Re: [Wikimedia-l] Paid translation

[James Salsman]

If the Foundation Endowment paid for translations of articles across
Wikipedias, it would still be like a Foundation Grant in terms of the
legal effect on the DMCA safe harbor provisions and the practical
effect on whether mistakes could bring the Foundation into disrepute.

Maybe the Foundation could pay for translations, as long as a much
smaller independent third party was reviewing them for fidelity and
freedom from bias under conditions where a group of people are trying
to confound the paid reviewers by including a constant but small
proportion of intentionally inaccurate and biased proposed
translations to make sure that the reviewer quality is sufficient.

If that doesn't work, then the independent third party anti-bias QA
organization could grow to do the translation, perhaps as a thematic
organization supported by both outside and less than half internal
Foundation grants.

Best regards,
Jim

On Sun, Mar 4, 2018 at 2:53 AM, WereSpielChequers
 wrote:
> Pine, there is one possible way to fund such translation in the future; The
> Foundation is building up an endowment. When that endowment has grown to
> the point where the annual return is sufficient to fund the Foundation,
> then you could re-purpose the annual fundraiser from collecting money to
> host Wikipedia, to collecting money to make Wikipedia available in other
> languages.
>
> If I'm correct in thinking that part of the problem for many of our widely
> spoken languages with weak wikipedias is that the more educated people who
> speak those languages are more likely to contribute edits in what is to
> them a  higher status or more language or one more useful to their career,
> then maybe we should test using fundraiser  type advertising to ask our
> English readers in places like India to translate articles from English to
> Indic languages.
>
> In some parts of the world where incomes are generally very low and
> financial donations reflect that perhaps we have little to lose by shifting
> now from asking for funds to asking for content donations, especially in
> the language of that area.
>
> WereSpielChequers
>
>>
>>
>>
>> --
>>
>> Message: 2
>> Date: Sat, 3 Mar 2018 18:13:38 -0800
>> From: Pine W 
>> To: Wikimedia Mailing List 
>> Subject: Re: [Wikimedia-l] Paid translation
>> Message-ID:
>> 

Re: [Wikimedia-l] Paid translation

[Peter Southwood]

Difficult, yes. Impossible, no. 
Part of the problem can be that some Wikipedias are somewhat fussy about the 
language you use. There are users who object to anyone using a word not 
approved by some authority, but cannot suggest what to do when there is no such 
word. 
Cheers,
Peter

-Original Message-
From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On Behalf Of 
Amir E. Aharoni
Sent: 04 March 2018 15:59
To: Wikimedia Mailing List
Subject: Re: [Wikimedia-l] Paid translation

Yes, I mentioned something like this in one of my emails in this thread.

Every language goes through a period of creating terminology. Some languages 
successfully create native words (Icelandic is a famous example), some 
languages are fine with taking foreign words (um, English took a lot from 
Latin, Greek and other languages), some are a mix (Russian). You can never say 
"it's *impossible* to write about science in this language"; you can, at most, 
say "it's *difficult* to write about science in this language *today".

People who speak a language that had already overcome this problem must 
remember that their language didn't always have this terminology. That's one of 
the reasons why the resolution "just learn our language instead of investing in 
your own" may be practical, but isn't very fair.

People who speak a language that hadn't yet overcome this must remember that 
it's a challenge, but not a blocker. A translator who cares about their 
language can overcome this with some ingenuity and resourcefulness.
(Teaser: I'm about to publish a blog post soon that talks about one language 
that is doing it now with considerable success.)


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

2018-03-04 15:22 GMT+02:00 Peter Southwood :

> Part of the problem may be that the vocabulary is lacking. It is very 
> difficult to explain a concept in one language when you know the words 
> only in another language, and it would be considered original research 
> by some Wikipedias to make up words for the job. I have struggled with 
> translations into Afrikaans, which has a reasonably extensive 
> technical vocabulary, and good electronic dictionary systems,, but 
> many concepts familiar to me in my fields of interest just do not have 
> Afrikaans words (yet).
> Cheers,
> Peter
>
> -Original Message-
> From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On 
> Behalf Of WereSpielChequers
> Sent: 04 March 2018 11:54
> To: Wikimedia Mailing List
> Subject: Re: [Wikimedia-l] Paid translation
>
> Pine, there is one possible way to fund such translation in the 
> future; The Foundation is building up an endowment. When that 
> endowment has grown to the point where the annual return is sufficient 
> to fund the Foundation, then you could re-purpose the annual 
> fundraiser from collecting money to host Wikipedia, to collecting 
> money to make Wikipedia available in other languages.
>
> If I'm correct in thinking that part of the problem for many of our 
> widely spoken languages with weak wikipedias is that the more educated 
> people who speak those languages are more likely to contribute edits 
> in what is to them a  higher status or more language or one more 
> useful to their career, then maybe we should test using fundraiser  
> type advertising to ask our English readers in places like India to 
> translate articles from English to Indic languages.
>
> In some parts of the world where incomes are generally very low and 
> financial donations reflect that perhaps we have little to lose by 
> shifting now from asking for funds to asking for content donations, 
> especially in the language of that area.
>
> WereSpielChequers
>
> >
> >
> >
> > --
> >
> > Message: 2
> > Date: Sat, 3 Mar 2018 18:13:38 -0800
> > From: Pine W 
> > To: Wikimedia Mailing List 
> > Subject: Re: [Wikimedia-l] Paid translation
> > Message-ID:
> > 

Re: [Wikimedia-l] Paid translation

[John Erling Blad]

Using a term from another language while creating an article and then later
localizing that term isn't that difficult, and should not be described as
impossible. What it does although identifies a problem with our current
production system; it is easy to move an article, but it is not easy to
make terms referring to that article or concept consistent.

On Sun, Mar 4, 2018 at 2:58 PM, Amir E. Aharoni <
amir.ahar...@mail.huji.ac.il> wrote:

> Yes, I mentioned something like this in one of my emails in this thread.
>
> Every language goes through a period of creating terminology. Some
> languages successfully create native words (Icelandic is a famous example),
> some languages are fine with taking foreign words (um, English took a lot
> from Latin, Greek and other languages), some are a mix (Russian). You can
> never say "it's *impossible* to write about science in this language"; you
> can, at most, say "it's *difficult* to write about science in this language
> *today".
>
> People who speak a language that had already overcome this problem must
> remember that their language didn't always have this terminology. That's
> one of the reasons why the resolution "just learn our language instead of
> investing in your own" may be practical, but isn't very fair.
>
> People who speak a language that hadn't yet overcome this must remember
> that it's a challenge, but not a blocker. A translator who cares about
> their language can overcome this with some ingenuity and resourcefulness.
> (Teaser: I'm about to publish a blog post soon that talks about one
> language that is doing it now with considerable success.)
>
>
> --
> Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
> http://aharoni.wordpress.com
> ‪“We're living in pieces,
> I want to live in peace.” – T. Moore‬
>
> 2018-03-04 15:22 GMT+02:00 Peter Southwood :
>
> > Part of the problem may be that the vocabulary is lacking. It is very
> > difficult to explain a concept in one language when you know the words
> only
> > in another language, and it would be considered original research by some
> > Wikipedias to make up words for the job. I have struggled with
> translations
> > into Afrikaans, which has a reasonably extensive technical vocabulary,
> and
> > good electronic dictionary systems,, but many concepts familiar to me in
> my
> > fields of interest just do not have Afrikaans words (yet).
> > Cheers,
> > Peter
> >
> > -Original Message-
> > From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> > Behalf Of WereSpielChequers
> > Sent: 04 March 2018 11:54
> > To: Wikimedia Mailing List
> > Subject: Re: [Wikimedia-l] Paid translation
> >
> > Pine, there is one possible way to fund such translation in the future;
> > The Foundation is building up an endowment. When that endowment has grown
> > to the point where the annual return is sufficient to fund the
> Foundation,
> > then you could re-purpose the annual fundraiser from collecting money to
> > host Wikipedia, to collecting money to make Wikipedia available in other
> > languages.
> >
> > If I'm correct in thinking that part of the problem for many of our
> widely
> > spoken languages with weak wikipedias is that the more educated people
> who
> > speak those languages are more likely to contribute edits in what is to
> > them a  higher status or more language or one more useful to their
> career,
> > then maybe we should test using fundraiser  type advertising to ask our
> > English readers in places like India to translate articles from English
> to
> > Indic languages.
> >
> > In some parts of the world where incomes are generally very low and
> > financial donations reflect that perhaps we have little to lose by
> shifting
> > now from asking for funds to asking for content donations, especially in
> > the language of that area.
> >
> > WereSpielChequers
> >
> > >
> > >
> > >
> > > --
> > >
> > > Message: 2
> > > Date: Sat, 3 Mar 2018 18:13:38 -0800
> > > From: Pine W 
> > > To: Wikimedia Mailing List 
> > > Subject: Re: [Wikimedia-l] Paid translation
> > > Message-ID:
> > > 

Re: [Wikimedia-l] Paid translation

[Amir E. Aharoni]

Yes, I mentioned something like this in one of my emails in this thread.

Every language goes through a period of creating terminology. Some
languages successfully create native words (Icelandic is a famous example),
some languages are fine with taking foreign words (um, English took a lot
from Latin, Greek and other languages), some are a mix (Russian). You can
never say "it's *impossible* to write about science in this language"; you
can, at most, say "it's *difficult* to write about science in this language
*today".

People who speak a language that had already overcome this problem must
remember that their language didn't always have this terminology. That's
one of the reasons why the resolution "just learn our language instead of
investing in your own" may be practical, but isn't very fair.

People who speak a language that hadn't yet overcome this must remember
that it's a challenge, but not a blocker. A translator who cares about
their language can overcome this with some ingenuity and resourcefulness.
(Teaser: I'm about to publish a blog post soon that talks about one
language that is doing it now with considerable success.)


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

2018-03-04 15:22 GMT+02:00 Peter Southwood :

> Part of the problem may be that the vocabulary is lacking. It is very
> difficult to explain a concept in one language when you know the words only
> in another language, and it would be considered original research by some
> Wikipedias to make up words for the job. I have struggled with translations
> into Afrikaans, which has a reasonably extensive technical vocabulary, and
> good electronic dictionary systems,, but many concepts familiar to me in my
> fields of interest just do not have Afrikaans words (yet).
> Cheers,
> Peter
>
> -Original Message-
> From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On
> Behalf Of WereSpielChequers
> Sent: 04 March 2018 11:54
> To: Wikimedia Mailing List
> Subject: Re: [Wikimedia-l] Paid translation
>
> Pine, there is one possible way to fund such translation in the future;
> The Foundation is building up an endowment. When that endowment has grown
> to the point where the annual return is sufficient to fund the Foundation,
> then you could re-purpose the annual fundraiser from collecting money to
> host Wikipedia, to collecting money to make Wikipedia available in other
> languages.
>
> If I'm correct in thinking that part of the problem for many of our widely
> spoken languages with weak wikipedias is that the more educated people who
> speak those languages are more likely to contribute edits in what is to
> them a  higher status or more language or one more useful to their career,
> then maybe we should test using fundraiser  type advertising to ask our
> English readers in places like India to translate articles from English to
> Indic languages.
>
> In some parts of the world where incomes are generally very low and
> financial donations reflect that perhaps we have little to lose by shifting
> now from asking for funds to asking for content donations, especially in
> the language of that area.
>
> WereSpielChequers
>
> >
> >
> >
> > --
> >
> > Message: 2
> > Date: Sat, 3 Mar 2018 18:13:38 -0800
> > From: Pine W 
> > To: Wikimedia Mailing List 
> > Subject: Re: [Wikimedia-l] Paid translation
> > Message-ID:
> > 

Re: [Wikimedia-l] Paid translation

[Peter Southwood]

Part of the problem may be that the vocabulary is lacking. It is very difficult 
to explain a concept in one language when you know the words only in another 
language, and it would be considered original research by some Wikipedias to 
make up words for the job. I have struggled with translations into Afrikaans, 
which has a reasonably extensive technical vocabulary, and good electronic 
dictionary systems,, but many concepts familiar to me in my fields of interest 
just do not have Afrikaans words (yet).
Cheers,
Peter

-Original Message-
From: Wikimedia-l [mailto:wikimedia-l-boun...@lists.wikimedia.org] On Behalf Of 
WereSpielChequers
Sent: 04 March 2018 11:54
To: Wikimedia Mailing List
Subject: Re: [Wikimedia-l] Paid translation

Pine, there is one possible way to fund such translation in the future; The 
Foundation is building up an endowment. When that endowment has grown to the 
point where the annual return is sufficient to fund the Foundation, then you 
could re-purpose the annual fundraiser from collecting money to host Wikipedia, 
to collecting money to make Wikipedia available in other languages.

If I'm correct in thinking that part of the problem for many of our widely 
spoken languages with weak wikipedias is that the more educated people who 
speak those languages are more likely to contribute edits in what is to them a  
higher status or more language or one more useful to their career, then maybe 
we should test using fundraiser  type advertising to ask our English readers in 
places like India to translate articles from English to Indic languages.

In some parts of the world where incomes are generally very low and financial 
donations reflect that perhaps we have little to lose by shifting now from 
asking for funds to asking for content donations, especially in the language of 
that area.

WereSpielChequers

>
>
>
> --
>
> Message: 2
> Date: Sat, 3 Mar 2018 18:13:38 -0800
> From: Pine W 
> To: Wikimedia Mailing List 
> Subject: Re: [Wikimedia-l] Paid translation
> Message-ID:
> 

Re: [Wikimedia-l] Paid translation

[John Erling Blad]

It should also be possible for an editor to let the payment go back to
foundation. This would probably be the case for many users in industrial
countries.

Perhaps it wasn't clear enough but the interface to manage translations
would be for someone other than the involved translators, aka a third
person within the local community, to accept the translation as valid and
good enough. After it is ticked off as "done" further payment of that
specific article will stop.

On Sun, Mar 4, 2018 at 11:27 AM, John Erling Blad  wrote:

> You guys are making the whole idea way to complex. There should be no
> editorial board. That goes against the whole wiki-way of doing things.
> There should be no additional foundation, that makes the whole idea
> unmanageable. It will also cost way more than the gain.
>
> Make thing DarnSimple™! A single list covering all universally valid
> topics that a true encyclopedia should cover. Leave it to the translator to
> chose which source article to use, as this creates the best opportunity to
> find translators. Allow other editors to join in after publication, but do
> respect the primary translators effort. Split the payment in one for the
> initial translation, and one for the followup edits. Cap them to avoid
> bloated articles.
>
> Make a DarnSimple™ interface to manage the translations, where the only
> action is for some identified user to tick of translated articles when they
> reach a certain threshold. In another interface the translator must
> identify himself with sufficient details to make the payment possible. This
> should be an optional part of the usual configuration of an account. All
> persons involved in the editing should have a split, but no payment will be
> done before the account for each editor reaches some threshold.
>
> Make the core list big enough to create a real encyclopedia, but small
> enough that there are room for local additions. There should probably be
> some way to specify local articles, like municipalities, important authors,
> and politicians. A good test is whether such additional articles makes
> sense in neighboring countries or languages. If it isn't possible to
> describe such things in a generic way they should probably be left out. I'm
> not sure if it should be possible to exclude articles, but I guess it will
> be an issue for some languages. Think Armenian genocide, which is
> problematic for some countries.
>
> A small single-book encyclopedia is about 60-70k articles, so lets say
> such a list would cover 25% of this. That would be a list of 15k articles.
> There are perhaps 50 Wikipedias that are large enough to be sustainable,
> and still small enough to miss articles on such a list. That would imply
> 750k articles,  thus plenty of articles for those that would like to
> translate one! Lets say this project is spread over 10 years with a cap on
> each article at 2x USD 10, then it would cost about USD 1500k each year. I
> believe that would be manageable. (Quite frankly I doubt it would be
> possible to find many enough translators, so this will never reach the
> proposed levels!)
>
___
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

[John Erling Blad]

You guys are making the whole idea way to complex. There should be no
editorial board. That goes against the whole wiki-way of doing things.
There should be no additional foundation, that makes the whole idea
unmanageable. It will also cost way more than the gain.

Make thing DarnSimple™! A single list covering all universally valid topics
that a true encyclopedia should cover. Leave it to the translator to chose
which source article to use, as this creates the best opportunity to find
translators. Allow other editors to join in after publication, but do
respect the primary translators effort. Split the payment in one for the
initial translation, and one for the followup edits. Cap them to avoid
bloated articles.

Make a DarnSimple™ interface to manage the translations, where the only
action is for some identified user to tick of translated articles when they
reach a certain threshold. In another interface the translator must
identify himself with sufficient details to make the payment possible. This
should be an optional part of the usual configuration of an account. All
persons involved in the editing should have a split, but no payment will be
done before the account for each editor reaches some threshold.

Make the core list big enough to create a real encyclopedia, but small
enough that there are room for local additions. There should probably be
some way to specify local articles, like municipalities, important authors,
and politicians. A good test is whether such additional articles makes
sense in neighboring countries or languages. If it isn't possible to
describe such things in a generic way they should probably be left out. I'm
not sure if it should be possible to exclude articles, but I guess it will
be an issue for some languages. Think Armenian genocide, which is
problematic for some countries.

A small single-book encyclopedia is about 60-70k articles, so lets say such
a list would cover 25% of this. That would be a list of 15k articles. There
are perhaps 50 Wikipedias that are large enough to be sustainable, and
still small enough to miss articles on such a list. That would imply 750k
articles,  thus plenty of articles for those that would like to translate
one! Lets say this project is spread over 10 years with a cap on each
article at 2x USD 10, then it would cost about USD 1500k each year. I
believe that would be manageable. (Quite frankly I doubt it would be
possible to find many enough translators, so this will never reach the
proposed levels!)
___
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

[WereSpielChequers]

Pine, there is one possible way to fund such translation in the future; The
Foundation is building up an endowment. When that endowment has grown to
the point where the annual return is sufficient to fund the Foundation,
then you could re-purpose the annual fundraiser from collecting money to
host Wikipedia, to collecting money to make Wikipedia available in other
languages.

If I'm correct in thinking that part of the problem for many of our widely
spoken languages with weak wikipedias is that the more educated people who
speak those languages are more likely to contribute edits in what is to
them a  higher status or more language or one more useful to their career,
then maybe we should test using fundraiser  type advertising to ask our
English readers in places like India to translate articles from English to
Indic languages.

In some parts of the world where incomes are generally very low and
financial donations reflect that perhaps we have little to lose by shifting
now from asking for funds to asking for content donations, especially in
the language of that area.

WereSpielChequers

>
>
>
> --
>
> Message: 2
> Date: Sat, 3 Mar 2018 18:13:38 -0800
> From: Pine W 
> To: Wikimedia Mailing List 
> Subject: Re: [Wikimedia-l] Paid translation
> Message-ID:
>