Wikipedia:Village pump (technical)/Archive 73

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Village pumps: PolicyTechnicalProposals (persistent)Miscellaneous

Contents

Yet another Java applet request[edit]

Recently a page I've been involved with received a request, from WikiProject Physics, for some additional content. As a result I spent a couple of weeks developing an interactive java applet. This was based on some old published work and was designed to illustrate some of the physics involved.

However when I tried to upload the applet I was very surprised and a little disappointed to learn that Wikipedia does not support applets. I made some enquiries and was directed to these pages. On searching for 'applet' I learnt that there have been earlier requests, i.e. Archive 50/Question 16, but very little progress.

As far as I can tell resistance comes from:

1. Worry about security. But Java applets are run in a sandbox for just this reason.
2. Worry about loading times and cpu use. But this is the same problem as with sound and other media and could be solved in the same way via a javascript box.
3. The people involved in the discussion know little about java applets and that "people would have to know about java". I do not see that this is any different from knowing how to create mp3 files, pdf files or whatever. The wikipedia web pages on Java applet and Java (programming language) are a start.

There was also a suggestion of using Wikiversity which I followed up - but this turned out to have the same applet support as Wikipedia. I then found a link to a wiki/applet project (which I thought was overkill for the changes needed) but even this seems to have run out of steam.

So I am wondering if the question should be put in another form i.e.:

1. Has the applet option been discussed at a formal meeting of wikipedia in recent years?
2. If so, is the written report of the meeting available? Were experts on java applets and representatives of the physics community present at the meeting?
3. How can the question be raised officially at a future meeting?

Regards, David Webb (talk) 09:52, 24 March 2010 (UTC)

Answering your questions in order:
  1. I doubt it, I don't think there's ever been a huge demand for it.
  2. See #1.
  3. If you want a feature request in MediaWiki, see https://bugzilla.wikimedia.org. Granted, I don't see this getting implemented on WMF sites.
^demon[omg plz] 10:16, 24 March 2010 (UTC)
Mediawiki should atleast support flash (does it?)... You can make stuff ass good as Java in Flash, and its much easier, too. FLash would be a good addition. It should be sandboxed with no external file access... and maybe only admins can upload to stop people from uploading ads. ManishEarthTalkStalk 12:24, 24 March 2010 (UTC)
Flash is much more resource-intensive than Java, and doesn't come with built-in Java security. Plus, anything that can be made in Flash can be made more easily (programming-wise) in Java. There is a small need for Java, but no need whatsoever for Flash. OrangeDog (τε) 13:24, 24 March 2010 (UTC)
Comment: I'm a Java programmer and I used to do flash, and I personally feel that for and interactive thingamabob which has encyclopedia value, Flash makes for easy graphics and interactivity is much easier. In Java, graphics are harder and coding gets heavy for applets. But you're right about the resource stuff. ManishEarthTalkStalk 15:03, 24 March 2010 (UTC)
And most importantly, Flash is not an open technology. JAVA wasn't either long ago, but now it is (GPL now). Though there still might be some patent issues. Most of all though, it's a matter of who will do the work I think. jmol has been discussed many times before for instance. Last time it was seriously proposed User:User:Ilmari Karon found a JavaScript injection security issue within minutes. Those are the kinds of problems. Also, we have to remember a few other things. Applets don't print for instance, so there would have to be some discussion on that. And applets don't have thumbnails, so what do we present to users before something is loaded ? There is one general applet loader extension that could be used for this. We of course already use the oggplayer extension which loads the Cortado applet. Other extensions that load applets (though not installed here) are IRC chat, GeoGebra, RDFEditor, X3d and many others. —TheDJ (talkcontribs) 14:18, 24 March 2010 (UTC)

This looks like the start of a good discussion but I would like more advice on how to continue. Given that Moore's Law is still operating it is odd that Wikipedia is not thinking more about the use of java applets to convey information. I am sure that once the facility was available this huge community would find good uses for it.

On the question of Flash or Java I would plump for Java on the basis that it is open source and not two difficult to learn. I found some good cheat sheets on the web and was able to do what I wanted, including bug hunting, in about two weeks starting from scratch. As a died in the wool fortran programmer that wasn't too bad. Re printing, the javascript box could have an associated jpeg representing the applet - the real applet only appearing once it had been clicked on. For anything else the user could copy and print the screen - but I sure that if asked by Wikipedia Oracle/Sun would suggest something better that worked.

Re bugzilla - the bugzilla site is talking about bugs and it is not obvious that anyone would take notice of my present request. Instead would I do better getting WikiProject Physics involved? What other routes are there for getting progress? David Webb (talk) 22:09, 24 March 2010 (UTC)

What is needed here is a multitude of things.
  1. Demand
  2. Code that actually makes this work
  3. A request to bugzilla to review said code and eventually deploy it.
And unfortunately not all those things come together at the same time. Remember that the software is opensource and mostly a volunteer effort. So if this is something that you want, you will need to evangelize, find a programmer and get Tim Starling to review the code and approve it for Wikipedia operations. I would always make sure by asking in bugzilla first, if there is a chance AT ALL if this will ever be approved for use on Wikipedia. That might save you some wasted effort. —TheDJ (talkcontribs) 22:42, 24 March 2010 (UTC)
Thanks. When you say 'code that actually makes this work' and 'finding a programmer' - do you mean changes to the wikipedia server code? For that I presume the best route is via Bugzilla or at a Wikipedia meeting. For everything else I would expect the contributers to do the programming. David Webb (talk) 18:14, 25 March 2010 (UTC)

Just throwing an idea out: it doesn't sound like there will ever be applets on Wikipedia, but would there be a possibility of another Foundation sister project, perhaps even a new one, for free-content audio/visual displays like presentations and applets that can be used to augment 'pedia content? This still would be limited to GPL-type content (Java for example), and likely there would need to be deeper disclaimers that the Foundation assumes no damages for malicious additions. (Of course, we could make it a requirement there that all such applications, since GPL, include the source code as back-end pages, thus allowing for review of malicious claims). Just an idea, as I see the usefulness of more interactive tools for some subjects, just not on 'pedia proper. --MASEM (t) 22:16, 24 March 2010 (UTC)

We could have a set of high-level trusted users who have disclosed their identity and are good enough at Java to review the code. That might save WMF the time of reviewing it. ManishEarthTalkStalk 03:32, 25 March 2010 (UTC)
I would hope this would only be necessary for a year or two until trust in Java Applets was established. David Webb (talk) 18:14, 25 March 2010 (UTC)

Downtime[edit]

If the admin log is unavailable to you, use Twitter.

Wikipedia is down for a lot of folks in Europe atm. A problem with the DNS entries or something. More news will follow on http://techblog.wikimedia.org and secure continues to to work for the obsessed. —TheDJ (talkcontribs) 17:45, 24 March 2010 (UTC)

Seems to be affecting me too (eh?)... Another solution is to place entries in your local hosts file...
208.80.152.2    en.wikipedia.org
208.80.152.3    upload.wikimedia.org
...yes - I'm an addict. –xenotalk 17:47, 24 March 2010 (UTC)
and the explanation: http://techblog.wikimedia.org/2010/03/global-outage-cooling-failure-and-dns/TheDJ (talkcontribs) 17:55, 24 March 2010 (UTC)
...of course, attempting to communicate this to folks affected by placing it on the affected site is probably an exercise in futility... –xenotalk 18:38, 24 March 2010 (UTC)
Geek.
But you'll also need
208.80.152.118 bits.wikimedia.org
:) Amalthea 18:48, 24 March 2010 (UTC)
Well en.wikipedia seems to be fixed, but I'm not getting any css or js back yet. Or indeed, edit conflict notices. OrangeDog (τε) 18:50, 24 March 2010 (UTC)
At least this means my computer isn't just going crazy. All I can get is the article-frame, everything else has disappeared. I've changed the header, though - I'm in the U.S. and affected.  :( *sigh* --Philosopher Let us reason together. 19:30, 24 March 2010 (UTC)
Additionally, the secure server appears to be completely down. --Philosopher Let us reason together. 19:43, 24 March 2010 (UTC)
No doubt from the additional load. –xenotalk 19:44, 24 March 2010 (UTC)
Ah, just noticed at the server admin log, it says the secure server got too much extra traffic, so they turned it off. --Philosopher Let us reason together. 19:48, 24 March 2010 (UTC)
...and en.wikipedia starts working fine as I switch computers. Hmm.... --Philosopher Let us reason together. via alternate account 20:01, 24 March 2010 (UTC)
Your Operating system and browser might be caching as well. so that might explain it a bit. —TheDJ (talkcontribs) 20:34, 24 March 2010 (UTC)
I couldn't access Wikipedia (nor some of its sister projects) at all until about 20:02 UTC. Waltham, The Duke of 20:28, 24 March 2010 (UTC)
Still can't get bits.wikimedia in the UK :( OrangeDog (τε) 21:34, 24 March 2010 (UTC)
Have you tried flushing your dns cache/editing your hosts file? Ale_Jrbtalk 21:39, 24 March 2010 (UTC)
Yes, I've manually configured the IP. Just letting you know that my regular DNS (216.146.35.35) still doesn't have bits.wikimedia.org, just in case the information is useful to anyone who can do something about it. OrangeDog (τε) 23:56, 24 March 2010 (UTC)
I've flushed my DNS, changed the DNS and cleared the cache but still not getting the "normal" Wiki. Bidgee (talk) 02:12, 25 March 2010 (UTC)

Europe? I'm not in Europe. I'm in North Carolina. I turned on my computer at home at around 19:00 and every time I tried to access Wikipedia, nothing happened. Around 22:00 I finally saw a page but was unable get results if I clicked on anything. While doing other stuff, I got one of the pages to come up after waiting and waiting and waiting, clicked again, did other stuff while waiting and waiting and waiting for another page, and only by 23:00 was Wikipedia operating anywhere close to normally. But I had other stuff to do.Vchimpanzee · talk · contributions · 13:29, 25 March 2010 (UTC)

Please read it all. The switch after the europe outage was what caused the global outage. —TheDJ (talkcontribs) 20:48, 25 March 2010 (UTC)
Oh, I see. There's a link. Well, the news made Mike Huckabee's show!Vchimpanzee · talk · contributions · 22:12, 25 March 2010 (UTC)

Subpages[edit]

Need expert opinion here of someone who understands one of the following: transcluded templates, subpages. Why: article Vitamin B Vitamin D contains 220+ references, many of which were formatted as {{cite doi}} and {{cite pmid}} templates. A crash in one of those entries effectively removed the whole reference list (which was not evident how to fix, at least to me). After that, I changed all {{cite doi}} and {{cite pmid}} templates to {{cite journal}}, believing this would stabilize the article. My argument is that a large number of {{cite doi}} templates transcludes many subpages/templates which is undesirable (?), and this is the question I am asking here. Sorry for incoherence. Materialscientist (talk) 23:40, 24 March 2010 (UTC)

I'm not sure I understand the question, but I can say this. if you view the source on any page there is a diagnostic tool - NewPP limit report. on the page in question it reads like this
NewPP limit report
Preprocessor node count: 15117/1000000
Post-expand include size: 214550/2048000 bytes
Template argument size: 114792/2048000 bytes
Expensive parser function count: 8/500
on a random page - Ulverscroft_Priory it reads like this:
NewPP limit report
Preprocessor node count: 1279/1000000
Post-expand include size: 8459/2048000 bytes
Template argument size: 3245/2048000 bytes
Expensive parser function count: 0/500
it's clear that your page is an expensive page to load, though not a horribly expensive one, but if you want to know whether your change was an improvement, take a page out of the history with the old revision and compare the numbers. --Ludwigs2 20:30, 25 March 2010 (UTC)
Thanks. How can we get to "NewPP limit report" function? The question was when does the page load approaches a level of discomforting an average reader? Two citation styles were alternately used on that page: (i) "cite journal" places all references into the main article, thus increasing its size; (ii) "cite doi/pmid" moves individual references out of the body into individual subpages - in case of this article, the was size almost halved by this, but. (a) An error in one subpage crashed the main page (sure any ref can do that, but such cases are more difficult to debug). (b) I'm not sure whether 200+ of subpages are not causing additional problems I don't know. Materialscientist (talk) 23:16, 25 March 2010 (UTC)
I've been editing that page: Vitamin D. By using the ref templates that have been brought into use by User:Citation bot and reworking of other links with Reflinks, the post-expand include size is kept within reason. The other templates that were being used, were what was causing the limits to be exceeded. See Wikipedia:Template limits. To see the report for any given page, use the view source command of your browser and look rather towards the end. Note that you can do this during an edit-preview. Also, this edit seemed on-track to bringing the other fork of that page under the limits. In the end, the standard templates that these tools bring into use are the way to go; anything else is pissing into the wind. Cheers, Jack Merridew 00:18, 26 March 2010 (UTC)

Double upload[edit]

Moved from Help Desk:

Why does the files that I upload keep doubling itself? Is there a bug or something? This problem only occur when I use Firefox. Please view here & here for example. I have no problem uploading file on commons using firefox. It only happens on this wiki. Please help. Arteyu ? Blame it on me ! 00:20, 24 March 2010 (UTC)

I'm sorry, I have no idea what you mean. Both images appear normal to me. (I'm using Firefox 3.0.15). --ColinFine (talk) 08:25, 24 March 2010 (UTC)
The file history indicates the files were uploaded twice in the same minute. I don't know the cause. They only appear once at Special:Contributions/Arteyu. You could try Wikipedia:Village pump (technical). PrimeHunter (talk) 15:00, 24 March 2010 (UTC)

The double upload still happens to two new files that I've uploaded just now, here and here. If it's a bug, I would need someone to fix it as soon as possible. Thank you. Arteyu ? Blame it on me ! 11:42, 25 March 2010 (UTC)

The only reason I can think of would be that you're somehow starting the upload twice - doubleclicking instead of singleclicking, for example. The uploads are simultaneous, and don't appear to be different in any way. The descriptions differ, as with File:Torquay United FC.svg (the more "recent" image has three summary headings shown, but only one on the page - the "older" image has only one summary heading), and I don't know what could cause that. It looks harmless - just an extra edit on your contribs, really - but it's still a curious technical glitch. UltraExactZZ Said ~ Did 13:07, 25 March 2010 (UTC)
I forgot to tell you guys that I've also got a problem with the page edit notice, which can be seen twice on top of page when I tried to edit it, see here for example [1]. Well, I solved both errors anyway by clicking on the "Restore all default settings" panel on "My Preferences" and refreshed the edit page. I went through all the gadgets again after that to know which of them made the error. And I found that it is caused by an editing gadget called "wikiEd". Think someone should fix this bug out so that it won't happen again to other users. Also I want to thank Algebraist and Avicennasis for helping me out on the freenode IRC channel. Arteyu ? Blame it on me ! 14:27, 25 March 2010 (UTC)
wikEd purposely creates the second editnotice below the preview. Discussion continued here. Gary King (talk) 14:52, 25 March 2010 (UTC)

Incidentally I believe the upload mechanism should disregard null uploads similarly to null edits, i.e. take no (database) action if the file you submit is identical to the current version. ―AoV² 20:08, 25 March 2010 (UTC)

MediaWiki:Viewsource[edit]

On a related note to the above discussion about MediaWiki:Protectedpagetext, I propose that the "edit this page" button should always say "edit this page". It should not become "View Source" on a page the user doesn't have permission to edit.

Reason:

  1. in most cases not having permission to edit is temporary or relatively easily overcome by getting autoconfirmed;
  2. "View Source" unnecessarily weakens the message that anyone can edit. This is a particular issue because many high-profile pages are more-or-less indefinitely semi-protected, and unloggedin users visiting these pages are not getting the "anyone can edit" message.
  3. clicking on "edit this page" provides further information about editing, discussing, and making edit requests, which is not at all obvious from "View Source".

This change can be made easily, I think, by editing MediaWiki:Viewsource. Rd232 talk 11:50, 25 March 2010 (UTC)

Ok, but make the change so it only applies to the article namespace. The Mediawiki: and WP: namespace are anyways stuff that isn't part of the encyclopedia, so it doesn't weaken the "anyone can edit" message. Anyways, these namespaces are hardly seen by new users except WP:5P and the other welcome pages. ManishEarthTalkStalk 13:01, 25 March 2010 (UTC)
Well, maybe, but the button is more than just "view source", it's also links to protection logs and the new edit request button. So long as AoV's point below is addressed, it might as well be the same across namespaces. Rd232 talk 14:17, 25 March 2010 (UTC)
Uggh, I assure you discarding information (namely whether the user can or cannot edit the page) with a change like this will upset more users than it will help. It′s bad enough this state of the button does not take “cascading protection” into account, see bugzilla:11700. Of course being an admin who can edit everything I wouldn′t expect you to understand. ―AoV² 13:45, 25 March 2010 (UTC)
Protected pages should generally have protection templates which indicate protection status at least with a top-right padlock icon. I take your point though that if people are used to seeing protection status from this change (edit -> view source), then maybe some form of indication in the tab should be maintained. It needs to be visible but non-invasive, so maybe just an asterisk after it, i.e. |Edit this page*|  ? Rd232 talk 14:17, 25 March 2010 (UTC)
A line through it at least but using completely different words would be much better. Making it say “edit” when one cannot edit is no less inane than showing the “move” tab when one cannot move. If you want to change “view source” to something more people will understand, that′s fine and dandy, but please keep it easily distinguishable from other buttons. ―AoV² 14:30, 25 March 2010 (UTC)
I would note that on the opposite, there are often people asking in the helpdesk "where is the edit button". So it creates as much confusion as that it provides information I'm guessing. Also I'm not sure if it is possible to use parserfunctions and magicwords in the tabs, to have context dependent contents. —TheDJ (talkcontribs) 14:50, 25 March 2010 (UTC)
I think that illustrates my point rather well :) I don't see why parserfunctions/magic words can't be used in the tabs (caching?), but I don't quite see what you'd do with it if you can. As I see it, consistency is a basic point of user interface design, and as a result when functions are not available for whatever reason (but in the context the function still makes sense) then this often highlighted in some way, without changing the text - often grayed out. But since the button is still usefully clickable, we should do something else. I think the asterisk solution is a good one, because the button does lead people into ways they can edit the page (even if it means registering first, requesting unprotection, or making an edit request), but it indicates a nonstandard process. Rd232 talk 15:10, 25 March 2010 (UTC)

I suppose the best approach overall would be to make “edit” and “view source” two completely different &actions. Then it would not matter whether the “edit” button appears in a disabled state, or does not appear at all—and the “view source” button could appear regardless of edit-ability (pointing to &action=viewsource or some-such). ―AoV² 20:04, 25 March 2010 (UTC)

Links from an independant wiki to Wikipedia[edit]

Hi guys and guyesses. First time I post here, I hope I do it correctly.

On a wiki, using MediaWiki but not connected to Wikimedia, I need to make links to Wikipedia/Commons pages with automatic conversion (made by a template) of spaces in titles to underscores. In other words, I'm searching for some code able to convert the title as it appears at the top of Wikipedia/Commons page to a correctly encoded URL.

I tried {{urlencode:, {{anchorencode:, {{fullurl:, but none seems to generate a correct URL.

An example is how to convert Wikipedia:Village pump (technical) to http://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29, when you are not on a Wikimedia's wiki?

Thanks in advance for your help. --GaAs (d) 18:04, 25 March 2010 (UTC)

As I don't want to lose your time,
which all three don't work, and
  • {{fullurl:
works only here, as it adds the local site prefix. --GaAs (d) 18:15, 25 March 2010 (UTC)
You will have to include the namespace Wikipedia: to make a link to this page.
I don't know for sure whether localurl will do the same on your wiki. PrimeHunter (talk) 19:22, 25 March 2010 (UTC)

You may wish to try mw:Manual:Interwiki to allow links like [[w:Village pump (technical)|foobar]] instead of the “external” link syntax above. This will handle all necessary url-encoding. ―AoV² 19:58, 25 March 2010 (UTC)

I knew this answer. But it's no possible to implement it at the moment (for "political" reasons). If anyone has another idea... --GaAs (d) 21:58, 25 March 2010 (UTC)
What does the localurl example produce on your wiki? PrimeHunter (talk) 03:11, 26 March 2010 (UTC)

Dumb question about the symbol menu

I know this comes up every so often but I can't think of a search term for the archives. When I change the pick in the drop-down menu under the edit window, say from "Symbols" to "Latin", the pickable special characters don't change. My monobook.js is fully commented out. Can anyone heave a sigh and tell me what I'm doing wrong? Franamax (talk) 19:43, 25 March 2010 (UTC)

What browser/OS? I'm jumping back and forth between all the options (and specifically between Symbols and Latin) and it's behaving correctly for me. EVula // talk // // 19:58, 25 March 2010 (UTC)
D'ohh! IE8 v8.0.6001.18702 on fully patched WinXP MCE SP3. And the PC case is black. ;) Franamax (talk) 20:01, 25 March 2010 (UTC)
Do you have cookies blocked ? Or perhaps you have some Gadgets enabled in your preferences ? —TheDJ (talkcontribs) 20:43, 25 March 2010 (UTC)

Stuck RfC re actor pages and their hard-coded formatting

nb: admin review requested at WP:AN, which garnered a comment that this would be a useful page to seek input at.

An RfC that involves a lot of technical issues is rather stuck at:

It would be helpful if uninvolved folks could review and suggest an appropriate route forward? It's a long read; maaf.

There are on the order of 30,000 pages involved, so it would be good to get on it. Sincerely, Jack Merridew 22:44, 25 March 2010 (UTC)

Stable Toolserver link breakage

FYI you might notice that a lot of our toolserver links are giving errors right now. It seems the old stable.toolserver.org server was shut down, and apparently no one around here follows the toolserver list or no one bothered to mention it in this forum at least. I'm fixing the links as fast as I can. —TheDJ (talkcontribs) 00:17, 26 March 2010 (UTC)

I do (ACC), it's just hard to keep track where people put links, WP:ACC was updated a while ago (yes, I'm too lazy to query extlinks). Q T C 00:25, 26 March 2010 (UTC)
OK, I think i fixed the most important cases now. There are still a lot of transcluded uses in archives, but they don't really matter much. Geohack was the most glaring issue. —TheDJ (talkcontribs) 00:31, 26 March 2010 (UTC)
My guess is that the largest number of broken links came from the coords-related URL? I mentioned this awhile ago (I think in an IRC channel) and would have fixed it myself, but I've got no superpowers here. Oh well. Though, really, unless there's some issue with redirecting, stable.toolserver.org should still be an alias/redirect/whatever. The intention was to have a stable environment, after all, and it's poor form to needlessly change or break a URI without implementing a redirect. --MZMcBride (talk) 00:33, 26 March 2010 (UTC)
See JIRA TS-535. Stable GeoHack URL should continue to work per JIRA TS-368 so no fixing is required. However, if your changing them anyway consider the possibly using the new Short URL. — Dispenser 02:21, 26 March 2010 (UTC)

RefToolbar script for new toolbar ready for testing

I was planning on waiting until the weekend, but with the above announcement, I think sooner may be better than later. The current RefToolbar script does not work on the enhanced toolbar created by the Usability Initiative. refToolbar 2.0 is a complete rewrite of the script designed based on the new toolbar and using lessons learned from the old script. Unfortunately, the script was only able to work on Wikimedia sites since about 24 hours ago when the dialog features of the new toolbar were turned back on, so I haven't been able to test it extensively myself, so I could use some brave beta testers to work out any bugs and provide feedback. The new script is more customizable than the old one, though it does not yet support some of the new features added by Apoc2400 in the "refToolbarPlus" script. Please leave any comments, bug reports (making sure to note what browser you are using), and feature requests on the script's talk page. Mr.Z-man 03:17, 26 March 2010 (UTC)

Abnormally fast GIFs

Resolved: not a problem with Firefox, but rather with all other browsers and some GIF creation programs. A Greasemonkey script has been created to fix affected files. PleaseStand (talk) 20:24, 22 March 2010 (UTC)

File:Animated Kaleidoscope.gif GIFs like the one left have been animating abnormally fast lately. Is there a reason why? NERDYSCIENCEDUDE (✉ msgchanges) 22:41, 20 March 2010 (UTC)

Apparently the Earth is spinning faster. Carcharoth (talk) 23:39, 20 March 2010 (UTC) Clearly I meant your computer is 'spinning' faster, not the Earth... (see below) Carcharoth (talk) 01:10, 21 March 2010 (UTC)
  • I have in the past made animated gifs that chundered along at a suitable speed and found to my horror that they raced on other machines. The reason was that I'd specified a frame rate that was faster than could be rendered at the clock speed of the machine I made them on. So they bumped along at the maximum speed they could be rendered which looked fine. On running them on a faster machine, the true speed that I'd set in the frame rate became evident. Reediting them to reduce the specified frame rate brought the animation speed to the same in all machines that were fast enough. The effect was particularly noticeable on large pixel-dimension images; length of animation didn't slow them much. Relevant? Trev M 00:58, 21 March 2010 (UTC)

I can't help with the precise technical issue (other than "oh dude, you're blowing my mind, it's actually the screen revolving around the image" :) but to touch on a pet peeve: why don't we have a Stop button for animated gif's? They're like an over-talkative dating partner - really interesting to start with, but eventually intensely annoying. Has some backing software for a kill-switch ever been mooted? Franamax (talk) 01:51, 21 March 2010 (UTC)

Maybe because many (most?) browsers stop GIF animations if you hit "escape"? --jpgordon::==( o ) 02:17, 21 March 2010 (UTC)
Eek! The Earth stopped moving. <phew> I hit refresh and things started spinning/bouncing again! :-) Carcharoth (talk) 12:21, 21 March 2010 (UTC)
Firefox can stop animated GIFs by hitting escape. Also, the Bouncy Wikipedia Logo has also been bouncing abnormally fast. NERDYSCIENCEDUDE (✉ msgchanges) 03:28, 21 March 2010 (UTC)
Firefox and IE have the esc control for gifs. Chrome doesn't. ManishEarthTalkStalk 07:51, 21 March 2010 (UTC)
Pressing Alt in Chrome pauses them for me. But only until you scroll or do something on the page. Not sure if this is an intentional feature by Google. - Kingpin13 (talk) 14:07, 21 March 2010 (UTC)
Ooh thanks! ManishEarthTalkStalk 14:57, 21 March 2010 (UTC)
Seems like alt pauses everything, including mouseovers, hovers, tooltips, and flash. Really useful! Thanks again! ManishEarthTalkStalk 15:01, 21 March 2010 (UTC)
They already have a key-sequence? Nice. :) I suppose I'm just prejudiced against animated GIF's, seeing as how one animation on Red blood cell would take a dial-up user around 7 minutes to load. Franamax (talk) 03:49, 21 March 2010 (UTC)
Huh, now that I'm editing on an iPhone, the GIFs animate normally. Maybe it's my computer. NERDYSCIENCEDUDE (✉ msgchanges) 03:59, 21 March 2010 (UTC)

In SVG animations, such as [2], timing issues are easier to correct. Unfortunately I can′t display it here as a non-moving PNG conversion would appear in place of it, but if I wanted to slow it down, I′d need only change the dur="3s" attribute to a higher number using a text-editor.

One also could manipulate it with javascript. Pasting something like this in the address bar while viewing the SVG directly will slow it by 5×:

javascript:for(i = 0; i < (a = document.getElementsByTagName("*")).length; i++) a[i].setAttribute("dur", "15s");

However, javascript cannot access the frames of a GIF or otherwise its animation. Sites which appear able to do this are in most cases changing the src attribute of the <img> tag (causing the browser to load pixels from a different file), or otherwise aren′t using GIFs at all. ―AoV² 04:34, 21 March 2010 (UTC)

Viewing the above Bouncy Wikipedia Logo on this page in Chrome 4.1.249.1036, Firefox 3.6, and IE 7.0.5730.13 show that Chrome and IE7 are both 2.6 sec per bounce, but Firefox 3.6 is 1.1 sec per bounce. More than twice the speed. --Redrose64 (talk) 14:01, 21 March 2010 (UTC)
I think it's Firefox that's causing the problem. The animations view correctly in Safari. NERDYSCIENCEDUDE (✉ msgchanges) 14:25, 21 March 2010 (UTC)
I also think it's Firefox, so I reported it to Mozilla as bug 553968. PleaseStand (talk) 03:26, 22 March 2010 (UTC)
At least I got a quick response since it turns out that the problem had already been reported (six years ago). (And I thought that I had discovered yet another Firefox 3.6 bug.) Firefox is actually behaving correctly, while others are not and treat delays less than 5 cs as 10 cs delays. So the Firefox developers are basically saying that for compatibility, websites should not use animated GIFs that have delays less than 5 cs (greater than 20 frames/s). I wrote a Greasemonkey script which you should be able to use to make any other files that are too fast the same speed in all browsers. Use the script, right-click the image and do Save Image As, and upload the new version to Wikipedia if appropriate. I just updated the two images shown above and these should now run at the same speed for everyone who sees them. PleaseStand (talk) 20:24, 22 March 2010 (UTC)
On a side note, I much prefer the Bouncy bouncing along at 1.1 seconds per bounce. --Izno (talk) 03:50, 22 March 2010 (UTC)
Note also that someone uploaded a faster version in January (but not sure what time scale was being referenced). --Splarka (rant) 08:29, 22 March 2010 (UTC)
Yes, I also noticed that, but nevertheless, the animation should run at the same speed in all browsers. Remember that the bouncy logo is included in the widely used welcome template {{w-screen}}. I find the faster bounce to be annoying and distracting. PleaseStand (talk) 10:53, 22 March 2010 (UTC)
I notice that the rate is now 1.9 sec/bounce in Firefox, while remaining at 2.6 sec/bounce in Chrome and IE7. I've recently upgraded from Firefox 3.6 to 3.6.2; maybe these are related? --Redrose64 (talk) 10:08, 26 March 2010 (UTC)

FWIW - One Possible Way To STOP GIF Animation on Wikipedia (and related sites?) is to Press ALT+SPACEBAR together, ESC To Remove DropDown Menu, ALT To Resume - Seems To Work In The GOOGLE CHROME BROWSER (Version 28.0.1500.95 m) (Windows XP) (08/16/2013) - In Any Case - Enjoy! :) Drbogdan (talk) 14:00, 17 August 2013 (UTC)

Help using CSS?

I'm trying to perfect my Monobook CSS skin, and I'm having some difficulty. The skin is located here, and I'm trying to figure out how to get things like templates, tooltips, etc. to keep their default background color. Right now, all the backgrounds are transparent, causing problems with tooltips and templates with colored backgrounds. Could someone who is good at CSS take a look at it? Hmmwhatsthisdo (talk) 23:54, 22 March 2010 (UTC)

Actually, as of now, I have a set background on all of them, but I don't know if something like "#default" exists. Hmmwhatsthisdo (talk) 00:35, 23 March 2010 (UTC)

You just need to make *{background-color: transparent !important; color: lightgrey !important;} much more specific / less broad. ¦ Reisio (talk) 05:43, 23 March 2010 (UTC)

Well, I figured out how to have them not blend with each other by using td {background-color: #313131 !important;}

, but this messes up the main page. Any ideas? Hmmwhatsthisdo (talk) 23:05, 23 March 2010 (UTC)

I'd suggest making background color transparent on a more specific basis rather than making it the default. I'd say that that's how to do it properly. Gary King (talk) 03:09, 27 March 2010 (UTC)

Technical question regarding mms://

Moved from WP:AN

My editing mostly covers radio and television stations and I and my fellow editors are finding that there are several stations (don't have an exact number) whose live stream links begin with mms:// which does not compute into a link on Wikipedia. If it begins in http:// it links no problem, but mms:// does not. How would one suggest what should be an easy change and only affect pages that link to webstreams (particually radio and television station pages). Thanks...NeutralHomerTalk • 09:11, 24 March 2010 (UTC)

You'll need to get community consensus to add 'mms://' to $wgUrlProtocols and then file a request on Bugzilla for a site configuration change, linking to the consensus. ^demon[omg plz] 10:22, 24 March 2010 (UTC)
Or we can just change the defaults. If it's useful, I don't see why we wouldn't add it. The Wikipedia article on Microsoft Media Server says it's a Microsoft technology, but also that there's now a publicly available spec, and multiple open-source implementations. As long as there are no scary side effects (there's at least one IM URL protocol where clicking a link can send a message with no further user interaction, IIRC), I don't see why we wouldn't add it. worldwind:// and svn:// are supported by default, although both are non-standard. —Aryeh Gregor (talk • contribs) 16:27, 24 March 2010 (UTC)
I'm all for erradicating mms indefinitely, but it is relatively common, and I don't think it would hurt to add it to the defaults, possibly making a lot of people happy. —TheDJ (talkcontribs) 17:19, 24 March 2010 (UTC)
I don't really care either way, we can add it if other people think it's worth it. I just figured people wouldn't by default, so I gave the usual "get consensus then ask the shells" speech ^demon[omg plz] 21:30, 24 March 2010 (UTC)
So is this consensus or do I need to go elsewhere and get others opinions? Never actively tried to get something added to Wikipedia, so I don't know the process. - NeutralHomerTalk • 04:57, 25 March 2010 (UTC)
  • We link to the homepage of each station. How many stations fail to link to their live streams from their own websites? There is no encyclopaedic purpose to be served by linking to their live streams as well as their own websites, as far as I can see. Guy (Help!) 14:46, 24 March 2010 (UTC)
    • Guy, most of the pages are actively updated, so there is a pretty good chance if you click on a livestream link, it is going to the actual live stream and is going to work. I still would like to see it done. - NeutralHomerTalk • 03:54, 25 March 2010 (UTC)

I added mms:// to MediaWiki defaults in r64220. No idea when it will be live on Wikipedia, might be months from now. Probably not terribly useful for Wikipedia, but it might be useful for other wikis, especially non-Wikimedia ones. —Aryeh Gregor (talk • contribs) 19:08, 26 March 2010 (UTC)

Thanks! You rock! - NeutralHomerTalk • 22:49, 26 March 2010 (UTC)

Dialogs are back in Usability Beta

Users of Safari/Chrome/Firefox/Opera might note that dialogs have returned for inserting links, tables and doing Search and Replace. Users who have disabled this in the past might have to go to Special:Preferences and enable it in the section Editing -> Beta Features. —TheDJ (talkcontribs) 00:08, 25 March 2010 (UTC)

Plus the navigable TOC in editing in Prefs>Editing>Labs features. I wonder when they'll roll out the template resizing... I've tried it out at the sandboxes and it's great! ManishEarthTalkStalk 03:18, 26 March 2010 (UTC)
Huh? The navigable TOC isn't showing... ManishEarthTalkStalk 03:21, 26 March 2010 (UTC)
Plus it blocked keyboard shortcuts... ManishEarthTalkStalk 03:22, 26 March 2010 (UTC)
The navigable TOC is still disabled. It depends on the iframe editor (which was accidently enabled yesterday) which still has too many issues to deploy. —TheDJ (talkcontribs) 19:28, 26 March 2010 (UTC)

Citing tool

I've been having problems with the "insert citation" tool over the last few days. Sometimes it works fine, but it has crashed firefox a couple of times when I click to add the citation (very annoying!), at other times it adds the reference at the top of the edit box rather than where I've clicked for it to be added. In this edit it place ref tags at the top and bottom of the article but didn't add the reference at all. It's doing the same things on different computers too. Has anyone else had problems/can anyone suggest how I might fix it? Thanks Smartse (talk) 13:29, 25 March 2010 (UTC)

Are you using WikEd? I have had the same problem with the sig button when using WikEd - either not working, adding my sig at the top of the page - similar weirdness to yours. Also had unpredictable Firefox crashes too. Disabling WikEd seems to fix it, but it's not a solution. Dang it just did it again! – ukexpat (talk) 03:34, 26 March 2010 (UTC)
Yea I use WikiEd but too, the benefits of it outweight the annoyance caused by this but it would be good if someone could fix it. When I add citations, the form stays up and filled in instead of collapsing as it should do. Smartse (talk) 13:15, 26 March 2010 (UTC)

Upcoming interface changes starting April 5

The usability Beta will be going live very soon. Please read the blogposts foundation and by the techteam. —TheDJ (talkcontribs) 01:27, 26 March 2010 (UTC)

Oh, noes.  fetchcomms 02:34, 26 March 2010 (UTC)
Uh oh. VPT-meltdown day approaches. OrangeDog (τε) 12:26, 26 March 2010 (UTC)
Fasten your seatbelts—it's gonna be a bumpy night! – ukexpat (talk) 19:43, 26 March 2010 (UTC)

Movement in the text box

I am encountering an issue (on IE8) while editing pages (and sections of pages) which contain enough text as to require scrolling the text box: the view (but not the text cursor) jumps around (one jump per action) when I move the mouse cursor over the edit summary box, the "Show preview" button, or the various links ("verifiable", "what's this?", "Cancel", "Editing help"), and when I click the "Show preview" button. I have no idea what could be causing the issue, but it must be fairly recent (I didn't encounter this when editing yesterday). Is anyone else having the same issue? Thanks, -- Black Falcon (talk) 17:46, 24 March 2010 (UTC)

Yes, I am also experiencing this problem, starting this morning. It's not the first time it's happened, and it's usually gone away within a day or so. Pats1 T/C 19:01, 24 March 2010 (UTC)

Someone must be playing with the JS again. ¦ Reisio (talk) 00:27, 25 March 2010 (UTC)

There also appeared to be changes to the size of the text in the "save page" etc. buttons at the buttom. Also to the size of the links in the top-right. Pats1 T/C 14:37, 25 March 2010 (UTC)

Is having the same problem and its on its 3rd day now. --> Halmstad, Charla to moi 13:36, 25 March 2010 (UTC)

Me too, thought it was just me. Also running Internet Exploder 8. I noticed the text box scrolls down one line for every character I type, or other action, until the cursor is at the bottom of the edit window. So one does not notice typing on the last line, but otherwise, this is a real annoyance. Could somone back out the JavaScript change please? W Nowicki (talk) 21:37, 25 March 2010 (UTC)
Mine isnt jumping down its jumping up either i click with the mouse or writte something, its especialy annoying since it prevents me from highlighting text in longer texts with the mouse. --> Halmstad, Charla to moi 00:46, 26 March 2009 (UTC)

Thank you for surfacing this issue. We appreciate it and we will be working to resolve this issue ASAP.--Parul Vora (talk) 03:10, 26 March 2010 (UTC)

We were accidentally serving beta users the (currently a bit unstable) iframe instead of the normal textarea because the code I wrote to prevent this from happening was broken :( Fixed this a few minutes ago. --Catrope (talk) 13:18, 26 March 2010 (UTC)
The problem has not yet been fixed, as far as I can tell. Typing any text in the textbox automatically scrolls the line you are editing to the bottom of the viewable area of the text book. Anyone else still having this problem? Pats1 T/C 05:23, 27 March 2010 (UTC)
Speaking of the beta - each time I go to try it (over a period of five+ months), there is always a problem with the "article" and "discussion" etc. links at the top. They aren't treated as links, but a background image, so you can't right click to open them in a new tab/window in IE8. Pats1 T/C 00:36, 27 March 2010 (UTC)

This jumping around almost completly prevents me from editing on wikipedia, the smalest thing i do makes it jump up, can't use only the keybord. --> Halmstad, Charla to moi 17:13, 26 March 2010 (UTC)

I have opened bugzilla:22983 to track this problem —TheDJ (talkcontribs) 00:34, 28 March 2010 (UTC)

Toolserver Opt-in Requirement

I'm guessing this must be an extremely recent thing, because I don't remember this being there yesterday. My Toolserver edit count has boxes now stating:

"Graphs have been set to opt-in as a result of a single complaint resulting in what is essentially a cease-and-desist letter from the toolserver admins. If you want to see graphs, please create User:Silver seren/EditCounterOptIn.js with any content. Alternatively, you can create meta:User:Silver seren/EditCounterGlobalOptIn.js to opt-in across all Wikimedia wikis."

Whatever's going on here, it sounds ridiculous. But I guess that's something for lawyers to work out. Anyhow, how exactly do I set up this Meta page so that it will start displaying everything again? I know absolutely nothing about code. Nothing. Beyond the basic features I use in making edits and such, I don't know how to do anything related to the mechanics. Can someone help me walk through this (or maybe do it for me?), so I can get this working again? And I sincerely doubt i'm going to be the only user with such a problem. Thanks in advance. SilverserenC 08:48, 25 March 2010 (UTC)

Create this page User:Silver_seren/EditCounterOptIn.js and write whatever in it. There is a discussion about this in X!'s userpage. --JokerXtreme (talk) 09:52, 25 March 2010 (UTC)
(e/c)Several editcounters have had opt-in requirements for some data, in my experience. In any case, what you have to do is very simple - just click this link: Special:Mypage/EditCounterOptIn.js and create the new page - it doesn't matter what you put on the page (hence the "any content" above).
If you want to use it on other projects, not just the English Wikipedia, click meta:Special:Mypage/EditCounterOptIn.js and do the same thing. Basically, they're just asking you to verify that you actually want the Edit Counter to show your edits - not unlike what websites do when they have to verify your e-mail address. BTW - I've used the Special:Mypage version of it so anyone can click on the link, not just Silver seren.--Philosopher Let us reason together. 09:57, 25 March 2010 (UTC)
We've put together a tswiki:List of things to ask WM-DE over edit counter and other weird German laws. Please feel free to correct anything there as I didn't have time to proof read it. Depending on the answer we may have to band edit counters on the Toolserver. German law does not have to be self consistent it just has to "make sense". Example: while you can legally take street view photos and post them to the internet, while anything like Google's street view is illegal. — Dispenser 11:02, 25 March 2010 (UTC)

Would blanking said page disable the edit-counter, or would one need to delete it? ―AoV² 10:13, 27 March 2010 (UTC)

Mixed content: HTTP mixed with HTTPS

When using Wikipedia in SSL mode, I keep getting warnings from Firefox that all the pages I visit on Wikipedia have mixed content. I tracked it down to two links that are present in every page: link rel="apple-touch-icon" href="http://en.wikipedia.org/apple-touch-icon.png" and a style="background-image: url(http://upload.wikimedia.org/wikipedia/en/b/bc/Wiki.png);". These links are forced to non-SSL (HTTP) transfer and Firefox complains that the page is mixed secure and non-secure items. How can I set Wikipedia to always give HTTPS links and thus remove the annoying warnings? Note: That I know I can simply turn off the warning in Firefox but I don't want to do that since that will lower my security. I prefer to keep the warning so that I know when something bad is happening. It is preferable if Wikipedia can be set to not generate mixed content. HumphreyW (talk) 23:03, 25 March 2010 (UTC)

This is not possible. See also bugzilla:18496. Currently Wikipedia does not have the resources to give priority to this problem. —TheDJ (talkcontribs) 23:06, 25 March 2010 (UTC)
Do you mean it is not possible because of lack of resources, or that is is not possible because of technical reasons, or perhaps both? HumphreyW (talk) 23:12, 25 March 2010 (UTC)
Well technically, it requires making sure that some servers (the media server and the CSS/JS/icon server) get an SSL layer and that this properly works with the main servers. Then that all has to be implemented, configured, tested whatever. So that is a lot of work, that only few people can do. —TheDJ (talkcontribs) 23:28, 25 March 2010 (UTC)
Good, so it is just a technical problem with a known solution. So when has it been scheduled to be fixed? HumphreyW (talk) 07:27, 26 March 2010 (UTC)
It isn't. One day in the future when someone gets really really bored, he might do it. —TheDJ (talkcontribs) 11:53, 26 March 2010 (UTC)
So is there some way to have the priority of this problem raised? It seems to me that a secure connection is needed for a number of reasons. To stop my cookies leaking to third parties. To solve the problem with the transparent proxies at my ISP. To avoid my government/employer/man-in-the-middle tracking my usage. etc. To me these reasons are important. How can I convince the Wikipedia people of the same? It is frustrating to see that it requires someone to be "really really bored". It doesn't appear to be very encouraging.
Most sites with HTTPS simply use the same URLs for both secure and non-secure access. I am curious to know why Wikipedia decided to use an entire set of different URLs when using the HTTPS protocol? HumphreyW (talk) 12:10, 26 March 2010 (UTC)
Yes, it's possible for the priority to be raised. No, it probably won't happen. Wikipedia doesn't have any particularly good reason to support SSL at all, which is why the implementation is so hacky. It's unlikely anyone is going to want to go to the effort of setting up a MITM attack to steal your Wikipedia login or spy on your browsing habits. As such, other things are much more important. I'm sure it will get fixed someday. —Aryeh Gregor (talk • contribs) 18:59, 26 March 2010 (UTC)
And to answer your question as to "why don't we have SSL certs for all the domains?" I think the original reason was cost...it wasn't worth the money to buy certificates for all the top-level domains, so they got one for secure.wikimedia.org. I don't think cost is so much the issue anymore, as now it's a matter of implementation. But like Aryeh says, it's not ultra-high on the priority list. ^demon[omg plz] 11:45, 27 March 2010 (UTC)
I thought that once the main domain was paid for then all sub-domains are essentially free (or at least of very low cost) with just logging in to the cert issuers site and generating a new cert automatically based upon previously registered credentials. Anyhow, I guess it doesn't matter how the certs are issued since the main problem seems to be inertia, time management and motivation, rather than finances. HumphreyW (talk) 12:17, 27 March 2010 (UTC)
As a temporary measure to avoid the problem I found that using the add-on NoScript there is an option under Advanced/HTTPS/Behavior/Force the following sites to use secure (HTTPS) connection: where I added upload.wikimedia.org. This forces Firefox to try using an SSL connection. The connection fails of course, but the warnings about mixed content are gone. The down side is that no pictures can be displayed when using HTTPS connections to Wikipedia. Not an ideal solution, but those annoying warnings and the potentially risky mixed content is gone. HumphreyW (talk) 10:26, 26 March 2010 (UTC)
Well, *.wikimedia.org has a certificate, so they just have to create https://upload.wikimedia.org (WHich they probably have reserved under *.wikimedia.org anyways). ManishEarthTalkStalk 12:29, 27 March 2010 (UTC)
And also to change the links so that HTTPS users get HTTPS link and HTTP users get HTTP links. HumphreyW (talk) 12:34, 27 March 2010 (UTC)
Shouldn't be hard... Will create a massive server lag (I think), but otherwise it requires a tiny change in coding for securewiki. ManishEarthTalkStalk 12:42, 27 March 2010 (UTC)

Looks like Gregory Maxwell determined that protocol-relative urls work fine on all clients tested in 2008. [3]AoV² 01:22, 27 March 2010 (UTC)

For what it's worth, there's still some places in the software that need cleanup. I did fix relative URLs in interwiki links just recently. Meaning if one day there's a https://en.wikipedia.org/ then the 'es' interwiki link can be '//es.wikipedia.org' for both sites and not require hacking to work right :) ^demon[omg plz] 11:43, 27 March 2010 (UTC)
We've got lots of SSL issues, by the look of it.wikitech: https://wikitech.wikimedia.org/view/ leads to an ssl site which has issued itself a certificate. Makes me wonder why the interwiki link leads to the https: page instead of http://wikitech.wikimedia.org/view/, which works fine. I've posted a message here, but no one's listening :(. ManishEarthTalkStalk 11:59, 27 March 2010 (UTC)
Update: Interwiki link fixed, though wikitech SSL still has an invalid certificate. ManishEarthTalkStalk 12:20, 27 March 2010 (UTC)

Edit box typing line jumps

Hi. I'm not sure when this new error was implemented, but today when I edit, the line that is being typed in automatically jumps to the bottom of the edit box if all the text being edited takes up more space than the height of the box. I do not like this feature, as it is distracting and easy for the editor to get lost. Is there any information avaliable on this recent development in editing? Thanks. ~AH1(TCU) 21:44, 26 March 2010 (UTC)

  • I've had the same problem. It has slowed in the last couple days and has stoped at present. Have no idea the cause Mlpearc MESSAGE 21:48, 26 March 2010 (UTC)
Just wanna let you guys know your not the only ones, read abit up under the headline Movement in the text box. --> Halmstad, Charla to moi 22:04, 26 March 2010 (UTC)
Is it the same problem described here: WP:Help desk#Editing window scrolls automatically if certain keys are used? – ukexpat (talk) 22:05, 26 March 2010 (UTC)
To Ukexpat, yes sounds the same. I reset all default preferences, cleared the browser a couple times, but I, by no means say this solves the problem. Mlpearc MESSAGE 22:16, 26 March 2010 (UTC)
The problem is apparently currently still apparent. ~AH1(TCU) 00:09, 27 March 2010 (UTC)
Yep. Definitely still a problem. As I commented in the Help Desk link provided above, it looks like an old problem has come come back. Despite User:Catrope saying "Fixed now", it is still a problem. While I don't have a fix, I do have more information: If you compare this screen capture to one I took a couple of weeks ago, you might notice the newer one has the "broken page icon" in the address bar. Somehow IE8 or Microsoft are convinced Wikipedia pages are now broken and should be viewed in "compatibility mode". I think either the Wikipedia software guys have made a change, or Microsoft have added Wikipedia.org to the compatibility list in C:\Program Files\Internet Explorer\iecompat.dll. Considering that my iecompat.dll dates from 18 Feb 2010 (ver: 8.0.6001.18902) and does not list wikipedia.org amongst the sites, I think this is more likely to be something the Wikipedia software guys have introduced within the past week. Astronaut (talk) 00:34, 28 March 2010 (UTC)
I wonder... a sitenotice has been running the past few days. perhaps the problem is there ? Then all pages sould have the "broken page icon". If it is just the edit page, then it is likely that the problem is with the code of the usability team. —TheDJ (talkcontribs) 00:39, 28 March 2010 (UTC)
Sitenotices in that past have not caused this. The annoying cursor jumping only appears in the edit box. As previously described: edit box jumps to put current line at the foot of the scroll region, and when popups would appear the edit box jumps up a few lines so the line with the current cursor position disappears off the bottom of the scroll region. Astronaut (talk) 00:57, 28 March 2010 (UTC)

I have opened bugzilla:22983 to track this problem —TheDJ (talkcontribs) 00:37, 28 March 2010 (UTC)

Thanks. Would be great if they could (temporarily) revert whatever was they set out to fix while they come up with another solution - the cursor jumping is getting really annoying (every time I drag the cursor back to the save button, the text box jumps a few lines - especially if I pass over a link that will make POPUPS leap into life). Astronaut (talk) 00:45, 28 March 2010 (UTC)
Did you check if this was a problem of popups btw (or another gadget of course) ? It's not really actively supported on IE. —TheDJ (talkcontribs) 00:51, 28 March 2010 (UTC)
I only noticed the possible popups link while typing here. I'll test turning it off, if it will catch up with my typing (is that a new problem with each character taking 0.5 sec to appear). Gahhh! Astronaut (talk) 00:57, 28 March 2010 (UTC)
Partial success when I turned off popups in my monobook.js (ie. not the gadget in preferences). The random scrolling when popups would appear has stopped, but the scroll region still annoyingly jumps to the last line when inputting text. Not so annoying when contributing to talk pages and such like because you normally add text to the end of a discussion, but a major annoyance when contributing to articles and you want to pop back and forth in the edit box or read the text in the paragraph below the cursor while typing. And I just spotted another thing, a random jump when you delete some selected text in the edit box.
FWIW, I have found popups really useful, particularly when checking history/diffs during anti-vandal work. It is one of the few addons that works with IE and has worked successfully for me for some years.
Oh, the slow typing has fixed itself - probably a temporary problem at my end :-) TBH, I don't see why the "broken page icon" has suddenly appeared. Something has changed very recently to cause these problem to appear. Astronaut (talk) 01:24, 28 March 2010 (UTC)

Skype formatting - topic reopened

Re-opening the topic mentioned here. Today I noticed a few posts with this from diverse editors so I think we may need the interface to filter out these text:

  • begin_of_the_skype_highlighting
  • end_of_the_skype_highlighting

Example edits:

  • In this one it appears to be in a hidden field so it isn't messing up the visible article but is likely affecting some other aspect of the infobox or other coding.

Looking for others now with AWB.  7  23:07, 26 March 2010 (UTC)

Two more examples with worse results (formatting ISBNs) [4] and [5]  7  23:16, 26 March 2010 (UTC)
Ok - I can't run AWB on my current machine but it looks like there are 70+ other occurrences of this out there if someone is willing to take a look.  7  23:24, 26 March 2010 (UTC)
I just corrected about 20 cases using [6] (not so useful right now as search lags the corrections already made). Sometimes it even appears twice for the same number.[7] --Rumping (talk) 00:38, 27 March 2010 (UTC)
I think we've gotten them all now - but may still be worth blocking the edits in the interface.  7  13:32, 27 March 2010 (UTC)
We might want to have an abusefilter looking at this ? Warning people before they save ? —TheDJ (talkcontribs) 00:14, 28 March 2010 (UTC)
Good idea TheDJ. I suggested it there.  7  02:25, 28 March 2010 (UTC)

Not working file redirect

I moved a number of files, but discovered that one of the redirects does not work: File:617044877897062.jpg. I do not know what happened. Ruslik_Zero 12:26, 27 March 2010 (UTC)

The lonk you've given is a "do not redirect" link which makes sure you won't be redirected when you visit the page. File:617044877897062.jpg should work ManishEarthTalkStalk 12:32, 27 March 2010 (UTC)
Thanks but I know. It does not work as an image link. See Fanny_and_Alexander. Ruslik_Zero 12:34, 27 March 2010 (UTC)
Wrong redirect syntax. You should have typed: #REDIRECT [[:File:617044877897062.jpg]] not #REDIRECT [[File:617044877897062.jpg]] (Notice the extra colon). Actually, that should be fixed. There's no reason someone will use #R syntax for a file(without the linking colon) except for a redirect. ManishEarthTalkStalk 12:44, 27 March 2010 (UTC)
Other redirects work fine. For example File:Enos_Throop.gif. In addition I did not create the redirect as it was created automatically. The system should probably create redirects with colon. Ruslik_Zero 12:49, 27 March 2010 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Maybe because it had no letters and only images? File a bug report. ManishEarthTalkStalk 13:54, 27 March 2010 (UTC)

Now it works without the colon too. Ruslik_Zero 14:07, 27 March 2010 (UTC)
This happens sometimes. A purge seems to clear the issue usually. Probably the redirect is accessed before the target has propagated on al the servers. So the server thinks it is a broken redirect and caches that result. —TheDJ (talkcontribs) 00:09, 28 March 2010 (UTC)

Create protection after page is created?

After deleting a file talk page that had been created for the third time, I salted it indefinitely with semiprotection, because all three creations were by IPs who simply added graffiti. If (for some odd reason) an autoconfirmed editor recreates the page for good reason, will that end the semiprotection, or will non-autoconfirmed editors still be unable to edit the page? I don't see anything discussing the subject at WP:SALT. Nyttend (talk) 14:07, 27 March 2010 (UTC)

The 'create' protection is separate from edit protection which will be set to allow all users after it is created. –xenotalk 17:03, 27 March 2010 (UTC)

Consensus building templates

I'd like to draw the community's attention to this Tfd:Wikipedia:Templates_for_discussion/Log/2010_March_27#Template:CB-support2. This concerns some other perennial (although not formally) request. Please post there, to keep the discussion in one place. --JokerXtreme (talk) 17:56, 27 March 2010 (UTC)

TFD discussion closed as speedy delete per WP:CSD#G4 and Wikipedia:Deletion review/Perennial requests#Template:Support. --BrownHairedGirl (talk) • (contribs) 18:24, 27 March 2010 (UTC)
I think that G4 does not apply in this case. G4 was probably made with article content in mind. Thiis is just one of the reasons I think it does not apply. I ask you to reconsider as I did in your talkpage. --JokerXtreme (talk)
If it were intended for articles only it would be an "A" criteria. –xenotalk 19:44, 27 March 2010 (UTC)

Template help - Internet Archive author

Resolved

Could I get some help getting {{Internet Archive author}} working, please. I've based it on {{gutenberg author}}, but when using it find it produces "George_Aaron_Barton" Works by George Aaron Barton at the Internet Archive. rather than the desired link. Example:

{{Internet Archive author | id=George_Aaron_Barton | name=George Aaron Barton}}

contrast with

Works by George Aaron Barton at Project Gutenberg

Thanks --Tagishsimon (talk) 21:07, 27 March 2010 (UTC)

got it - it just needed the query portion of the link urlencoded. --Ludwigs2 21:51, 27 March 2010 (UTC)
Thanks, Ludwigs2 - but there's still an issue. Looks like query= is being made into query%3D, which causes it to fail.
{{Internet Archive author | id=Verne_Jules_1828-1905 | name=Jules Verne}}
gives a url of
http://www.archive.org/search.php?query%3Dcreator%3A%22Verne_Jules_1828-190%22
which fails. A working URL is:
http://www.archive.org/search.php?query=creator%3A%22Verne_Jules_1828-1905%22
Tagishsimon (talk) 22:08, 27 March 2010 (UTC)
Sorry, put the URLENCODE in the wrong place. this seems to do it, though it looks like you may need to play with the search term a bit. --Ludwigs2 22:51, 27 March 2010 (UTC)
Thanks. Will do. --Tagishsimon (talk) 22:56, 27 March 2010 (UTC)
twiddled with it a bit more - think I go it--Ludwigs2 22:59, 27 March 2010 (UTC)
That's very promising indeed :) --Tagishsimon (talk) 23:01, 27 March 2010 (UTC)

How can I have the alt text changed on the wiki logo on the top left? Gnevin (talk) 23:16, 21 March 2010 (UTC)

You mean to change the logo? Or to change the alt text for the logo(the "Visit the main page")?Smallman12q (talk) 23:25, 21 March 2010 (UTC)
Change the alt text to "Main page" or something similar Gnevin (talk) 23:28, 21 March 2010 (UTC)
I think only developers can modify those texts (as in the section just above, about view source). A search in Special:AllMessages didn't return anything. Cenarium (talk) 01:19, 22 March 2010 (UTC)
Actually, it is in MediaWiki:Tooltip-n-mainpage or MediaWiki:Tooltip-n-mainpage-description. Ucucha 02:16, 22 March 2010 (UTC)
I think he wants an alt text, not a title (tooltip) text for those who have disabled images. ManishEarthTalkStalk 02:52, 22 March 2010 (UTC)
And no, it isn't. Its at MessagesEn.php (search 'tooltip-p-logo'), changed in rev 55515. ManishEarthTalkStalk 02:57, 22 March 2010 (UTC)

(edit conflict)Well, the logos not exactly something which you can add an alt to. The image is the background of an a element, which cannot have an alt attribute (You can easily see that the image is not directly shown by right-clicking the logo. The menu which will appear will not have options like "Save image", "Copy Image address", etc). The logo does have a title (tooltip) attribute, though, which can be changed through js. Here's how the HTML of the logo looks:

<div id="p-logo">
<a style="background-image: url(http://upload.wikimedia.org/wikipedia/en/b/bc/Wiki.png);" href="/wiki/Main_Page" title="Visit the main page">
</a>
</div>


If you want to change the tooltip text, add this bit of js to Mediawiki:Common.js (you need an admin to do this):

addOnloadHook(function(){
document.getElementById()
document.getElementById('p-logo').childNodes[0].title="TOOLTIP TEXT HERE"
 
})
<BR/>

On a side (but related) note, the MW devs really should make the logo into an <img> in an <a> with alt text instead of an <a> with a background:

<div id="p-logo">
<a href="/wiki/Main_Page" title="Visit the main page" style="text-align:center">
<img src="http://upload.wikimedia.org/wikipedia/en/b/bc/Wiki.png" alt="Some alt text here..."/>
</a>
</div>
<!--Yes, the image gets a teeny bit misaligned, but it's too small to notice unless the code is changed dynamically 
(On my comp, the image jumps a pixel or two to the left and three pixels up)-->


I've always wondered why the logo and the person in the personal portlet are backgrounds and not imgs with alts (I noticed this when I was trying to fix an error in the logo and was trying to find the commons page. The error has been fixed in some of the other wikilogos at commons, but this logo). Hope the devs can answer why they have to use a roundabout method which doesn't allow alt text... ManishEarthTalkStalk 02:52, 22 March 2010 (UTC)

Using CCS background-image property allows people to use a different image in place of the official Wikipedia logo. I don't believe there's any way to change an img's src using CSS, so this is how they make it happen. Reach Out to the Truth 17:44, 22 March 2010 (UTC)
By 'people', you probably mean account holders, who can change src through js anyways. Anyways why would someone want to change the logo src for themselves? ManishEarthTalkStalk 03:10, 23 March 2010 (UTC)
Changing it through CSS would be easier than JavaScript. As for why people would do that, people are crazy. :P Maybe they don't like the logo, or would like to replace it with a free image. Who knows? Reach Out to the Truth 02:02, 28 March 2010 (UTC)

As noted, the logo cannot have alt text because it's a CSS background-image, not an <img>. I don't know why Gabriel Wicke chose to use a CSS background instead of an image here. It could be changed, I guess, but there doesn't seem to be any strong reason one way or another. If I had to give it alt text, I'd give it empty alt text anyway, because you already have "From Wikipedia, the free encyclopedia" on every page, and there's no point in asking screen readers to read it twice.

If it were an image, you could still change it in pure CSS by simply hiding the image and creating a new one with background-image. So that's not a good reason.

If you want to change the tooltip (not alt text), currently "Visit the main page", you don't need to use JS. The correct message to alter is MediaWiki:Tooltip-p-logo. —Aryeh Gregor (talk • contribs) 18:22, 22 March 2010 (UTC)

Like I said above, no it isn't (visit the page, its deleted). Its at MessagesEn.php (pagesearch 'tooltip-p-logo'), changed in rev 55515. This means that you need to be a sysadmin to change this inconsequential stuff. Easier with js. ManishEarthTalkStalk 03:10, 23 March 2010 (UTC)
That file contains the default messages. They can be customized on a per-site basis by editing pages in the MediaWiki namespace. Deleting a page from the MediaWiki namespace causes MediaWiki to use the default messages defined by the software and its extensions. MediaWiki:tooltip-p-logo was deleted, so it uses the default set in MessagesEn.php. That's how it works. Look at the text under the log snippet. That's the default message from MessagesEn.php. Reach Out to the Truth 03:33, 23 March 2010 (UTC)
To clarify, creating MediaWiki:Tooltip-p-logo will override the default settings in MessagesEn.php OrangeDog (τε) 13:13, 23 March 2010 (UTC)
Reach Out to the Truth and OrangeDog are correct. Creating the page will change the message's contents as actually used on the wiki. JS would be much more complicated. —Aryeh Gregor (talk • contribs) 18:03, 23 March 2010 (UTC)
Withdrawn in light of Wikipedia talk:Manual of Style#Alt text Gnevin (talk) 15:57, 28 March 2010 (UTC)

What's up with the geogrouptemplate?

Any ideas about what's going wrong with {{GeoGroupTemplate}}? In the last couple of days, any time I select "Map of all coordinates from Google", I get a "File not found at http://toolserver.org/~para...." message, while when I select the Bing option, I get a message of "This content is not available. It may have been deleted by the author." Selecting the other three options on the list, I get a "Can't download this page" popup for the GeoRSS option and the following error for KML:

The server encountered an internal error and is unable to complete your request at this time. If the problem persists, please contact the owner of the tool you are trying to use and inform them of this error, quoting the following information:
Request host: toolserver.org
Request path: GET /~para/cgi-bin/kmlexport
The owner of this tool is: para [at] toolserver [dot] org.

HTTP server at toolserver.org - ts-admins [at] toolserver [dot] org

Only the "map of all microformated coordinates" option is working; it brings up a proper map. Nyttend (talk) 13:49, 24 March 2010 (UTC)

It was due to toolserver software upgrades on that day. More details at Wikipedia talk:WikiProject Geographical coordinates#GeoGroupTemplate problem.3F. --Para (talk) 12:33, 28 March 2010 (UTC)

Thumbnail expand image button improve

I know that this should go to WP:VPR, but I wanted to get a technical opinion first. I'm proposing to tweak image thumbnails so that clicking on them or the "expand image" link will, instead of opening the image page at the File: namespace, will gray out the page and float an expanded version of the image on the page, and that image will link to the File: namespace. For a better explanation of what I mean, see here (click on the image). THis technology is already used for Google Buzz and you see it on many blogs, too. Is this feasable for WP? (I think it can be done through js...will try). ManishEarthTalkStalk 12:07, 25 March 2010 (UTC)

Uggh, I do hope you′d provide a way to disable it. It′s bad enough I had to write this script which disregards irrelevant links like Smiley.svg and restores direct access to the file descriptions in namespace six. ―AoV² 13:28, 25 March 2010 (UTC)
I once started working on such a Javascript gadget based on lightbox. But never really got it off the ground. I do think it is a good idea however. —TheDJ (talkcontribs) 14:52, 25 March 2010 (UTC)
Disabling shouldn't be too hard (prefs). Visitors click on thumbs to see a bigger version, not a description page (The description is usually captioned anyways). This will also allow the user to see a bigger version of images than shown on description pages. For example, on Colaba#Present, there's a panorama. A user will click it to get a bigger version, and will be disappointed to see the (relatively) tiny image at File:Colaba_panorama.jpg. The user will most probably overlook the "Full resolution" button. We should make thumbs open in a superimposed layer, with maximum image size, with some space for links to the File description page and the full resolution version. This will make viewing images easy for them, and they'll be wowed by the effect, too. ManishEarthTalkStalk 15:13, 25 March 2010 (UTC)
A way to disable it is a must. I'm not the only one who hates lightboxing. In the Wikipedia case, however, it should not be attempted at all unless you also can display the license information associated with the image. If you want to see some really nifty zooming stuff, go to the Commons and enable the ZoomViewer gadget. It's still experimental, but looks very promising. Lupo 16:04, 25 March 2010 (UTC)
I think this is a "reader" UI improvement, but, as an "editor" it would be a disimprovement. As an editor, clicking on the images take the user to the page where the image can be edited, its history can be viewed and its content talked about (which is good). As a reader, I recall form when I first visited Wikipedia, that is quite unexpected. In most cases, as a reader, having images that link anywhere is unexpected.
The question then is thus: is this a "reader" website or an "editor" website. There is an argument to say that this is an "editor" website (so leave things as they are). There is an argument too that says, in purist wiki philosophy, it is both a "reader" and an "editor" website - but in such philosophy, readers are editors, not the other way around (so leave things as they are). There is no argument to say that this is "reader" website, other websites do that (so leave things as they are).
That said, the file page is a visually awful. More could be done to make it prettier and more usable to readers (and editors).
Bear in mind too that the file page contains attribution information, which is necessary to link to in some circumstances for copyright reasons. A lightbox would need to contain that information in order to be copyright compliant in all circumstances. --RA (talk) 16:33, 25 March 2010 (UTC)
It would be very easy to add a link under the lightbox to the full description page, and even to take the author info from the description page (perhaps even the license with some clever parsing). And editors can be accommodated by using an alt or shift click action to the image that would take you immediately to the page description. —TheDJ (talkcontribs) 20:40, 25 March 2010 (UTC)
Not to mention that - for better or for worse - the vast, vast majority of visitors are most definitely not editors: they're readers. Generally, if something makes things better for readers (which I actually think this might do) to the (minimal) detriment to editors, it is A Good Thing (in my opinion, anyway). Ale_Jrbtalk 21:25, 25 March 2010 (UTC)

Perhaps you could make this happen only when one clicks the “enlarge” icon:

<a href="/wiki/File:Foobar.jpg" class="internal" title="Enlarge">
<img src="http://bits.wikimedia.org/skins-1.5/common/images/magnify-clip.png" alt="" height="11" width="15">
</a>

(by adding onClick="overwhelmWindow()" or whatever to the above). ―AoV² 21:41, 25 March 2010 (UTC)

Really however we should ensure that the target content of each link remains bookmarkable in some easy fashion. The dynamic loading mechanisms of many sites unfortunately do not conform to that. ―AoV² 21:49, 25 March 2010 (UTC)

Most people don't see the enlarge icon. Anyways, I do agree that image enlarging is pretty useless for editors, just as links to the image description page are useless to readers. But disabling it shouldn't be hard thru prefs. And we seem to already have the coding for making Web 2.0 windows, as is visible by the beta link dialog (I think they've used jQuery). I might start the coding in a week or so... ManishEarthTalkStalk 03:14, 26 March 2010 (UTC)
There's also User:Zocky/Picture Popups, but that doesn't work in IE6 so its not included as a gadget. — Dispenser 02:43, 26 March 2010 (UTC)

Obtaining image dimensions

Could someone tell me how to fetch the height and width of a full resolution image on WP? Thx, ManishEarthTalkStalk 03:13, 26 March 2010 (UTC)

Me too, that way I can simplify Template:Multiple image/doc#Horizontal placement: matching image heights. --Redrose64 (talk) 11:43, 26 March 2010 (UTC)
like this. Jus replace the image title with whatever you want. And whatever you do, remember that we have some 3000+ pixels wide images. You might want to put a limit on a request. —TheDJ (talkcontribs) 19:26, 26 March 2010 (UTC)
To me, that looks like some XML where somebody has manually filled in the dimensions. What I need is a method by which the domensions may be obtained automatically, so that a template can be incorporated directly into wikicode. Let's call them {{fileheight}} and {{filewidth}}, which might be used as follows:
{{multiple image
 | footer    = The two images are both rendered 100 pixels high
 | align     = right

 | image1    = Donkey 1 arp 750px.jpg
 | width1    = {{#expr: (100 * {{filewidth|File:Donkey 1 arp 750px.jpg}} / {{fileheight|File:Donkey 1 arp 750px.jpg}}) round 0}}
 | alt1      = A donkey
 | caption1  = A donkey, not square and originally 536 pixels high

 | image2    = Rainbow_trout.png
 | width2    = 100
 | alt2      = A trout
 | caption2  = A fish, square and originally 300 pixels high
}}
Given that File:Donkey 1 arp 750px.jpg is 750 × 536 pixels, this will perform the calculation {{#expr: (100 * 750 / 536) round 0}} which yields 140. --Redrose64 (talk) 12:38, 28 March 2010 (UTC)
I meant by AJAX. Sorry for not specifying :P ManishEarthTalkStalk 13:12, 28 March 2010 (UTC)
Thanks DJ. ManishEarthTalkStalk 13:13, 28 March 2010 (UTC)
constrained to 100px exactly.

Looks like you only want to constrain the height without regard to the width. That′s easy enough, just do something like [[Image:Foobar.fmt|9000x100px]] (where 9000 can be any number large enough to imply an aspect ratio greater than that of the original, see picture at right which renders at 32×100).

Now if you want to get into more esoteric things like making calculations based on the original size you′d need to fetch this information using javascript after the page loads (barring parser-functions/variables to expose it directly). Let me know if you actually want some code for the latter. ―AoV² 13:21, 28 March 2010 (UTC)

Naah, I'll do it myself when I get the time. And could you please read the message on your talk page. It'll prove quite useful. ManishEarthTalkStalk 13:25, 28 March 2010 (UTC)

Volume stuck with OGG

Please see Template talk:Listen#Increasing the width of this box and Template talk:Listen#Volume stuck with OGG. The problems are probably related but not identical. bugzilla:22538 was raised concerning the first, but could an expert check the second issue and maybe expand on the bug description or raise another one? This problem affects Firefox users and in my opinion this requires more than a "Normal enhancement" because it effectively disables the volume control, which I am sure was working before. -84user (talk) 04:36, 26 March 2010 (UTC)

I think this is a bug in Firefox. See also [8]. —TheDJ (talkcontribs) 12:11, 26 March 2010 (UTC)

<-- Firefox users, try adjusting the volume of this video!

I now think so too. I can reproduce the bug just by dragging a local file into a Firefox 3.6.2 window. Thank you for the mozilla bugzilla:533201 link, that describes the problem exactly. I added some screen recordings to Template talk:Listen#Volume stuck with OGG (not really the right place for it now ...). The problem also occurs with video elements if the video is short enough or narrow enough, for example see File:Check64x48bwrg16pal_25fps.avi.026v10.stereo3levclicks.ogv included above. -84user (talk) 00:47, 29 March 2010 (UTC)

Cascading protection

Is there a way to add cascading protection to a template, but not to protect its documentation subpage? For example, could noinclude sections be excluded from the protection? — Martin (MSGJ · talk) 11:29, 27 March 2010 (UTC)

Well, you can put the main contents in a separate template, which is cascadeprotected, then transclude that to the main template page which is protected (non-cascading). Thus the doc stays unprotected. ManishEarthTalkStalk 11:51, 27 March 2010 (UTC)
I have always thought that the cascade protection does not influence templates that are transcluded inside a <noinclude> block. See {{!}}. Ruslik_Zero 12:33, 27 March 2010 (UTC)
The page Template:! is not itself cascade protected. It is within the cascade of the main page, but the documentation is obviously not transcluded there. Martin is asking whether it is possible to apply cascading protection to a template page and yet exclude the documentation; the answer to which is no. Manishearth's suggestion is the only viable method. Happymelon 10:49, 28 March 2010 (UTC)
Thanks Manishearth, that might work well. — Martin (MSGJ · talk) 17:22, 28 March 2010 (UTC)

How to list subpages and subdirectories?

Is there any tool or other procedure to view the subdirectories and the subfiles rooted at a given WP page? For example, my page has a section containing a list of links to subpages that I created. How can I list all subpages, including pages other people may have created? David Spector 13:13, 27 March 2010 (UTC)

See Special:PrefixIndex. Ruslik_Zero 13:14, 27 March 2010 (UTC)
Indeed - can be transcluded as well: {{Special:PrefixIndex/User:David spector}}. –xenotalk 17:00, 27 March 2010 (UTC)

Thanks, it works! Gotta love WP: it has everything. David Spector 17:15, 27 March 2010 (UTC)

To list subpages
addOnloadHook( function () {
  addPortletLink("p-tb", wgServer+wgArticlePath.replace("$1", "Special:PrefixIndex/"+wgPageName+"/"), "Subpages", "t-subpages", "See all subpages of this page");
});
To delete subpages
  • Add tl|db-user to any of your pages to request deletion.

---— Gadget850 (Ed) talk 09:29, 28 March 2010 (UTC)

Sukhoi/HAL FGFA

Is Sukhoi/HAL FGFA (with a slash) correctly named? If not, what should it be? (There's an article at Sukhoi.) Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 22:32, 27 March 2010 (UTC)

Subpages are intentionally disabled in mainspace for this reason (note the lack of a link back to Sukhoi at the top). I don't see any issue with the name being the way it is. --Shirik (Questions or Comments?) 07:42, 28 March 2010 (UTC)
Depends only upon whether the sources use this name. Cf. Face/Off which is the proper name of a feature film (good) and Trentino-Alto Adige/Südtirol which is nobody′s proper name for anything (bad). ―AoV² 09:31, 28 March 2010 (UTC)

When I go to http://wikimediafoundation.org/wiki/Special:ContributionHistory/en it says at the top Mr. [name redacted] 00:00, 4 December 2010 USD 50.00 and that don't make no sense cuz it's 28 March and not 4 December!!!! —Preceding unsigned comment added by 76.230.211.69 (talk)

Barring a data-entry error, perhaps he deliberately paid using a post-dated check. ―AoV² 17:35, 28 March 2010 (UTC)

Strange redirect

For some reason Template:Image other (which is a redirect to Template:File other) does not redirect me when I visit the page. Other template redirects do not work like this so I'm a bit confused. — Martin (MSGJ · talk) 17:23, 28 March 2010 (UTC)

Odd. A null edit might fix it, if one is an admin. ―AoV² 17:29, 28 March 2010 (UTC)
Okay I just tried it, and it behaved even more strangely. After a null edit, it became like this:
  1. REDIRECT Template:File other
and then I purged it and it went back to its previous behaviour. — Martin (MSGJ · talk) 17:51, 28 March 2010 (UTC)
Try deleting and recreating the template from scratch? Gary King (talk) 21:17, 28 March 2010 (UTC)
I think I fixed it. I'm not sure why what I did worked, though. Ucucha 23:17, 28 March 2010 (UTC)
By changing the way you write REDIRECT, you probably triggered the system to put the page on the redirect table again. Guess the original addition was lost due to being delayed or something. —TheDJ (talkcontribs) 23:25, 28 March 2010 (UTC)

Proposal for creation of Village pump (development)

There is a proposal for a village pump development, please comment at the talk page. Cenarium (talk) 00:29, 29 March 2010 (UTC)

This is development in the general sense developing ideas and proposals in response to problems; not software/technical development. Cenarium (talk) 02:46, 29 March 2010 (UTC)

You have new messages issue

Has anyone else had a problem with their message notification thing not linking to their talk page? Kevin Rutherford (talk) 00:40, 29 March 2010 (UTC)

You sure it was the actual notification, and not a prank? ♫ Melodia Chaconne ♫ (talk) 02:06, 29 March 2010 (UTC)
It's the real thing. I've had it happen for the last few days at least, although I remember it happening a few times before that. I click on the links and nothing happens. For some reason, I want to think it is my being in college, but I don't know why it would be weird on this IP. Kevin Rutherford (talk) 02:10, 29 March 2010 (UTC)

So where is it linking to? ¦ Reisio (talk) 04:27, 29 March 2010 (UTC)

Nowhere. I click on it and it doesn't even open a new page. It's as though there is no link but blue text substituting the link in its place. Kevin Rutherford (talk) 14:38, 29 March 2010 (UTC)
I experienced it months ago but only on the Main page. Others at the time had the same problem on the main page which apparently contained something that could interfere with the link functionality in the new messages banner. The html source had link code but it didn't work for some users. I don't have problems currently. Do you have the problem on all pages? Have you recently changed something in your account? Which browser do you use? I have Firefox. PrimeHunter (talk) 23:46, 29 March 2010 (UTC)
Actually it only appears on the main page. I'm using Internet Explorer with Windows 7. Kevin Rutherford (talk) 01:34, 30 March 2010 (UTC)

Proxy-Connection: keep-alive

A user reported that an article contained garbled text - some content had been replaced with "Proxy-Connection: keep-alive" and "Cache-Control: max-age=0".

I searched, and found it on a few other articles; [9] [10] [11] [12]

I wanted to mention it here, in case it might help pin down a bug or something.

It was previously mentioned at VPT here, but no result, as far as I can see.

Cheers,  Chzz  ►  00:42, 29 March 2010 (UTC)

P.S. User:Ap exhibit design/AldricPears Associates#External links looks like a severe case.  Chzz  ►  00:56, 29 March 2010 (UTC)
I've seen this many times while watching recent changes, Chzz. Initially, I suspected some sort of vandalism afoot. An IP I queried stated that it was a glitch with Firefox. Whether they were correct or just having me on, I don't know. In any event I've become wary of completely trusting what my browser is displaying. Regards Tiderolls 04:57, 29 March 2010 (UTC)
Looks like a broken proxy. Cacycle (talk) 20:02, 29 March 2010 (UTC)
If it happens often enough, there might be a case for creating an edit filter to tag/warn/block (or any combination thereof) such edits, which would be easy if the string in question is uniform enough. Opinions? {{Nihiltres|talk|edits|⚡}} 20:16, 29 March 2010 (UTC)
You can see this many places throughout the Internet: [13]. I don't think it's vandalism, but rather a web browser or proxy server issue. I don't know why it's happening though, but indeed, if it is problematic enough, an edit filter could disallow the edit, asking the user to go back and try to save the edit again. PleaseStand (talk) 23:52, 29 March 2010 (UTC)

Extra Space

Thereès extra space on Port_Adelaide_Football_Club. Can someone fix it.174.3.113.245 (talk) 05:01, 29 March 2010 (UTC)

I took a look and didn't notice anything. Can you be more specific? Tiderolls 05:07, 29 March 2010 (UTC)
There's an extra vertical space between the dab text and the intro text... some problem with the use or implementation of {{infobox australian football club}}. ¦ Reisio (talk) 06:22, 29 March 2010 (UTC)

I believe this is a side-effect of mixing html table code with wikitable code. I rewrote it to use html exclusively, as debugging the way MW parses the latter is simpler and easier. I couldn′t tell you where the problem was, only that I′ve fixed it. ―AoV² 13:33, 29 March 2010 (UTC)

I too have found some glitches mixing the two in {{NMI list item}}, and also found that I could fix it by using raw HTML. --BrownHairedGirl (talk) • (contribs) 14:01, 29 March 2010 (UTC)
You don't need to do this using html. The problem with these kinds of things is always the optional rows in the table. You need to be sure that all optional |- of the rows are within the if statements. Otherwise they cause such problems. —TheDJ (talkcontribs) 14:12, 29 March 2010 (UTC)
See diff. The primary problem was the |- after {{{founded}}}. —TheDJ (talkcontribs) 14:31, 29 March 2010 (UTC)

That′s actually more confusing as the {{!}}- represents both the closing </tr> and the opening <tr> element for the following row. It also multiplies the number of curly braces, template dependencies, and mandatory line-breaks unnecessarily. ―AoV² 14:42, 29 March 2010 (UTC)

Technically, it only presents the presence of a new row. MediaWiki itself decides where rows end, which is part of why wikitable markup is so deficient for advanced tables. Personally, i'm against using html code where it isn't absolutely required however. Mixing the two markup languages should be avoided wherever possible. (for reasons of reuse mostly). —TheDJ (talkcontribs) 16:21, 29 March 2010 (UTC)

"prototype" server problem?

Wikipedia is incredibly slow for me today. Firefox tells me its waiting for "prototype.wikimedia.org". Is there a server problem at the moment? Thanks. – ukexpat (talk) 16:21, 29 March 2010 (UTC)

What is it trying to retrieve from there ? Because that server is not supposed to be used by vanilla wikipedia. —TheDJ (talkcontribs) 16:22, 29 March 2010 (UTC)
I have absolutely no idea. I thought it may be related to WikEd, but disabling it doesn't improve things. It can take several minutes for the browser to get whatever its looking for from prototype. Maybe one of the other scripts I have installed interfaces with that server? – ukexpat (talk) 17:07, 29 March 2010 (UTC)
Check your error console for errors, if something crashes you might simply see the last status message before the crash. Cacycle (talk) 19:52, 29 March 2010 (UTC)
You load an awful lot of Javascript. Probably one of those tools is responsible. My Safari has an "Activity" window, which shows which elements of a page timeout. Most other browsers have something similar. It would help you track which file is trying to get, which can likely tell you which javascript extension is doing this. —TheDJ (talkcontribs) 22:50, 29 March 2010 (UTC)

I've just started having the same issue in the past few minutes - I'm using Firefox 3.6.2, I'm definitely not using the beta functionality, but I do have quite a bit of custom JS. I'm getting delayed enough to trigger the dreaded cascading edit conflict when I try to edit a busy page. Gavia immer (talk) 23:10, 29 March 2010 (UTC)

Default cursor location

The default cursor location when a Wikipedia page is displayed should be in the search box, yet it is not, which is puzzling, given that locating the cursor in the most likely input field is standard for websites. —Preceding unsigned comment added by 24.6.189.169 (talk) 19:09, 29 March 2010 (UTC)

See the FAQ entry at the top of the page. Happymelon 20:19, 29 March 2010 (UTC)

Template:Infobox Buddhist temple

Hi. I tried to add apushpinmap and coord section in like Wat Phra Kaew but it has messed up the alignment. I want it to be aligned like Sera Monastery. Can somebody please tweak the template to prrevent misalignment when a map is added? Dr. Blofeld White cat 19:48, 29 March 2010 (UTC)

Done. The pushpin map was placed in the first column, because it wasn't specified that it should span 2 columns. There is still a problem with coordinates information in that infobox however. It is listed 2 times in different ways, that might not be handy. —TheDJ (talkcontribs) 23:15, 29 March 2010 (UTC)

Page listed on "what links here" ... doesn't

Resolved: Intelligentsium 22:52, 29 March 2010 (UTC)

[WITHDRAWN] - The answer was: a link from a template which is collapsed by default, when transcluded into an article, will not show up either when searching the article nor in the Wiki source when the article is edited; yet it will show up in "what links here" for the target article. Wnt (talk) 21:56, 29 March 2010 (UTC)

Yep, Barack Obama transcludes navbox Template:Barack Obama which has a link to Bill Ayers presidential election controversy. Special:ExpandTemplates can help to track down transcluded links by transcluding the article itself, in this case using {{:Barack Obama}} as the input text (the colon ensures that the article is expanded rather than just the template of the same name). Alternatively, a collapsed link can be found by searching the HTML source code when the article is displayed in your browser. — Richardguk (talk) 22:12, 29 March 2010 (UTC)

CSS help needed for en.wikinews

Hi all. Can someone please do me the honors of adjusting my custom css at en.wikinews.org, so that LQT doesn't show any background color, e.g. White only, and no box around comments or messges please?

The current layout is ugly as hell, and I just need something which looks like it sorta used to, and i can't do that :)

Cheers BarkingFish Talk to me | My contributions 00:28, 30 March 2010 (UTC)

All of the backgrounds look white to me (at least for the LiquidThreads pieces, the Vector sidebar is a dark grey). For the borders and padding, add the following code to your .css subpage. If you use Monobook, it would be something like n:User:MZMcBride/monobook.css:
.lqt-post-wrapper {
  padding: 0 !important;
  border: none !important;
}
Then bypass your local cache and you'll be good to go. --MZMcBride (talk) 00:46, 30 March 2010 (UTC)

"Leave a message" template

I have just created {{subst:lam}} to save typing when leaving messages on new users' talk pages. Is there any way I could improve it? PleaseStand (talk) 03:07, 30 March 2010 (UTC)

You could have it include the whole sentence. The beginning is mostly fluff unless it includes a negation. "To reply, please ...", "Feel free to ..." - the meaning is basically independent of the beginning of the sentence. Mr.Z-man 03:42, 30 March 2010 (UTC)
To quote from my documentation: "This template is for use within talk page notices that are not template-generated [emphasis added]". Defeating the purpose by including an entire sentence? Now that would be ironic. PleaseStand (talk) 05:03, 30 March 2010 (UTC)
I presumed it was something that you would add at the end of a longer comment, not something where that would be the only thing that you're saying. Mr.Z-man 05:11, 30 March 2010 (UTC)
The main purpose is that the rest of the sentence can be varied. For example, if the user has tried leaving a message on your user page rather than your user talk page, you can write: "In the future, please {{subst:lam}} rather than my user page." If you prefer the ability to easily enter the entire sentence, can you come up with a good wording? You could use another template name such as {{lame}} (for Leave a Message, Extended). PleaseStand (talk) 05:44, 30 March 2010 (UTC)

Clearing undo history in Firefox

I have written a user script that allows for editing articles that suffer from citation template bloat (for lack of a better name). Unfortunately, I have found that it is possible to undo its changes improperly using the browser's undo function. That's not good for the user experience. If anyone here can offer any input on how to "clear" the previous undo history of a textbox (or otherwise resolve the issue), that would be helpful. Leave messages on User talk:PleaseStand/References segregator. PleaseStand (talk) 06:11, 30 March 2010 (UTC)

Global file usage

Do we need all those global file usage links at the bottom of commons-hosted files ? This seems like a lot of irrelevant links for readers and most users, and also makes file pages unnecessarily long, while many are already quite long and for navigability images should be as quick as possible to load - then close. The commons description page is linked anyway and we can add a link to Special:Globalusage in MediaWiki:Linkstoimage, which is certainly good enough for curious users. Cenarium (talk) 23:34, 26 March 2010 (UTC)

Should we file a bug, or not ? Cenarium (talk) 16:33, 27 March 2010 (UTC)
Can't say that I really see any problem. From what I can tell, it doesn't add significantly to the loading time of the page, since there's a cap on how many instances it shows; see File:Wikipedia-logo.png (very likely the most heavily used image in the WMF system) for an example. As for it making the page longer... so what? Aside from the fact that you can just hit the "end" key to get to the very bottom of any page, the only thing below Global Usage is the Metadata and Categories; the latter doesn't (usually) exist for Commons-only images. The Metadata section being displayed above the "Global file usage" and "File links" sections wouldn't be a bad idea, though... EVula // talk // // 16:48, 27 March 2010 (UTC)
This is one of the things that the multimedia usability project might look at at some point. There have been ideas to move all history and maintenance stuff like this into the page history file for instance. But it will be a while (if ever) before guillom gets around to that. There are more important issues to solve before he focuses on this. —TheDJ (talkcontribs) 00:12, 28 March 2010 (UTC)
Yes, I think the file pages need an overhaul. Readers looking at an image are... just looking for the image. They're some important additional information like copyright and co but the history and internal links sometimes take a lot of place, more than they should imo (so more capping needed for example), and the global file usage links are the last straw. It can increase load for users with old/weak computers, connexions or browsers, or with mobile phones, and image pages are the prototype of pages that are opened and closed quickly, so we should make this easy. That's usability and such so mostly in the hands of the WMF, but I don't see how those global usage links can be useful to us. Cenarium (talk) 22:58, 28 March 2010 (UTC)
There is a mention of moving it into WhatLinksHere, in Usability:File:GPaumier multimedia usability draft mock-ups Oct09.pdf. I disagree as image embedding and image description linking/transcluding are two different things. — Dispenser 04:06, 31 March 2010 (UTC)
I create a Gadget that allows you to 'show/hide' these elements. importScript('User:TheDJ/usagecollapse.js');TheDJ (talkcontribs) 14:09, 30 March 2010 (UTC)

Allowing users to see where they've logged in from

I think that users should be able to see the IP's that they've recently logged in as. Gmail has it image here. This will enable users to see where they've logged in from in the past and make sure that their account hasn't been hacked. Additionally, Gmail introduced an automatic warning system which warns you if your account was accesses from the other side of the globe, for users who don't check the IP table continuously More info here I don't think it should be hard to implement, a user can be allowed to checkuser himself (Maybe in a limited way, only see last month's IPs). Comments or thoughts? ManishEarthTalkStalk 01:44, 27 March 2010 (UTC)

Please see wmf:Privacy policyxenotalk 02:32, 27 March 2010 (UTC)
But privacy to oneself? Thats absurd. You and only you log in to your account, so its OK to make your data available to you. ManishEarthTalkStalk 04:22, 27 March 2010 (UTC)
I would prefer that WMF keep as little information as possible. And if your statement "You and only you log in to your account" is true, then your proposal is unnecessary. Per Z-Man below, this solution is in search of a problem. –xenotalk 16:54, 27 March 2010 (UTC)
I think that's a good idea, but still, if someone hacked into another person's account, that person's IP would be known (for example, if Y hacked into X's account, Y can find out what X's IP is). --Hadger 04:27, 27 March 2010 (UTC)
Then how 'bout we just calculate if the acct was accessed from far away. If the system detects that X logged in from Chicago, and then half an hour later X logs in from London, the system can block X, and tell him to check his email to unblock himself. In his email will be an unblock link. Anyways, if X's account is hacked, he will probably have greater things to worry about... This can also be extended to inactive accounts. If an account is inactive for more than a year, it will be blocked and the user can unblock it thru email. ManishEarthTalkStalk 04:47, 27 March 2010 (UTC)
The hacker might be your next door neighbour, or your work colleague, or your employee. There is nothing to guarantee that a hacker is going to be far away from you. HumphreyW (talk) 04:52, 27 March 2010 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Something's better than nothing... At the moment we have no protection. Protecting yourself from the 5 billion people who don't live near you is better than not being protected at all. Just hope that the remaining one billion don't try to hack you. ManishEarthTalkStalk 04:57, 27 March 2010 (UTC)

It would be even better to emailnotifyandblock a user regardless of the time difference. For example, If I edit from a different country, even if I do so after a week, I should be emailnotifiedandblocked. Frequent fliers should be given an option to turn this off except where the time difference is too little (logging in from different hemispheres within half an hour or so). ManishEarthTalkStalk 05:01, 27 March 2010 (UTC)
No thanks. Using proxies would then get you blocked. Some companies have strange routing setups and you can suddenly be using an IP from thousands of kilometres away without knowing it. And for people that don't register an email address there would be no way to unblock oneself. HumphreyW (talk) 05:17, 27 March 2010 (UTC)
Reply below ManishEarthTalkStalk 05:47, 27 March 2010 (UTC)

I've thought of this before. It's a good idea though. Maybe it could be setup on the toolserver and users log in to see their IPs, thus protecting them from others seeing their IPs. Kevin Rutherford (talk) 05:17, 27 March 2010 (UTC)

It could only protect other from seeing if the toolserver uses SSL connections. HumphreyW (talk) 05:20, 27 March 2010 (UTC)

Sounds like a solution in search of a problem to me. Mr.Z-man 05:28, 27 March 2010 (UTC)

Better to be safe than sorry... ManishEarthTalkStalk 05:31, 27 March 2010 (UTC)
So because I have no protection from ninjas rappelling from my roof and breaking through my 8th floor window, I should run out and put steel bars over my window just in case? Wikipedia accounts almost never get "hacked" because 99.94% of accounts have no special access that a hacker couldn't get just by creating their own account. There is virtually no personal information that can be stolen by hacking into a normal (non-admin) account. No evidence has yet been presented that A) Wikipedia accounts getting hacked is a serious problem and B) The current system for dealing with it (having a checkuser get the information) is failing somehow. Mr.Z-man 16:41, 27 March 2010 (UTC)

(edit conflict) Good idea. That way a hacker won't come to know your IP unless he logs into toolserver (which he probably won't know about). The tool should have a strict "3 logins per day" thing, though, after which you have to login thru a link in your email. The toolserver should also have a separate password. What about the autoblockandemailunblocklink proposal? ManishEarthTalkStalk 05:31, 27 March 2010 (UTC)

"autoblockandemailunblocklink" is a a bad idea. See my response above. HumphreyW (talk) 05:40, 27 March 2010 (UTC)
Oh I didn't see that. Keep it disabled by default. Let users with email enable it (I would), with some options like: "Would you like to autoblock if its accessed from somewhere else, or just recieve a notification", and "I'm changing my location, please do not bother me for the next x days" ManishEarthTalkStalk 05:45, 27 March 2010 (UTC)
...you lost me at "hackers logging into the toolserver". "lolwut?" was pretty much my reaction. I believe Mr.Z-man had it right: this is a solution in search of a problem. EVula // talk // // 16:55, 27 March 2010 (UTC)

Good Idea. I have seen this on a number of commercial sites. I guess wikipedia members doesn't care about security. There are a number of ways to scramble the IP and for privacy scrambling is a good idea so you don't see the IP but you can see if previous IP's are different to your current one. Regards, SunCreator (talk) 16:46, 28 March 2010 (UTC)

Correct, I would rather people spend time and/or money on useful projects than on securing something that in more than 99.9% of cases is under no threat of hacking. Scrambling the IP is pretty much just a token privacy measure. MediaWiki is open source software, so whatever scrambling algorithm is used would be public knowledge. You wouldn't be able to use a very good scrambling method either, or else it would be completely useless to people with a dynamic IP (i.e. almost everyone) as even a small variation in the IP would result in a completely different result. Mr.Z-man 04:25, 30 March 2010 (UTC)
Then how 'bout we just report the locations (country, maybe state), and notify if the location has changed. That shouldn't be hard, and it doesn't violate privacy. ManishEarthTalkStalk 04:33, 30 March 2010 (UTC)

I think the point people are trying to make (though not in so many words) is that the privacy policy is a double-edged sword; it protects your information from the hacker, and protects the hacker′s information from you. This is a sensible default because in a scenario like this where Alice passing Bob′s log-in screen implies that the server has exhausted available means to distinguish her from Bob. Restricting users to IPs “similar” to those from which they registered or last edited only introduces greater DoS and LTDR problems. ―AoV² 17:25, 28 March 2010 (UTC)

Wait... did someone use Toolserver and log in and password and storing IPs in the same sentence? Err...? Sorry guys. — The Earwig (talk) 04:01, 30 March 2010 (UTC)

Can we please close/archive this pointless discussion. Its not going to happen, for all the reasons above and more. It doesn't prevent accounts being hacked, which doesn't happen that often anyway, it will probably increase privacy issues, will require time spent building it which could be spent on much more useful things, and besides a major thing like this would probably require the foundations approval too, which isn't likely considering barely anybody here supports it. Go back to editing and stop wasting time here--Jac16888Talk 04:39, 30 March 2010 (UTC)

I think the general rule on village pumps is that discussions are not 'closed' or 'archived' (prematurely) they are just left to die a natural death and get archived by Misza. So, just stop responding =) Even if a bugzilla is filed, a developer will likely very swiftly close it as "WONTFIX". But I agree that further discussion of this proposal is just wasted breath. –xenotalk 13:36, 30 March 2010 (UTC)

SeaMonkey and images not displaying

Resolved

I am quite sure nothing got changed on my end, but suddenly (few days ago or so) my SeaMonkey 1.1.17 stopped displaying images (screenshot of the same problem). My IE and other browsers are fine, but I have this browser customized for my Wikipedia operations (updating the browser would result in many broken extensions, particularly for the new SM 2.0). Any ideas what I could try to get the images back other then updating (I am assuming something changed on the Wikimedia side)? A little more info: Wikipedia beta skin has the same issue; I cannot see images on any other WMF projects, but other MediaWikis I tried look fine, with the exception of Wikia front page (but it subwikis look fine). I tried refreshing the cache, deleting cookies, disabling the firewall, restarting the browser and the computer, to no effect. Suggestions welcome :) --Piotr Konieczny aka Prokonsul Piotrus| talk 16:49, 29 March 2010 (UTC)

It seems the browser has lost upload.wikimedia.org. This might be because it is using a DNS cache that is confused due to the problems from last week. I don't know how to solve this. Perhaps try to find out where SM stores all its caching information and delete it manually ? —TheDJ (talkcontribs) 16:53, 29 March 2010 (UTC)
My problem does seem to be connected to the DNS issues earlier, but I can access and view pictures like this one. I tried cleaning the cache and flushing my system cached DNS, so far this is not helping. PS. Updating SM to the newest 1.x build, the 1.1.19, did not help, neither. --Piotr Konieczny aka Prokonsul Piotrus| talk 19:32, 29 March 2010 (UTC)
Update: resolved. I traced the problem to upload.wikimedia.org being blacklisted somehow in Preferences->Images->Mange Image Permissions. --Piotr Konieczny aka Prokonsul Piotrus| talk 15:16, 30 March 2010 (UTC)

Disruptive usernames in the search box

Screenshot of examples

Hope this is the correct place... Is it possible to stop accounts that are blocked for violating the username policy from popping up in the search box? I don't think these type of promotional and offensive usernames should be seen when trying to search for articles. Thanks, Pyrrhus16 23:26, 29 March 2010 (UTC)

Get the page deleted or change your search options in Special:Preferences to not include the User: and User talk: namespaces. I know of no better option; both noindexed pages (e.g. User:PleaseStand/Sandbox) and redirects (e.g. User:PleaseStand/subsearch) do show up in the list. Note that the default is only to include articles in the AJAX search box. PleaseStand (talk) 00:05, 30 March 2010 (UTC)
Thanks for the response. I've changed the search options in My Preferences. Pyrrhus16 12:43, 30 March 2010 (UTC)

Why does it show user-pages when one has begun typing something other than “User:”? I miss the old behavior where results were both in alphabetical order and identical to the listing in Special:Prefixindex. Is that version still available somewhere as a gadget, or ought I write my own? ―AoV² 03:56, 30 March 2010 (UTC)

Because the user has changed the namespaces he wants to search by default in his preferences. —TheDJ (talkcontribs) 11:50, 30 March 2010 (UTC)

Is there a watchlist tool?

Is there a tool to show all my recent diffs for watched pages?

For example, suppose I edit an article; call this edit A. Then other editors make edits B then C to the same article. In my watchlist, the only change shown is C. Is there a tool that would show the intervening edit B in addition to C, both as their diffs? This would make it easy to see what changes have happened since I last ran the tool. David Spector (talk) 18:40, 30 March 2010 (UTC)

Go to your Preferences -> Watchlist -> "Expand watchlist to show all changes, not just the most recent". Gary King (talk) 18:44, 30 March 2010 (UTC)

Make it easier to submit edit requests

Currently, MediaWiki:Protectedpagetext (shown when users try to edit a page they don't have rights to) provides instruction on how to make an edit request for a (semi)protected page. Why not make it a bit easier and just give visitors a prominent button to push? We can use an InputBox to preload the necessary {{editsemiprotected}} / {{editprotected}} text, and provide an explanatory editintro too. The example below, using preload/editintro borrowed from the Article Wizard, shows how.


Good idea? It might encourage frivolous requests, but we'd only know the signal/noise ratio by trying.

On a related note, I think it would be a lot friendlier if (a) the Edit button didn't become View Source when an editor doesn't have permission to edit - it's a missed opportunity to encourage people to find out more about editing and especially how signing up can have benefits and enable editing of semiprotected pages. (b) I don't really see that the average unregistered users actually wants to see the source wikitext; this is probably just confusing. Ditch it and use the space for a bigger, friendlier explanation of protection. Provide a link to the source for those few who really want it. Rd232 talk 21:25, 21 March 2010 (UTC)

Strong support. I personally think this is brilliant! Thumbs up! Although, oppose the second idea about the view source button, which I have found useful many times. --JokerXtreme (talk) 21:44, 21 March 2010 (UTC)
Strong oppose the removal of the "view source" tab. Two reasons:
  • Sometimes, when explaining things to IP users, I have purposefully directed them to a given template, and suggested that they "view source", in order to see how something is done
  • I often use "view source" on protected templates to see what the actual parameters are, and what happens when certain combinations are used. The editors who maintain templates often forget to keep the documentation in synch. As a non-admin, if this were implemented, I would lose the ability to check why my template transclusions don't work as per the documentation.
--Redrose64 (talk) 22:37, 21 March 2010 (UTC)
If I understand it correctly, what he proposes isn't that you won't be able to look at the source code, just that it would require one more click. Svick (talk) 23:05, 21 March 2010 (UTC)
Correct. I'm not suggesting removing the ability to see the source, I'm suggesting hiding the source, because for most people it's a confusing distraction at their first point of entry. It could be collapsed on the same page (hidden by default), or there could be a link to the source on a separate page. Rd232 talk 23:07, 21 March 2010 (UTC)
Also, the argument applies mainly for articles. For protected templates, people who come along probably do actually want to see the source. Rd232 talk 23:18, 21 March 2010 (UTC)
Oppose part about hiding “View source”. I find it very useful, especially for protected templates. Also, the editnotice on semi-protected articles already explains that creating an account helps. Svick (talk) 23:05, 21 March 2010 (UTC)

Comment - Well I think it would be useful if it was easy to propose, for instance, a minor change on a protected article, without going through all the fuzz of creating a section on the talk page. --JokerXtreme (talk) 23:15, 21 March 2010 (UTC)

  • Strong adding an input box below the current text of MediaWiki:Protectedpagetext. Although users wishing to edit probably don't often get there in the first place, this surely happens from time to time and we should make it easier for them to request edits. Note that {{TALKPAGENAME}} should be used to make it work in general. Neutral on the rest; I agree with the general idea to make it easier to request edits on protected pages, maybe we should change the text which appears when pointing on view source from "you can view its source" to "you can view its source and request edits", though I think we'd need to ask devs for this. Cenarium (talk) 00:27, 22 March 2010 (UTC)
    Noting that the editprotected/editsemiprotected template should be contained in the preload. Cenarium (talk) 01:16, 22 March 2010 (UTC)
  • Strong support. adding the inputbox. It will help new editors unfamiliar with our policies to submit requests. ManishEarthTalkStalk (continued below)
  • Oppose. "view source" suggestion. "View source" is a useful button, and anyways, an explanation is given on the view source page. ManishEarthTalkStalk 02:32, 22 March 2010 (UTC)
    • I'm not talking about getting rid of the button, I'm suggesting making the button always say "Edit this page", to make it more inviting for people to click on and get an explanation that yes, they really can edit (but not this page right now). In addition, I'd try and make the explanation friendlier and more welcoming, and part of that might be not showing scary source text which you can't even edit: that seem offputting to me. In principle, there could be another Inputbox which uses the current article text to feed a sandbox, so that people can test editing (with a sufficiently clear explanation, this might work). Rd232 talk 08:06, 22 March 2010 (UTC)
      • I knew that it wasn't removing the button. Anyways, we have a nice header explaining why it can't be edited. ManishEarthTalkStalk 10:14, 22 March 2010 (UTC)
        • The MediaWiki:Protectedpagetext message isn't particularly nice for newbies (I might draft a redesign if I have time); and semi-protected pages often only have the small lock icon, so there's no nice message on the page itself. And "View Source" is not exactly reinforcing the "Anyone Can Edit" message. Rd232 talk 10:19, 22 March 2010 (UTC)
            • We already have a tooltip on it which is quite helpful and it pops up immediately ManishEarthTalkStalk 09:23, 23 March 2010 (UTC)
              • Yes, it pops up when you put the mouse over the button (also the tooltip just says "you can view the source" ...). What proportion of people do that? Why would they? The issue is sending a message that's right there. We do this as a matter of course with unprotected pages - I don't see we avoid doing this with protected pages. Rd232 talk 11:22, 23 March 2010 (UTC)
          • See also #Proposed implementation design by Rd232
  • Endorse the inputbox for submit an edit request - great idea. Far too often IPs suggest edits without using the editprotected template, and are thus disenfranchised. –xenotalk 17:54, 22 March 2010 (UTC)
  • Support, this could also work where an article is fully protected too. Mjroots (talk) 19:20, 22 March 2010 (UTC)
  • Support, a great idea. Can't imagine any drawbacks to implementing the make-a-request idea, whether for semiprotected pages or for fully protected pages. I have no opinion on the question of hiding the source-text link. Nyttend (talk) 23:35, 22 March 2010 (UTC)
  • Support, love the idea as well. Lets stick to the editform first, we can deal with the issue of the "view source" tab in a separate discussion. —TheDJ (talkcontribs) 23:56, 22 March 2010 (UTC)
  • Strong support. And I make a mental note to give a hand on the design as soon as I have some spare time. Also, the source code should be in a collapsible box. It has to stay efficient and easy to use for us power users. But beginners should notice it only when they need it. It's a good compromise. Dodoïste (talk) 01:49, 23 March 2010 (UTC)
    • That's a good idea. Can it be done without changing the software? Rd232 talk 11:22, 23 March 2010 (UTC)
  • Ditto what Nyttend said. Seraphim 02:15, 23 March 2010 (UTC)
  • Neutral - I might be interested in seeing how a temporary test implementation goes, because I'm worried that making it this easy will cause lots of trivial requests. Currently only people who are seriously concerned will go through the trouble of reading the instructions and making a request, and I think for protected pages only significant concerns should really warrant the attention and effort of an admin. So the commitment required to make a request, so to speak, was always sort of a built-in check against trivial requests. But with this button, anyone who wants to make any edit might use the button since it's so easy. So I'm not sure if this would turn out to be a good thing. Equazcion (talk) 22:33, 23 Mar 2010 (UTC)

Proposed implementation design by Rd232

Please don't vote here on the idea in general, this is about collaborating on a specific design

OK, so User:Rd232/protectedpagetext has a draft redesign, and includes a working preload/editintro for editsemiprotected. Rd232 talk 17:27, 22 March 2010 (UTC)

Looks nice, but what about fully protected pages? --JokerXtreme (talk) 17:32, 22 March 2010 (UTC)
That's not a problem. The same principles apply, I suggest we design them 1 at a time. —TheDJ (talkcontribs) 00:11, 23 March 2010 (UTC)
Probably a good idea, but I've already drafted that too, at User:Rd232/protectedpagetext/sub1. Rd232 talk 01:03, 23 March 2010 (UTC)
  • The thing is a tad tall this way. I suggest a similar layout as the Help desk. Just 2 columns with text, and then a new row with a request button. 80% of the people don't read that other text anyways. —TheDJ (talkcontribs) 00:11, 23 March 2010 (UTC)
    I don't quite get that. Can you draft it? Rd232 talk 01:03, 23 March 2010 (UTC)

I've improved the design a bit, not least using the trick from {{Feedback page}} where the inputbox parameters are plugged into a fullurl link, thus avoiding the format-scrambling inputbox. Rd232 talk 12:07, 23 March 2010 (UTC)

The format is much better, but it still would be nice to have a button. The edit proposal option now gets kinda lost in there. --JokerXtreme (talk) 12:50, 23 March 2010 (UTC)
How about an arrow icon (User:Rd232/protectedpagetext/sub1)? Or change things around more dramatically. Rd232 talk 13:47, 23 March 2010 (UTC)
Hm, I don't think an arrow there makes much sense. How about "you can submit an edit request by clicking the button on the bottom" or something? So that the more experienced users can find the button right away. --JokerXtreme (talk) 14:04, 23 March 2010 (UTC)
Why doesn't an arrow make sense there? It highlights the key action link. Your suggestion I think could be confusing. More showing and less telling would be helpful in this discussion - let's see what we're talking about; just edit my draft or create your own. Rd232 talk 14:15, 23 March 2010 (UTC)
Something like that:User:Rd232/protectedpagetext/sub2. The text may need rewriting though, but the layout should look something like this (I think). --JokerXtreme (talk) 14:51, 23 March 2010 (UTC)
I guess that's OK, but it slightly upsets the priority (unprotection should come last, I think). But it it's an improvement on the status quo either way, and it can always be revised later. What do people think? Rd232 talk 17:33, 23 March 2010 (UTC)
I like JokerXtreme's suggestion. —TheDJ (talkcontribs) 19:04, 23 March 2010 (UTC)
Agree, that one is probably my favorite out of both proposals. I have a feeling the button will "click" better with new users who are trying to figure out how to edit the page. I really don't think it upsets the priority either, because it's separate from everything else - i.e., it's not a bullet point. That's how I see it. — The Earwig (talk) 20:10, 23 March 2010 (UTC)
Fair enough. Let's give it a bit little longer for comments (24hrs?), and then somebody implement it please. Rd232 talk 22:16, 23 March 2010 (UTC)
I like it too. :) It's much better than the current one imo, well done people. Ale_Jrbtalk 22:25, 23 March 2010 (UTC)
Rd, what do you mean implement it? I thought it was already finished. It just needs the code from your "edit request" link. The templates probably need to change too. --JokerXtreme (talk) 22:51, 23 March 2010 (UTC)
I assume he means putting it here, so it actually shows up. :) Ale_Jrbtalk 23:00, 23 March 2010 (UTC)
Aaah, I see! Well, Rd, I guess you must have the honor :) Are those ok btw? {{editsemiprotected}}, {{editprotected}} Or do they need any change? --JokerXtreme (talk) 23:08, 23 March 2010 (UTC)
OK, I'll do it then, probably tomorrow morning (probably better anyway that I do it since I split the code across pages to make editing easier, and it needs reintegrating). I don't see that those templates need any changes. Rd232 talk 09:07, 24 March 2010 (UTC)

YesY Done And implemented - now live. Rd232 talk 11:34, 25 March 2010 (UTC)

and I see that it's already attracting improper requests --Redrose64 (talk) 00:12, 26 March 2010 (UTC)
It is too early to say, but I think the proper requests outweigh the improper. --JokerXtreme (talk) 08:20, 26 March 2010 (UTC)
Define "improper". It appears to be a good faith request relating to John_George_Adair#Adair.E2.80.99s_death, but in the wrong place (death template). Rd232 talk 09:41, 26 March 2010 (UTC)

Can something be done to encourage editors to use a more descriptive section heading rather than the generic "edit request"? — Martin (MSGJ · talk) 10:17, 26 March 2010 (UTC)

I guess that could be useful. Maybe it should be added in the tmbox here: [14]. I couldn't find the template for it. But, I think the heading should start with "Edit request -". --JokerXtreme (talk) 10:58, 26 March 2010 (UTC)
And it should be possible to add the date to the default heading so that if people don't type in their own heading, as least you don't get several identical headings on a page which stops the table of contents working. — Martin (MSGJ · talk) 12:48, 26 March 2010 (UTC)
The editintro templates are Template:Editsemiprotected/editintro and Template:Editprotected/editintro. — Martin (MSGJ · talk) 12:50, 26 March 2010 (UTC)
Sounds good. Headings in an "Edit request - (time, date) - " preform, that users will fill in. OK, we need to add a third advice and probably place it first. Something like "Please use a descriptive headline. Fill in further details about the subject of the edit request.". Just throwing this on the table. --JokerXtreme (talk) 13:36, 26 March 2010 (UTC)
The wording needs to be more clear and easily seen. Already there has been a double or triple in requests, most of them not correct. Emphasis needs to be made on presenting a reliable source to back the change up, as well as a "change X to Y"-format, which may not be clear enough for some people.  fetchcomms 00:21, 27 March 2010 (UTC)
Indeed. As probably the person who has actioned most semi's in the past few days (as I often do), I've seen about a five-fold increase in the number, but most of them have been crap. Blank requests, garbled junk, vandalism. A few have been partly comprehensible, but of them, most gave no source, and/or didn't make a specific request. I am all for the idea of making the requests easier, but please, could someone make the request thingy clearer - emphasize the need to say exactly what needs to change, and to give reliable sources. Preferably in 6-foot high flashing comic-sans. Ty.  Chzz  ►  08:04, 27 March 2010 (UTC)
What about increases in good requests? Rd232 talk 09:51, 27 March 2010 (UTC)
NB requests like this one may not be exactly what we want, but they are an opportunity to engage new users who might otherwise not understand and walk away. It's a chance to explain. Rd232 talk 09:58, 27 March 2010 (UTC)
Yes, we've had a few of those as well :) But mainly I seem to be telling people to discuss their proposals on the talk page first, which is especially important on fully protected pages. — Martin (MSGJ · talk) 09:56, 27 March 2010 (UTC)

Well, this was my concern about making the Request button too visible in MediaWiki:Protectedpagetext - it makes it more likely people will just jump to clicking the button to see what it does. I had the Request link more within the text (User:Rd232/protectedpagetext/sub) so people would actually have to read it to find that option. Also, it occurs to me that we should probably remove the button anyway on the main page, replacing it with a pointer to the tutorial perhaps. Rd232 talk 09:51, 27 March 2010 (UTC)

I think that will rather discourage edit requesting in general. No one will bother to read, I know I wouldn't. I think my proposal below addresses the concerns raised.--JokerXtreme (talk) 10:52, 27 March 2010 (UTC)

Editintro

I changed the template a little bit. I changed the icon and the instructions. Template:Editsemiprotected/editintroTEST. We could also bold the "follow instructions" here: MediaWiki:Protectedpagetext--JokerXtreme (talk) 09:07, 27 March 2010 (UTC)

Better, for sure. As I said, I'm all in favour of encouraging editing, and in making it easier. Hopefully they will read that notice; I have a feeling that lack of sources will be the biggest problem - but that's OK; only takes a minute to respond to it explaining why we need sources. We shall see.
I'd like to make a request of my own here: it would be great if the people supporting this could keep an eye on the requests (Category:Wikipedia_semi-protected_edit_requests; bot updates can be transcluded from {{User:VeblenBot/SPERtable}}) and help to action them.  Chzz  ►  15:06, 27 March 2010 (UTC)
Out of interest, I had a bit of a look. To my knowledge, we've had 44 requests since 25th; 9 of which have been successful. Notes in user:chzz/sper.  Chzz  ►  18:17, 27 March 2010 (UTC)
I see most of the rejected requests were not referenced. And to tell the truth, no one was asked to provide references in the instructions. Am I wrong? Additionally, some of the requests I declined, did not check for consensus first. We should address that too. --JokerXtreme (talk) 18:21, 27 March 2010 (UTC)
Personally, I think the consensus thing is a decision that the person dealing with it can make - and the request can be the start of a discussion; I don't see a problem with that. The main thing to drill in is the need for references - as always. Same as at WP:AFC, where I spend hours each day telling people what an RS is, what V really means, and that blogs and press-releases ain't enough for N. References, that's the key. To be honest we're bound to get some rubbish requests, no matter what; if we can somehow make it ultra-clear that refs are needed, that's the best chance we have to increase the good/bad ratio.  Chzz  ►  22:09, 27 March 2010 (UTC)
Should I publish my editintro draft for a start and see how that goes? --JokerXtreme (talk) 22:23, 27 March 2010 (UTC)
Sure, why not? I would suggest something in-between the current one and Template:Editsemiprotected/editintroTEST - I don't like the harsh red sign on the latter, I prefer the more friendly informational, but I'd like to see something like, perhaps, the "Please provide reliable sources when possible" highlighted?
Another idea is, how about 2 input boxes - one for their suggestion, and another for the reference?  Chzz  ►  22:39, 27 March 2010 (UTC)
How about the new version? Template:Editsemiprotected/editintroTEST. Maybe we should just add that sources are necessary in the request form, like here:


--JokerXtreme (talk) 23:38, 27 March 2010 (UTC)

I've adopted Joker's changes and adapted and expanded on them - see Template:Editsemiprotected/editintro. I've also customised MediaWiki:Protectedpagetext for the Main Page, since an edit request button there doesn't seem useful (check it out by clicking View Source when not logged in). Rd232 talk 01:39, 28 March 2010 (UTC)

Looks good, let's see how that goes. But, how about having "Edit request - {{currentuser}}" as the default heading? This would fix the problem with multiple ERs messing up the table of contents. --JokerXtreme (talk) 07:45, 28 March 2010 (UTC)
done, with date as well. Rd232 talk 15:17, 29 March 2010 (UTC)

Problem with apostrophes

A user tried to use this at 2010 NCAA Men's Division I Basketball Tournament and was sent to edit Talk:2010 NCAA Men&. It reproduces easily. Celestra (talk) 03:39, 29 March 2010 (UTC)

I couldn't fix this using inputbox; it ought to work with either TALKPAGENAME or TALKPAGENAMEE (Help:Magic_words), but neither does. Using the direct link trick with fullurl, it does work, so I've switched to that. Maybe someone else can sort it better. Rd232 talk 15:16, 29 March 2010 (UTC)

Requests relating to non-existent pages

We are getting quite a lot of requests on salted pages. Quite often, the page has been through an AfD and {{editprotected}} is not appropriate. I think perhaps the link should not be displayed on these? — Martin (MSGJ · talk) 13:03, 30 March 2010 (UTC)

OK, done. Hid the whole table if the page doesn't exist but is protected, leaving the SALTed notice which points users to DRV. Rd232 talk 20:08, 30 March 2010 (UTC)

Huge ugly confusing box

Please revert the recent changes to this template. All they have done is take up twice as much space to display the exact same information, and in two boxes, one of them multi-column, rather than a single one. The result is an ugly confusing mess. There is nothing to stop you adding a link to make an edit request without all that junk. Gurch (talk) 23:30, 30 March 2010 (UTC)

Please be constructive. The aim is to provide a helpful link to editors. You could suggest an alternative method rather than asking for all changes to be reverted. — Martin (MSGJ · talk) 10:55, 31 March 2010 (UTC)
As Rd pointed out, there were many more changes than just changing the template display. The point of the two boxes was to make sure that newbies pay attention to it, and in fact, it works. If you have something better to propose, please do so. --JokerXtreme (talk) 11:18, 31 March 2010 (UTC)

Collapse template transclusion list

When you edit a page, or View Source on a page you don't have permission to edit, you get a list of transcluded templates (which is often quite long). I've never understood why we do this; it's never been any use to me and it must confuse newbies (see eg edit request on a template, clicked through from article).

Proposal: use a {{hat}}-style approach (with an appropriate message of course) to collapse that list by default (i.e. people wishing to see it need to click [show]). Rd232 talk 15:31, 29 March 2010 (UTC)

  • If this is done please allow it to be shown-by-default with some CSS magic. Thanks, –xenotalk 15:47, 29 March 2010 (UTC)
    • Of course. Incidentally I've no idea how my proposal can exactly be implemented; but I'm sure we can figure it out. Rd232 talk 15:54, 29 March 2010 (UTC)
      • I'm working on something User:TheDJ/usagecollapse.js, but it is far from finished. —TheDJ (talkcontribs) 12:21, 30 March 2010 (UTC)
        • Now hide template lists, file usage and global file usage, when you want it to. Uses a cookie to remember your last action. —TheDJ (talkcontribs) 14:07, 30 March 2010 (UTC)
          • That's great, thanks. But the primary motivation for this is not confusing newbies, so the show/hide option needs to be available by default for all users, and the list hidden by default. I guess it needs to be added to MediaWiki:common.js? Shall I suggest that there? Rd232 talk 08:03, 31 March 2010 (UTC)

On this topic one frequent annoyance to me is that the same template list is not visible during a section-edit. I know the least astonishing behavior would be to list only the templates upon which that section depends, but listing them unfilteredly from the `templatelinks` table would be better than no information. Perhaps I′ll write a gadget for this. ―AoV² 08:47, 31 March 2010 (UTC)

7 accts in 24 hrs

Resolved

Is our accountcreation blocker working correctly? The one that allows a user to make only 6 accounts in 24 hours? I'm not an accountcreator, which means if I create six accounts within 24 hours, I'm blocked from creating more. I've started at WP:ACC, and looking at my user creation log, it seems that I have made 11 accounts in the 24 hr range March 29, 15:45-March 30, 15:45 (UTC). Even if we consider that the software might consider a day as a UTC cay, not any arbitrary range of 24 hrs, I've still created 7 accounts on March 30 (UTC). How did I bypass the accountcreation block? Thanks, ManishEarthTalkStalk 10:59, 30 March 2010 (UTC)

My guess is that your IP must get changed by your ISP from time to time - if you had the same IP, you would only be able to create 6 accounts from 0000UTC-235959.99UTC, especially as your count for today is 11 accounts created. -- PhantomSteve/talk|contribs\ 13:38, 30 March 2010 (UTC)
Or you used different computers. Ruslik_Zero 14:57, 30 March 2010 (UTC)
Didn't change comps, and I didn't turn off my router. I do have a rotating IP, but my comp was on the whole time (I was studying my notes). Do IP's rotate when you're using them? ManishEarthTalkStalk 02:39, 31 March 2010 (UTC)
They could...depends on your ISP provider.Smallman12q (talk) 23:29, 31 March 2010 (UTC)
Oooh. I didn't know that... Thanks! ManishEarthTalkStalk 03:25, 1 April 2010 (UTC)

Best way to save watchlist items for more than the max. 7 days?

What's the best way to save watchlist items for more than the maximum of 7 days? It will probably either require use of the RSS feed or the API. I'd prefer using a service that's already available rather than write my own. Also, the solution should not require human intervention; i.e., it should download new watchlist items automatically. I was thinking of using Google Reader as it archives items pretty far back, but I can't recall if it only fetches new items when you visit it, which would require human intervention and is therefore not an option for me. Gary King (talk) 17:20, 30 March 2010 (UTC)

The maximum is in fact 30 days, see Special:Watchlist [15]. Were you wanting something to list (in a file) every passing edit for a set of pages or only to display “recent changes” in some window other than your web browser? ―AoV² 07:01, 31 March 2010 (UTC)
Good Lord you just saved me a lot of time. Thanks! Gary King (talk) 18:03, 31 March 2010 (UTC)

Diff bugs

I suppose this is a waste of time. I keep begging for the stupid bugs in Wikipedia's diff generation to be fixed, but nothing ever happens. I despair. Is it just me?

Typical example of what I mean:

http://en.wikipedia.org/w/index.php?title=Amir_Khan_%28boxer%29&action=historysubmit&diff=353069392&oldid=348362877

Look at the vast amounts of red ink where, in fact, nothing at all has changed.

Please don't bother replying with hard luck stories about how difficult this stuff is to code. These are FUNDAMENTAL, GLARING BUGS, not esoteric niceties. Rant over for now, but I would so love it if you could help get this fixed. 86.152.242.9 (talk) 01:26, 31 March 2010 (UTC).

Agreed. I suggest using wikEdDiff for a better diff tool. You'll need an account to use it, though, of course. Gary King (talk) 01:43, 31 March 2010 (UTC)
@IP_Address It's open source, send patches, if you so clearly understand how this is done. —TheDJ (talkcontribs) 02:41, 31 March 2010 (UTC)
Indeed, if they are FUNDAMENTAL and GLARING then you should be able to correct them yourself. OrangeDog (τε) 12:13, 31 March 2010 (UTC)
Total nonsense, OrangeDog, and you know it! 81.156.127.170 (talk) 23:57, 31 March 2010 (UTC).

Weird

So this was on Category:Wings of the United States Air Force for around 12 hours, "The following 193 pages are in this category, out of 197 total. This list may not reflect recent changes." It's not all that bad except for the fact that there is no second page. Any ideas? I'm going to finish emptying the category right now, but it's really weird that it's saying that. Kevin Rutherford (talk) 17:42, 31 March 2010 (UTC)

Long help desk archive page won't load

[16] It doesn't even have "history" or "edit" at the top, or my name or links to my talk page and contributions. Usually it just takes a while, but this time it only loads the portion up to Oct. 10, 2006 and it appears to be at the bottom. Regardless of how many times I try. But I know I've seen the page as far as Oct. 15.Vchimpanzee · talk · contributions · 18:17, 31 March 2010 (UTC)

Fixed. I'm curious about why it did that, though. Normally it doesn't just quit like that.Vchimpanzee · talk · contributions · 18:23, 31 March 2010 (UTC)

It's probably just the size of the page. It looks like there was an error that resulted in many of threads being duplicated (between 2.287 and 2.511, although a few have additional comments); this is a possible explanation for the length of that archive. snigbrook (talk) 23:54, 31 March 2010 (UTC)

Script help

I use User:Ais523/catwatch.js, its incredibly useful to me for things like attack pages and requests for unblock etc, but because I'm in the UK and we're forward an hour now, new pages in the categories i'm watching appear are listed as being an hour earlier than they are to me, so I keep missing them. Is there anyway to change that in my vector.js?--Jac16888Talk 22:00, 31 March 2010 (UTC)

Special:Preferences → Date and time → Time offset. ---— Gadget850 (Ed) talk 22:59, 31 March 2010 (UTC)
But that would change my whole watchlist. Is there anyway to change it so the category items are listed an hour later?--Jac16888Talk 00:58, 1 April 2010 (UTC)

Any talk page

Why is it that if you edit a talk page when logged out you can only see the edit box, whilst logged in you can see the whole page? This applies to "edit this page". This is definitely a recent change. IMO for the worst. Simply south (talk) 20:51, 27 March 2010 (UTC)

Whoa, that is new. I can't imagine that being of any benefit at all. Gavia immer (talk) 21:58, 27 March 2010 (UTC)

I don't understand the problem description. More info required. screenshots might help. —TheDJ (talkcontribs) 00:05, 28 March 2010 (UTC)
This is something that can be enabled in preferences, if "Show preview on first edit" (in the "Editing" section) is selected. snigbrook (talk) 01:09, 28 March 2010 (UTC)

For those who don't know...

Comparison of Viking ship tp.png Comparison of thriller tp.png Comparison of AFD tp.png

Where was the discussion on these changes (a bit ironic eh?) Simply south (talk) 01:14, 28 March 2010 (UTC)

Possibly an option in your preferences. Take a look at the editing tab and see if the tick is on "Show preview on first edit". Astronaut (talk) 01:32, 28 March 2010 (UTC)
Okay, possibly it's just so old that I've forgotten there was ever a choice. That's the same as "new", right? Gavia immer (talk) 02:18, 28 March 2010 (UTC)
No, you've got it wrong Astronaut and Snigbrook. Logged out\IP editors do not get preferences. Look at the pics again and look especially at the top of the screen. Simply south (talk) 18:42, 28 March 2010 (UTC)
And logged in users DO. Look at your settings, like Astronaut described. The NORMAL behavior is to have the edit box when you click edit, and not, to have both the preview and the editbox. —TheDJ (talkcontribs) 23:09, 28 March 2010 (UTC)
In the past it has been both even when not in user. Simply south (talk) 15:43, 1 April 2010 (UTC)

Twinkle vs. spam blacklist

I tagged Lakhotia Computer Centre for speedy deletion as a copyvio, but it wasn't easy. Twinkle said that tagging the article was complete, but I didn't see the tag on the article! So I tried adding it manually, thinking that it was just an intermittent glitch in Twinkle. To my surprise, the URL the content came from was on the spam blacklist! No wonder why the article hadn't earlier been deleted with most of the other "large unwikified new articles". PleaseStand (talk) 06:44, 30 March 2010 (UTC)

Please report twinkle issues to WT:TW/BUGS. –xenotalk 13:40, 30 March 2010 (UTC)
I'm thinking that the easiest fix is actually to change {{db-g12}} to nowiki external links. I don't want to break the somewhat complex template (though not nearly as much as {{dated prod}}) though; I would prefer that someone with more template experience make the change. But wait...is it even possible to nowiki a template parameter from within the template? Yes it is. Use {{#tag:nowiki|{{{url}}}}}. PleaseStand (talk) 18:13, 30 March 2010 (UTC)
Posted at Template talk:Db-meta#Nowiki the URL in Db-g12 with the proposed changes. PleaseStand (talk) 06:33, 2 April 2010 (UTC)

A bit of CSS signature help...

Well, I have spent sometime reading, and have tried to make my self a signature. What I have come up with is this:

A p3rson

However, I get an "Invalid raw signature, check HTML tags" error when I attempt to use it. I have it at ~253 characters, did I miss something?

P.S. - The full code is available at my sandbox.  A  p3rson  03:02, 31 March 2010 (UTC)
This should do it. Gary King (talk) 03:26, 31 March 2010 (UTC)

Maybe this instead? :D :p

A p3rson]]

¦ Reisio (talk) 12:02, 31 March 2010 (UTC)

NB, blinking signatures are disallowed per Wikipedia:SIG#Appearance_and_color╟─TreasuryTagcabinet─╢ 17:44, 31 March 2010 (UTC)
Mmm, but according to WP:POLICY, a policy, both policies and guidelines are meaningless because of WP:NOTSTATUTE, a policy; and even if they weren't, they'd be meaningless because of WP:IGNORE, a policy; and even if they weren't… % ¦ Reisio (talk) 18:42, 31 March 2010 (UTC)
You clearly misunderstand WP:IGNORE. ╟─TreasuryTagsecretariat─╢ 06:49, 1 April 2010 (UTC)
just musing, but it's probably best not to argue policy on a matter that does not help to improve the encyclopedia. --Ludwigs2 19:57, 31 March 2010 (UTC)
I think the blink tag tends to incite violence. It is dangerous, do not use it. Gary King (talk) 20:05, 31 March 2010 (UTC)
Was just mockery, anyways. ¦ Reisio (talk) 20:12, 31 March 2010 (UTC)

Table rendering

Can somebody tell me why this table isn't rendering?

{| {{table}}
| align="center" style="background:#f0f0f0;"|'''Conference'''
| align="center" style="background:#f0f0f0;"|'''Year'''
| align="center" style="background:#f0f0f0;"|'''Date Held'''
| align="center" style="background:#f0f0f0;"|'''Venue'''
| align="center" style="background:#f0f0f0;"|'''Host'''
| align="center" style="background:#f0f0f0;"|'''Theme'''
| align="center" style="background:#f0f0f0;"|'''Programme Chairs'''
| align="center" style="background:#f0f0f0;"|'''Conference Chairs:'''
| align="center" style="background:#f0f0f0;"|'''Site'''
|-
| WWW1||1994||May 25-27||Geneva, Switzerland/France||Cern||?||?||?||?
|-
| 
|}

I got it using http://excel2wiki.net/index.php where I pasted

Conference      Year    Date Held       Venue   Host    Theme   Programme Chairs        Conference Chairs:      Site
WWW1    1994    May 25-27       Geneva, Switzerland/France      Cern    ?       ?       ?       ?

Any help would be appreciated!Smallman12q (talk) 23:26, 31 March 2010 (UTC)

It can be fixed by removing the {{table}}, which doesn't appear to have any purpose. snigbrook (talk) 23:45, 31 March 2010 (UTC)
That works! Was a template updated...I always used to include the {{table}} template...do you know if there are broken tables because of this?Smallman12q (talk) 23:51, 31 March 2010 (UTC)
I've never seen the template before, but it looks like combining the two types of table causes problems. The {{table}} template isn't transcluded in a large number of pages so maybe you could identify where it was used in Special:WhatLinksHere. snigbrook (talk) 00:01, 1 April 2010 (UTC)
Editors should use the updated WP-specific tool: http://excel2wiki.net/wikipedia.php. See Template talk:Table#Who uses this template? for more information. Flatscan (talk) 03:58, 1 April 2010 (UTC)
See also: Commons:Convert tables and charts to wiki code --Timeshifter (talk) 09:47, 1 April 2010 (UTC)

{{table}} has always been useless in this case. It was never intended to be used like that. It doesn't do anything without any parameters, so the template was recently changed changed to spit out an error if it lacked parameters. But if you put the template where you did, the error message breaks the table and the error is never seen. Oops. That may be difficult to detect and fix, because as far as I can tell not even categories will work there. Reach Out to the Truth 18:09, 1 April 2010 (UTC)

output of OS mapping API in wikipedia

Today (1st April) the UK government agency the Ordnance Survey (OS) has released a new API & totally changed the licencing agreements which should allow the use of the best UK mapping data free for non-commercial purposes. It would be great to be able to use this to illustrate UK place & geography articles. Further info & OS OpenSpace® Developer Agreement are available. I have generated a map (of Cheddar) without any markers or routes which can be included, just to test the inclusion in wikipedia. I have stripped out the header code "<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">" closing "</html>" as they suggest if embedding the map in another application & put it on a page in my sandbox - User:Rodw/Sandbox/Cheddar. I am unable to get it to display the map. Any help appreciated. NB if editing it you must leave the API Key <script type="text/javascript" src="http://openspace.ordnancesurvey.co.uk/osmapapi/openspace.js?key=832E76B2051F6AB8E0405F0ACA6011DE"></script> alone as this tells the OS that it can be run on http://en.wikipedia.org/.— Rod talk 20:30, 1 April 2010 (UTC)

It is not possible to embed just any html into Wikipedia. In order to support it like this, you either need to export it to an image, or you should create a MediaWiki extension. But I would like to point out that Wikipedia does not accept or use material that is allowed only for non-commercial purposes. So in terms of Wikipedia usage, this material is useless. —TheDJ (talkcontribs) 20:53, 1 April 2010 (UTC)
I have no idea how to create a mediawiki extension - I just think these maps could be a really useful addition to UK geography articles. The phrase non-commercial is mine - not from their rules. On checking it doesn't say that - where would be the best place to get advice on the suitability (or otherwise) of using this API on wikipedia? I have already put requests for help/opinion on Wikipedia_talk:WikiProject_Maps#Use_of_UK_Ordnance_Survey_new_API & Wikipedia:Media_copyright_questions#Use_of_UK_Ordnance_Survey_new_API. — Rod talk 21:34, 1 April 2010 (UTC)
Problems that I see
  1. "Usage of the API is subject to Usage Limits and system capacity", yet we it seems we are not allowed to duplicate the service and provide our own resources to host it either.
  2. "This Agreement conveys a limited licence for You to use the API Key, Ordnance Survey Data, Developer Documentation and trade mark OS OpenSpace solely for the purposes of creating, delivering and maintaining a Web Application" so that means the material is not usable for books or other non-web application uses. This is a sever usage restriction
  3. "You [sic. developer] may create Derived Data, and You may permit End User's to create Derived Data, in connection with Your Web Application. In the event that You or any End User creates Derived Data, such Derived Data shall be owned by Us" YUCK !!! you subsequently (5.4.2) get a license for your own work, which they can revoke if they stop liking you.
  4. (6.2) "Your Web Application must not fall into any one or more of the categories listed below." again a usage restriction
  5. (6.6) "Your Web Application must not be undertaken for, or in connection with, nor must it result in any Unacceptable Financial Gain. For the avoidance of doubt, this means that You are not permitted (nor may You permit others to) charge End Users any subscription or other fee for accessing and using the whole or any part of Your Web Application or the Ordnance Survey Data." commercial usage not allowed
  6. (6.8) "You must ensure that a copy of the EULA is made accessible to End Users through a hypertext link at the bottom of each page of Your Web application." that will NEVER happen on Wikipedia, we don't allow such things.
  7. (6.9) "You may only allow End Users to print a maximum of ten (10) paper copies, no greater than A4 (625 cm2) in size of any screen shot." Seriously ? that's not even close to being compatible with Wikipedia
I'm stopping here, because I think the point is clear, but there is plenty more where that came from. —TheDJ (talkcontribs) 21:59, 1 April 2010 (UTC)
Having said all that. The data itself is licensed per the OpenData license and mostly available for download. It's just the API that has a crazy set of unneeded restrictions. —TheDJ (talkcontribs) 22:14, 1 April 2010 (UTC)
As far as I'm concerned, any product that hires a gazillion lawyers to draw up something as unreadable as this, isn't worthy of using the name "Open" btw. —TheDJ (talkcontribs) 21:05, 1 April 2010 (UTC)
We can already use OpenStreetMap maps. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 22:46, 1 April 2010 (UTC)

Could be helpful

It would be helpful for editors with hundreds of articles on their watchlist, if the list was automatically adjusted (shortened) as they process it. What I mean is that when they tick diff on an entry on their watchlist, that particular notification would disappear from the list as they went off to see the change they were being alerted to, (but of course any subsequent change would appear later on the list). This would automatically decrease the clutter of the user's watchlist. Perhaps it could be added as an option in users' watchlist preferences. Comments? Moriori (talk) 00:59, 2 April 2010 (UTC)

I thought something like this would be useful, so I created a Windows application for myself that does something similar: Desktop Watchlist. Svick (talk) 01:48, 2 April 2010 (UTC)
On the IMSLP, they have it set to bold any new changes, but this affects all the edits of the article, not just the ones changed since you last visited. Of course, you also have to refresh your list. ♫ Melodia Chaconne ♫ (talk) 02:13, 2 April 2010 (UTC)

Templates on articles and talk pages to avoid perennial issues

At the Harry Potter articles, editors keep changing the title Harry Potter and the Philosopher's Stone (correct, it's a Brtish book & author) to Harry Potter and the Sorcerer's Stone (US market only). How can I show a banner at articles and talk pages to tell the world which is the correct title - I'm fed up with reverting. In talk pages, I'd like the banner to fix at the top even if the page is archived. --Philcha (talk) 06:04, 1 April 2010 (UTC)

Page can be move protected, and we could add an editintro warning about the title I guess. —TheDJ (talkcontribs) 11:29, 1 April 2010 (UTC)
AFAIK move protection works only for the article Harry Potter and the Philosopher's Stone. Unfortunately the title of the 1st book appears in many articles and talk pages - possibly all, because of the infobox - so all these articles and their talk pages need banners. --Philcha (talk) 16:52, 1 April 2010 (UTC)
People are going to change it with or without the big ugly banner at the top of the article. --OnoremDil 16:53, 1 April 2010 (UTC)
But when they do, at least there's something to refer to, rather than starting the debate from scratch each time. I've seen {{tmbox}} used in talk page headers for this purpose, e.g.
PL290 (talk) 17:19, 1 April 2010 (UTC)
Something on the talk page, sure. Hidden comment or editnotice on the article, sure. Notices at the top of the articles, like now appear on Harry Potter and Harry Potter and the Chamber of Secrets, I don't like it. Also, between those two articles, the name was changed once in the past month. I can't think of any other article where that type of notice is at the top of the page. --OnoremDil 17:34, 1 April 2010 (UTC)
Ew. Yeah, messages like that should not be in the article itself. The talk page and the editnotice is the correct place for that. Reach Out to the Truth 18:10, 1 April 2010 (UTC)
See current User_talk:Philcha. If there's no banner on the article to minimise the number of incorrect edits, I won't bother fixing the title of the 1st book - which appears in the articles about the books and on Harrty Potter. --Philcha (talk) 22:52, 2 April 2010 (UTC)
Reach Out to the Truth, see User_talk:Philcha --Philcha (talk) 23:37, 2 April 2010 (UTC)

IPA font rendering in Firefox

Hi,

It seems that the .IPA class doesn't work correctly in Firefox 3.6 on Windows. See Wikipedia talk:IPA#Wrong rendering on Firefox 3.6 on XP.

Can this be fixed without making users edit their private CSS?

Thanks in advance. --Amir E. Aharoni (talk) 12:30, 2 April 2010 (UTC)

secure.wikimedia.org Proxy Error 502

I'm getting a lot of these over the last three days. I am signing off secure.wikimedia.org and signing in to en.wikipedia.org in order to accomplish anything. Any others with this problem? Is there a better place to determine the status of secure.wikimedia.org? Many thanks, Darrell_Greenwood (talk) 18:43, 26 March 2010 (UTC)

Me too. It happens when I try to get diffs, edit (it'll make the edit but not return the proper screen), and even when I'm just reading a page. I've had to go to the unsecure server too. RlevseTalk 19:16, 26 March 2010 (UTC)
Yes, the same for me. It seems to happen with any large page, one that has a lot of data to load. Clearing my browser cache does not help. It seems to have started just after the server failure a few days ago, and is very annoying. Have the secure servers not yet been fully restored? --Tryptofish (talk) 21:02, 26 March 2010 (UTC)
I concur, it seems to mainly happen on large pages, and the problem started with the failure noted here "Update 21:32 UTC: Our SSL gateway, secure.wikimedia.org, was disabled due to overload issues, but is now back up." Thanks for your comments, as they make it unlikely that the continuing problem is mine. (Parenthetical remark, whilst troubleshooting, I came across a comment that European users can bypass the Amsterdam servers when those servers are down by logging into secure.wikipedia.org because it is in Florida. If enough people are still doing that... overload?) I haven't been able to find a web status indication for secure.wikipedia.org. Darrell_Greenwood (talk) 21:33, 26 March 2010 (UTC)
OK, but can something be done to fix this now? RlevseTalk 23:59, 26 March 2010 (UTC)
The problem is continuing for me. I've left a message at User talk:Tim Starling. It's not really clear to me where one can report what appears to be a hardware problem. --Tryptofish (talk) 21:38, 27 March 2010 (UTC)

In case it helps, here are further details of the 502 error message: Apache/2.2.8 (Ubuntu) mod_fastcgi/2.4.6 PHP/5.2.4-2ubuntu5.7wm1 with Suhosin-Patch mod_ssl/2.2.8 OpenSSL/0.9.8g Server at secure.wikimedia.org Port 443. --Tryptofish (talk) 21:49, 27 March 2010 (UTC)

bugzilla:22982 was opened to track this issue. —TheDJ (talkcontribs) 00:29, 28 March 2010 (UTC)
Thank you, TheDJ. --Tryptofish (talk) 00:47, 28 March 2010 (UTC)
Yes, I am getting the same error:
Proxy Error

The proxy server received an invalid response from an upstream server.
The proxy server could not handle the request POST 
/wikipedia/en/w/index.php.

Reason: Error reading from remote server

Apache/2.2.8 (Ubuntu) mod_fastcgi/2.4.6 PHP/5.2.4-2ubuntu5.7wm1 with 
Suhosin-Patch mod_ssl/2.2.8 OpenSSL/0.9.8g Server at secure.wikimedia.org 
Port 443

--Ancheta Wis (talk) 00:36, 28 March 2010 (UTC)


Ditto. Over the last few days I've been getting a 502 on https://secure.wikimedia.org/wikipedia/en. I'm not experiencing it on http://en.wikipedia.org. It happens on both Safari and Firefox, survives a reboot, is intermitted, and doesn't seem to affect the Main Page. One article I've seen it on consistently is Ireland.

Full response:

Proxy Error

The proxy server received an invalid response from an upstream server.
The proxy server could not handle the request GET 
/wikipedia/en/wiki/Ireland.

Reason: Error reading from remote server

Apache/2.2.8 (Ubuntu) mod_fastcgi/2.4.6 PHP/5.2.4-2ubuntu5.7wm1 with 
Suhosin-Patch mod_ssl/2.2.8 OpenSSL/0.9.8g Server at secure.wikimedia.org 
Port 443

--RA (talk) 20:42, 28 March 2010 (UTC)


I don't know if this helps or confuses, but I get the error message:

"The proxy server received an invalid response from an upstream server. The proxy server could not handle the request GET /wikipedia/en/wiki/3rd_Battalion_3rd_Marines."

when I am on the secure server, *but* when I go to the English server and try to edit the page, I get the same error. I have gotten this error on numerous pages on both the secure and English server.Thomas R. Fasulo (talk) 00:03, 29 March 2010 (UTC)

  • Me too, happens intermittently. Started last few days. ► RATEL ◄ 12:09, 30 March 2010 (UTC)

According to the Bugzilla report, linked above, they are in the process of setting up a new server that, when installed, will hopefully fix the problem. --Tryptofish (talk) 17:03, 30 March 2010 (UTC)

At least for me, this now seems to be fixed. --Tryptofish (talk) 17:53, 31 March 2010 (UTC)

It's not happening as often, but I got it once today so far. RlevseTalk 20:39, 3 April 2010 (UTC)
One edit completed without error messages, and it got all the way through. Thank you to the server people. --Ancheta Wis (talk) 21:43, 3 April 2010 (UTC)

Issue with switchs and colors

See Book:Calgary for the problem. |text-color= isn't treated correctly when a hex-color is given. See {{saved book}} and {{pp-book-cover}} for the templates used to generate the text. Headbomb {talk / contribs / physics / books} 17:41, 2 April 2010 (UTC)

I fixed it with a &nbsp; hack for the moment. I'll need to look more carefully as to why it's happening (because it shouldn't be). --Ludwigs2 18:11, 2 April 2010 (UTC)
{{#if: #ee2222| #ee2222|CornFlowerBlue}}seems to introduce a linebreak for some reason. Strange... —TheDJ (talkcontribs) 18:16, 2 April 2010 (UTC)
The problem is this switch statement snippet: {{#switch:{{lc:{{{cover-color|}}} }} |black={{{text-color|black}}} |{{{text-color|}}} }}. for instance, if you test just this, like so:
AAA{{#switch:{{lc:{{{cover-color|black}}} }} |black={{{text-color|#96e}}} |{{{text-color|#e69}}} }}BBB
what you get is

AAA

  1. 96eBBB
Notice how (for some reason) the SWITCH statement is adding a carriage return, which then forces the #-sign to get interpreted as a list element. seems like a bug to me. adding the non-breaking spaces fixes it by keeping the #-sign from being the first character on the line, but it is a hack. --Ludwigs2 18:38, 2 April 2010 (UTC)
&nbsp; breaks things since it's not allowed in the css. &#32; works however. Thanks for the help. Headbomb {talk / contribs / physics / books} 01:41, 3 April 2010 (UTC)
I think this is another instance of bug 12974http://bugzilla.wikimedia.org/show_bug.cgi?id=12974. Svick (talk) 18:47, 2 April 2010 (UTC)

I′ll let you consider the following and draw your own conclusion. ―AoV² 21:45, 2 April 2010 (UTC)

input output html
{| style="background:{{#ifeq:a|a|#ffaaaa|green}};"
| pink 
| style="background:{{#switch:z|x = yellow| z = #aaaaff}};" | blue
|}
  1. ffaaaa;"
pink style="background:
  1. aaaaff;" | blue
<ol>
<li>ffaaaa;"</li>
</ol>
<table>
<tr>
<td>pink</td>
<td>style="background:
<ol>
<li>aaaaff;" | blue</li>
</ol>
</td>
</tr>
</table>
<table style="background:{{#ifeq:a|a|#ffaaaa|green}};">
<tr>
<td> pink </td>
<td style="background:{{#switch:z|x = yellow| z = #aaaaff}};"> blue </td>
</tr>
</table>
pink blue
<table style="background: #ffaaaa;">
<tr>
<td>pink</td>
<td style="background: #aaaaff;">blue</td>
</tr>
</table>
hmmm... the conclusion I draw from that is that the template should be rewritten with HTML rather than wikitext. is that what you were aiming at? smile happy to do it, if so... --Ludwigs2 05:15, 3 April 2010 (UTC)

That probably would be safer than the predescribed hack which relies upon the software changing &#32; to a literal space, but not doing it too early. ―AoV² 06:50, 3 April 2010 (UTC)

That doesn't explain why these parserfunctions introduce linebreaks in the first place in this case though. —TheDJ (talkcontribs) 13:19, 3 April 2010 (UTC)
bugzilla:23033TheDJ (talkcontribs) 13:30, 3 April 2010 (UTC)

Help reverse redlinks

Rich put together a great list of derm ICD9 codes at Wikipedia:WikiProject Medicine/Dermatology task force/ILDS-ICD. However, many times the redlinks utilize a comma, like Balanitis, amoebic. Therefore, I wanted to know if someone would help reverse all the redlinks with commas. So, with the previous example, the redlink would be changed to Amoebic balanitis. Thanks in advance! ---kilbad (talk) 17:29, 1 April 2010 (UTC)

Working on it automatically using AWB (assumptions will be made about capitalisation which you may want to check). All links, blue or red, will also be affected. - Jarry1250 [Humorous? Discuss.] 14:48, 3 April 2010 (UTC)

Oh good, next you can create [[Lastname, Firstname]] redirects for all biographies. ―AoV² 21:15, 3 April 2010 (UTC)

Help with Template:Drugbox needed

In {{Drugbox}}, the "chembox validation field" (the very last line in transclusions where it shows, e. g. in Metformin, saying "[check mark] (what is this?) (verify)" has a light grey background in Firefox, but an eerie looking black background in IE 8. Does anyone know how this can be fixed? Thanks in advance, ἀνυπόδητος (talk) 16:55, 3 April 2010 (UTC)

I don't have IE to test on, but the only difference I see here is that the other cells use style="backgound-color: #xxx" while the last cell used bgcolor="#xxx" as a table cell attribute. maybe convert the last to a style attribute as well? --Ludwigs2 17:16, 3 April 2010 (UTC)
If that doesn't work, try to make a small screenshot and put it on http://imagebin.ca so that we can see the problem. —TheDJ (talkcontribs) 17:18, 3 April 2010 (UTC)
Thanks, that does it! --ἀνυπόδητος (talk) 17:29, 3 April 2010 (UTC)
I don't see the change in the template - did you test that on your machine? if so, let me know and I'll add an {{editprotected}} request. see it now.--Ludwigs2 23:02, 3 April 2010 (UTC)

Create article quality and importance fields in the database

Article quality is presently detected by looking at article talk pages (see User:Pyrospirit/metadata). I wonder if we might instead make quality and importance database fields, either in the page table or in quality/importance table(s) linked to that table, and add functions to detect that data (e.g. Article::getQuality(), or Title::getArticleImportance, etc.) so that extensions and such can more easily make use of it. Rather than assigning quality and importance values via edits to the talk page, quality and importance could be set on the article page itself via a button/drop-down menu in a similar manner as, for instance, protection or categorization is done, and modifications to this metadata would be logged. (Note that protected_titles has its own table.) The talk page WikiProject templates could then populate this data using magic words such as {{ARTICLEQUALITY}} and {{ARTICLEIMPORTANCE}}. Measures similar to those used by Wikipedia:ESSAY_C/C/A#Score might be used.

We are perhaps used to thinking of notability as a true/false dichotomy because our deletion debates operate within that paradigm, but really notability is a spectrum on which many levels could be set in accordance with various criteria. Deletion debates do result in a boolean measure of article importance being added to the database (i.e. deleted or undeleted). But a more nuanced approach could be in order. Quality and importance are vital attributes of an article, because article length, completeness, sourcing, style, notability, etc. help determine an article's overall current usefulness and how much attention the article deserves/requires. I.e., if you only have an hour a day to devote to editing, you might want to focus your attention on high-importance, low-quality articles. The fact that WikiProjects go around grading articles by quality and importance, and that we plaster Wikipedia maintenance templates to warn readers of article problems and notability issues and to help us categorize and spot articles in need of improvement or other action, shows that there is some recognition of these metrics' importance.

So then, what exact use could be made of these fields? Maybe Special:RecentChanges could filter articles by quality and importance; some users might prefer to only keep an eye on articles above a certain threshold of importance. I am thinking in particular of this being useful in a hyper-inclusionist encyclopedia that allows lots of content that would be below Wikipedia's current standards for inclusion. Suppose there are a bunch of articles on garage bands and such. Some users might deem those subjects to be so non-notable that they don't even care if those articles get vandalized, so they would filter those articles out of their recent changes. On the other hand, a user might wish to search out articles that are both high-importance and low-quality and concentrate his efforts on those. A reader could set minimum standards of importance and quality in his user preferences, and wikilinks could turn different styles (much as article titles turn different colors with Pyrospirit's gadget) or be deactivated entirely (i.e. turn into black plaintext) based on whether or not an article meets his preferred thresholds of quality and importance.

It might be argued, But then people will edit war over what an article's importance and quality rating should be. I see no reason to expect this. We do not see much edit warring over quality and importance assessments under our current system even though we do make those assessments; and we have orderly processes such as WP:FAC, WP:FAR, and WP:GA for deciding whether to promote or demote articles from certain tiers of quality. I have never seen a user storm off angrily from the project or go on a rampage over something that happened in those processes. The major disputes occur at XFD and WP:DRV, but the need for most deletions could be obviated in a hyper-inclusionist encyclopedia in which information were graded and filtered by user preference.

It might be further argued, But if people don't pay much attention to the low-quality articles, and filter them out of what they see, then those articles will never get the love they need to improve. Again, though, when combined with the importance metric, this problem can be obviated, because people can focus their efforts on high-importance, low-quality articles. Articles on low-importance subjects such as garage bands, because few people know or care about them, will tend to not receive many edits, and to stay in a low-quality state. This is no big deal; the encyclopedia's reputation will not suffer much from their presence, because who is going to google those subjects, find those articles, and be disgusted with our coverage of them? If a lot of people are googling a subject and reading the article on that subject, then that article is probably going to receive more edits from that increased readership and thereby improve, and someone may even bump its importance level up (if for instance it gets slashdotted), leading to further attention from those who otherwise would have filtered it out. I think the built-in self-correction mechanisms of this model can provide for all of these problems to solve themselves.

I am not going to Bugzilla yet with this, because I still haven't worked out all the details of how this could/should work. I think it should be tested on another wiki first. E.g., we are doing some soul-searching over at Libertapedia over what we want to include, and the question arises, Does any harm come from having non-notable content on the project? The only harm is that it gets in the way; it could clog RecentChanges and such. But if we can filter it out, it might not matter, and we would be freed from having to police the site and always be sitting in judgment as to what is notable enough to include and what isn't. I think it could reduce the potential for drama and bad feelings, if no one has to have their (otherwise well-written and in conformity with the encyclopedia's style standards) article deleted for lack of notability.

I welcome any thoughts on this proposal and any insights that can help refine the details. Tisane (talk) 17:31, 30 March 2010 (UTC)

tl;dr, but I should note that importance (and to a lesser degree, class) is subjective among wikiprojects. –xenotalk 17:33, 30 March 2010 (UTC)
Good point. A starting point for the quality scale could be some of the attributes listed at, for instance, Wikipedia:Featured article criteria and Wikipedia:Good article criteria. Still, it is noteworthy that people don't seem to get in too many conflicts about the assessments, until deletion becomes a possibility. For instance, at FAC, people will make objections, others will address the objections by editing the article or point out why the objection is non-actionable, and the process is pretty civil. Why do you suppose that is? I speculate it is because the spectre of imminent deletion isn't looming over people's heads, and any decision made is pretty reversible, whereas it can be pretty hard to reverse a deletion; for one thing, the content is removed and collaboration stops after a deletion. Tisane (talk) 17:40, 30 March 2010 (UTC)
You're right, there is very little warring when it comes to assessing an article, even the most popular ones. That's because class/importance of an article is not really important; it's just a way to indicate the status of an article, for a particular WikiProject. According to your proposal, though, the class and importance of an article would increase, which might lead to conflict, especially regarding what articles would be considered of "Top" importance. Anyway, the first step in this would be for an extension to MediaWiki to be created, then test it, then we can see exactly how it would work. Your proposal sounds promising, but also probably a bit idealistic and it assumes that people will actually spend more time working on high-importance articles than low- ones. Have a look at WP:CORE and WP:VITAL, as well as WP:FA and you will see that that is not the case. Gary King (talk) 17:44, 30 March 2010 (UTC)
For someone who wants to publicize a certain topic, quality assessment decisions like FAC are very important. Consider, for instance, how the readership of Mendip Hills went from about 100 a day to 18,200 after it hit the main page. Yet FAC is still a pretty-laid back process, despite all that is riding on FA status.
There is a de facto quality measure, as least as far as completeness in concerned, in that the length of an article seems to affect its google ranking, and the sheer quantity of keywords makes a long article likely to show up in more searches. Redundant material tends to be eliminated by editors, and unsourced info tends to be removed, so any long article must have a lot of good stuff, may be Google's theory.
I agree that core articles are in bad shape. Play (theatre), for instance, has always been in abysmal shape. I think it is because people do use Wikipedia for promotion of their pet causes/organizations/interests, and there is less incentive to write about such general, well-known subjects than a particular subset that the person perceives as yearning for more attention - e.g., a particular play, rather than plays in general. That's just the economics of the situation. But maybe we can bend economics to our advantage. If articles with higher quality ratings get more readers (as they already do to some extent, as in the main page and Google examples above), then people will have an incentive to improve, if not the core articles, at least the articles that pertain to their interests. That seems to have worked well in incentivizing people to raise articles to FA standards. (I think a lot of people pursue GA as just a step along the way to FA.) Tisane (talk) 18:08, 30 March 2010 (UTC)
Hitting the main page does, of course, increase views for an article, but only for that one day. Then the article goes back to stagnating like it was before. Most visits to Wikipedia come from web search results, and not from visiting the Main Page then clicking around. FAC is definitely not a laid-back process. Most of the people who don't have problems getting through the process have already done it before. Google rankings are barely affected by keywords; they are well known for not focusing on keywords, but rather on inbound links as they have been doing since they were founded. Ultimately, people will pretty much always edit whatever they want, regardless of how important an article has been subjectively determined to be. Also, general articles like the one you provide are very hard to write. It's easier to write about something concrete like an actual historic event or book. Gary King (talk) 18:20, 30 March 2010 (UTC)
As I said, WikiProjects will usually agree on the class, but importance is a horse of a different colour. While Toronto is of top-importance to WikiProject Toronto, it is considered high-importance to WikiProject Cities whereas WikiProject Foo may consider it only 'low' importance. –xenotalk 18:14, 30 March 2010 (UTC)


I run the WP 1.0 bot. The bot does keep importance and quality information in a database on toolserver. If there is a need to get this information in an automated way, I can probably implement an API to allow tools to query the data.

There are a few issues with conflicting ratings:

  • Importance ratings will vary wildly between projects. For example, an article might be very high importance for WikiProject Alabama but very low priority for WikiProject Geography.
  • Quality ratings are usually less varies. But a few projects have A-class checklists, and so an article might be rated B by them even if it is rated A by others. There are also B-class checklists.
  • Some projects use non-standard ratings like "Bottom-importance" or "B+-class".

I will not be following this thread so please contact me separately if there is something you'd like to see in the WP 1.0 bot. — Carl (CBM · talk) 18:29, 30 March 2010 (UTC)

(ec) Setting an importance for an arcticle across Wikipedia as compared to the current system of setting an importance per wikiproject is not likely to be agreed apon easy. Why have one importance rating rather then the existing system? Regards, SunCreator (talk) 18:34, 30 March 2010 (UTC)

Since deletion is already a sitewide importance decision that override WikiProject consensus, part of the idea is to replace the steep cliff of notable enough for inclusion vs. non-notable enough to merit deletion, with a more finely-graded slope for determining what level of obscurity an article should be relegated to. This could make decisions less acrimonious, as in close cases all that is riding on it is an upgrade/downgrade between, say, Start-Class and C-Class, or Bottom-Importance and Low-Importance. This is kinda analogous to the reasoning adopted by the U.S. Sentencing Commission in setting up a Sentencing Table with 43 overlapping levels; they figured it would prevent steep cliffs that would cause a lot of legal wrangling and appeals over whether, say, the defendant possessed 49 grams of crack or the 50 needed to push him over the edge into a much harsher sentence.

Even if it were just a binary notable/non-notable dichotomy, a field for importance differentiation could allow for some filtering and thereby make it possible to dispense with deletions of content that some people find useful but the bulk of people would rather not see. The goal is to minimize deletion, and even to eliminate it with the exception of that which is legally required to be deleted, and to minimize the harmful effects of such a policy. The question has to be asked, "Why do we want to delete non-notable stuff?" It's because we don't want it hanging around the wiki. But if you can avoid having it interfering with what you're trying to accomplish, it's as good as if it weren't there. So in that way, everyone gets what they want (or most of what they want, anyway). Tisane (talk) 18:52, 30 March 2010 (UTC)

This is now bug 23016. Tisane (talk) 10:41, 31 March 2010 (UTC)
No, it's because if every non-notable concept was allowed to split into its own article, instead of remaining as part of a larger, notable concept, then the encyclopaedia would devolve into a category/outline tree containing only poorly sourced (if at all) stubs and spam. OrangeDog (τε) 12:21, 31 March 2010 (UTC)
Notability is only indirectly related to the importance ratings used by Wikiprojects. It is entirely possible for an article to be about an obviously notable topic to be rated low-importance by a project simply because it isn't strongly connected to the the project. Michael Jackson, for example, is an obviously notable person, however WikiProject California considers the article to be of low importance with regard to the state of California. Assessment isn't related to notability at all. Assessment is a measure of article quality. There are plenty of high-quality articles about barely notable subjects and plenty of stubs about extremely important topics. Proposing this as an end run around notability is going to attract much more opposition then just as a technical change to improve metadata. Mr.Z-man 16:12, 31 March 2010 (UTC)
I'm not sure there's any point to having an importance field for each article (as opposed to the current scheme in which each WikiProject assigns its own measure of importance) except to help implement an alternative to deletion of non-notable articles. WikiProjects can probably agree on quality metrics more readily, though.
@OrangeDog: If we get rid of WP:N but keep W:V, might that not take care of the "poorly sourced (if at all)" pages problem? Also, are stubs worse than no article at all? Lastly, notability policy is not necessarily needed in order to ensure consolidation of stubs into a larger article; there are notable subjects that nonetheless have no potential to grow beyond a stub, and we therefore merge and redirect. The same can be done with non-notable subjects. A list of garage bands in Detroit, for instance, is not necessarily a bad thing, if each item is sourced. Tisane (talk) 13:28, 1 April 2010 (UTC)
The problem is that you're trying to turn what could be almost a no-brainer technical change (I can't really imagine why anyone would oppose a system that can cleanly and completely replace the current template-based system for wikiproject tagging - if that's all it does) into a massive, sweeping policy change that would basically deprecate Wikipedia:Notability and all the more specific notability guidelines, would probably turn Wikipedia:What Wikipedia is not into a shell of what it is now, would require major revision to Wikipedia:Deletion policy, Wikipedia:Criteria for speedy deletion, Wikipedia:Proposed deletion, and the Wikipedia:Articles for deletion system, possibly require changes to other policies (any policy that mentions deletion), and then create a whole new policy on the use of this new system as well. You saw how much opposition there was for PWD, do you honestly think you're going to get support for all this? Mr.Z-man 01:08, 2 April 2010 (UTC)
Well, it's envisaged as being a long-term plan. PWD has also morphed into a long-term plan. Tisane (talk) 16:42, 4 April 2010 (UTC)

Addition not replacement

I was originally resistance to this idea, but having given it some thought, the idea is looking more attractive. The problem in the way it is presented here in the above discussion as a replacement for the WikiProject tags. What could work I believe is to have these new database fields in ADDITION to the WikiProject tags. Then you don't have an issue of forcing combined information into one field or changing all the bot stuff that currently exists etc. A more minor point is about the word 'importance'. This almost definitely wants to be checked with previous discussion because some WikiProjects will not(and do not) use the word importance. {{WikiProject Economics}} and {{WikiProject British Royalty}} use the word 'priority' for example. And while an addition of a new database field may not be WikiProject related, the objections to the use of 'importance' will no doubt eventually become a valid point. Regards, SunCreator (talk) 17:15, 4 April 2010 (UTC)

Logging in

I can browse and edit wikipedia as an IP address, but for the last several weeks, if I try to log in, then it says "Log in successful" but the page won't load, and then wikipedia becomes completely inaccessible. Once this happens, I try clearing all the cookies and cache but there is still no connection to wikipedia. I try switching to a different browser but once wikipedia won't load then it won't load via any browser. The only thing that seems to work is: logging in, closing the tab, wait for about 10 minutes then try wikipedia again. Why is this happening? Why the 10 min wait?

Sometimes switching the broadband on and off again in order to get a new dynamically assigned IP address works, but not always. I've tried turning off all firewall and anti-virus software but the problem persists even then. The problem is erratic - sometimes the login works fine. I'm from the UK on a BT broadband. Is there a problem between Wikipedia and BT? When this happens I can still access the rest of the internet and use a web-proxy to browse Wikipedia (but not edit). BT says it must be something at Wikipedia causing this. Is Wikipedia blocking connections from BT? Is BT secretly blocking access to Wikipedia? 86.149.236.34 (talk) 12:00, 3 April 2010 (UTC)

What do you mean by “the page won't load”? Do you get any error? Or is the loading taking forever? Svick (talk) 12:42, 3 April 2010 (UTC)
I mean it takes forever, and if I close the tab and try to load any other wikipedia page then it comes up with a conection error message. 86.149.237.193 (talk) 14:26, 3 April 2010 (UTC)
It is also possible that you have CSS or JS installed that is broken. An administrator can help you if you give your username. If you don't want to do that here, I suggest going into IRC. Do "!admin I need help" and wait until an administrator is available to help you. —TheDJ (talkcontribs) 13:26, 3 April 2010 (UTC)
I've started new usernames. The same problem persists with any new username. 86.149.237.193 (talk) 14:26, 3 April 2010 (UTC)
Can you try and see what happens if you use something like the german Wikipedia? And try using Wikinews. —TheDJ (talkcontribs) 15:29, 3 April 2010 (UTC)
OK, the problem just occurred again. So I tried the german wikipedia and wikinews and neither were available. I just got the error message: "Unable to connect Firefox can't establish a connection to the server at www.wikinews.org." Although both were perfectly available via a web-proxy. This time the problem happened not when I tried logging in, but when I wanted to use the wikipedia search box and realised javascript was off, so I switched javascript on in the web browser and hit page reload, and this is when the problem started. Takes forever to load and then all wikipedia (and it seems wikimedia) sites come up with the error message "Unable to connect Firefox can't establish a connection to the server at ..." So I switched javascript off again and tried to reload but still got the same problem. After a couple of minutes it started working again - so not as long a wait as when logging in, but the same sort of problem. 86.149.237.225 (talk) 18:54, 3 April 2010 (UTC)
That is really strange. The fact that you have no such issues using a proxy, is indicative that the problem really is with BT. An explanation why all servers are affected might be that they all use the server bits.wikimedia.org to host their Javascript and CSS. I strongly encourage you to take this up with BT and just push their helpdesk to make sure you have someone who knows what they are talking about, instead of just the simple help desk guys. —TheDJ (talkcontribs) 12:31, 4 April 2010 (UTC)
Well it wouldn't be the first time BT has covertly interfered with internet traffic. I wouldn't be surprised if BT is doing some packet-monitoring that is slowing everything down to the point that it doesn't work at all. If this is down to BT traffic-shaping malfunctions or deliberate traffic blocking then I don't expect BT will admit to it. According to the article on Phorm, they originally denied conducting clandestine trials to profile the information customers were browsing. 81.132.172.8 (talk) 17:46, 4 April 2010 (UTC)

Video

Resolved: Video re-encoded at 400x300 resolution using Firefogg, which fixed the problem. PleaseStand (talk) 15:35, 4 April 2010 (UTC)

I've uploaded a video and added it to Steinway & Sons in the section "Expansion". When the video is not playing there is no picture, just a red cross and the text "Steinway piano - Duo-Art.ogg". Do somebody know what is wrong? Rerumirf (talk) 20:28, 3 April 2010 (UTC)

First thing that comes to my mind is that some part of Mediawiki thinks that all .ogg files are audio, so it doesn't generate the thumbnail. Try reuploading it with the .ogv extension. Svick (talk) 20:35, 3 April 2010 (UTC)
I've uploaded it as a ".ogg" file because the video in the article Sun also is a ".ogg" file. I don't think that ".ogg" is a problem but thank you for helping anyway. Rerumirf (talk) 20:43, 3 April 2010 (UTC)
Generation of the thumbnail simply failed for some reason. I'm not sure if there is a way to deal with this, other than reuploading. If the file still doesn't have a thumbnail, it might be that the file has a problem that trips up the thumbnail generator. —TheDJ (talkcontribs) 21:13, 3 April 2010 (UTC)
Hmm, when trying with native HTML5 player + XiphQT, I notice that the file is not seekable using that software. If the thumbgenerator has that same problem, it would be logical that no thumbnail can be created. The file needs to be remuxed. —TheDJ (talkcontribs) 21:17, 3 April 2010 (UTC)
– I don't know what "native HTML5 player + XiphQT" is. Rerumirf (talk) 21:24, 3 April 2010 (UTC)
Not really relevant. The point is that if there is one player that doesn't understand the video, there is a likely possibility that there is more software that doesn't understand the video. Implying that this video was not properly created according to the ogg/theora specifications. —TheDJ (talkcontribs) 21:28, 3 April 2010 (UTC)
– At Commons the video has the same problem: no picture but a red cross and some text. Rerumirf (talk) 21:27, 3 April 2010 (UTC)
What program did you use to generate the ogg? If it's just a matter of re-encoding the file then I'm sure someone will be able to help. A bug can also be filed to get the problem fixed at source. Chris Cunningham (not at work) - talk 22:59, 3 April 2010 (UTC)
The video is made in "Windows Movie Maker". After making the video I changed the file to a ".ogg" on www.media-convert.com. Rerumirf (talk) 00:55, 4 April 2010 (UTC)
Which is probably the problem I suspect you now have an wmv file in an ogg container. You need to convert it to a Theora file.©Geni 01:33, 4 April 2010 (UTC)
No, the file is using the Theora+Vorbis codecs or else it would not have played in Firefox (native HTML5 rather than Java player). The Java player also does not support the WMV codec. PleaseStand (talk) 03:34, 4 April 2010 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────Re-encoded video using Firefogg, scaled down from 640x480 to 400x300. The scaled-down version is now in the article, and its thumbnail does appear. For future reference, use, as recommended by commons:Help:Converting video#Ogg-Theora - Multi-platform solutions, Miro Video Converter or Firefogg, preferably on the original file you got from Windows Movie Maker. Both use the same ffmpeg2theora encoder program and both should work on both Windows and Mac. The Firefogg option, although it requires that you use the Mozilla Firefox web browser when encoding the video, is particularly useful in that it will automatically downscale your video to reduce file size, according to its default settings. The Miro converter has no option to do so. As a note: I noticed when looking at the Codec Information of the original file that the file was encoded using libavcodec, known for producing broken files. PleaseStand (talk) 05:04, 4 April 2010 (UTC)

That last part is a rather useless remark, since libavcodec is the key element of ffmpeg, and used by all of the software you listed above (as well as most codec packs actually) —TheDJ (talkcontribs) 12:23, 4 April 2010 (UTC)
True. It's actually libavformat that does the multiplexing in VLC, the software program I meant to refer to. PleaseStand (talk) 15:35, 4 April 2010 (UTC)
Thank you PleaseStand, as I see it you have fixed the problem. Rerumirf (talk) 12:48, 4 April 2010 (UTC)

section references possible?

This may be beyond the software's capabilities, but is there anyway to convince the {{reflist}} template (or some equivalent) to use only the references in a given section? I ran into this concern on a talk page, where I wanted to display only the references for a particular section of text under review (not every reference that was on the talk page). --Ludwigs2 20:16, 4 April 2010 (UTC)

See Template:Reflist#Multiple uses. {{Reflist-talk}} is designed for talk pages and explicitly closes the reference list. ---— Gadget850 (Ed) talk 20:21, 4 April 2010 (UTC)
ah, that's cool... thanks. smile --Ludwigs2 21:08, 4 April 2010 (UTC)

Talk page comments on same indent level are not vertically separated as much as other comments—bug?

I've noticed for a while that if I add a talk page comment on same indent level as the one above it, I must add a linespace first, or the rendering doesn't vertically separate those two comments as much as it does comments on different levels of indentation. This discrepancy can produce the impression that the lower comment is a continuation of the higher, because the line spacing is the same as when a line wraps within a paragraph. The two different vertical spacings can be seen in the example at Help:Talk_page#Indentation (second table, George's reply to Jane is too high: compare it with her reply to John above). In that example, it doesn't matter too much because the comments are so short they don't reach the right margin. But typically, comments do reach the right margin, and, depending how long the last line is, at a glance it's not always obvious that there's a break between two comments in this scenario. Is this a wikimedia bug? Or perhaps a css issue? I expect there might be a way to fix my css to make it look different to me, but obviously any fix should really produce the correct rendering for the default user, not just those who fix their own view. Anyone know a global fix for this? (Same in FF, Opera and IE, btw.) PL290 (talk) 15:22, 1 April 2010 (UTC)

I think it's working as intended.
So that people can seperate thoughts on different lines.
Use a line break to seperate yourself from the poster above you. –xenotalk 15:27, 1 April 2010 (UTC)
Interesting that you consider the feature to allow that; I'd have said it afflicts it in just the same way! Those are paragraphs too (and may of course extend for more than a single line), so paragraph spacing (as opposed to wrapped line spacing) would be appropriate there too. So you think we should merely update Help:Talk_page#Indentation and its examples to advise adding an extra linebreak in this scenario? PL290 (talk) 16:36, 1 April 2010 (UTC)
Putting the blank line in between seems to work for me (this post is the evidence). I do not think MediaWiki has "real" indentation. What most people call indentation is a special case of the definition list syntax where there is no defined term, not semantically correct. (Defined terms are lines starting with semicolons.) The closest we could have to "real" talk page indentation (without changing MediaWiki) would be to create a template similar to User:PleaseStand/Tpm. The first parameter is the number of half-tabs to indent by; the second (2=) is the message. Use <p> to start a new paragraph. The result has a line separating each post to a talk page. Of course all this is avoided in LiquidThreads, but when would that be implemented? PleaseStand (talk) 19:27, 1 April 2010 (UTC)
LQT is already enabled on some Wikimedia wikis, such as mediawiki.org (where it's optional page-by-page). I'd be mildly surprised if it weren't enabled optionally on most of the wikis within a year, although performance might need some work for the biggest ones. —Aryeh Gregor (talk • contribs) 18:46, 2 April 2010 (UTC)

If this really bothers anyone, you could change the CSS to adjust spacing however you like. The : feature isn't really meant for indenting, though, it's meant for definition lists, like:

Item 1 
Description/definition/etc.
Item 2 
Description/definition/etc.
Item 3 
Description/definition/etc.

The spacing is meant for this case, not for talk page discussions. —Aryeh Gregor (talk • contribs) 18:46, 2 April 2010 (UTC)

See post above ("What most people call indentation..."); I already noticed that. Definition lists have some correct uses, for example, on my user page, and in glossaries. They are mostly used incorrectly though. And by the way, the CSS (to "fix" the formatting where a blank line is not left) is dd { margin-top: 1em; }, but that is far from perfect. It breaks multi-paragraph comments up because of the semantic incorrectness. PleaseStand (talk) 04:35, 3 April 2010 (UTC)

Perhaps we could have lines beginning with “:” render as some other element in the absence of a <dt> (line beginning with “;”). I might have used something like:

div.threaded div.threaded {
        margin-left:2em;
        }
such that
Foo
:Bar
::Baz
becomes
<div class="threaded">Foo
<div class="threaded">Bar
<div class="threaded">Baz
</div>
</div>
</div>
becomes
Foo
Bar
Baz

AoV² 11:32, 5 April 2010 (UTC)

That's what I was thinking of, but it would require changing MediaWiki. That is still a good idea. Note that "indentation" is not limited to threaded discussions: some articles used it (as of the March 12 database dump, 30028 mainspace pages) in hatnotes instead of the proper templates (see Wikipedia:Village pump (technical)/Archive 57#Dablink for section hatnotes). Also, the initial post in a discussion often does not begin with a colon, so this would change the way discussion pages work, albeit slightly. The main concern would be whether old discussions would be too badly broken. If that is a concern, pages from before the change would have to be updated.
Alternatively, it might be possible to put a tag on all new talk pages to trigger the behavior, such as wrapping each thread in <thread></thread> tags. That last option has the advantage that no colon before the initial post would be necessary, exactly as you describe above. PleaseStand (talk) 01:11, 6 April 2010 (UTC)

Error on reversion

Resolved: Seems to be OK now Rodhullandemu 21:53, 5 April 2010 (UTC)

I have a persistent but intermittent error message "Reverting page: couldn't grab element "editform", aborting, this could indicate failed response from the server", and can see no reason for it. Any ideas? Cheers. Rodhullandemu 23:47, 4 April 2010 (UTC)

You have this with Twinkle revert or something ? What action are you performing ? —TheDJ (talkcontribs) 23:57, 4 April 2010 (UTC)
I have Twinkle here and a shedload of other stuff. This is a recent issue, however. Rodhullandemu 00:02, 5 April 2010 (UTC)
But if it says "reverting page", you have to have clicked something. What action are you performing? —TheDJ (talkcontribs) 00:15, 5 April 2010 (UTC)
I have options on viewing a diff to "[rollback (AGF)] || [rollback] || [rollback (VANDAL)]"; it's mostly on AGF rollback that I get this message. Rodhullandemu 00:20, 5 April 2010 (UTC)
That's WP:TWINKLE. Suggest you complain there. Ale_Jrbtalk 16:56, 5 April 2010 (UTC)

WTH happened to Twinkle?

Resolved

A user notified me about a Twinkle revert to an article in which I didn't assume good faith and marked the edit as vandalism. Just now, I notice that the Twinkle tabs at the top aren't visible anymore. Did a user globally remove my rights to the program (and Friendly), or is it some technical issue? Schfifty3 22:58, 5 April 2010 (UTC)

Sorry, that was my fault. —TheDJ (talkcontribs) 23:14, 5 April 2010 (UTC)
Thanks. The only question now, though, is: How do I reactivate it? Schfifty3 23:17, 5 April 2010 (UTC)
WP:BYPASS should fix any remaining issues. If it doesn't, do please let me know. —TheDJ (talkcontribs) 23:33, 5 April 2010 (UTC)

::::Do I just go to my monobook page and hit the * (purge) tab? I'm still in a confused mode, sorry. :-) Schfifty3 23:39, 5 April 2010 (UTC) Got it, never mind. Finally working! Schfifty3 23:43, 5 April 2010 (UTC)

Subst bug

I ran into a rather confusing bug recently,[17] where editing one section (#60) caused the deletion of many preceding sections and the pasting of my signature at the end of a previous entry. A person could easily miss such a problem, imagining that a small edit in one section is too simple to go wrong. What I know is that it involves an unpaired {{subst: present in the saved article text, which becomes activated and substituted in the course of a later edit. What confuses me is that the text I added did not contain }}, which is the only way I've managed to activate it so far when fooling with the archived version preview.

A related issue is presented at Wikipedia:Village_pump_(technical)/Archive_60#.26lt.3Bref.3E_and_subst: where the subst remains in a reference section.

Because of the potential for loss of content or perhaps even recriminations over deletions, I think it might be useful if someone set a bot to seek and destroy loose "subst"s that are not between pre or nowiki tags. Wnt (talk) 23:40, 5 April 2010 (UTC)

This is so weird

Image:Mafe SN.JPG is not showing up on african cuisine. Someome please fix.174.3.123.220 (talk) 00:38, 6 April 2010 (UTC)

OK, wow, that is weird. It shows up in preview, but in the article itself it's just text. --Golbez (talk) 00:49, 6 April 2010 (UTC)
Aha, someone found the problem; a missing ] earlier in the article. --Golbez (talk) 00:50, 6 April 2010 (UTC)
Yes, fixed in [18]. You must have made a preview of a section edit after the section with the error. PrimeHunter (talk) 00:53, 6 April 2010 (UTC)
(edit conflict)
OK. I'm fixing it as we speak. What's wrong is that you linked it to the wrong Wikipedia namespace ("Image"). There is no "Image" namespace, only "File". I'm not even sure why the others worked, it may have just been a system fluke. Oh, and I fixed the heading. Go ahead, say I have OCD. Hmmwhatsthisdo (talk) 00:55, 6 April 2010 (UTC)
"Image:" is a former name for "File:", and "Image:" still works. The fix was to add a missing ']'. You must have clicked edit after the fix was made and thought it was your change which fixed it. PrimeHunter (talk) 00:59, 6 April 2010 (UTC)

Dynamic IP

When using the linked RIR tools, is there anyway to confirm that the queried IP is a dynamic? Or is there no way?  A p3rson  01:33, 6 April 2010 (UTC)

Nuke?

While attempting to find a speedy deletion tag on an old WP:VP/PR archive, I encountered a discussion about enabling mw:Extension:Nuke. Has this ever been activated here? I've never heard of it before, and I don't know how to look at an extension that has been implemented. Nyttend (talk) 01:51, 6 April 2010 (UTC)

Special:Nuke. I don't know what appears there, though, I'm not an admin. Gary King (talk) 01:55, 6 April 2010 (UTC)
(edit conflict) Yeah, Special:Nuke is enabled here. I remember this AN discussion rather well. — The Earwig (talk) 01:57, 6 April 2010 (UTC)
Special:Version lists the enabled extensions. --MZMcBride (talk) 02:06, 6 April 2010 (UTC)

New village pump discussed

See Wikipedia talk:Village pump (development). The name may change. ... "Idea lab" or "Think tank" or "Idea Workshop" or something. --Timeshifter (talk) 19:14, 6 April 2010 (UTC)

Duplicate section names issue

I'm working on an article with duplicate subsection names, where for example I have "Middle East" under "History" and "Middle East" under "Current status." When editing the latter section, after I am finished editing and save my changes, I am returned instead to the first section! —Khin2718 18:43, 6 April 2010 (UTC)

Probably a known issue and only a problem when using section links from the watchlist or the return-to magic. The table of contents will appropriately append a "_2" to the second section when clicking from there. –xenotalk 18:45, 6 April 2010 (UTC)
Nevertheless, this should be avoided wherever possible, as it means that anchored links from other articles may go to the wrong section. Consider retitling one or both headers to avoid this. Chris Cunningham (not at work) - talk 11:04, 7 April 2010 (UTC)

Gadgets not working

Some of gadgets that can be enabled in Special:Preferences do not appear to be currently working, such as the UTC clock in the upper right corner and the [edit] button to edit the lead section. I am running the Vector skin with Firefox 3.6 on Mac OS X 10.6.2. NERDYSCIENCEDUDE (✉ messagechanges) 22:30, 6 April 2010 (UTC)

Also, Twinkle and Friendly aren't showing up underneath the down triangle when I hover over it. NERDYSCIENCEDUDE (✉ messagechanges) 22:38, 6 April 2010 (UTC)
Huh, now it's working again. NERDYSCIENCEDUDE (✉ messagechanges) 23:33, 6 April 2010 (UTC)

Bots and Logging In

(Crosspost from WP:BON): Please note, about 10-15 minutes ago there was a security change to the login system (Which is now live on all WMF wikis), which will break change the current login systems (API users included), For more information see: bugzilla:23076. Peachey88 (Talk Page · Contribs) 00:43, 7 April 2010 (UTC)

This is causing AutoWikiBrowser not to work. --Auntof6 (talk) 01:19, 7 April 2010 (UTC)
AWB will need to be updated. — Carl (CBM · talk) 01:31, 7 April 2010 (UTC)
Fix was already pre-prepared for this. Just need to get my desktop machine back up online and I'll get AWB sorted. Reedy 08:20, 7 April 2010 (UTC)

about that search-box suggestor

Suppose an article′s title is Föøbåŕ and you search for Foobar. The search-box will “suggest” Föøbåŕ for you, sure—but it is useless to determine whether Foobar exists as a redirect, or not at all. I know this because I typed Loys Delteil in the search-box and it suggested Loÿs Delteil, which is cool enough, but see… I first thought it determined this rough equivalence by resolving a redirect, but after striking the “enter” key I found that none existed.

Note: for the sake of discussion, refrain from creating that redirect quite yet.

I referred then to redirects which I knew existed (such as GdanskGdańsk) and the results were indistinguishable from the one above. It seems the only case in which the titles appearing below the search-box provide clear information about multiple page-names which decompose to the same base-letters is when they exist separately as proper articles (e.g. Socrates and Sócrates, not that that′s a terrific idea either).

This and the way the titles tend to defy alphabetical order (and gravity) have me thinking about creating sort of a low-tech replacement gadget which behaves in a predictable fashion, so let me know if somebody doesn′t have a working version already. ―AoV² 12:11, 6 April 2010 (UTC)

You can just use the address bar... OrangeDog (τε) 20:16, 6 April 2010 (UTC)
The suggestions were created to be most useful for average users to detect the most likely thing they want to type in. Thus, as you noticed they are not very well suited for finding redirects, or search the full alphabetical listing. You can use Special:Allpages and Special:PrefixIndex for that (redirects are in italics). Before this search suggest has been implemented a couple of tools existed that used those, one of them was User:Zocky/AutoComplete.js. --rainman (talk) 00:27, 7 April 2010 (UTC)

Here I came up with something short enough to be comprehensible: User:AoV2/suggestor.js. I suppose the key differences are that it uses Prefixindex, it sorts in alphabetical order, and does not float. Let me know if anyone else has a use for non-fuzzy lookup. ―AoV² 05:39, 9 April 2010 (UTC)

Namespace templates

I was wondering if a template similar to commons:Template:Namespace and commons:Template:Namespaces could be useful here. I recently noted that {{Uncategorized}} still uses a #switch function for this, so it might be handy to use a template there. After searching for templates, I only found {{Pagetype}} (which is however not the same), so such a template doesn't appear to have been created yet. --The Evil IP address (talk) 16:45, 6 April 2010 (UTC)

All those templates do is provide a slightly different output than the magicword {{NAMESPACE}}... The parserFunction would still be required. –xenotalk 16:49, 6 April 2010 (UTC)
{{pagetype}} fulfils essentially the same function; because of its different history (spun out of {{WPBannerMeta}}) it has a slightly different syntax and default, but it is still used and usable in templates, see for instance {{db-meta}}. While the behaviour of pagetype could probably be tweaked to be more useful in such cases, I'm not sure if adding another parallel template would be helpful. Happymelon 19:40, 6 April 2010 (UTC)
We already have loads of templates that perform near-identical functions. What's the harm in adding more? :P I just noticed {{pagetype}} doesn't have any categories. What would be appropriate here, Category:Namespace manipulation templates perhaps? Reach Out to the Truth 01:38, 7 April 2010 (UTC)
Because you end up with a mass of poorly-maintained, poorly-documented and hard to find templates. If ever possible, add parameters to existing templates. OrangeDog (τε) 22:46, 8 April 2010 (UTC)

Import XML dump

I am looking for a mechanism to apply edits to pages from a Special:Export XML dump file. What I am looking for is not the same as Special:Import, which is oriented toward importing pages with revision histories from other wikis. What I want is to take the page text from the XML dump for each page and apply it as an edit to the page it was created from.

The use case I have in mind is as follows:

  • Use Special:Export to download a number of pages (most recent revision only) as an XML file.
  • Modify the text of the pages in the XML file, using whatever tools are appropriate.
  • Apply the changes back to the Wiki from the modified XML files.

Is such a tool available? -- JPMcGrath (talk) 12:43, 7 April 2010 (UTC)

I don't think it is. Why do you want to do it exactly this way? Isn't the API sufficient? Or using AWB? In the case of AWB, you can either write a module (in C# or VB.NET) or use some external tool directly. Svick (talk) 16:29, 7 April 2010 (UTC)
I have made some systematic changes to large numbers of articles with AWB, using an external AWK script to perform the changes. In order to test the scripts, I created a wrapper that processes an Export dump, feeding each page to the script. By the time I actually performed the changes with AWB, I had run all the pages through the process multiple times and created diff files and other reports of the changes to ensure they were correct.
After all of the testing and verification, I had to run the changes through AWB, which took several hours when I recently made a change on 559 pages. I had to look at each individual change and click "save" for each one. If you have ever done that, you know how mind-numbing it can be, and how easy it is to not really examine each change. And the result is somewhat less secure, since if the page had been changed since I created the last dump, I would not know it.
A dump-upload tool as I conceive it would be more efficient and more secure. It would verify that the page had not been modified since the dump was made, so there is no possibility that new changes would create a situation that was not planned for.
From the above, you can see that AWB does not satisfy the need. I think that the API is exactly what would be needed to implement the tool. It looks like it would be rather simple, assuming some of the available libraries work as well as it looks like they should.
Although this does not sound like the traditional view of a bot, in that the changes are all reviewed by a human before they are made, it still would use the same access methods that a bot uses. It also would be a potent tool, if someone chose to abuse it, so I am thinking it would make sense to bring the idea to WP:Bag before writing such a tool. Does that make sense to you?
-- JPMcGrath (talk) 19:10, 7 April 2010 (UTC)
I think AWB is still the easiest way. I don't know how you loaded your changed articles into AWB (copy-paste?), but AWB can be quite easily customized, so it can, for example, load the article text from XML file, check the date of last revision (if it doesn't match with the one in XML, skip and log it) and save it. Automatic saving is enabled only if you are approved as a bot in AWB. I (or possibly someone else on WT:AWB) can help you if don't know how to program in .NET. I don't have experience with BAG, but their approval shouldn't be necessary if you actually review each change and if you don't do any mass changes. Svick (talk) 21:55, 7 April 2010 (UTC)
Why do you think AWB would be easier? I don't really see much there that would make it simpler than using a bot library and the API calls. What specific advantages would outwiegh the additional costs and design constraints that come from working within a framework such as AWB? -- JPMcGrath (talk) 22:28, 8 April 2010 (UTC)
If you want to use your own code, you would have to actually write the code that communicates with the API and parses the answers. It's nothing difficult, but I think using already written code saves you time. And you can either use AWB directly with a custom module or write your own application using classes from WikiFunctions.dll. Svick (talk) 22:55, 8 April 2010 (UTC)
Most non-ancient bot frameworks have some method for using the API. There is a list of the various frameworks in different languages. Mr.Z-man 23:08, 8 April 2010 (UTC)

Spaces in URL?

The Official Charts Company finally released an archive, but it's got a horrible technical glitch: it actually expects space characters as a part of the URL, and will not allow "_" or "+" to be used instead. That means we can't create links, because [http://www.theofficialcharts.com/artist/_/Cheryl Cole] expands a link named "Cole" to http://www.theofficialcharts.com/artist/_/Cheryl. I can get around this in {{singlechart}} by skipping the urlencode function and requiring the user to pass in "%20", but that's pretty ugly, and it makes the URL title display with the %20 as well. Test version is at User:Kww/singlechart, with the test driver at User talk:Kww/singlechart. Can anyone figure out a way to do this?—Kww(talk) 21:09, 7 April 2010 (UTC)

Characters can be unsafe for a number of reasons... The space character is unsafe because significant spaces may disappear and insignificant spaces may be introduced when URLs are transcribed or typeset or subjected to the treatment of word-processing programs... All unsafe characters must always be encoded within a URL.
I don't believe there is currently a solution other than the method you suggest. {{urlencode}} and friends were not designed with such a high level of abject stupidity in mind... :P Happymelon 21:32, 7 April 2010 (UTC)
Is there at least a way to change "Cheryl%20Cole" back to "Cheryl Cole"? I can make things work (albeit kludgily) if I can accomplish that.—Kww(talk) 21:56, 7 April 2010 (UTC)
You could use one parameter for each part of the name, something like {{chart template|Cheryl|Cole}} and in the template [http://url.to.charts/{{{1}}}%20{{{2}}} {{{1}}} {{{2}}}]. Svick (talk) 22:11, 7 April 2010 (UTC)
Hard to make general purpose: What works for "Cheryl Cole" won't work for "Bonzo Dog Doo-Dah Band", and the game is over by the time you get to "The Ogden Edesl Wahalia Blues Ensemble and Mondo Bizzario Band" (who fortunately never had a hit in the UK).—Kww(talk) 22:15, 7 April 2010 (UTC)
(ec)How about emailing or filling in the feedback on the site. It's a new design of the site, they may not yet know of the technical problems spaces in URL's create. It also may get fixed in the coming days. Regards, SunCreator (talk) 22:17, 7 April 2010 (UTC)
I've already done that, but I'm not holding my breath. I've gone through similar issues with Stefan Hung, and, while he is sympathetic, I still can't create a link to Beyoncé Knowles on his site because of non-standard URL formatting issues.—Kww(talk) 22:23, 7 April 2010 (UTC)
As initially described, this would be easy if {{#replace:{{#urlencode:{{{1}}}}}|+|%20}} were allowed. It isn't though; a proposal to enable StringFunctions was rejected since ParserFunctions were not intended to become a general-purpose programming language. Apparently it was thought of to use Lua as a more flexible template programming language, although it would require a compiled Lua binary or PHP extension to work. PleaseStand (talk) 23:12, 7 April 2010 (UTC)
bugzilla:22474 --MZMcBride (talk) 02:36, 8 April 2010 (UTC)
Thanks for the information about that addition to the urlencode function that allows using PHP's rawurlencode function instead of urlencode. Of course it hasn't made it to the live site yet; this change is only 2.5 hours old as I am writing this message. PleaseStand (talk) 02:57, 8 April 2010 (UTC)
This works fine... Cheryl Cole at theofficialcharts.com. Why would you ever want a bare url in an article? OrangeDog (τε) 11:33, 8 April 2010 (UTC)
The problem is that {{singlechart}} can't generate that for the UK in the same way that it does for other countries, because there's no way to pass in the "artist" parameter and generate "Cheryl%20Cole" for the URL and "CHERYL COLE - The Official Charts Company" for the page title. The documentation at {{singlechart}} can be a little intimidating, but taking a look at Beautiful (Christina Aguilera song)#Charts and certifications makes how it works for other archives pretty obvious, and I want to add support for this one.—Kww(talk) 14:53, 8 April 2010 (UTC)

Is rev:64726 what you were looking for? Once that is live here, {{urlencode:Cheryl Cole|PATH}} should result in Cheryl%20Cole. Reach Out to the Truth 19:29, 8 April 2010 (UTC)

It will certainly work for my purposes. How can I find out a timeline for actual implementation?—Kww(talk) 22:54, 8 April 2010 (UTC)

api watchlist in json format

Despite being quite firmly logged-in, I′m getting the message “foo({"error":{"code":"wlnotloggedin","info":"You must be logged-in to have a watchlist"}})” (but only when I nominate a call-back function). I thought this might be for security reasons, but noticed it still fails after I provide appropriate values for wluser and wltoken. So, what′s the trick? ―AoV² 06:58, 8 April 2010 (UTC)

"callback - If specified, wraps the output into a given function call. For safety, all user-specific data will be restricted."
You cannot use callback for anything but anonymous actions. This is because it would open huge XSS security flaws: with callback, you can load a remote api call as a <script src="remote link"> that executes on load (letting malicious sites send api actions as you, via your browser). The way around it is to get the page via AJAX (which typically cannot send requests across domains), and then stick the responseText into an eval(), eg: eval('callback(' + foo.responseText + ')');. --Splarka (rant) 08:03, 8 April 2010 (UTC)
PS: It does work for me with wltoken and wlowner: api.php?action=query&format=json&callback=foo&list=watchlist&wlowner=Splarka&wltoken=XXXXXXXXXXXX gives foo({"query":{"watchlist":[{"pageid":3252662,"revid":354706456,"ns":4,"title":"Wikipedia:Village pump (technical)"}... --Splarka (rant) 08:10, 8 April 2010 (UTC)

Okay, got that to work but… I don′t suppose there′s any quick way to set the same watchlist token on all wikis? See, I was going to try making something which combines watchlists from several projects onto the same page. ―AoV² 08:44, 8 April 2010 (UTC)

REL and CLASS attributes in links and images[edit]

In order to improve our emitted microformats, it would be useful to be able to add rel attributes to internal (including File:) and external links; and class attributes to links and images; none of which are possible at the moment. Where's the best place to raise this issue, and how should it best be approached? Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 09:36, 8 April 2010 (UTC)

For the time being, both of these requests can be supported by cooking up inline templates for the purpose. In the long run, the [[file:]] container is extensible, so it would presumably be easy enough to add class= support. Internal links are another matter; possibly by altering the behaviour when a link is piped not once but twice? Currently the output inserts the last two "arguments" with a literal pipe between them, as piping|piper shows. Chris Cunningham (not at work) - talk 09:49, 8 April 2010 (UTC)
Thanks. Can you give an example of such an inline template, or how one would work, please? Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:05, 8 April 2010 (UTC)

Exposing the rel attribute would be foolhardy to any extent which allows negating nofollow.

But perhaps the html-tidy or whate′er can be configured to render:

<span class="foo" style="font-size:larger;">[[Bar]]</span>

as:

<a href="/wiki/Bar" title="Bar" class="foo" style="font-size:larger;">Bar</a>

It already does something similar with <font> tags when the link is the only member (actually it turns them inside-out). However, removing the extra tag-layer would cure the common problem where the link looks like crap because its text-color and underline-color do not match. This might be the most versatile approach short of allowing arbitrary <a> tags. ―AoV² 10:29, 8 April 2010 (UTC)

Presumably the parser can be modified to insert a default class/rel into the <a> element when generating the html, use of html tidy is not required and would make things more confusing. Alternatively, some javascript could be used to modify all the relevant elements when the page loads, or a new php module that does the same on server-side. Comment by User:OrangeDog
the rel or class values needed vary (from a limited set, for microformats) depending on the circumstances. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:02, 8 April 2010 (UTC)
There were long discussions about support for these kind of things in January on the wikitech list. The discussion is far from concluded :D —TheDJ (talkcontribs) 13:29, 8 April 2010 (UTC)
Thanks; but that appears to be a discussion of RDFa and microdata - similar to, but not the same as - microformats; and not of rel and class attributes. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 14:19, 8 April 2010 (UTC).
True, but it mostly comes from the idea to implement rel-license. (CC uses rel-license +RDFa). There are likely more related bugzilla tickets, but this is one of themTheDJ (talkcontribs) 14:38, 8 April 2010 (UTC)
It would certainly be possible for certain values to be prohibited/ stripped-out on rendering. Your surrounding-span idea is clever. 'Nofollow' is evil, as used on Wikipedia, BTW. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 12:00, 8 April 2010 (UTC)

There was an attempt a while ago to allow direct insertion of <a> tags. I don't know what happened to that; Tim Starling didn't like the implementation and it seems not to work on trunk, so I guess it was either reverted, or turned off by default. For images, we could pretty easily add new syntax to image inclusions, like we did for alt and title, although that code is a complete mess last I checked.

Somehow it's not surprising to me that this is being requested by someone who uses a vCard as a signature.  :P —Aryeh Gregor (talk • contribs) 16:49, 8 April 2010 (UTC)