Someone in the thread on friendliness mentioned that categories are
always in one language (usually english). Well still a long way from
fixing the issue, perhaps if we allowed unrestricted
{{DISPLAYTITLE:...}}, combined with the {{int: hack, that'd allow
better translatable categories. (of course you'd only be able to use
the actual category name in [[category:Foo]] links. I suppose one
could use a bot to automatically change links to redirect categories
to their canonical name, but then we're getting really really hacky).
Anyways, just a thought.
cheers,
bawolff
Food for thought.
---------- Forwarded message ----------
From: Michel Vuijlsteke <wikipedia(a)zog.org>
Date: 22 February 2011 16:29
Subject: Re: [Foundation-l] Friendliness (was: Missing Wikipedians: An Essay)
To: Wikimedia Foundation Mailing List <foundation-l(a)lists.wikimedia.org>
On 22 February 2011 14:14, Yaroslav M. Blanter <putevod(a)mccme.ru> wrote:
>
> > We have to make a profound choice in the culture here:
> > 1) we continue with the whacking and scaring the newbies away (content
> > priority #1, people #2), or
> > 2) we embrace the newbies and we let some spam through (people priority
> #1,
> > content #2).
> >
> > So far we are steadily moving along the first route. I believe, it is
> time
> > we switch the priorities. People are important. It's the people who will
> be
> > creating content in the future, and not the other way around. Wikipedia
> > will
> > inevitably fail without participation. And content... we are already the
> > largest and the best...
> >
> > Renata
>
> To me it sounds too much black and white. Indeed, there are points you
> better not stumble across as an editor: engaging into battles over disputed
> content (like Middle East conflict), writing articles on smth with disputed
> notability, pushing POV or not getting immediately the image upload rules.
> But I assume this is a relatively minor fraction of editors (though of
> course it still represents a problem). I can not recall that I ever got any
> templates in my articles (I have written over 500 of them since 2007),
> except for a couple of times from a bot that there are no links to the
> article, and that I ever got any angry comments from admins/other editors
> concerning the articles I have written.
>
I don't think it has to be as obviously annoying as slathering templates all
over pages or wikilawyering the newbies away -- it's often much more subtle
how content/data seems to be considered more important than people.
One interaction I encountered recently is typical. Michiel Hendryckx, one of
Belgium's best-known photographers, started uploading fairly
high-resolution, good quality images to Wikipedia (well, Commons) on 3 July
2010. Stuff like this 1983 Chet Baker portrait:
http://commons.wikimedia.org/wiki/File:Chet675.jpg
The first message on his talk page was a request to confirm his identity
(which he did).
The second message was a complaint by Nikbot (no valid license for one
particular image). A couple of hours later, at 10:51 on 4 July, the next
message is from CategorizationBot, asking Hendryckx to add categories to his
images.
The third message, not six hours later, was this:
*Please categorize our images !!!*
You already have been asked by a bot to categorize your images. Therefore I
don't understand why you keep on uploading images without categories.
Uploading images without categorizing them doesn't make sense. Only
categorized images can be found!
I'm pretty sure the user in question meant really well, but *this* is what
that focusing on content over people means to me. It's in the small things,
the interactions that experienced Wikipedians take in their stride, but that
can end up scaring people away.
It's like the last message on Hendryckx' talk page, dated 1 February 2011: a
notification that one if this images is listed at commons:deletion requests,
and to "please do not take the deletion request personally... thank you!".
Follow the link to the discussion (
http://commons.wikimedia.org/wiki/Commons:Deletion_requests/File:Van_istend…):
turns out the requester couldn't see the image. His/her first action was to
nominate the image for deletion. Took about three hours for someone to
confirm that no, the image works perfectly fine for them, and about five
hours for the original person to close the deletion request ("thanks").
Again: content over people. No personal interaction with the photographer,
no message on the photographer's talk page after the deletion request was
closed, nothing. The last interaction Hendryckx had on Commons -- on 19
February, almost three weeks after the deletion request was closed -- was a
baffled question (
http://commons.wikimedia.org/wiki/Commons_talk:Deletion_requests/File:Van_i…),
asking what on Earth is wrong with the image, and that he'd like to at least
know why it needed to be deleted.
Again, I'm sure the user in question meant really well again, but here too:
content over people. Drive-by templating, shoot first, don't ask questions,
don't even provide feedback, trust people will read every last word in the
templates, etc.
Michel Vuijlsteke
_______________________________________________
foundation-l mailing list
foundation-l(a)lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
"users from Germany are the root of the problem" (Lars Aronsson, Commons-l,
27 Feb 2011)
Wow. I never expected to read such a phrase on Commons-l. Though I hate to
possibly fulfil Godwin's law, when reading Aronsson's above cited claim, the
infamous Nazi slogan "The Jews are our misfortune" (sv: "Judarna är vår
olycka") came to my mind.
Mr. Aronsson obviously hasn't understood the Wiki system: if you think
something can (or needs to) be improved, just DO IT or make a substantial
proposal at the right place, but without spitting your contempt in the face of
those who have done the pioneer work (even, if somewhat lousy).
Though Mr. Aronsson's second focus seems to be on German admins, it may
have slipped his attention that default message templates used for notifying
uploaders that one of their uploads is considered copyvio, derivative-work,
missing permission, missing source or missing a license, etc. were not
necessarily written by the admins who use them on a daily basis. But if Mr.
Aronsson prefers, I can ask all 48 de-native-tagged admins on Commons to stop all
their admin-work until we get Mr. Aronsson's personal approval.
The problem Commons is facing since a while, is massively missing
admin-power/resources.
On February 23rd, Commons had 9 mio files. (read on [[:de:WP:Kurier]])
Today (Febr 28th), Commons has 9.35 mio files. (as of
[[Special:Statistics]])
That means, in 5 days 350.000 uploads needed to be checked for detecting
blatant and not so overt copyvios, for attack images/pages, for personality
rights violations, for useless bullshit, for clearly promotional material,
for missing source entry, for missing license, for missing permission (when
uploader not identical to author), etc. etc.
Yes, we have 270 admins on our list. Surely all are doing valuable work.
But rather few are active in the dirty work of upload patrol (To get a surely
incomplete impression:
http://toolserver.org/~vvv/adminstats.php?wiki=commonswiki_p&tlimit=15768000). Why? Though upload patrol may keep Commons
existing (instead of being closed down by WMF after getting the 10.000th DMCA
takedown-request), it doesn't make you any friends. You get angry reactions from
clueless, careless or reckless uploaders, you are personally insulted, you
even get death threaths, your userpage is vandalized, etc. but you NEVER get
a thank-you from WMF or any institutional body of WMF or Commons (except
from a few single users or fellow sysops). I'm not really asking for the
latter, but mentioning it may help to understand the "constructive" impact of your
rant.
Túrelio
(COI disclosure: admin on Commons, who is de-native, but has never written
a default message template; Disclaimer: speaking strictly for myself, not
for a non-existing de-admin cabal)
In einer eMail vom 28.02.2011 02:52:19 Westeuropäische Normalzeit schreibt
commons-l-request(a)lists.wikimedia.org:
> Message: 4
> Date: Sun, 27 Feb 2011 20:30:00 +0100
> From: Lars Aronsson <lars(a)aronsson.se>
> Subject: Re: [Commons-l] Fwd: [Foundation-l] "
> To: Wikimedia Commons Discussion List <commons-l(a)lists.wikimedia.org>
> Message-ID: <4D6AA638.7090607(a)aronsson.se>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>
> Magnus Manske wrote MediaWiki, Commons was suggested
> by Erik M?ller, and the Toolserver is German. The Germans
> have contributed more than most to the Wikimedia projects,
> especially in software and technology. Not to mention
> the huge archive photo donations and Wikipedia Academy,
> pioneered by the German chapter. However, the German
> Wikipedia has a different set of standards, more strict rules
> for inclusion and notability, and more speedy deletions. This
> adds to the "focus on content, rather than people", and when
> this is described as a problem on Commons, what I can see
> is that users from Germany are the root of the problem.
>
> Spanish or Norwegian admins on Commons are not the
> problem, as far as I can see, despite using the same
> ..
>
>
> --
> Lars Aronsson (lars(a)aronsson.se)
> Aronsson Datateknik - http://aronsson.s
e
http://www.ajmj.fr/php/annuaire/pageFicheAnnuaire.php?id=354
if my french still serves me
- Stephane Gorrias
- Mandataire Judiciaire
- Près la Cour D'appel De Versailles
- Tribunal principal :
- Année d'inscription : 2003
- SCP : SCP BECHERET-THIERRY-SENECHAL-GORRIAS
- Courriel : stephane.gorrias(a)ajmj.fr
-
- Domicile Professionnel
- 1 Place Boieldieu
- 75002 PARIS
- Télécopie : 01 40 28 06 70
these are the contact details
On 28 February 2011 06:44, Jimmy Wales <jwales(a)wikia-inc.com> wrote:
> Happy to help if I can, but it does seem that the photos are physical
> objects and therefore possibly expensive to store/digitize. If someone can
> help me find the email address of the decision maker there, I'm happy to
> open or lend support to a conversation with them.
>
> It isn't clear from the blog whether they own only the physical objects or
> also the copyrights.
>
>
>
> On 2/26/11 12:28 PM, Gnangarra wrote:
>
>> http://blog.melchersystem.com/2011/02/25/the-fire-this-time/
>>
>> 12 Million Photographs are to be destroyed because liquidators cant find
>> a buyer?
>>
>> these arent just random landscape photos, these are photos are from the
>> fench news agency Sygma
>>
>> Wikimedia-france, Jimbo, any chance they could find their way to Commons
>> rather than just being destroyed and lost for ever.......
>>
>> --
>> GN.
>> Photo Gallery: http://gnangarra.redbubble.com
>> Gn. Blogg: http://gnangarra.wordpress.com
>>
>
>
--
GN.
Photo Gallery: http://gnangarra.redbubble.com
Gn. Blogg: http://gnangarra.wordpress.com
Related to the discussion of size and bitrate of uploaded video
discussions on wikivideo-l, I thought it would be nice to share the
latest efforts on the TimedMediaHandler extension.
This extension will help make the size and bitrate choices less of an
issue, and will just be a matter of uploading the highest quality
version you can. The extension auto transcodes to a few different
derivative formats.
Its best illustrated by features overview page: ( best viewed with
Firefox 4 )
http://prototype.wikimedia.org/timedmedia/Main_Page
This extension standardises a lot of features of the mwEmbed player
gadget, like pop-up gallery videos, temporal media fragments, timed
text, iframe embed video sharing, html5 standard "video", "source" and
"track" page output, etc.
The bottom clip illustrates a key feature of this extension with a real
commons "media of the day" clip from a few days back. JJ Harrison
uploaded a very nice, very encyclopaedic HD nature clip, but at
1920x1088 and 13mbs it fails to playback almost any time some tires to
play it :( With this extension we get the 'right' resolution given the
embed size, and have easy access to switch streams as you go into
fullscreen if your computer / connection supports it.
There are still a few resource loader integration issues to work out
before more widely sharing these efforts, but I thought I would share it
on these list to get some early feedback.
I have filed a tracking bug
https://bugzilla.wikimedia.org/show_bug.cgi?id=27699 to keep track of
its progress toward deployment. If you file related related bugs or
feature requests, you can tag them with that tracking bug.
Also got started on a mediawiki.org extension page:
http://www.mediawiki.org/wiki/Extension:TimedMediaHandler
peace,
michael
2011/2/22 Michel Vuijlsteke <wikipedia(a)zog.org>:
> One interaction I encountered recently is typical. Michiel Hendryckx, one of
> Belgium's best-known photographers, started uploading fairly
> high-resolution, good quality images to Wikipedia (well, Commons) on 3 July
> 2010. Stuff like this 1983 Chet Baker portrait:
> http://commons.wikimedia.org/wiki/File:Chet675.jpg
>
> The first message on his talk page was a request to confirm his identity
> (which he did).
>
> The second message was a complaint by Nikbot (no valid license for one
> particular image). A couple of hours later, at 10:51 on 4 July, the next
> message is from CategorizationBot, asking Hendryckx to add categories to his
> images.
This is where it starts. Thousands of our users have their first
interactions with a bot or with a user leaving a template. We're
unlikely to alter our practice to completely abandon bots and talk
page templates (although we can improve our software to give more
direct user feedback which makes bots and automated messages
unnecessary, e.g. for something like missing categories), but while
we're still using them, we really need to pay more attention to what
they are saying.
IMO every single Wikimedia project would benefit from dedicated
community effort to 1) catalog the most widely used templates on talk
pages, 2) systematically improve them with an eye on the impact they
can have on whether people feel their work is valued and the
environment in which they're contributing is a positive and welcoming
one. This is something that anyone can help with, right now.
The messages left by CategorizationBot are an example of the issues
with our current and approach. There's only very limited
acknowledgment of the user's valuable contribution, there's no
explanation what the message represents (is it a warning, a reminder,
what?), there's no invitation to turn off the message if it's not
wanted, there's immediate and unexplained use of jargon like "image
description page", and the overall message sounds like "You've done
something wrong, please fix it and ask for help if you need it". All
of this drives towards rules-compliance and against shared ownership
of community norms and practices.
Specifically with regard to this message, I've left some suggestions
for improvement here:
http://commons.wikimedia.org/wiki/Template_talk:Please_link_images/en
I'm not saying that the suggested changes are a vast improvement, but
I think that's the kind of conversation we need to be having.
Obviously we don't want fake friendliness and personality in our bots
and templates, but at the same time, I think we should strike a tone
and use language that's consistent with the culture we want to create.
--
Erik Möller
Deputy Director, Wikimedia Foundation
Support Free Knowledge: http://wikimediafoundation.org/wiki/Donate
Hello,
There are some amazing and educational videos in MIT's OpenCourseWare
collection, and it's been on my mind for a while that none of them are
on Wikipedia, even though the copyright holders (the profs) often
would love for them to be.
Over the past weeks, I've been working with Peter Kaufman (Intelligent
Television), Ben Moskowitz (OVA), and some of MIT's OpenCourseWare
team to identify a few videos that could be split up into useful
sections to illustrate math and science articles on Wikipedia.
You can see a few examples from Prof. Walter Lewin's physics courses here:
http://en.wikipedia.org/wiki/Newton%27s_laws_of_motion
OCW is interested in running a small project, with clips from ~100
course videos, to figure out how we can make this work on a larger
scale, and what the interest and response will be. If this is
successful, we could add thousands of great clips to commons.
(Professors own the rights to their videos, and the default (c) on OCW
is CC-NC, so each prof must explicitly release their videos under
CC-SA before they can be used in articles.)
Working on this project - my first work with video for awhile - raised
a few questions, below.
OCW wants a failproof way to instruct people to set up their browsers
so that our media player works. Of the MIT staff who tried it, 3 of 10
had problems until they installed another browser or fiddled
around.(!)
Q1: Is there a page that says "choose your OS below, follow the link
to download the lates browser version, and the player will work" ?
Q2: Do we have data on the % of our visitors for whom the video-player
doesn't work properly? (to answer the question I got twice today:
"will all readers actually be able to use these videos?")
Q3: Do we have historical stats on the # of media files in Commons by
filetype or mediatype?
I am looking for a Boston-local ambassador who can work with Peter
(whose staff offered to do the clip-selection and transcoding for this
pilot) and the university (which will reach out to a few more
professors to find interest) to step through the process a few times,
from choosing suitable clips and important science articles needing
illustration, through to sending a permissions email to OTRS.
Q4: can we start offering transcoding automatically, for people who
upload non-ogg formats? Dailymotion seems to do this flawlessly,
perhaps we can learn from their toolchain.
Q5: why is the link to the permissions email still so hard to find?
Is there a new snazzy upload form that people can be pointed to that
lets uploaders say:
- "this file is by FOO who releases it under license L" ...
- "send an email to FOO through this form, reminding them to confirm
the license release"
Q6: do we still have that 100MB file size limit? can we change this to 500MB?
Q7: people often need access to raw high-res media: for restoration,
manipulating full-size animation frames, or editing HD video. these
can be a few GB in size. Is there any plan to set up a
quarantine/scratch space where these files can be uploaded and shared?
Thanks for any pointers and answers, including to relevant threads
that I may have missed,
Sam.
--
Samuel Klein identi.ca:sj w:user:sj +1 617 529 4266
On Sun, Feb 13, 2011 at 10:21 PM, <pbk(a)intelligenttv.com> wrote:
> Thanks, Sam. Another opportunity here may be for the community to help
> define new citation standards for time-based media. There are no commonly
> accepted practices yet - still - for how to write and punctuate references
> to film and sound in captions, footnotes, and bibliographies.
>
> -- Peter.
Very true. I look forward to having a page at
http://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style_(multimedia)
... That could start with a list of nonstandard and unaccepted
practices in print, and the current templates in use on Wikipedia[1].
SJ
[1] For instance, on en:wp -
http://en.wikipedia.org/wiki/Template:Cite_videohttp://en.wikipedia.org/wiki/Template:Cite_episode
Greetings,
As you may know, the Wikimedia teach team has started to upgrade
MediaWiki on some wikis. MediaWiki is the software that runs all
Wikimedia wikis.
The most visible change for Wikimedia users will be the deployment of
ResourceLoader [1].
ResourceLoader optimizes the use of JavaScript in MediaWiki, speeding up
its delivery by compressing it sometimes, and cutting down on the amount
of unused JavaScript that gets delivered to the browser in the first
place.
The installation of ResourceLoader may cause compatibility issues with
existing JavaScript code.
Trevor Parscal and Roan Kattouw, the main developers of ResourceLoader,
will be available on IRC [2] on Monday, February 14th, at 18:00 (UTC)
[3], to answer questions and help fix issues related to ResourceLoader.
*If you maintain JavaScript code on your home wiki, please attend.*
Don't wait until your wiki's JavaScript is all broken.
Please spread this information as widely as possible; it's critical to
reach as many local JavaScript maintainers as possible.
Logs of the session will be published publicly.
[1] http://www.mediawiki.org/wiki/ResourceLoader
[2] http://meta.wikimedia.org/wiki/IRC_office_hours
[3] All timezones: http://ur1.ca/3819u
--
Guillaume Paumier
Product manager - Wikimedia Foundation
Support free knowledge: http://donate.wikimedia.org