have a problem.

Todd

On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood <
peter.southw...@telkomsa.net> wrote:

> Those who pay get to select what is translated.
> Cheers,
> Peter
>
> -Original Message-
> From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> Behalf Of Jean-Philippe Béland
> Sent: 24 February 2018 16:55
> To: Wikimedia Mailing List
> Subject: Re: [Wikimedia-l] Paid translation
>
> I think the request for such projects should come from the concerned
> language projects, same for the list of articles. If not, in my simple
> opinion, it is a form of coloniasm again.
>
> Jean-Philippe Béland
> Vice President, Wikimedia Canada
>
>
> On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com> wrote:
>
> > Should have added that the remaining points are somewhat less
> > interesting in this context. Preloading a set of articles is a bad
> > idea, the translators should be able to chose for themselves. Articles
> > should also be pretty broad, not very narrow technical or medical, ie
> > vertical articles, as the number of editors that can handle those will
> be pretty small.
> >
> > In particular: Do not believe you can turn a teanslator into a new
> editor!
> > You can although turn an existing editor into a translator.
> >
> > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <jeb...@gmail.com>
> > wrote:
> >
> > > 1) You must start with high quality content and thus all articles
> > > are
> > >> extensively improved before being proposed for translation.
> > >
> > >
> > > Note that to much pressure on "quality" can easily kill the project.
> > >
> > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > more
> > >> efficient than handing around word documents. Would love to see
> > >> that
> > tool
> > >> improved further such as having it support specific lists of
> > >> articles
> > that
> > >> are deemed ready for translation by certain groups. Would also love
> > >> the tool to have tracking metrics for these types of projects.
> > >
> > >
> > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > >
> > > 4) We used volunteer translators mostly associated with our partner
> > >> Translators Without Borders. One issue we found was that languages
> > >> in which their are lots of translators such as French, Spanish, and
> > >> Italian there is often already at least some content on many of the
> > >> topics in question.
> > The
> > >> issue than becomes integration which needs an expert Wikipedia. And
> > >> for languages in which we have little content there are often few
> > >> avaliable volunteers.
> > >
> > >
> > > I used projects below 65k articles as an example, as the chance of
> > > competing articles are pretty low.
> > >
> > > 5) With respect to "paying per word" the problem is this would
> > > require
> > >> significant checks and balances to make sure people are taking the
> > >> work seriously and not simple using Google translate for the 70 or
> > >> so
> > languages
> > >> in which it claims to work. We often had translations undergo a
> > >> second review and the volunteers at TWB have to pass certain tests
> > >> to be accepted.
> > >
> > >
> > > I'n my original email I wrote "verified good translators". It is as
> > > simple as "Has the editor contributed other articles at the project?"
> > >
> > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <jmh...@gmail.com>
> wrote:
> > >
> > >> We learned a few things during the medical translation project
> > >> which started back in 2011:
> > >>
> > >> 1) You must start with high quality con

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Peter Southwood

Those who pay get to select what is translated. 
Cheers,
Peter

-Original Message-
From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On Behalf Of 
Jean-Philippe Béland
Sent: 24 February 2018 16:55
To: Wikimedia Mailing List
Subject: Re: [Wikimedia-l] Paid translation

I think the request for such projects should come from the concerned language 
projects, same for the list of articles. If not, in my simple opinion, it is a 
form of coloniasm again.

Jean-Philippe Béland
Vice President, Wikimedia Canada


On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeb...@gmail.com> wrote:

> Should have added that the remaining points are somewhat less 
> interesting in this context. Preloading a set of articles is a bad 
> idea, the translators should be able to chose for themselves. Articles 
> should also be pretty broad, not very narrow technical or medical, ie 
> vertical articles, as the number of editors that can handle those will be 
> pretty small.
>
> In particular: Do not believe you can turn a teanslator into a new editor!
> You can although turn an existing editor into a translator.
>
> On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <jeb...@gmail.com>
> wrote:
>
> > 1) You must start with high quality content and thus all articles 
> > are
> >> extensively improved before being proposed for translation.
> >
> >
> > Note that to much pressure on "quality" can easily kill the project.
> >
> > 3) The "Content Translation" tool developed by the WMF made efforts 
> > more
> >> efficient than handing around word documents. Would love to see 
> >> that
> tool
> >> improved further such as having it support specific lists of 
> >> articles
> that
> >> are deemed ready for translation by certain groups. Would also love 
> >> the tool to have tracking metrics for these types of projects.
> >
> >
> > Didn't mention ContentTranslation, but it should be pretty obvious.
> >
> > 4) We used volunteer translators mostly associated with our partner
> >> Translators Without Borders. One issue we found was that languages 
> >> in which their are lots of translators such as French, Spanish, and 
> >> Italian there is often already at least some content on many of the 
> >> topics in question.
> The
> >> issue than becomes integration which needs an expert Wikipedia. And 
> >> for languages in which we have little content there are often few 
> >> avaliable volunteers.
> >
> >
> > I used projects below 65k articles as an example, as the chance of 
> > competing articles are pretty low.
> >
> > 5) With respect to "paying per word" the problem is this would 
> > require
> >> significant checks and balances to make sure people are taking the 
> >> work seriously and not simple using Google translate for the 70 or 
> >> so
> languages
> >> in which it claims to work. We often had translations undergo a 
> >> second review and the volunteers at TWB have to pass certain tests 
> >> to be accepted.
> >
> >
> > I'n my original email I wrote "verified good translators". It is as 
> > simple as "Has the editor contributed other articles at the project?"
> >
> > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <jmh...@gmail.com> wrote:
> >
> >> We learned a few things during the medical translation project 
> >> which started back in 2011:
> >>
> >> 1) You must start with high quality content and thus all articles 
> >> are extensively improved before being proposed for translation.
> >>
> >> 2) A lot of languages want "less" content than is present on EN WP. 
> >> Thus we moved to just improving and suggesting for translation the 
> >> leads of the English articles.
> >>
> >> 3) The "Content Translation" tool developed by the WMF made efforts 
> >> more efficient than handing around word documents. Would love to 
> >> see that
> tool
> >> improved further such as having it support specific lists of 
> >> articles
> that
> >> are deemed ready for translation by certain groups. Would also love 
> >> the tool to have tracking metrics for these types of projects.
> >>
> >> 4) We used volunteer translators mostly associated with our partner 
> >> Translators Without Borders. One issue we found was that languages 
> >> in which their are lots of translators such as French, Spanish, and 
> >> Italian there is often already at

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Michael Snow

I think the experience I've had with translating matches up well with 
the conclusions James has outlined. Even though I'm more likely to 
translate content into English rather than out of English, the 
principles still hold.


Trying to produce a translation without quality content in the original 
article is a frustrating and pointless exercise for the translator. 
Unless the original meets certain standards, it would be better and 
easier to write the article from scratch in the "destination" language 
and translate it back to the "source" language.


Assuming we have a good article in the original language, I definitely 
encourage translators to use editorial judgment in what they carry over. 
Focusing on the lead section is one possible approach. In general, 
because we are trying to translate information and not literature, we 
should have different priorities. It is more important that the 
translation maintain fidelity to the facts than to the language and 
structure of the article. Sometimes it makes sense to pass over certain 
details, even a beginning-to-end translation might come out a bit 
condensed. As one reason for this, making some details accessible to the 
cultural audience in the new language can at times require a fair amount 
of elaboration, more than may be ideal for the context under discussion. 
The best approach to use is one of adaptation as much as translation.


I don't have strong feelings about whether a paid model will work, or 
work better than purely volunteer activity, but I would be open to 
seeing a trial. The essential thing is that we find translators who can 
understand and apply standards of quality in their work, much like we 
would expect if they were editors writing entirely new articles.


--Michael Snow

On 2/24/2018 5:26 AM, James Heilman wrote:

We learned a few things during the medical translation project which
started back in 2011:

1) You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.

2) A lot of languages want "less" content than is present on EN WP. Thus we
moved to just improving and suggesting for translation the leads of the
English articles.

3) The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool
improved further such as having it support specific lists of articles that
are deemed ready for translation by certain groups. Would also love the
tool to have tracking metrics for these types of projects.

4) We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which
their are lots of translators such as French, Spanish, and Italian there is
often already at least some content on many of the topics in question. The
issue than becomes integration which needs an expert Wikipedia. And for
languages in which we have little content there are often few avaliable
volunteers.

5) With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work
seriously and not simple using Google translate for the 70 or so languages
in which it claims to work. We often had translations undergo a second
review and the volunteers at TWB have to pass certain tests to be accepted.

6) I hired a coordinator for the translation project for a couple of years.
The translators at TWB did not want to become Wikipedians or learn how to
use our systems. The coordinator created account like TransSW001 (one for
each volunteer) and preloaded the article to be translated into Content
Translation. They than gave the volunteer translator the user name and
password to the account.

7) Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This
includes articles on the 440 medications that are on the WHO Essential
List. We have worked a bit in some 100 languages. The efforts have resulted
in more than 5 million works translated and integrated into different
Wikipedias. The coordinator has unfortunately moved on to his real job of
teaching high school students.

8) The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single
handedly translated nearly all 1,000 leads into Odia a language spoken by
40 million people in Eastern India. The amazing thing is that for many of
these topics this is the first and only information online about it. Google
translate does not even claim to work in this language. Our partnerships
with WMTW and medical school in Taipai continue to translate into Chinese.
There the students translate and than their translations are reviewed by
their profs before being posted. They translate in groups using hackpad to
make it more social.

I am currently working to re invigorate the project :-)
James

On Sat,

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Gnangarra

this would be a good practical exercise to develop for WiR / WikiEd
programs in universities where they can engage with International Students
and local students studying additional languages as means of learning the
written nuances of the individual languages.  Any funding would be better
utilised in enabling such programs where the flow on impact is more
likely{fact} to be lasting.  Though I can see value in using a gift/reward
system for technically disadvantaged communities like the case presented
about Swahili .The focus would need to be on basic health, hygiene,
biology, science topics rather than more social or political topics.

On 25 February 2018 at 01:08, Vi to  wrote:

> I'll reply to the most recent email just for laziness.
>
> I'm doubtful for a series of reasons, most of were already expressed in a
> better way by others:
> *a remuneration in terms of quantity will weaken the quality of
> translations unless there's a strong mechanism of quality verification
> requiring a quantity of resources comparable to translations themselves;
> *articles are the result of a long process which reflects cultural identity
> of different communities, I'm not confident with transferring them to a
> different "weaker" cultures. My usage of "weaker" adjective only focuses
> about the strength of a cultural presence on the Internet;
> *articles to be translated are at high risk of reflecting the cultural
> identity (and biases) of the Western culture;
> *finally I think paid translators would hardly turn into stable
> Wikipedians.
>
> IMHO some paid editing may be better exploited in order to digitalise texts
> of unrepresented cultures (wikisource) or preserving their vocabularies
> (wiktionary).
>
> Also those languages which are secondary for all their speakers should be
> dealt with in a different fashion. I, for one, am a native speaker of
> specific variant of Sicilian, Sicilian is a secondary language to any of
> its speakers. Honestly, I'd find pointless to read the biography of
> Leonardo da Vinci in Sicilian while I can find thousands of books about him
> in Italian. Also I find this kind of translation creates a fake "literary"
> language totally detached from reality: there's no "encaustic painting" in
> Sicilian, still a Sicilian article about Leonardo will invent one.
>
> As a general principle we should always collect, rather than create,
> knowledge.
>
> Vito
>
> 2018-02-24 16:30 GMT+01:00 John Erling Blad :
>
> > My reply can be read as a bit more harsh than intended, it was merely a
> > statement about my present experience about translators in general.
> >
> > The problem with lack of contributors (and translators) in a specialized
> > area is that there is a small community, and within this community some
> > kind of selection is made. Each time a selection is repeated the
> remaining
> > group shrinks. Specialize the selection sufficiently many times and there
> > will be no contributors (or translators) left. It is simply a game of
> > probabilities. Thus, to make such a project work it must have a
> > sufficiently broad scope for the articles. Articles about public health
> > services will probably work even for a pretty small language group, but
> > specialized medical articles might create a problem. But then you find
> > a retired
> > orthopedic surgeon like Subas Chandra Rout…
> >
> > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman  wrote:
> >
> > > I agree with John that it is very difficult to turn a translator into a
> > new
> > > editor. I also agree with Jean-Philippe that it is key to have
> > involvement
> > > of the local projects and preferable if they lead the efforts. Of the
> > > languages we worked in only one explicitly requested not to be
> involved /
> > > have translations from TWB.
> > >
> > > James
> > >
> > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> > > wrote:
> > >
> > > > You can turn it around; give added credits for translations from
> small
> > > > language projects and into the larger ones, that is a lot more
> > > interesting
> > > > than strictly translating from the larger language projects.
> > > >
> > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > > > jpbel...@wikimedia.ca
> > > > > wrote:
> > > >
> > > > > I think the request for such projects should come from the
> concerned
> > > > > language projects, same for the list of articles. If not, in my
> > simple
> > > > > opinion, it is a form of coloniasm again.
> > > > >
> > > > > Jean-Philippe Béland
> > > > > Vice President, Wikimedia Canada
> > > > >
> > > > >
> > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad  >
> > > > wrote:
> > > > >
> > > > > > Should have added that the remaining points are somewhat less
> > > > interesting
> > > > > > in this context. Preloading a set of articles is a bad idea, the
> > > > > > translators should be able to chose for

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Vi to

I'll reply to the most recent email just for laziness.

I'm doubtful for a series of reasons, most of were already expressed in a
better way by others:
*a remuneration in terms of quantity will weaken the quality of
translations unless there's a strong mechanism of quality verification
requiring a quantity of resources comparable to translations themselves;
*articles are the result of a long process which reflects cultural identity
of different communities, I'm not confident with transferring them to a
different "weaker" cultures. My usage of "weaker" adjective only focuses
about the strength of a cultural presence on the Internet;
*articles to be translated are at high risk of reflecting the cultural
identity (and biases) of the Western culture;
*finally I think paid translators would hardly turn into stable Wikipedians.

IMHO some paid editing may be better exploited in order to digitalise texts
of unrepresented cultures (wikisource) or preserving their vocabularies
(wiktionary).

Also those languages which are secondary for all their speakers should be
dealt with in a different fashion. I, for one, am a native speaker of
specific variant of Sicilian, Sicilian is a secondary language to any of
its speakers. Honestly, I'd find pointless to read the biography of
Leonardo da Vinci in Sicilian while I can find thousands of books about him
in Italian. Also I find this kind of translation creates a fake "literary"
language totally detached from reality: there's no "encaustic painting" in
Sicilian, still a Sicilian article about Leonardo will invent one.

As a general principle we should always collect, rather than create,
knowledge.

Vito

2018-02-24 16:30 GMT+01:00 John Erling Blad :

> My reply can be read as a bit more harsh than intended, it was merely a
> statement about my present experience about translators in general.
>
> The problem with lack of contributors (and translators) in a specialized
> area is that there is a small community, and within this community some
> kind of selection is made. Each time a selection is repeated the remaining
> group shrinks. Specialize the selection sufficiently many times and there
> will be no contributors (or translators) left. It is simply a game of
> probabilities. Thus, to make such a project work it must have a
> sufficiently broad scope for the articles. Articles about public health
> services will probably work even for a pretty small language group, but
> specialized medical articles might create a problem. But then you find
> a retired
> orthopedic surgeon like Subas Chandra Rout…
>
> On Sat, Feb 24, 2018 at 4:04 PM, James Heilman  wrote:
>
> > I agree with John that it is very difficult to turn a translator into a
> new
> > editor. I also agree with Jean-Philippe that it is key to have
> involvement
> > of the local projects and preferable if they lead the efforts. Of the
> > languages we worked in only one explicitly requested not to be involved /
> > have translations from TWB.
> >
> > James
> >
> > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> > wrote:
> >
> > > You can turn it around; give added credits for translations from small
> > > language projects and into the larger ones, that is a lot more
> > interesting
> > > than strictly translating from the larger language projects.
> > >
> > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > > jpbel...@wikimedia.ca
> > > > wrote:
> > >
> > > > I think the request for such projects should come from the concerned
> > > > language projects, same for the list of articles. If not, in my
> simple
> > > > opinion, it is a form of coloniasm again.
> > > >
> > > > Jean-Philippe Béland
> > > > Vice President, Wikimedia Canada
> > > >
> > > >
> > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad 
> > > wrote:
> > > >
> > > > > Should have added that the remaining points are somewhat less
> > > interesting
> > > > > in this context. Preloading a set of articles is a bad idea, the
> > > > > translators should be able to chose for themselves. Articles should
> > > also
> > > > be
> > > > > pretty broad, not very narrow technical or medical, ie vertical
> > > articles,
> > > > > as the number of editors that can handle those will be pretty
> small.
> > > > >
> > > > > In particular: Do not believe you can turn a teanslator into a new
> > > > editor!
> > > > > You can although turn an existing editor into a translator.
> > > > >
> > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
> jeb...@gmail.com>
> > > > > wrote:
> > > > >
> > > > > > 1) You must start with high quality content and thus all articles
> > are
> > > > > >> extensively improved before being proposed for translation.
> > > > > >
> > > > > >
> > > > > > Note that to much pressure on "quality" can easily kill the
> > project.
> > > > > >
> > > > > > 3) The "Content Translation" tool developed by the WMF made
> efforts
> > > > more
> > > > > >> efficient

Re: [Wikimedia-l] Paid translation

My reply can be read as a bit more harsh than intended, it was merely a
statement about my present experience about translators in general.

The problem with lack of contributors (and translators) in a specialized
area is that there is a small community, and within this community some
kind of selection is made. Each time a selection is repeated the remaining
group shrinks. Specialize the selection sufficiently many times and there
will be no contributors (or translators) left. It is simply a game of
probabilities. Thus, to make such a project work it must have a
sufficiently broad scope for the articles. Articles about public health
services will probably work even for a pretty small language group, but
specialized medical articles might create a problem. But then you find
a retired
orthopedic surgeon like Subas Chandra Rout…

On Sat, Feb 24, 2018 at 4:04 PM, James Heilman  wrote:

> I agree with John that it is very difficult to turn a translator into a new
> editor. I also agree with Jean-Philippe that it is key to have involvement
> of the local projects and preferable if they lead the efforts. Of the
> languages we worked in only one explicitly requested not to be involved /
> have translations from TWB.
>
> James
>
> On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad 
> wrote:
>
> > You can turn it around; give added credits for translations from small
> > language projects and into the larger ones, that is a lot more
> interesting
> > than strictly translating from the larger language projects.
> >
> > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> > jpbel...@wikimedia.ca
> > > wrote:
> >
> > > I think the request for such projects should come from the concerned
> > > language projects, same for the list of articles. If not, in my simple
> > > opinion, it is a form of coloniasm again.
> > >
> > > Jean-Philippe Béland
> > > Vice President, Wikimedia Canada
> > >
> > >
> > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad 
> > wrote:
> > >
> > > > Should have added that the remaining points are somewhat less
> > interesting
> > > > in this context. Preloading a set of articles is a bad idea, the
> > > > translators should be able to chose for themselves. Articles should
> > also
> > > be
> > > > pretty broad, not very narrow technical or medical, ie vertical
> > articles,
> > > > as the number of editors that can handle those will be pretty small.
> > > >
> > > > In particular: Do not believe you can turn a teanslator into a new
> > > editor!
> > > > You can although turn an existing editor into a translator.
> > > >
> > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad 
> > > > wrote:
> > > >
> > > > > 1) You must start with high quality content and thus all articles
> are
> > > > >> extensively improved before being proposed for translation.
> > > > >
> > > > >
> > > > > Note that to much pressure on "quality" can easily kill the
> project.
> > > > >
> > > > > 3) The "Content Translation" tool developed by the WMF made efforts
> > > more
> > > > >> efficient than handing around word documents. Would love to see
> that
> > > > tool
> > > > >> improved further such as having it support specific lists of
> > articles
> > > > that
> > > > >> are deemed ready for translation by certain groups. Would also
> love
> > > the
> > > > >> tool to have tracking metrics for these types of projects.
> > > > >
> > > > >
> > > > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > > > >
> > > > > 4) We used volunteer translators mostly associated with our partner
> > > > >> Translators Without Borders. One issue we found was that languages
> > in
> > > > >> which
> > > > >> their are lots of translators such as French, Spanish, and Italian
> > > there
> > > > >> is
> > > > >> often already at least some content on many of the topics in
> > question.
> > > > The
> > > > >> issue than becomes integration which needs an expert Wikipedia.
> And
> > > for
> > > > >> languages in which we have little content there are often few
> > > avaliable
> > > > >> volunteers.
> > > > >
> > > > >
> > > > > I used projects below 65k articles as an example, as the chance of
> > > > > competing articles are pretty low.
> > > > >
> > > > > 5) With respect to "paying per word" the problem is this would
> > require
> > > > >> significant checks and balances to make sure people are taking the
> > > work
> > > > >> seriously and not simple using Google translate for the 70 or so
> > > > languages
> > > > >> in which it claims to work. We often had translations undergo a
> > second
> > > > >> review and the volunteers at TWB have to pass certain tests to be
> > > > >> accepted.
> > > > >
> > > > >
> > > > > I'n my original email I wrote "verified good translators". It is as
> > > > > simple as "Has the editor contributed other articles at the
> project?"
> > > > >
> > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman

Re: [Wikimedia-l] Paid translation

I agree with John that it is very difficult to turn a translator into a new
editor. I also agree with Jean-Philippe that it is key to have involvement
of the local projects and preferable if they lead the efforts. Of the
languages we worked in only one explicitly requested not to be involved /
have translations from TWB.

James

On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad  wrote:

> You can turn it around; give added credits for translations from small
> language projects and into the larger ones, that is a lot more interesting
> than strictly translating from the larger language projects.
>
> On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <
> jpbel...@wikimedia.ca
> > wrote:
>
> > I think the request for such projects should come from the concerned
> > language projects, same for the list of articles. If not, in my simple
> > opinion, it is a form of coloniasm again.
> >
> > Jean-Philippe Béland
> > Vice President, Wikimedia Canada
> >
> >
> > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad 
> wrote:
> >
> > > Should have added that the remaining points are somewhat less
> interesting
> > > in this context. Preloading a set of articles is a bad idea, the
> > > translators should be able to chose for themselves. Articles should
> also
> > be
> > > pretty broad, not very narrow technical or medical, ie vertical
> articles,
> > > as the number of editors that can handle those will be pretty small.
> > >
> > > In particular: Do not believe you can turn a teanslator into a new
> > editor!
> > > You can although turn an existing editor into a translator.
> > >
> > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad 
> > > wrote:
> > >
> > > > 1) You must start with high quality content and thus all articles are
> > > >> extensively improved before being proposed for translation.
> > > >
> > > >
> > > > Note that to much pressure on "quality" can easily kill the project.
> > > >
> > > > 3) The "Content Translation" tool developed by the WMF made efforts
> > more
> > > >> efficient than handing around word documents. Would love to see that
> > > tool
> > > >> improved further such as having it support specific lists of
> articles
> > > that
> > > >> are deemed ready for translation by certain groups. Would also love
> > the
> > > >> tool to have tracking metrics for these types of projects.
> > > >
> > > >
> > > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > > >
> > > > 4) We used volunteer translators mostly associated with our partner
> > > >> Translators Without Borders. One issue we found was that languages
> in
> > > >> which
> > > >> their are lots of translators such as French, Spanish, and Italian
> > there
> > > >> is
> > > >> often already at least some content on many of the topics in
> question.
> > > The
> > > >> issue than becomes integration which needs an expert Wikipedia. And
> > for
> > > >> languages in which we have little content there are often few
> > avaliable
> > > >> volunteers.
> > > >
> > > >
> > > > I used projects below 65k articles as an example, as the chance of
> > > > competing articles are pretty low.
> > > >
> > > > 5) With respect to "paying per word" the problem is this would
> require
> > > >> significant checks and balances to make sure people are taking the
> > work
> > > >> seriously and not simple using Google translate for the 70 or so
> > > languages
> > > >> in which it claims to work. We often had translations undergo a
> second
> > > >> review and the volunteers at TWB have to pass certain tests to be
> > > >> accepted.
> > > >
> > > >
> > > > I'n my original email I wrote "verified good translators". It is as
> > > > simple as "Has the editor contributed other articles at the project?"
> > > >
> > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman 
> > wrote:
> > > >
> > > >> We learned a few things during the medical translation project which
> > > >> started back in 2011:
> > > >>
> > > >> 1) You must start with high quality content and thus all articles
> are
> > > >> extensively improved before being proposed for translation.
> > > >>
> > > >> 2) A lot of languages want "less" content than is present on EN WP.
> > Thus
> > > >> we
> > > >> moved to just improving and suggesting for translation the leads of
> > the
> > > >> English articles.
> > > >>
> > > >> 3) The "Content Translation" tool developed by the WMF made efforts
> > more
> > > >> efficient than handing around word documents. Would love to see that
> > > tool
> > > >> improved further such as having it support specific lists of
> articles
> > > that
> > > >> are deemed ready for translation by certain groups. Would also love
> > the
> > > >> tool to have tracking metrics for these types of projects.
> > > >>
> > > >> 4) We used volunteer translators mostly associated with our partner
> > > >> Translators Without Borders. One issue we found was that languages
> in
> > > >> which
> > > >>

Re: [Wikimedia-l] Paid translation

You can turn it around; give added credits for translations from small
language projects and into the larger ones, that is a lot more interesting
than strictly translating from the larger language projects.

On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland  wrote:

> I think the request for such projects should come from the concerned
> language projects, same for the list of articles. If not, in my simple
> opinion, it is a form of coloniasm again.
>
> Jean-Philippe Béland
> Vice President, Wikimedia Canada
>
>
> On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad  wrote:
>
> > Should have added that the remaining points are somewhat less interesting
> > in this context. Preloading a set of articles is a bad idea, the
> > translators should be able to chose for themselves. Articles should also
> be
> > pretty broad, not very narrow technical or medical, ie vertical articles,
> > as the number of editors that can handle those will be pretty small.
> >
> > In particular: Do not believe you can turn a teanslator into a new
> editor!
> > You can although turn an existing editor into a translator.
> >
> > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad 
> > wrote:
> >
> > > 1) You must start with high quality content and thus all articles are
> > >> extensively improved before being proposed for translation.
> > >
> > >
> > > Note that to much pressure on "quality" can easily kill the project.
> > >
> > > 3) The "Content Translation" tool developed by the WMF made efforts
> more
> > >> efficient than handing around word documents. Would love to see that
> > tool
> > >> improved further such as having it support specific lists of articles
> > that
> > >> are deemed ready for translation by certain groups. Would also love
> the
> > >> tool to have tracking metrics for these types of projects.
> > >
> > >
> > > Didn't mention ContentTranslation, but it should be pretty obvious.
> > >
> > > 4) We used volunteer translators mostly associated with our partner
> > >> Translators Without Borders. One issue we found was that languages in
> > >> which
> > >> their are lots of translators such as French, Spanish, and Italian
> there
> > >> is
> > >> often already at least some content on many of the topics in question.
> > The
> > >> issue than becomes integration which needs an expert Wikipedia. And
> for
> > >> languages in which we have little content there are often few
> avaliable
> > >> volunteers.
> > >
> > >
> > > I used projects below 65k articles as an example, as the chance of
> > > competing articles are pretty low.
> > >
> > > 5) With respect to "paying per word" the problem is this would require
> > >> significant checks and balances to make sure people are taking the
> work
> > >> seriously and not simple using Google translate for the 70 or so
> > languages
> > >> in which it claims to work. We often had translations undergo a second
> > >> review and the volunteers at TWB have to pass certain tests to be
> > >> accepted.
> > >
> > >
> > > I'n my original email I wrote "verified good translators". It is as
> > > simple as "Has the editor contributed other articles at the project?"
> > >
> > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman 
> wrote:
> > >
> > >> We learned a few things during the medical translation project which
> > >> started back in 2011:
> > >>
> > >> 1) You must start with high quality content and thus all articles are
> > >> extensively improved before being proposed for translation.
> > >>
> > >> 2) A lot of languages want "less" content than is present on EN WP.
> Thus
> > >> we
> > >> moved to just improving and suggesting for translation the leads of
> the
> > >> English articles.
> > >>
> > >> 3) The "Content Translation" tool developed by the WMF made efforts
> more
> > >> efficient than handing around word documents. Would love to see that
> > tool
> > >> improved further such as having it support specific lists of articles
> > that
> > >> are deemed ready for translation by certain groups. Would also love
> the
> > >> tool to have tracking metrics for these types of projects.
> > >>
> > >> 4) We used volunteer translators mostly associated with our partner
> > >> Translators Without Borders. One issue we found was that languages in
> > >> which
> > >> their are lots of translators such as French, Spanish, and Italian
> there
> > >> is
> > >> often already at least some content on many of the topics in question.
> > The
> > >> issue than becomes integration which needs an expert Wikipedia. And
> for
> > >> languages in which we have little content there are often few
> avaliable
> > >> volunteers.
> > >>
> > >> 5) With respect to "paying per word" the problem is this would require
> > >> significant checks and balances to make sure people are taking the
> work
> > >> seriously and not simple using Google translate for the 70 or so
> > languages
> > >> in which it claims to work. We often had

Re: [Wikimedia-l] Paid translation

Thank you James for this detailed feedback. It is very interesting.

JP

On Sat, Feb 24, 2018 at 8:27 AM James Heilman  wrote:

> We learned a few things during the medical translation project which
> started back in 2011:
>
> 1) You must start with high quality content and thus all articles are
> extensively improved before being proposed for translation.
>
> 2) A lot of languages want "less" content than is present on EN WP. Thus we
> moved to just improving and suggesting for translation the leads of the
> English articles.
>
> 3) The "Content Translation" tool developed by the WMF made efforts more
> efficient than handing around word documents. Would love to see that tool
> improved further such as having it support specific lists of articles that
> are deemed ready for translation by certain groups. Would also love the
> tool to have tracking metrics for these types of projects.
>
> 4) We used volunteer translators mostly associated with our partner
> Translators Without Borders. One issue we found was that languages in which
> their are lots of translators such as French, Spanish, and Italian there is
> often already at least some content on many of the topics in question. The
> issue than becomes integration which needs an expert Wikipedia. And for
> languages in which we have little content there are often few avaliable
> volunteers.
>
> 5) With respect to "paying per word" the problem is this would require
> significant checks and balances to make sure people are taking the work
> seriously and not simple using Google translate for the 70 or so languages
> in which it claims to work. We often had translations undergo a second
> review and the volunteers at TWB have to pass certain tests to be accepted.
>
> 6) I hired a coordinator for the translation project for a couple of years.
> The translators at TWB did not want to become Wikipedians or learn how to
> use our systems. The coordinator created account like TransSW001 (one for
> each volunteer) and preloaded the article to be translated into Content
> Translation. They than gave the volunteer translator the user name and
> password to the account.
>
> 7) Were are we at now? There are currently just over 1,000 leads of
> articles that have been improved and are ready for translation. This
> includes articles on the 440 medications that are on the WHO Essential
> List. We have worked a bit in some 100 languages. The efforts have resulted
> in more than 5 million works translated and integrated into different
> Wikipedias. The coordinator has unfortunately moved on to his real job of
> teaching high school students.
>
> 8) The project continues but at a slower pace than before. The Wikipedian
> and retired orthopedic surgeon Subas Chandra Rout has basically single
> handedly translated nearly all 1,000 leads into Odia a language spoken by
> 40 million people in Eastern India. The amazing thing is that for many of
> these topics this is the first and only information online about it. Google
> translate does not even claim to work in this language. Our partnerships
> with WMTW and medical school in Taipai continue to translate into Chinese.
> There the students translate and than their translations are reviewed by
> their profs before being posted. They translate in groups using hackpad to
> make it more social.
>
> I am currently working to re invigorate the project :-)
> James
>
> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad 
> wrote:
>
> > This discussion is going to be fun! =D
> >
> > A little more than seventy Wikipedia-projects has more than 65k articles,
> > the remaining two hundred or so are pretty small.
> >
> > What if a base set of articles were opened for paid translators? There
> are
> > several lists of such base sets. We have both the thousand articles from
> > "List of articles every Wikipedia should have"[1] and and the ten
> thousand
> > articles from the expanded list[2].
> >
> > Lets say verified good translators was paid about $0.01 per word (about
> $1
> > for a 1k-article) for translating one of those articles into another
> > language, with perhaps a higher pay for contributors in high-cost
> > countries. The pay would also have to be higher for languages that lacks
> > good translation tools.
> >
> > I believe this would be an _enabling_ activity for the communities, as
> > without a base set of articles it won't be possible to build a community
> at
> > all. By not paying for new articles, and only translating well-referenced
> > articles, some of the disputes in the communities could be avoided.
> Perhaps
> > we should also identify good source articles, that would be a help.
> > Translated articles should be above some minimum size, but they does not
> > have to be full translations of the source article.
> >
> > A real problem is that our existing lists of good articles other projects
> > should have is pretty much biased towards Western World, so they need a
> lot
> > of

Re: [Wikimedia-l] Paid translation

I think the request for such projects should come from the concerned
language projects, same for the list of articles. If not, in my simple
opinion, it is a form of coloniasm again.

Jean-Philippe Béland
Vice President, Wikimedia Canada


On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad  wrote:

> Should have added that the remaining points are somewhat less interesting
> in this context. Preloading a set of articles is a bad idea, the
> translators should be able to chose for themselves. Articles should also be
> pretty broad, not very narrow technical or medical, ie vertical articles,
> as the number of editors that can handle those will be pretty small.
>
> In particular: Do not believe you can turn a teanslator into a new editor!
> You can although turn an existing editor into a translator.
>
> On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad 
> wrote:
>
> > 1) You must start with high quality content and thus all articles are
> >> extensively improved before being proposed for translation.
> >
> >
> > Note that to much pressure on "quality" can easily kill the project.
> >
> > 3) The "Content Translation" tool developed by the WMF made efforts more
> >> efficient than handing around word documents. Would love to see that
> tool
> >> improved further such as having it support specific lists of articles
> that
> >> are deemed ready for translation by certain groups. Would also love the
> >> tool to have tracking metrics for these types of projects.
> >
> >
> > Didn't mention ContentTranslation, but it should be pretty obvious.
> >
> > 4) We used volunteer translators mostly associated with our partner
> >> Translators Without Borders. One issue we found was that languages in
> >> which
> >> their are lots of translators such as French, Spanish, and Italian there
> >> is
> >> often already at least some content on many of the topics in question.
> The
> >> issue than becomes integration which needs an expert Wikipedia. And for
> >> languages in which we have little content there are often few avaliable
> >> volunteers.
> >
> >
> > I used projects below 65k articles as an example, as the chance of
> > competing articles are pretty low.
> >
> > 5) With respect to "paying per word" the problem is this would require
> >> significant checks and balances to make sure people are taking the work
> >> seriously and not simple using Google translate for the 70 or so
> languages
> >> in which it claims to work. We often had translations undergo a second
> >> review and the volunteers at TWB have to pass certain tests to be
> >> accepted.
> >
> >
> > I'n my original email I wrote "verified good translators". It is as
> > simple as "Has the editor contributed other articles at the project?"
> >
> > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman  wrote:
> >
> >> We learned a few things during the medical translation project which
> >> started back in 2011:
> >>
> >> 1) You must start with high quality content and thus all articles are
> >> extensively improved before being proposed for translation.
> >>
> >> 2) A lot of languages want "less" content than is present on EN WP. Thus
> >> we
> >> moved to just improving and suggesting for translation the leads of the
> >> English articles.
> >>
> >> 3) The "Content Translation" tool developed by the WMF made efforts more
> >> efficient than handing around word documents. Would love to see that
> tool
> >> improved further such as having it support specific lists of articles
> that
> >> are deemed ready for translation by certain groups. Would also love the
> >> tool to have tracking metrics for these types of projects.
> >>
> >> 4) We used volunteer translators mostly associated with our partner
> >> Translators Without Borders. One issue we found was that languages in
> >> which
> >> their are lots of translators such as French, Spanish, and Italian there
> >> is
> >> often already at least some content on many of the topics in question.
> The
> >> issue than becomes integration which needs an expert Wikipedia. And for
> >> languages in which we have little content there are often few avaliable
> >> volunteers.
> >>
> >> 5) With respect to "paying per word" the problem is this would require
> >> significant checks and balances to make sure people are taking the work
> >> seriously and not simple using Google translate for the 70 or so
> languages
> >> in which it claims to work. We often had translations undergo a second
> >> review and the volunteers at TWB have to pass certain tests to be
> >> accepted.
> >>
> >> 6) I hired a coordinator for the translation project for a couple of
> >> years.
> >> The translators at TWB did not want to become Wikipedians or learn how
> to
> >> use our systems. The coordinator created account like TransSW001 (one
> for
> >> each volunteer) and preloaded the article to be translated into Content
> >> Translation. They than gave the volunteer translator the user name and
> >>

Re: [Wikimedia-l] Paid translation

colonialism *


On Sat, Feb 24, 2018 at 9:55 AM Jean-Philippe Béland 
wrote:

> I think the request for such projects should come from the concerned
> language projects, same for the list of articles. If not, in my simple
> opinion, it is a form of coloniasm again.
>
> Jean-Philippe Béland
> Vice President, Wikimedia Canada
>
>
> On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad  wrote:
>
>> Should have added that the remaining points are somewhat less interesting
>> in this context. Preloading a set of articles is a bad idea, the
>> translators should be able to chose for themselves. Articles should also
>> be
>> pretty broad, not very narrow technical or medical, ie vertical articles,
>> as the number of editors that can handle those will be pretty small.
>>
>> In particular: Do not believe you can turn a teanslator into a new editor!
>> You can although turn an existing editor into a translator.
>>
>> On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad 
>> wrote:
>>
>> > 1) You must start with high quality content and thus all articles are
>> >> extensively improved before being proposed for translation.
>> >
>> >
>> > Note that to much pressure on "quality" can easily kill the project.
>> >
>> > 3) The "Content Translation" tool developed by the WMF made efforts more
>> >> efficient than handing around word documents. Would love to see that
>> tool
>> >> improved further such as having it support specific lists of articles
>> that
>> >> are deemed ready for translation by certain groups. Would also love the
>> >> tool to have tracking metrics for these types of projects.
>> >
>> >
>> > Didn't mention ContentTranslation, but it should be pretty obvious.
>> >
>> > 4) We used volunteer translators mostly associated with our partner
>> >> Translators Without Borders. One issue we found was that languages in
>> >> which
>> >> their are lots of translators such as French, Spanish, and Italian
>> there
>> >> is
>> >> often already at least some content on many of the topics in question.
>> The
>> >> issue than becomes integration which needs an expert Wikipedia. And for
>> >> languages in which we have little content there are often few avaliable
>> >> volunteers.
>> >
>> >
>> > I used projects below 65k articles as an example, as the chance of
>> > competing articles are pretty low.
>> >
>> > 5) With respect to "paying per word" the problem is this would require
>> >> significant checks and balances to make sure people are taking the work
>> >> seriously and not simple using Google translate for the 70 or so
>> languages
>> >> in which it claims to work. We often had translations undergo a second
>> >> review and the volunteers at TWB have to pass certain tests to be
>> >> accepted.
>> >
>> >
>> > I'n my original email I wrote "verified good translators". It is as
>> > simple as "Has the editor contributed other articles at the project?"
>> >
>> > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman 
>> wrote:
>> >
>> >> We learned a few things during the medical translation project which
>> >> started back in 2011:
>> >>
>> >> 1) You must start with high quality content and thus all articles are
>> >> extensively improved before being proposed for translation.
>> >>
>> >> 2) A lot of languages want "less" content than is present on EN WP.
>> Thus
>> >> we
>> >> moved to just improving and suggesting for translation the leads of the
>> >> English articles.
>> >>
>> >> 3) The "Content Translation" tool developed by the WMF made efforts
>> more
>> >> efficient than handing around word documents. Would love to see that
>> tool
>> >> improved further such as having it support specific lists of articles
>> that
>> >> are deemed ready for translation by certain groups. Would also love the
>> >> tool to have tracking metrics for these types of projects.
>> >>
>> >> 4) We used volunteer translators mostly associated with our partner
>> >> Translators Without Borders. One issue we found was that languages in
>> >> which
>> >> their are lots of translators such as French, Spanish, and Italian
>> there
>> >> is
>> >> often already at least some content on many of the topics in question.
>> The
>> >> issue than becomes integration which needs an expert Wikipedia. And for
>> >> languages in which we have little content there are often few avaliable
>> >> volunteers.
>> >>
>> >> 5) With respect to "paying per word" the problem is this would require
>> >> significant checks and balances to make sure people are taking the work
>> >> seriously and not simple using Google translate for the 70 or so
>> languages
>> >> in which it claims to work. We often had translations undergo a second
>> >> review and the volunteers at TWB have to pass certain tests to be
>> >> accepted.
>> >>
>> >> 6) I hired a coordinator for the translation project for a couple of
>> >> years.
>> >> The translators at TWB did not want to become Wikipedians or learn how
>> to
>> >> use

Re: [Wikimedia-l] Paid translation

Should have added that the remaining points are somewhat less interesting
in this context. Preloading a set of articles is a bad idea, the
translators should be able to chose for themselves. Articles should also be
pretty broad, not very narrow technical or medical, ie vertical articles,
as the number of editors that can handle those will be pretty small.

In particular: Do not believe you can turn a teanslator into a new editor!
You can although turn an existing editor into a translator.

On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad  wrote:

> 1) You must start with high quality content and thus all articles are
>> extensively improved before being proposed for translation.
>
>
> Note that to much pressure on "quality" can easily kill the project.
>
> 3) The "Content Translation" tool developed by the WMF made efforts more
>> efficient than handing around word documents. Would love to see that tool
>> improved further such as having it support specific lists of articles that
>> are deemed ready for translation by certain groups. Would also love the
>> tool to have tracking metrics for these types of projects.
>
>
> Didn't mention ContentTranslation, but it should be pretty obvious.
>
> 4) We used volunteer translators mostly associated with our partner
>> Translators Without Borders. One issue we found was that languages in
>> which
>> their are lots of translators such as French, Spanish, and Italian there
>> is
>> often already at least some content on many of the topics in question. The
>> issue than becomes integration which needs an expert Wikipedia. And for
>> languages in which we have little content there are often few avaliable
>> volunteers.
>
>
> I used projects below 65k articles as an example, as the chance of
> competing articles are pretty low.
>
> 5) With respect to "paying per word" the problem is this would require
>> significant checks and balances to make sure people are taking the work
>> seriously and not simple using Google translate for the 70 or so languages
>> in which it claims to work. We often had translations undergo a second
>> review and the volunteers at TWB have to pass certain tests to be
>> accepted.
>
>
> I'n my original email I wrote "verified good translators". It is as
> simple as "Has the editor contributed other articles at the project?"
>
> On Sat, Feb 24, 2018 at 2:26 PM, James Heilman  wrote:
>
>> We learned a few things during the medical translation project which
>> started back in 2011:
>>
>> 1) You must start with high quality content and thus all articles are
>> extensively improved before being proposed for translation.
>>
>> 2) A lot of languages want "less" content than is present on EN WP. Thus
>> we
>> moved to just improving and suggesting for translation the leads of the
>> English articles.
>>
>> 3) The "Content Translation" tool developed by the WMF made efforts more
>> efficient than handing around word documents. Would love to see that tool
>> improved further such as having it support specific lists of articles that
>> are deemed ready for translation by certain groups. Would also love the
>> tool to have tracking metrics for these types of projects.
>>
>> 4) We used volunteer translators mostly associated with our partner
>> Translators Without Borders. One issue we found was that languages in
>> which
>> their are lots of translators such as French, Spanish, and Italian there
>> is
>> often already at least some content on many of the topics in question. The
>> issue than becomes integration which needs an expert Wikipedia. And for
>> languages in which we have little content there are often few avaliable
>> volunteers.
>>
>> 5) With respect to "paying per word" the problem is this would require
>> significant checks and balances to make sure people are taking the work
>> seriously and not simple using Google translate for the 70 or so languages
>> in which it claims to work. We often had translations undergo a second
>> review and the volunteers at TWB have to pass certain tests to be
>> accepted.
>>
>> 6) I hired a coordinator for the translation project for a couple of
>> years.
>> The translators at TWB did not want to become Wikipedians or learn how to
>> use our systems. The coordinator created account like TransSW001 (one for
>> each volunteer) and preloaded the article to be translated into Content
>> Translation. They than gave the volunteer translator the user name and
>> password to the account.
>>
>> 7) Were are we at now? There are currently just over 1,000 leads of
>> articles that have been improved and are ready for translation. This
>> includes articles on the 440 medications that are on the WHO Essential
>> List. We have worked a bit in some 100 languages. The efforts have
>> resulted
>> in more than 5 million works translated and integrated into different
>> Wikipedias. The coordinator has unfortunately moved on to his real job of
>> teaching high school students.
>>
>> 8) The

Re: [Wikimedia-l] Paid translation

>
> 1) You must start with high quality content and thus all articles are
> extensively improved before being proposed for translation.


Note that to much pressure on "quality" can easily kill the project.

3) The "Content Translation" tool developed by the WMF made efforts more
> efficient than handing around word documents. Would love to see that tool
> improved further such as having it support specific lists of articles that
> are deemed ready for translation by certain groups. Would also love the
> tool to have tracking metrics for these types of projects.


Didn't mention ContentTranslation, but it should be pretty obvious.

4) We used volunteer translators mostly associated with our partner
> Translators Without Borders. One issue we found was that languages in which
> their are lots of translators such as French, Spanish, and Italian there is
> often already at least some content on many of the topics in question. The
> issue than becomes integration which needs an expert Wikipedia. And for
> languages in which we have little content there are often few avaliable
> volunteers.


I used projects below 65k articles as an example, as the chance of
competing articles are pretty low.

5) With respect to "paying per word" the problem is this would require
> significant checks and balances to make sure people are taking the work
> seriously and not simple using Google translate for the 70 or so languages
> in which it claims to work. We often had translations undergo a second
> review and the volunteers at TWB have to pass certain tests to be accepted.


I'n my original email I wrote "verified good translators". It is as simple
as "Has the editor contributed other articles at the project?"

On Sat, Feb 24, 2018 at 2:26 PM, James Heilman  wrote:

> We learned a few things during the medical translation project which
> started back in 2011:
>
> 1) You must start with high quality content and thus all articles are
> extensively improved before being proposed for translation.
>
> 2) A lot of languages want "less" content than is present on EN WP. Thus we
> moved to just improving and suggesting for translation the leads of the
> English articles.
>
> 3) The "Content Translation" tool developed by the WMF made efforts more
> efficient than handing around word documents. Would love to see that tool
> improved further such as having it support specific lists of articles that
> are deemed ready for translation by certain groups. Would also love the
> tool to have tracking metrics for these types of projects.
>
> 4) We used volunteer translators mostly associated with our partner
> Translators Without Borders. One issue we found was that languages in which
> their are lots of translators such as French, Spanish, and Italian there is
> often already at least some content on many of the topics in question. The
> issue than becomes integration which needs an expert Wikipedia. And for
> languages in which we have little content there are often few avaliable
> volunteers.
>
> 5) With respect to "paying per word" the problem is this would require
> significant checks and balances to make sure people are taking the work
> seriously and not simple using Google translate for the 70 or so languages
> in which it claims to work. We often had translations undergo a second
> review and the volunteers at TWB have to pass certain tests to be accepted.
>
> 6) I hired a coordinator for the translation project for a couple of years.
> The translators at TWB did not want to become Wikipedians or learn how to
> use our systems. The coordinator created account like TransSW001 (one for
> each volunteer) and preloaded the article to be translated into Content
> Translation. They than gave the volunteer translator the user name and
> password to the account.
>
> 7) Were are we at now? There are currently just over 1,000 leads of
> articles that have been improved and are ready for translation. This
> includes articles on the 440 medications that are on the WHO Essential
> List. We have worked a bit in some 100 languages. The efforts have resulted
> in more than 5 million works translated and integrated into different
> Wikipedias. The coordinator has unfortunately moved on to his real job of
> teaching high school students.
>
> 8) The project continues but at a slower pace than before. The Wikipedian
> and retired orthopedic surgeon Subas Chandra Rout has basically single
> handedly translated nearly all 1,000 leads into Odia a language spoken by
> 40 million people in Eastern India. The amazing thing is that for many of
> these topics this is the first and only information online about it. Google
> translate does not even claim to work in this language. Our partnerships
> with WMTW and medical school in Taipai continue to translate into Chinese.
> There the students translate and than their translations are reviewed by
> their profs before being posted. They translate in groups using hackpad to
> make it more social.

Re: [Wikimedia-l] Paid translation

Articles about LGBT topics would be great! Similarly topics about important
women in the third world. I would also like a focus on articles about
primary health. Perhaps also agriculture.

I'm not sure if it is wise to move this out into chapters, keep it simple,
but perhaps community groups should be able to make direct feedback.

On Sat, Feb 24, 2018 at 2:04 PM, Lane Rasberry 
wrote:

> Excellent idea.
>
> I have the idea that the WMF invests $10,000 in the developing world to
> recruit $1000 of volunteer labor.
>
> We need to be realistic about the relative costs of doing Western-style,
> rich country outreach in all economies. In the past, the strategy has been
> to fund the recruitment of volunteers and avoid hiring content producing
> staff no matter the outcomes, cost, or impact.
>
> In the spectrum of the average income of individuals in some places, it is
> obviously easier and more impactful to hire someone with a masters degree
> to outright produce content than to pay for a program which will recruit
> volunteers.
>
> Obviously, the WMF cannot and will never pay for content. However, I think
> that we need to make it easier for Wikimedia chapters, community groups,
> and partner organizations to hire paid contributors. Translation is the
> most obvious place to start because having base content in an encyclopedia
> is the foundation for demonstrating the legitimacy and value of Wikimedia
> projects. Funding should go from WMF to chapters to paid staff for content.
>
> To make this project a go we would need to have a conversation about what
> sort of content is a priority for translation. I have a draft of an idea
> for prioritizing content for translation.
> 
> The idea is that for any given field of study, subject matter experts
> identify 99 articles in that field which they deem and come to consensus as
> priorities for having a global conversation in that field. So for example,
> if a group funds translation of LGBT+ content, then we would need to
> develop a canon of LGBT topics to which everyone in the world would have
> access. I have no idea how to choose topics, but fewer than 100 is probably
> not enough and more than 100 is probably too much for an all-languages
> translation project. I could use some help drafting guidelines for how to
> make priorities for what to translate given limited resources.
>
>
> On Sat, Feb 24, 2018 at 7:51 AM, John Erling Blad 
> wrote:
>
> > This discussion is going to be fun! =D
> >
> > A little more than seventy Wikipedia-projects has more than 65k articles,
> > the remaining two hundred or so are pretty small.
> >
> > What if a base set of articles were opened for paid translators? There
> are
> > several lists of such base sets. We have both the thousand articles from
> > "List of articles every Wikipedia should have"[1] and and the ten
> thousand
> > articles from the expanded list[2].
> >
> > Lets say verified good translators was paid about $0.01 per word (about
> $1
> > for a 1k-article) for translating one of those articles into another
> > language, with perhaps a higher pay for contributors in high-cost
> > countries. The pay would also have to be higher for languages that lacks
> > good translation tools.
> >
> > I believe this would be an _enabling_ activity for the communities, as
> > without a base set of articles it won't be possible to build a community
> at
> > all. By not paying for new articles, and only translating well-referenced
> > articles, some of the disputes in the communities could be avoided.
> Perhaps
> > we should also identify good source articles, that would be a help.
> > Translated articles should be above some minimum size, but they does not
> > have to be full translations of the source article.
> >
> > A real problem is that our existing lists of good articles other projects
> > should have is pretty much biased towards Western World, so they need a
> lot
> > of adjustments. Perhaps such a project would identify our inherit bias?
> >
> > [1]
> > https://meta.wikimedia.org/wiki/List_of_articles_every_
> > Wikipedia_should_have
> > [2]
> > https://meta.wikimedia.org/wiki/List_of_articles_every_
> > Wikipedia_should_have/Expanded
> > ___
> > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> > wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> > wiki/Wikimedia-l
> > New messages to: Wikimedia-l@lists.wikimedia.org
> > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> > 
>
>
>
>
> --
> Lane Rasberry
> user:bluerasberry on Wikipedia
> 206.801.0814
> l...@bluerasberry.com
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
>

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread James Salsman

>...  make sure people are taking the work seriously and not
> simply using Google translate

People are likely to start with Google Translate whether they are
taking the translation seriously or not, so it would still help if we
could get Google to provide numeric per-word translation confidence
scores.

So please star the request for those at:
https://issuetracker.google.com/issues/73830349

Thank you!

Best regards,
Jim

On Sat, Feb 24, 2018 at 7:18 AM, James Heilman  wrote:
> One further case, some of the translations we did into Swahili had funding
> associated with them. Few people in the country have easy access to a
> computer and cellphones are not as suitable for translation work. Basically
> TWB has a brick and mortar translation center in Nairobi with computers.
> They have staff that keep an eye on the center. People were recruited,
> provided instruction, provided access to the computers, and provided cell
> phone credits for their involvement. What they worked on helped them
> develop a CV.
>
> James
>
> On Sat, Feb 24, 2018 at 6:30 AM, James Heilman  wrote:
>
>> Meant to write "more than 5 million words translated". Apologies.
>>
>> James
>>
>> On Sat, Feb 24, 2018 at 6:26 AM, James Heilman  wrote:
>>
>>> We learned a few things during the medical translation project which
>>> started back in 2011:
>>>
>>> 1) You must start with high quality content and thus all articles are
>>> extensively improved before being proposed for translation.
>>>
>>> 2) A lot of languages want "less" content than is present on EN WP. Thus
>>> we moved to just improving and suggesting for translation the leads of the
>>> English articles.
>>>
>>> 3) The "Content Translation" tool developed by the WMF made efforts more
>>> efficient than handing around word documents. Would love to see that tool
>>> improved further such as having it support specific lists of articles that
>>> are deemed ready for translation by certain groups. Would also love the
>>> tool to have tracking metrics for these types of projects.
>>>
>>> 4) We used volunteer translators mostly associated with our partner
>>> Translators Without Borders. One issue we found was that languages in which
>>> their are lots of translators such as French, Spanish, and Italian there is
>>> often already at least some content on many of the topics in question. The
>>> issue than becomes integration which needs an expert Wikipedia. And for
>>> languages in which we have little content there are often few avaliable
>>> volunteers.
>>>
>>> 5) With respect to "paying per word" the problem is this would require
>>> significant checks and balances to make sure people are taking the work
>>> seriously and not simple using Google translate for the 70 or so languages
>>> in which it claims to work. We often had translations undergo a second
>>> review and the volunteers at TWB have to pass certain tests to be accepted.
>>>
>>> 6) I hired a coordinator for the translation project for a couple of
>>> years. The translators at TWB did not want to become Wikipedians or learn
>>> how to use our systems. The coordinator created account like TransSW001
>>> (one for each volunteer) and preloaded the article to be translated into
>>> Content Translation. They than gave the volunteer translator the user name
>>> and password to the account.
>>>
>>> 7) Were are we at now? There are currently just over 1,000 leads of
>>> articles that have been improved and are ready for translation. This
>>> includes articles on the 440 medications that are on the WHO Essential
>>> List. We have worked a bit in some 100 languages. The efforts have resulted
>>> in more than 5 million works translated and integrated into different
>>> Wikipedias. The coordinator has unfortunately moved on to his real job of
>>> teaching high school students.
>>>
>>> 8) The project continues but at a slower pace than before. The Wikipedian
>>> and retired orthopedic surgeon Subas Chandra Rout has basically single
>>> handedly translated nearly all 1,000 leads into Odia a language spoken by
>>> 40 million people in Eastern India. The amazing thing is that for many of
>>> these topics this is the first and only information online about it. Google
>>> translate does not even claim to work in this language. Our partnerships
>>> with WMTW and medical school in Taipai continue to translate into Chinese.
>>> There the students translate and than their translations are reviewed by
>>> their profs before being posted. They translate in groups using hackpad to
>>> make it more social.
>>>
>>> I am currently working to re invigorate the project :-)
>>> James
>>>
>>> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad 
>>> wrote:
>>>
 This discussion is going to be fun! =D

 A little more than seventy Wikipedia-projects has more than 65k articles,
 the remaining two hundred or so are pretty small.

 What if a base set of articles

Re: [Wikimedia-l] Paid translation

One further case, some of the translations we did into Swahili had funding
associated with them. Few people in the country have easy access to a
computer and cellphones are not as suitable for translation work. Basically
TWB has a brick and mortar translation center in Nairobi with computers.
They have staff that keep an eye on the center. People were recruited,
provided instruction, provided access to the computers, and provided cell
phone credits for their involvement. What they worked on helped them
develop a CV.

James

On Sat, Feb 24, 2018 at 6:30 AM, James Heilman  wrote:

> Meant to write "more than 5 million words translated". Apologies.
>
> James
>
> On Sat, Feb 24, 2018 at 6:26 AM, James Heilman  wrote:
>
>> We learned a few things during the medical translation project which
>> started back in 2011:
>>
>> 1) You must start with high quality content and thus all articles are
>> extensively improved before being proposed for translation.
>>
>> 2) A lot of languages want "less" content than is present on EN WP. Thus
>> we moved to just improving and suggesting for translation the leads of the
>> English articles.
>>
>> 3) The "Content Translation" tool developed by the WMF made efforts more
>> efficient than handing around word documents. Would love to see that tool
>> improved further such as having it support specific lists of articles that
>> are deemed ready for translation by certain groups. Would also love the
>> tool to have tracking metrics for these types of projects.
>>
>> 4) We used volunteer translators mostly associated with our partner
>> Translators Without Borders. One issue we found was that languages in which
>> their are lots of translators such as French, Spanish, and Italian there is
>> often already at least some content on many of the topics in question. The
>> issue than becomes integration which needs an expert Wikipedia. And for
>> languages in which we have little content there are often few avaliable
>> volunteers.
>>
>> 5) With respect to "paying per word" the problem is this would require
>> significant checks and balances to make sure people are taking the work
>> seriously and not simple using Google translate for the 70 or so languages
>> in which it claims to work. We often had translations undergo a second
>> review and the volunteers at TWB have to pass certain tests to be accepted.
>>
>> 6) I hired a coordinator for the translation project for a couple of
>> years. The translators at TWB did not want to become Wikipedians or learn
>> how to use our systems. The coordinator created account like TransSW001
>> (one for each volunteer) and preloaded the article to be translated into
>> Content Translation. They than gave the volunteer translator the user name
>> and password to the account.
>>
>> 7) Were are we at now? There are currently just over 1,000 leads of
>> articles that have been improved and are ready for translation. This
>> includes articles on the 440 medications that are on the WHO Essential
>> List. We have worked a bit in some 100 languages. The efforts have resulted
>> in more than 5 million works translated and integrated into different
>> Wikipedias. The coordinator has unfortunately moved on to his real job of
>> teaching high school students.
>>
>> 8) The project continues but at a slower pace than before. The Wikipedian
>> and retired orthopedic surgeon Subas Chandra Rout has basically single
>> handedly translated nearly all 1,000 leads into Odia a language spoken by
>> 40 million people in Eastern India. The amazing thing is that for many of
>> these topics this is the first and only information online about it. Google
>> translate does not even claim to work in this language. Our partnerships
>> with WMTW and medical school in Taipai continue to translate into Chinese.
>> There the students translate and than their translations are reviewed by
>> their profs before being posted. They translate in groups using hackpad to
>> make it more social.
>>
>> I am currently working to re invigorate the project :-)
>> James
>>
>> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad 
>> wrote:
>>
>>> This discussion is going to be fun! =D
>>>
>>> A little more than seventy Wikipedia-projects has more than 65k articles,
>>> the remaining two hundred or so are pretty small.
>>>
>>> What if a base set of articles were opened for paid translators? There
>>> are
>>> several lists of such base sets. We have both the thousand articles from
>>> "List of articles every Wikipedia should have"[1] and and the ten
>>> thousand
>>> articles from the expanded list[2].
>>>
>>> Lets say verified good translators was paid about $0.01 per word (about
>>> $1
>>> for a 1k-article) for translating one of those articles into another
>>> language, with perhaps a higher pay for contributors in high-cost
>>> countries. The pay would also have to be higher for languages that lacks
>>> good translation tools.
>>>
>>> I believe

Re: [Wikimedia-l] Paid translation

Meant to write "more than 5 million words translated". Apologies.

James

On Sat, Feb 24, 2018 at 6:26 AM, James Heilman  wrote:

> We learned a few things during the medical translation project which
> started back in 2011:
>
> 1) You must start with high quality content and thus all articles are
> extensively improved before being proposed for translation.
>
> 2) A lot of languages want "less" content than is present on EN WP. Thus
> we moved to just improving and suggesting for translation the leads of the
> English articles.
>
> 3) The "Content Translation" tool developed by the WMF made efforts more
> efficient than handing around word documents. Would love to see that tool
> improved further such as having it support specific lists of articles that
> are deemed ready for translation by certain groups. Would also love the
> tool to have tracking metrics for these types of projects.
>
> 4) We used volunteer translators mostly associated with our partner
> Translators Without Borders. One issue we found was that languages in which
> their are lots of translators such as French, Spanish, and Italian there is
> often already at least some content on many of the topics in question. The
> issue than becomes integration which needs an expert Wikipedia. And for
> languages in which we have little content there are often few avaliable
> volunteers.
>
> 5) With respect to "paying per word" the problem is this would require
> significant checks and balances to make sure people are taking the work
> seriously and not simple using Google translate for the 70 or so languages
> in which it claims to work. We often had translations undergo a second
> review and the volunteers at TWB have to pass certain tests to be accepted.
>
> 6) I hired a coordinator for the translation project for a couple of
> years. The translators at TWB did not want to become Wikipedians or learn
> how to use our systems. The coordinator created account like TransSW001
> (one for each volunteer) and preloaded the article to be translated into
> Content Translation. They than gave the volunteer translator the user name
> and password to the account.
>
> 7) Were are we at now? There are currently just over 1,000 leads of
> articles that have been improved and are ready for translation. This
> includes articles on the 440 medications that are on the WHO Essential
> List. We have worked a bit in some 100 languages. The efforts have resulted
> in more than 5 million works translated and integrated into different
> Wikipedias. The coordinator has unfortunately moved on to his real job of
> teaching high school students.
>
> 8) The project continues but at a slower pace than before. The Wikipedian
> and retired orthopedic surgeon Subas Chandra Rout has basically single
> handedly translated nearly all 1,000 leads into Odia a language spoken by
> 40 million people in Eastern India. The amazing thing is that for many of
> these topics this is the first and only information online about it. Google
> translate does not even claim to work in this language. Our partnerships
> with WMTW and medical school in Taipai continue to translate into Chinese.
> There the students translate and than their translations are reviewed by
> their profs before being posted. They translate in groups using hackpad to
> make it more social.
>
> I am currently working to re invigorate the project :-)
> James
>
> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad 
> wrote:
>
>> This discussion is going to be fun! =D
>>
>> A little more than seventy Wikipedia-projects has more than 65k articles,
>> the remaining two hundred or so are pretty small.
>>
>> What if a base set of articles were opened for paid translators? There are
>> several lists of such base sets. We have both the thousand articles from
>> "List of articles every Wikipedia should have"[1] and and the ten thousand
>> articles from the expanded list[2].
>>
>> Lets say verified good translators was paid about $0.01 per word (about $1
>> for a 1k-article) for translating one of those articles into another
>> language, with perhaps a higher pay for contributors in high-cost
>> countries. The pay would also have to be higher for languages that lacks
>> good translation tools.
>>
>> I believe this would be an _enabling_ activity for the communities, as
>> without a base set of articles it won't be possible to build a community
>> at
>> all. By not paying for new articles, and only translating well-referenced
>> articles, some of the disputes in the communities could be avoided.
>> Perhaps
>> we should also identify good source articles, that would be a help.
>> Translated articles should be above some minimum size, but they does not
>> have to be full translations of the source article.
>>
>> A real problem is that our existing lists of good articles other projects
>> should have is pretty much biased towards Western World, so they need a
>> lot
>> of adjustments. Perhaps such a

Re: [Wikimedia-l] Paid translation

We learned a few things during the medical translation project which
started back in 2011:

1) You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.

2) A lot of languages want "less" content than is present on EN WP. Thus we
moved to just improving and suggesting for translation the leads of the
English articles.

3) The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool
improved further such as having it support specific lists of articles that
are deemed ready for translation by certain groups. Would also love the
tool to have tracking metrics for these types of projects.

4) We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which
their are lots of translators such as French, Spanish, and Italian there is
often already at least some content on many of the topics in question. The
issue than becomes integration which needs an expert Wikipedia. And for
languages in which we have little content there are often few avaliable
volunteers.

5) With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work
seriously and not simple using Google translate for the 70 or so languages
in which it claims to work. We often had translations undergo a second
review and the volunteers at TWB have to pass certain tests to be accepted.

6) I hired a coordinator for the translation project for a couple of years.
The translators at TWB did not want to become Wikipedians or learn how to
use our systems. The coordinator created account like TransSW001 (one for
each volunteer) and preloaded the article to be translated into Content
Translation. They than gave the volunteer translator the user name and
password to the account.

7) Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This
includes articles on the 440 medications that are on the WHO Essential
List. We have worked a bit in some 100 languages. The efforts have resulted
in more than 5 million works translated and integrated into different
Wikipedias. The coordinator has unfortunately moved on to his real job of
teaching high school students.

8) The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single
handedly translated nearly all 1,000 leads into Odia a language spoken by
40 million people in Eastern India. The amazing thing is that for many of
these topics this is the first and only information online about it. Google
translate does not even claim to work in this language. Our partnerships
with WMTW and medical school in Taipai continue to translate into Chinese.
There the students translate and than their translations are reviewed by
their profs before being posted. They translate in groups using hackpad to
make it more social.

I am currently working to re invigorate the project :-)
James

On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad  wrote:

> This discussion is going to be fun! =D
>
> A little more than seventy Wikipedia-projects has more than 65k articles,
> the remaining two hundred or so are pretty small.
>
> What if a base set of articles were opened for paid translators? There are
> several lists of such base sets. We have both the thousand articles from
> "List of articles every Wikipedia should have"[1] and and the ten thousand
> articles from the expanded list[2].
>
> Lets say verified good translators was paid about $0.01 per word (about $1
> for a 1k-article) for translating one of those articles into another
> language, with perhaps a higher pay for contributors in high-cost
> countries. The pay would also have to be higher for languages that lacks
> good translation tools.
>
> I believe this would be an _enabling_ activity for the communities, as
> without a base set of articles it won't be possible to build a community at
> all. By not paying for new articles, and only translating well-referenced
> articles, some of the disputes in the communities could be avoided. Perhaps
> we should also identify good source articles, that would be a help.
> Translated articles should be above some minimum size, but they does not
> have to be full translations of the source article.
>
> A real problem is that our existing lists of good articles other projects
> should have is pretty much biased towards Western World, so they need a lot
> of adjustments. Perhaps such a project would identify our inherit bias?
>
> [1]
> https://meta.wikimedia.org/wiki/List_of_articles_every_
> Wikipedia_should_have
> [2]
> https://meta.wikimedia.org/wiki/List_of_articles_every_
> Wikipedia_should_have/Expanded
> ___
> Wikimedia-l mailing

Re: [Wikimedia-l] Paid translation

2018-02-24 Thread Lane Rasberry

Excellent idea.

I have the idea that the WMF invests $10,000 in the developing world to
recruit $1000 of volunteer labor.

We need to be realistic about the relative costs of doing Western-style,
rich country outreach in all economies. In the past, the strategy has been
to fund the recruitment of volunteers and avoid hiring content producing
staff no matter the outcomes, cost, or impact.

In the spectrum of the average income of individuals in some places, it is
obviously easier and more impactful to hire someone with a masters degree
to outright produce content than to pay for a program which will recruit
volunteers.

Obviously, the WMF cannot and will never pay for content. However, I think
that we need to make it easier for Wikimedia chapters, community groups,
and partner organizations to hire paid contributors. Translation is the
most obvious place to start because having base content in an encyclopedia
is the foundation for demonstrating the legitimacy and value of Wikimedia
projects. Funding should go from WMF to chapters to paid staff for content.

To make this project a go we would need to have a conversation about what
sort of content is a priority for translation. I have a draft of an idea
for prioritizing content for translation.

The idea is that for any given field of study, subject matter experts
identify 99 articles in that field which they deem and come to consensus as
priorities for having a global conversation in that field. So for example,
if a group funds translation of LGBT+ content, then we would need to
develop a canon of LGBT topics to which everyone in the world would have
access. I have no idea how to choose topics, but fewer than 100 is probably
not enough and more than 100 is probably too much for an all-languages
translation project. I could use some help drafting guidelines for how to
make priorities for what to translate given limited resources.

On Sat, Feb 24, 2018 at 7:51 AM, John Erling Blad  wrote:

> This discussion is going to be fun! =D
>
> A little more than seventy Wikipedia-projects has more than 65k articles,
> the remaining two hundred or so are pretty small.
>
> What if a base set of articles were opened for paid translators? There are
> several lists of such base sets. We have both the thousand articles from
> "List of articles every Wikipedia should have"[1] and and the ten thousand
> articles from the expanded list[2].
>
> Lets say verified good translators was paid about $0.01 per word (about $1
> for a 1k-article) for translating one of those articles into another
> language, with perhaps a higher pay for contributors in high-cost
> countries. The pay would also have to be higher for languages that lacks
> good translation tools.
>
> I believe this would be an _enabling_ activity for the communities, as
> without a base set of articles it won't be possible to build a community at
> all. By not paying for new articles, and only translating well-referenced
> articles, some of the disputes in the communities could be avoided. Perhaps
> we should also identify good source articles, that would be a help.
> Translated articles should be above some minimum size, but they does not
> have to be full translations of the source article.
>
> A real problem is that our existing lists of good articles other projects
> should have is pretty much biased towards Western World, so they need a lot
> of adjustments. Perhaps such a project would identify our inherit bias?
>
> [1]
> https://meta.wikimedia.org/wiki/List_of_articles_every_
> Wikipedia_should_have
> [2]
> https://meta.wikimedia.org/wiki/List_of_articles_every_
> Wikipedia_should_have/Expanded
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 

-- 
Lane Rasberry
user:bluerasberry on Wikipedia
206.801.0814
l...@bluerasberry.com
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

Re: [Wikimedia-l] Paid translation

I should probably say that I don't believe our present "lists of articles
every Wikipedia should have" are really good. I believe the lists should
reflect what people from different places actually reads, or try to read,
but normalized to a global perspective. That is also a necessity if the
purpose is to create local communities in the different languages. Think
globally, act locally!

On Sat, Feb 24, 2018 at 1:51 PM, John Erling Blad  wrote:

> This discussion is going to be fun! =D
>
> A little more than seventy Wikipedia-projects has more than 65k articles,
> the remaining two hundred or so are pretty small.
>
> What if a base set of articles were opened for paid translators? There are
> several lists of such base sets. We have both the thousand articles from
> "List of articles every Wikipedia should have"[1] and and the ten thousand
> articles from the expanded list[2].
>
> Lets say verified good translators was paid about $0.01 per word (about $1
> for a 1k-article) for translating one of those articles into another
> language, with perhaps a higher pay for contributors in high-cost
> countries. The pay would also have to be higher for languages that lacks
> good translation tools.
>
> I believe this would be an _enabling_ activity for the communities, as
> without a base set of articles it won't be possible to build a community at
> all. By not paying for new articles, and only translating well-referenced
> articles, some of the disputes in the communities could be avoided. Perhaps
> we should also identify good source articles, that would be a help.
> Translated articles should be above some minimum size, but they does not
> have to be full translations of the source article.
>
> A real problem is that our existing lists of good articles other projects
> should have is pretty much biased towards Western World, so they need a lot
> of adjustments. Perhaps such a project would identify our inherit bias?
>
> [1] https://meta.wikimedia.org/wiki/List_of_articles_
> every_Wikipedia_should_have
> [2] https://meta.wikimedia.org/wiki/List_of_articles_
> every_Wikipedia_should_have/Expanded
>
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,

[Wikimedia-l] Paid translation