Well, lately my posts here are falling on deaf ears. But since this
one is more technical maybe it will get some attention. :P
Bug 1710: Ability to watch all articles in a category
http://bugzilla.wikimedia.org/show_bug.cgi?id=1710
(By 'watch', I assume this means 'receive notification when members
are added to/removed from the category', rather than 'put all members
on my watchlist automatically')
Anyone who works with the category system would be well aware how much
this is needed, so that the categories can be properly maintained. It
doesn't seem like an official MediaWiki solution is going to show up
any time soon. I had an idea for a solution:
Make an article page, maybe for category X call it 'X watchlist' or 'X
changes log', that you can put on your watchlist. Have a Bot somehow
monitor the category (like, look at the category and compare the names
of members to previous check) and report any differences on the
article. Then by watching the article you will effectively get to find
out any changes within the category.
The bot would only need to check the category like once a day, I would
say. And possibly you would want to not allow this for very very large
categories like [[:Category:Public domain]] or [[:Category:Linguistic
maps]] or whatever. But for country-level cats and below it should be
OK.
So... thoughts? Please fwd to wikitech-l if appropriate & if you're on
that list.
I look forward to you techy-types nutting this one out for me. :)
thanks,
Brianna [[user:pfctdayelise]]
Please help us stalk newbies!
http://commons.wikimedia.org/wiki/User:Orgullobot/Welcome_log
When they are new is our best chance to catch them and correct their
mistakes, before they upload 200 images under a wrong license and no
categories... these are the two most important mistakes. Uploading
stuff under unacceptable/wrong/no license or no source, and not
categorising good images.
Orgullobot automatically welcomes around 50-70 new users per day, who
have made at least one edit or upload. EugeneZelenko and I cannot
check this many people. As we know it takes 2 seconds to upload a
copyvio and 2 months to get rid of it... but the earlier we can catch
them the quicker they can be deleted. Be ruthless!
So all you need to do is just check their contribs and if they have
made any mistakes, correct them (if you want) and tell them how to
correct them and what they did wrong. e.g. post {{please link images}}
when they should categorise. Use NSD and {{copyvio}}, tell them
there's no originality in screenshots, there's no fair use, use
CommonsHelper to transfer images...etc. Once you've done this remove
their name from the log.
If, by some miracle, they don't make any mistakes, you may want to
congratulate them, or else just move onto the next person. :)
Note YOU DON'T HAVE TO BE AN ADMIN TO DO THIS! You just have to have a
decent understanding of what's OK and what's not, and what the correct
procedure is.
Hope to see some new eyes on this log soon!
Brianna
Recently I was asked [1] to delete the obsolete WeatherBot images.
(Background: Over most of 2005, User:WeatherBot uploaded about 25000
images depicting current temperature worldwide. WeatherBot is now
defunct, and the old temperature maps are no longer used. Deletion
was discussed in a bunch of places on the Commons over many months.)
Today, I started deleting the old maps, with a bot. My bot archives
each image before deleting, so if it makes a mistake, any image can
easily be restored. I deleted 2284 images (in 88 minutes, or 2.3
seconds per). Then Duesentrieb pointed out [2] that by flooding the
deletion log, I was breaking his CommonsTicker tool.
Duesentrieb suggested I should post to the mailing list before
continuing. (My bot stopped running automatically as soon as he
posted his comment to my talk page.)
At full speed, the remaining 23000 images will take about 15 hours to delete.
Does anyone else have a tool that watches the delete log? If you'll
be affected by me deleting 23000 images, please let me know. I'm open
to suggestions. Two possible ideas: I could run the bot in 3-hour
chunks at a particular time each day. I could also throttle the bot
slightly. As slow as 1 image every 10 seconds would still get the job
done in 3 days.
If no one says otherwise, I'll plan to restart the bot next Saturday,
June 7, and allow it to run at full speed until it's done.
-- Commons:User:Dbenbenn
Links:
[1]: http://commons.wikimedia.org/wiki/User_talk:Dbenbenn#Deleting_WeatherBot_im…
[2]: http://commons.wikimedia.org/wiki/User_talk:Dbenbenn#Weather_Images
[*]: http://commons.wikimedia.org/wiki/User:WeatherBot
Hello all
I have activated CommonsTicker [1] for a first live test at the German
Wikipedia [2] (have a look at the wiki text of the actual entries in
[3], you don't need to speak German for that). An announcement about it
is on the Village Pump [4]. After a short test run, I will set up ticker
pages on other projects on request (in cooperation with a local admin).
CommonsTicker monitors Common's RC feed and records deletion and
replacement of images, as well as deletion tags being placed on
description pages. For each project with a ticker page, CommonsTicker
then determines which images are used on that project, and posts a
notice for the local community.
I hope this will help to integrate Commons more closely with the
Wikipedia communities; I hope this will get people more involved with
commons, and gets us some help unlinking images that have to be deleted,
etc.
So, please have a look...
-- Daniel
PS: please forward this message to the mailing lists of wikimedia
projects that may be interested in a CommonsTicker page.
[1] http://meta.wikimedia.org/wiki/User:Duesentrieb/CommonsTicker
[2] http://de.wikipedia.org/wiki/Wikipedia:CommonsTicker
[3] http://de.wikipedia.org/wiki/Wikipedia:CommonsTicker/Liste
[4]
http://commons.wikimedia.org/wiki/Commons:Village_pump#CommonsTicker:_the_G…
--
Homepage: http://brightbyte.de
I've started adding captions to the Gray's anatomy images, but there are
just too many.
I have had a bot prepare links and tags with the appropriate captions at:
http://tools.wikimedia.de/~magnus/gray.html
Ayn help would be appreciated. Best you start somewhere in the list at
random, to avoid double effort.
I'll run the bot again in a few days to see what's left. Then I also can
list the missing Gray images.
Magnus
Hi All,
Apologies in advance if this has been brought up already, but is there any
specific reason wikipedia doesn't use book references as a source of income?
Personally, and for many others I'm sure, this would make wikipedia a lot
more convenient, as a lot of times I'm looking up the same books I buy on
Amazon as topics I'm looking up on wikipedia. For example, I recently bought
Proofs From The Book, referenced here:
http://en.wikipedia.org/wiki/Schroeder-Bernstein_theorem
Had there been a direct link, I would have saved myself about 10 seconds of
opening an Amazon tab and searching for the book, and made Wikipedia a buck
or two.
It would also make good sense for there to be a strong correlation between
repeat customers on Amazon and frequent visitors on wikipedia (I know I am
both...).
Cheers,
Andrey
Hello,
The ability to block a user from uploading only would be unbelievably
useful on the Commons. Blocking users alltogether is a bit silly for
(suspected) copyright violators/repeat offenders, since we want them
to still be able to add copyright info to image description pages.
EugeneZelenko just raised it again on the VP but I thought I would
raise it here since maybe a few more 'dev-types' read the mailing
list.
>From my extremely limited knowledge of MediaWiki, it seems like the
solution would be to create a usergroup 'bannedupload', and the only
different permission they have is ['upload'] = false.
But I don't know who has permission to change user groups (ie
add/remove people). Probably Bureaucrats by default since they make
sysops and bots. We would actually want this permission for
administrators, since they can block users, blocking uploads is a
lesser action. Seems like that might be trickier...
Any solution or comments are welcome here and/or at
http://bugzilla.wikimedia.org/show_bug.cgi?id=4995 .
Brianna
Here's a somewhat longer answer to Brianna's message :
http://mail.wikipedia.org/pipermail/foundation-l/2006-May/006995.html
I have recorded in a table, the activity of the village pump on Commons,
the number of administrators, and the date of last edit of the main page
in their language for the various language communities. Here's the link :
http://commons.wikimedia.org/wiki/User:Teofilo/communities
My conclusion is that although some communities seem to have been
completely inactive in 2006, most of them show at least some small
activity in 2006.
>Help us harass the developers for automatic translation of templates,
>automatic language selection based on browser settings and/or
>drop-down menu for language choice on the main page, for adoption of
>Duesentrieb's proposed category translation scheme via interwikis.
>These are just a few of the proposals that we dream of being
>implemented but have no idea if it will ever happen.
Do you have links ? On bugzilla, maybe ? I have never used the voting
system on bugzilla. Maybe I could, but please tell me which are the
relevant bugs.
Teofilo
http://commons.wikimedia.org/wiki/User_talk:Teofilo
Brianna
http://mail.wikipedia.org/pipermail/foundation-l/2006-May/006995.html
« (fr) Commons:Bistro Maintained, but not very active (6 topics).
»
Actually the French Bistro has around 5 new topics per month. The
preceding months have been recently archived, and this is why it
looks so short.
I like your idea of classifying which language community is active
or not. I'll try to write a longer answer to your interesting point of
view.
Teofilo