Re: [Wikitech-l] New feature: tool edit

2015-02-23 Thread Bináris
2015-02-11 14:02 GMT+01:00 Ricordisamoa ricordisa...@openmailbox.org:

 Keep in mind that it isn't always easy to tell 'tool' and 'bot' edits
 apart. Several scripts can perform actions whose degree of automation
 varies widely.
 For my part, I make most of my semi-automated edits using my bot's
 account, but many users also have separate 'flood' accounts for use with
 Wikidata Game https://tools.wmflabs.org/wikidata-game/ and similar
 tools.


Definitely this is the point. In enwiki's environment the word bot is
usually meant as a fully automated tool, while other communities treat it
differently. Let's see a major utility of Pywikibot, replace.py. This is
equally prepared for automatic and semiautomatic mode, and some tasks may
be solved automatically, while others -- above all spelling corrections --
manually. This still means a very high speed rate of editing but it is
human-controlled.
If I use it in manual mode, it is a tool, and when I see it working well,
and at a given point I choose a (replace all) instead of y (yes,
replace actual occurance), it suddenly becomes a bot?

I think these tool-assisted edits like AWB are essentially bot edits with
human contribution: high speed, huge amount of edits in a short time that
may be misused before anybody notices. Either they flood recent changes or
if they are hidden, they are very hard to notice in case of a mistake and
even harder to undo. Therefore the right of using AWB is equal to the right
of using PWB and should require a highly trusted user in my opinion.

That does not mean I am against a new group (which still means that every
community may use or not use it); that means I don't see any important
difference between bot and tool account.
___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] New feature: tool edit

2015-02-23 Thread Petr Bena
I don't believe that users would actually use it if this permission
was so hard to obtain as bot flag is (on english wiki). If there was
such a huge complex bureaucratic process for this, most of users would
just keep doing semi-automated edits as regular edits. The summary of
differences between flags:

* Bot edits are usually more trusted and are evaluated and reviewed by
different people. That means bot edits can be safely ignored by most
of users. This wouldn't really apply for semi-automated edits made by
users. They are still humans and they make mistakes, they should be
reviewed by admins, other users etc at some point. But most of regular
users can safely ignore them.
* Bot flags is hard to obtain, usually a matter of weeks. Tool flag
shouldn't be any harder than getting rollback permissions.
* Bot flag applies for all edits, tool flag should be used only for some edits.
* Bots are robots (non-thinking processes) that work fully
automatically. Tool edits would be made by people. There is a
difference between these 2 BUT should this difference be visible?
(this is actually a question)

In case that there is no need to differentiate between bot edits and
automated edits made by users, let's rename bot to automated edit
in bot flag (and rename whole bot flag) using different letter (b -
a) and let's make it possible for trusted users to flag their edits as
automated edit even without requirement to be in bot group. Eg.
bots would still have higher API limits, regular users not, but both
trusted and bots could mark their edits as automated.

IMHO I don't think we need to be make a distinction between bots and
people. Bots should have bot in their username which makes it simple
to see if edit was made by robot or human and in both cases the edits
are automated.

On Mon, Feb 23, 2015 at 11:23 AM, Bináris wikipo...@gmail.com wrote:
 2015-02-11 14:02 GMT+01:00 Ricordisamoa ricordisa...@openmailbox.org:

 Keep in mind that it isn't always easy to tell 'tool' and 'bot' edits
 apart. Several scripts can perform actions whose degree of automation
 varies widely.
 For my part, I make most of my semi-automated edits using my bot's
 account, but many users also have separate 'flood' accounts for use with
 Wikidata Game https://tools.wmflabs.org/wikidata-game/ and similar
 tools.


 Definitely this is the point. In enwiki's environment the word bot is
 usually meant as a fully automated tool, while other communities treat it
 differently. Let's see a major utility of Pywikibot, replace.py. This is
 equally prepared for automatic and semiautomatic mode, and some tasks may
 be solved automatically, while others -- above all spelling corrections --
 manually. This still means a very high speed rate of editing but it is
 human-controlled.
 If I use it in manual mode, it is a tool, and when I see it working well,
 and at a given point I choose a (replace all) instead of y (yes,
 replace actual occurance), it suddenly becomes a bot?

 I think these tool-assisted edits like AWB are essentially bot edits with
 human contribution: high speed, huge amount of edits in a short time that
 may be misused before anybody notices. Either they flood recent changes or
 if they are hidden, they are very hard to notice in case of a mistake and
 even harder to undo. Therefore the right of using AWB is equal to the right
 of using PWB and should require a highly trusted user in my opinion.

 That does not mean I am against a new group (which still means that every
 community may use or not use it); that means I don't see any important
 difference between bot and tool account.
 ___
 Wikitech-l mailing list
 Wikitech-l@lists.wikimedia.org
 https://lists.wikimedia.org/mailman/listinfo/wikitech-l

___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] Global user pages deployed to all wikis

2015-02-23 Thread Helder .
On Fri, Feb 20, 2015 at 6:19 PM, phoebe ayers phoebe.w...@gmail.com wrote:
 p.s. re: babel, my absolute favorite use is on Wikidata, where you can
 add babel templates to your userpage and then get the appropriate
 fields to add stuff in that language :) I love showing that to people
 who are getting started on Wikidata.

Will that still work if a user moves his babel config to his global
user page on Meta-wiki, considering that the categorization will not
work on Wikidata?
https://phabricator.wikimedia.org/T90398

___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] Global user pages deployed to all wikis

2015-02-23 Thread Helder .
On Fri, Feb 20, 2015 at 10:07 AM, Gerard Meijssen
gerard.meijs...@gmail.com wrote:
 Hoi,
 Babel templates are replaced by the #Babel functionality... The only
 problem I have with the Babel functionality on Meta is that they decided to
 have everything in Green..

 However check it out on my profile.
 Thanks,
  GerardM

There is one problem, though: the babel categorization does not work.
https://phabricator.wikimedia.org/T90398

Best regards,
Helder

___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

[Wikitech-l] ignoring composer.lock

2015-02-23 Thread Amir E. Aharoni
Hi,

Should composer.lock be added to .gitignore?

It may be different for different extensions. In ContentTranslation we
currently only have:
require: {
php: =5.3.0,
composer/installers: =1.0.1
},

I don't know much about Composer, but it looks like nothing more than the
bare minimum. Should composer.lock be version-controlled in such a
situation?

--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
‪“We're living in pieces,
I want to live in peace.” – T. Moore‬
___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] E-mail login to wiki - needs feedback

2015-02-23 Thread Jeremy Baron
On Feb 23, 2015 12:06 PM, Lars Aronsson l...@aronsson.se wrote:
 It would be possible to just say sorry, login by e-mail
 is not possible for you; please login by username.

No, that isn't possible. We can't reveal existence or non-existence of an
account with an address. If there's more than one with a given address and
we throw that error message then we've revealed something we can't.

Multiple accounts match response should be identical to wrong password
response and identical to no such email/username response.

-Jeremy
___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] E-mail login to wiki - needs feedback

2015-02-23 Thread Lars Aronsson

On 02/20/2015 05:52 PM, devunt wrote:

We should consider some edge cases like:

* More than two accounts with exactly same email and password.
- In this case, which account should be chosen for logged-in? Maybe
account selector could be one of the answers.


Is there any indication how common that situation is?
It would be possible to just say sorry, login by e-mail
is not possible for you; please login by username.
But you also run into the question if f...@gmail.com
and f...@gmail.com are the same e-mail address or
two different ones. If there is a verified account with
f...@gmail.com, can that user login as
f...@gmail.com?


--
  Lars Aronsson (l...@aronsson.se)
  Aronsson Datateknik - http://aronsson.se



___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] ignoring composer.lock

2015-02-23 Thread Legoktm

On 02/23/2015 08:34 AM, Amir E. Aharoni wrote:

Hi,

Should composer.lock be added to .gitignore?


IMO yes. I had removed it from some extensions in the past IIRC.


It may be different for different extensions. In ContentTranslation we
currently only have:
require: {
 php: =5.3.0,
 composer/installers: =1.0.1
},

I don't know much about Composer, but it looks like nothing more than the
bare minimum. Should composer.lock be version-controlled in such a
situation?


If you are using composer to install extensions, I would say yes, it 
should be ignored. In that case the extension is basically a library, 
and composer will just ignore the lock file when installing, so it ends 
up being misleading[1].


If you are using composer to manage development dependencies like phpcs 
or phpunit, then maybe, though I personally would prefer to ignore it.


[1] https://getcomposer.org/doc/02-libraries.md#lock-file

-- Legoktm

___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] Announcing service-runner, a startup module / supervisor for node services

2015-02-23 Thread Dan Garry
On 23 February 2015 at 16:46, S Page sp...@wikimedia.org wrote:

 I'm so excitoid


S, you just made my day!

Dan

-- 
Dan Garry
Associate Product Manager, Mobile Apps
Wikimedia Foundation
___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] Announcing service-runner, a startup module / supervisor for node services

2015-02-23 Thread S Page
On Mon, Feb 23, 2015 at 1:48 PM, Gabriel Wicke gwi...@wikimedia.org wrote:

 Service-runner [1] is a small module that we moved out of restbase. It
 generalizes some simple start-up, monitoring and supervision facilities
 ...


I'm surprised there isn't something like this already in nodejs that you
get for free when you use forever[6] to run a node command. Did you
consider forever-service [7] ?  It sounds similar:


1. Make an universal service installer across various Linux distros
and other OS.
2. Automatically configure other useful things such as Logrotation
scripts, port monitoring scripts etc.
3. Graceful shutdown of services as default behaviour.

 It's _great_ that https://github.com/wikimedia/service-runner#see-also
mentions similar packages, I made a pull request to add forever-service to
the README though I didn't compare its features.

For small (third party) installs with limited memory, we also added the
 capability to cleanly run multiple services in a single process.


Yay. Should MediaWiki-Vagrant use this, or does it only benefit when you're
running more -oids than just Parsoid?

I'm so excitoid

[6] https://github.com/foreverjs/forever
[7] https://github.com/zapty/forever-service
--
=S Page  WMF Tech writer
___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l

Re: [Wikitech-l] If you hear about 'hackathon buddies'...

2015-02-23 Thread Quim Gil
If you are looking for a Wikimania Hackathon buddy, check
https://wikimania2015.wikimedia.org/wiki/Hackathon/Buddies

On Thu, Feb 19, 2015 at 3:57 PM, Quim Gil q...@wikimedia.org wrote:

 Yesterday we opened the travel request process for Wikimedia Foundation
 employees in Engineering and Product willing to participate at the
 Wikimedia Hackathon or Wikimania. There is no public link, but you can
 follow this task at https://phabricator.wikimedia.org/T89355

 In this process, we are asking WMF employees to find a hackathon buddy
 with the sole requirement of not being another WMF employee. In fact, in
 the registration for the hackathons we will request the same to all
 participants.

 https://www.mediawiki.org/wiki/Hackathons#Pairing_buddies

 This means that non-WMF contributors might receive a request from a WMF
 employee to be hackathon buddies. This also means that if you are planning
 to participate in any of these events (and especially if you plan to
 request travel sponsorship to Lyon) you will be encouraged to find a buddy
 as well.

 It's going to be fun.  :) And no worries, we will help making connections
 to whoever needs that help.


-- 
Quim Gil
Engineering Community Manager @ Wikimedia Foundation
http://www.mediawiki.org/wiki/User:Qgil
___
Wikitech-l mailing list
Wikitech-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikitech-l