Wikipedia:Village pump (technical)/Archive 13

TeX limitations in MediaWiki?

I seem to have encountered a problem with rendering matrices (see here). The matrix I try to input appears to be too large: if I remove a few columns it works all right. I tried to search MediaZilla but haven't found anything useful. StevenDH (talk) 17:05, 29 December 2007 (UTC)

The following latex code has the same issue:
\documentclass[12pt]{article}
\usepackage{amsmath}
\begin{document}
\[
\begin{bmatrix}
1 & 2 & 3 & 4 & 5 & 6 & 7 & 8 & 9 & 10 & 11 & 12 & 13 & 14 & 15\\
\end{bmatrix}
\]
\end{document}
So I think this is a latex issue, not a mediawiki issue. — Carl (CBM · talk) 20:17, 29 December 2007 (UTC)
Indeed, it seems that the 'bmatrix' environment has a limit of ten columns. I solved it by using ordinary arrays, the result is also prettier. Maybe interesting to mention on Help:Displaying a formula? StevenDH (talk) 23:54, 29 December 2007 (UTC)

Creation of similar usernames

Does anyone know if there's been a change in the way the software deals with similar usernames at account creation? I'm wondering how the account Tasko 0 could have been created when there was already an account called Tasco 0. It was my understanding that a new username which differed by only one character from an existing one could only be created by a sysop. WjBscribe 21:46, 29 December 2007 (UTC)

Anti-spoof prevents the creation of visually similar usernames, e.g. replacing I (captial "eye") and l (lowercase "elle"), or O (capital "oh") with 0 (numeral zero). Single character differences are still allowed provided the resulting names are visually dissimilar. Hence the phonetically similar but visually distinctive Tasko vs. Tasco has never been blocked. Dragons flight (talk) 22:11, 29 December 2007 (UTC)
It would lead to many false positives, such as "face" and "fake". 1 != 2 22:16, 29 December 2007 (UTC)
I don't know who created this account. And what's more even weirder is that the user redirected the userpage to mine. Check here. Tasc0 It's a zero! 00:01, 30 December 2007 (UTC)

Redirect page in Google Earth

If you go to 48°40′04.00″N 124°49′23.78″W / 48.6677778°N 124.8232722°W / 48.6677778; -124.8232722 on Google Earth there is an unlabelled purple dot. When you click on it you get a redirect to Non-breaking space. I checked all the redirects but none appear to have any coordinates in them. Any ideas? CambridgeBayWeather (Talk) 13:55, 23 December 2007 (UTC)

Sounds like a bug in the Google Earth software. [[ ]] (that is, [[ ]]) redirects to Non-breaking space, which is a common character that could be inserted into a link target by mistake. AmiDaniel (talk) 01:46, 26 December 2007 (UTC)
Thanks. CambridgeBayWeather (Talk) 06:42, 26 December 2007 (UTC)
Google shows placemarks for users with the English user interface even when the English article doesn't have coordinates, but one of the interwikied articles do, so the problem may be outside the English Wikipedia. Since they get the data from database dumps and don't say where and when the data is from, it could be anywhere really, or even fixed already, so it's almost impossible for us to find the source. Anyway, the "Contact Wikipedia" link on the left side of the page has a section for reporting problems such as this: Wikipedia:Contact us/Article problem/Google Earth. --Para (talk) 19:22, 30 December 2007 (UTC)

Could somebody help to rename a picture

there are local picture Image:Sophia Loren.jpg that disturb using same name of picture from commons, could somebody help to rename this local picture to some other name, thanks--Musamies (talk) 06:31, 30 December 2007 (UTC)

Download it, reupload it under a different name then request deletion of the old copy. It is not possible to move images using the move page function. MER-C 11:00, 30 December 2007 (UTC)

Change expert-subject template to allow reference to multiple Wikiprojects?

Hello, I'm interested in finding someone who can change the expert-subject template to refer to more than one Wikiproject. I am currently working on the Biphobia article, and there are two expert-subject templates in use, asking for help from Wikiproject LGBT studies, and Wikiproject Discrimination. It would look a lot better if the formatting could be changed so that one expert-subject template could list both Wikiprojects, or even more than two. Can anyone help with this? Thanks! Photouploaded (talk) 15:50, 30 December 2007 (UTC)

Odd behavior in quotes

 
odd behavior

This seems to illustrate something very odd. A quote pushed off to the right side proceeds to walk straight over a section splitter, and what's more, shows the section splitter right under it! It was taken from Commonwealth realms using Firefox (newest version); is there anything that can be done about it? It's unsightly! ~user:orngjce223 how am I typing? 19:14, 30 December 2007 (UTC)

I've adjusted the quote's formatting so it is displayed above the section, instead of beside it. Picaroon (t) 21:54, 30 December 2007 (UTC)
I was hoping for a more general fix (i.e. using a background div for the quote), but this'll probably do. ~user:orngjce223 how am I typing? 22:05, 30 December 2007 (UTC)
Yes, I realize it's not optimal. I'm trying other skins, and, as can be expected, the ones without horizontal lines coupled with sections (eg cologne blue, classic) have no such problem; in chick, meanwhile, the same problem occurs. Would a div placed in {{rquote}} that would hide horizontal lines the quote is running over cause any downsides? Picaroon (t) 22:15, 30 December 2007 (UTC)
The problem is the template. It has the background set to inherit, which in most cases is transparent. EdokterTalk 01:39, 31 December 2007 (UTC)

Notes in page history when page protection expires?

Would it possible to enable a feature (I've been thinking about this for maybe two minutes now, so bear with the vagueness) which makes notes in the page history to show when a temporary protection expired? The issue of whether some edits to a page were made when the page was protected or not has come up in an arbitration case; while it is possible to figure out when the first edit made after the expiration of protection occurred by counting the hours, that is not the easiest thing to do. I'm not sure what form a page-history notification that protection has expired would take, but the implementation of such a thing strikes me as likely to be beneficial in cases like this, and there are no immediately apparent downsides. Thoughts? Picaroon (t) 21:13, 27 December 2007 (UTC)

There is a bot that removes protection tags from pages when they expire (see User:DumbBOT/Protection); this gives an informal reminder e.g. on a watchlist that a protected page is open for editing. This wouldn't be that good for something like an Arbcom case since there would be a delay between the protection expiring and the tag being removed. I think the only way to find the expiry time 'properly' is to work it out from looking at expiry times, as you said. Tra (Talk) 22:02, 27 December 2007 (UTC)
I believe the issue is that nothing "happens" when protection expires; the test for whether the page is protected will start to come back false, but nothing changes in the database. Otherwise it would be easy to fix the problem that when one type of protection expires, the previous type of protection is lost, even if the previous protection is indefinite. This should be fixed at the same time as the problem you mention. — Carl (CBM · talk) 03:11, 28 December 2007 (UTC)
Actually, I think there is some sort of "lazy purge" of expired restrictions from the database. If the function could be modified to add entries to the page history and protection log, this would work fine. Perhaps this would warrant an enhancement request for includes/Title.php Tuvok[T@lk/Improve me] 04:30, 28 December 2007 (UTC)
That is an option, although it might be confusing for users that the message appears not when the protection expires but at some later point in time. — Carl (CBM · talk) 15:19, 31 December 2007 (UTC)

Use of unicode in <math>...</math> environment?

I tried to use unicode characters (such as α, β, →, ...) in a <math> ... </math> environment and texvc failed to parse the expression. I have read that texvc should support unicode with the cjk-latex package. Is the package installed by default at Wikipedia? Shall I do something special to tell that I need the package for unicode support? Thanks in advance for your help. —Preceding unsigned comment added by Hugo Herbelin (talkcontribs) 11:53, 29 December 2007 (UTC)

Well, as a work around, \alpha, \beta, and \rarr in <math> produce:  . MilesAgain (talk) 12:21, 29 December 2007 (UTC)
Yes, I know that there is a workaround. I just say that it would be more comfortable to be able to write (and to read in the editor!), say <math>((α → β) → α) → α</math> rather than <math>((\alpha \rarr \beta) \rarr \alpha) \rarr \alpha</math>.
Shall I understand from your answer that cjk-latex for texvc is not supported in Wikipedia and that there is no plan to support it? —Preceding unsigned comment added by Hugo Herbelin (talkcontribs) 14:11, 29 December 2007 (UTC)
Make it a feature request? I wonder if you could bother say User:Taw about this. O:-) --Kim Bruning (talk) 15:01, 29 December 2007 (UTC)
Texvc has many limitations; it is not a full LaTeX system in any way. But since it works well enough, I don't expect to see any development on it until MathML becomes more widely available. Even then, it will be a while before the TeX system is updated here. I just don't think it's a high priority for developer time. On the other hand, if someone wrote the code to accept Unicode letters in texvc, and it was no less efficient than the current texvc code, it could be sent to the devs for review. — Carl (CBM · talk) 15:03, 29 December 2007 (UTC)
I didn't check myself that it works correctly with texvc, but it seems that the only necessary modification to have utf8 support is to call latex with an extra \usepackage[utf8]{inputenc} in the document. At worst, if one follows this page, texvc does not need to be modified. Only its local installation has.
Thanks for the comments. I will also try to better understand the expectable "roadmap" with MathML. Hugo Herbelin (talk) 17:12, 29 December 2007 (UTC)
Hmm. The TODO file for texvc includes "CJK support". Let me look into that. — Carl (CBM · talk) 17:17, 29 December 2007 (UTC)
The problem appears to be that the lexer in texvc, which reads the input before it is ever send to latex, is not smart enough to handle multibyte characters. The lexer is written using ocamllex - I don't think that supports multibyte characters at all. So it seems like it would be necessary to rewrite the lexer. I don't know whether blahtexml supports utf8, but it has a better chance. — Carl (CBM · talk) 17:38, 29 December 2007 (UTC)
Here are some news... I installed texvc on my gentoo box and here is how it works to get unicode:
  • First, you need the packages latex-unicode, which contains ucs.sty, and cjk-latex installed
  • Then, texvc accepts utf-8 only in mbox's, so you have to write something like <math>3 \mbox{→} 4</math>.
  • And, texvc does not accept utf-8 letters, so that → is ok, but α is not.
Regarding latex-unicode, it does not seem to be installed in Wikipedia, so no hope to make unicode working even partially here.
Regarding the unicode letters restriction, there is a very easy patch to texvc which makes it works.
Regarding the needed mbox, we can certainly change the lexer a bit so that the mbox is not needed. If User:Taw is not going to do it, I can try, if he wishes, to do something. Hugo Herbelin (talk) 12:34, 30 December 2007 (UTC)
How did you get texvc to create a PNG for input like <math>\mbox{α}</math>? — Carl (CBM · talk) 14:39, 30 December 2007 (UTC)
Do you mean: what is the patch to get <math>\mbox{α}</math> working with texvc? If this is the question, the answer is: replace the line
UTF8 -> "\\usepackage{ucs}\n\\usepackage[utf8x]{inputenc}\n"
of file math/texutil.ml of the mediawiki distribution by
UTF8 -> "\\usepackage[mathletters]{ucs}\n\\usepackage[utf8x]{inputenc}\n"
and recompile (with make). Hope it helps. Hugo Herbelin (talk) 15:45, 31 December 2007 (UTC)

<syntaxhighlight lang=""> tag

This seems to be in widespread use but is it really ready?

  • It does not seem to parse very well. For example, here it does not seem to even recognise that keywords or strings are embedded in comments.
  • It does not seem to be documented in any help I can find in Wikipedia. I have had to traipse as far as mw:Extension:SyntaxHighlight_GeSHi still to find little to help a simple but curious editor.
  • The colours it chooses, or are chosen for it, I don't know which, are, being very polite, not quite up to the usual Wikipedia standard of tasteful harmony. Are these adjustable?

Is this even the right place to talk about it? Bill F (talk) 00:51, 31 December 2007 (UTC)

Yeah, this'd be the right place, I think. I've had issues with it as well. Not sure if it's the extension itself, or, HTML_tidy, but, one of the two really mis-renders complex regexes. SQLQuery me! 12:41, 31 December 2007 (UTC)
Yes, I tried using it for User:Nihiltres/Click-to-ImageMap, but it only highlighted the first couple of lines and then left the rest of the script red when I previewed it. It's plainly broken. :( Nihiltres{t.l} 15:03, 31 December 2007 (UTC)

Week 1-2007??

Can somebody look at the macro values for the week number and year? It is used in the Portal:Space exploration/Picture featured picture mechanism and it is now set to "Week 1, 2007"... That does not sound right to me ;-) Awolf002 (talk) 04:08, 31 December 2007 (UTC)

I fixed it with {{ISOYEAR|{{CURRENTYEAR}}|{{CURRENTMONTH}}|{{CURRENTDAY}}}}.--Patrick (talk) 09:03, 31 December 2007 (UTC)
That looks great! I will use that in other places, as well. Thanks! Awolf002 (talk) 12:20, 31 December 2007 (UTC)

Remove extra 0 in date template usage

Is there a way for the age templates to automatically remove an extra zero in the date? See the article Anja Langer for an example. Birth= 03 June instead of 3 June. I am not expert enough to try it myself. Don't even know what template to edit. {{Birth date and age}} {{Birth date}} or {{Age}}? Garion96 (talk) 19:21, 30 December 2007 (UTC)

I just edited the box to replace "03" with "3", that seems to have fixed it. Corvus cornixtalk 21:31, 30 December 2007 (UTC)
True, that does help. However I encountered the same extra 0 on many other articles, so I was hoping a fix in the templates would solve it instead of a lot of AWB edits. Garion96 (talk) 21:42, 30 December 2007 (UTC)
Maybe a bot request? -- John Broughton (♫♫) 22:08, 30 December 2007 (UTC)
For anyone wanting to make the change, one way to go about it might be with the #time ParserFunction.
{{#time:j|2007-06-03}} produces 3
{{#time:j|2007-06-3}} produces 3
{{#time:j|June 03}} produces 3
{{#time:j|June 3}} produces 3
Depending which appears to be more efficient. The first pair needs all 3 parameters, but the second pair needs to use {{MONTHNAME}} first to get the "June" out. –Pomte 09:45, 31 December 2007 (UTC)
#time works for dates from 1 Jan 1970 only.--Patrick (talk) 11:00, 31 December 2007 (UTC)
I changed {{Birth date and age}} so that it removes a leading zero. However, if leading zeros in the template calls are intentional and meant to be shown we can revert the change.--Patrick (talk) 11:49, 31 December 2007 (UTC)
Seems to work great, thanks. I can't image an intentional reason for the extra zero. Do the other templates I mentioned also need to be changed? Garion96 (talk) 12:46, 31 December 2007 (UTC)
I changed {{Birth date}} too, for the other it is not needed.--Patrick (talk) 13:29, 31 December 2007 (UTC)
Again, thanks. Garion96 (talk) 11:33, 1 January 2008 (UTC)

What would happen if a bot tried to edit a fully protected page?

Say a user's talk page is full protected, or is salted, and Betacommandbot tried to inform that user of a non-free image. What would happen?--h i s s p a c e r e s e a r c h 19:20, 31 December 2007 (UTC)

The edit would not succeed, since BetacommandBot's username does not have administrator rights. I don't know whether the bot would go on the the next page, or whether Betacommand would need to restart it. — Carl (CBM · talk) 19:32, 31 December 2007 (UTC)
I think that the bot would just continue on. You should just ask BC about this. - Rjd0060 (talk) 02:59, 1 January 2008 (UTC)

BCBot just ignores the page, and writes it to a log file as being un-able to tag. βcommand 20:46, 1 January 2008 (UTC)

Some way to deal with images screwing up page formatting?

 
Example text.

One of my pet projects is to put lots of highly descriptive images into wikipedia to balance text-heavy technical discussions. However, I have been finding that wikipedia's auto-flow rendering system frequently screws up when there are a multitude of large pictures but not enough vertical text to "balance" the height of the picture.

An illustration of the problem

 
<--An illustration..

Example text. Foo.

Example Heading 1

 
Heading 1 image

The problem is worse yet on widescreen computers, where even a large paragraph ends up stretching out into..

Example Heading 2

 
Heading 2 image

..very long lines of text. If you're reading this on an "average" resolution you probably won't..

Insert another group

 
Insert another..

Example Heading 3

 
Heading 3 image

..notice the effect so I am forcing it here with these short lines of text. The wrong image seems to be next to this section.

Example Heading 4

THIS SECTION HAS NO IMAGE BUT I SEE #3 TO THE RIGHT.

Since the text in these headings isn't "tall" enough to balance the images, the images have skewed out of alignment with the associated heading. This makes for a confusing article that is documenting steps of a process using many many images .

Tabling the issue

It's a horrible hack I'm sure but usually I end up throwing up my hands in frustration with the mess and the image misalignment, and I force the formatting with tiny embedded tables.
 
Table-forced 1
Most other editors seem to hate this, but I don't know what else to do, to keep an image aligned with the text to which it is associated. The built-in freeflow image/page formatting cannot be trusted to handle large amounts of illustrations properly without totally screwing up the page. I don't want to get in a revert war over whether a table is needed to fix this mess.
 
Table-forced 2

While I'm still a somewhat new editor, I'm of the opinion that about the only solution is some deep level rewriting of the code that handles how images are placed on pages. Or maybe some new code feature to force an image to stay aligned with, and inside, the section to which it is associated. DMahalko (talk) 14:11, 1 January 2008 (UTC)

A workaround solution for this is to add <br clear="right"/> at the end of any section with right-aligned picture(s) that are taller than the text in the section. - MTC (talk) 16:16, 1 January 2008 (UTC)
There are also templates {{clear}} and {{-}} that do the same thing, and might be less confusing to new editors. But I suspect the next complaint would be about the excess whitespace this produces; you have to choose one or the other, or else change things around to not use floated images and/or reduce the size of the images that are being used. BTW, it's not "Wikipedia's auto-flow rendering system", it's a basic part of HTML and CSS. Anomie 17:58, 1 January 2008 (UTC)
There's also {{imageframe}} that can be used to combine multiple pictures into a single frame that may help as well. --MASEM 18:44, 1 January 2008 (UTC)

McAfee Stadium

When I try to read the McAfee Stadium article, my browser closes a few seconds after the page finishes loading. What's going on? Dalekusa (talk) 17:12, 1 January 2008 (UTC)

Works for me (Firefox 2.0.0.11/Linux). The article redirects to McAfee Coliseum, which has a number of navigation templates at the bottom. Have you tried the direct link instead of the redirect? -- ReyBrujo (talk) 18:28, 1 January 2008 (UTC)
And what is your browser and operating system? Some browsers throw a temper and close themselves down if they spot an error. The article is standards compliant which suggests it could be related to your browser. x42bn6 Talk Mess 21:00, 1 January 2008 (UTC)
There are errors in this page, see Image:CSS errors McAfee Coliseum.jpg. - Erik Baas (talk) 23:26, 1 January 2008 (UTC)
I didn't manage to check the CSS because the W3 validator for it doesn't seem to work for me, but the XHTML is alright. x42bn6 Talk Mess 23:39, 1 January 2008 (UTC)

Script safe?

Hi, is this script you add to your monobook called twinkle safe?

importScript('User:AzaToth/twinkle.js');

AntiVanMan (talk) 03:48, 2 January 2008 (UTC)

Yes, see WP:TWINKLE for more information on it. Thank you, — xaosflux Talk 05:28, 2 January 2008 (UTC)

Problem with article name

I need to create an article called "Pilgrimage (album)" (by band "VUVR"). Bu there already exist an article "Pilgrimage (album)" by the band "Wishbone Ash". So I would create "Pilgrimage (VUVR album)". But if there will be "Pilgrimage (VUVR album)", there should also be "Pilgrimage (Wishbone Ash album)" instead of "Pilgrimage (album)". How can I rename "Pilgrimage (album)" to "Pilgrimage (Wishbone Ash album)" so that all the existing links work for the new name?Lykantrop (Talk) 12:11, 1 January 2008 (UTC)

Use the "move" tab at the top of the page, then update the backlinks and change the original into a disambiguation page. But maybe there should be an article on Vuvr/VUVR itself first? –Pomte 12:54, 1 January 2008 (UTC)
Pilgrimage (album) could be moved and the pages in Special:Whatlinkshere/Pilgrimage (album) updated. But does the VUVR album satisfy Wikipedia:Notability (music)? You appear to be about to create VUVR and could mention the album there. VUVR Pilgrimage only has 350 Google hits and "Wishbone Ash" pilgrimage has 25900. So it seems fair to me if Pilgrimage (album) by Wishbone Ash keeps that name and adds a hatnote to either VUVR or "Pilgrimage (VUVR album)". The VUVR album could also get an entry in Pilgrimage (disambiguation) so it would still be easy to find. PrimeHunter (talk) 13:00, 1 January 2008 (UTC)
Thanks men. As you see I am not that experienced yet. But the VUVR is ready - would you correct my english in there?... —Preceding unsigned comment added by Lykantrop (talkcontribs) 19:08, 2 January 2008 (UTC)

(If you're using FireFox, sorry for the assault on your eyes)

I have been using Internet Explorer 7 on Windows XP. I just tried viewing this under FireFox 2.0.0.5 and the formatting is horrificly worse yet with FireFox. The text bunches way up and even further out of alignment with the images. The table-forced images are aligned correctly but the table ends up ABOVE the "heading 3" image.

And where is the "heading 2" image? It seems to have completely disappeared from the article under Firefox. Yecccch. DMahalko (talk) 14:28, 1 January 2008 (UTC)

 
It is a known issue. There are some workarounds for that so that each image stays in their own edit version, but can't remember the page right now (maybe in the Help:Image section). -- ReyBrujo (talk) 18:30, 1 January 2008 (UTC)
According to the CSS specifications, the rendering where it "bunches way up and even further out of alignment with the images" is actually the correct behavior (IE7 is wrong). OTOH, Firefox is incorrect in allowing the table to overlap the float, which is where your "heading 2" image disappeared to: it's underneath your "table forced" images. Anomie 18:32, 1 January 2008 (UTC)

WP:BUNCHING MilesAgain (talk) 02:32, 3 January 2008 (UTC)

List_of_churches_in_Greater_Manchester

Could someone have a look at what the table is doing to the images in this page.List of churches in Greater Manchester under Firefox 2.0.0.11 and Konquerer when the screen setting is 1024X768. The markup is as suggested but it suggests that the default css is blown. Oh, happy new year!ClemRutter (talk) 22:44, 1 January 2008 (UTC)
Though I'm still new to using this {{clear}} / {{-}} template, it looks like the {{-}} is just in the wrong spot, in the See Also section rather than the Wigan section where the problem is. It should be rendering correctly now, but with huge whitespace.. Oh well. DMahalko (talk) 00:07, 2 January 2008 (UTC)
I found a different problem with that church article in the lead/intro section. If the font size is very small, that problematic image-sticking-out-of-section occurs. But you cannot insert a {{clear}} into the lead section of the article because then the table of contents doesn't sit properly on the left side, but moves way down to sit by itself below the image. Using {{clear}} will not fix that spot. DMahalko (talk) 00:28, 2 January 2008 (UTC)
You can use the __TOC__ magic word to place the TOC just above the {{clear}}, like this. Anomie 03:31, 2 January 2008 (UTC)
Thanks for trying guys. One picture solved, but 17 more still causing a distortion. This isn't my page, it just slipped onto my watchlist by accident. It is the general problem that interests me. I have worked on CMSs with CSS, and it is possible to do anything with CSS- if you are prepared for a couple of nervous breakdowns on the way. At the moment it appears that there is an incompatibility with Wiki table generation and the default right thumb illustration, the size of which is set by user preferences and defaults to 300 in :en.
Either, in a case like this one must force, a smaller right thumb overwriting default width attribute of the right thumb, --or-- in cases of clash, the image must left sticky to the generated table and over flow page right and invoke horizontal scrolling.
Now, looking at another way, when the page without illustration has a table, and the window width is reduced then the text in a table column is word wrapped, and the table becomes smaller.
When an infobox meets the toc box, the text in the toc box wordwraps so the infobox doesn't over write, till it is down to 10 characters or so, narrower and it [-[clear]-]s. When a picture meets a toc box, it just overwrites it- no attempt is made to word wrap down. And taking this one stage further, when a picture is in a infobox and it meets the tocbox, the infobox (with picture) acts as an infobox.
So what I imagine that needs to be done is to rewrite the default image css to act like a infobox css. Over to you guys.

ClemRutter (talk) 13:14, 2 January 2008 (UTC)

Ivisible vandalism

What, if there is a good article (more experienced users wrote the article and see the result as good) and there are users, which make there permanently small unconstructive edits and add or delete information completely subjective just as they like it, and decrease the quality of the article, but still are under the terms of the article so the security can't see that?Lykantrop (Talk) 19:33, 2 January 2008 (UTC)

Another knowledgeable and experienced editor will have to clean it up again. Luckily every page has a complete edit history that can help editors in looking trough the history of the articles. That can be really helpful.
P.S. This is not really the place to ask these questions. WP:HELPDESK is more appropriate. --TheDJ (talkcontribs) 20:23, 2 January 2008 (UTC)

Monobook error

There is an error message at my user page Monobook. This happened when i tried installing wikiEd. Can someone pls help me rectify this error ? sriks8 (talk) 07:31, 30 December 2007 (UTC)

Your monobook.js is currently completely empty. If you are referring to the big red box which states "If a message on your talk page led you here, please be wary of who left it", that isn't an error message. It is a standard warning that appears on the top of everyone's monobook.js. Raven4x4x (talk) 09:23, 30 December 2007 (UTC)
Shouldn't it be orange, then? MilesAgain (talk) 11:25, 3 January 2008 (UTC)

Sources

What, if somebody, who knows a lot in some area (a professional for example), writes down to some Wikipedia article about this area facts or information, that he knows "from his head", or uses some books or encyclopedias, that he has at home, so he can't show any sources or citations and some other user just deletes his contibutions as unreliable?Lykantrop (Talk) 19:33, 2 January 2008 (UTC)

Books can be sourced, please take a look at WP:CITE. Working from head is also fine, but than you should also not be surprised that some of that information might disappear again. --TheDJ (talkcontribs) 20:20, 2 January 2008 (UTC)
There are times that I have added informatin to articles that I already know, but I can do a quick search for the information and find a reliable source to back it up. -=# Amos E Wolfe talk #=- 11:01, 3 January 2008 (UTC)
Geographical location articles are a challenge. I can find a source to show the village of Trottiscliffe, Kent is pronounced Tros-le. But Chatham, is now Cha-um not Chat-tum as it was in 1982, when my source was published. The IPA crew- never provide a reference- and they get away with it whether wrong or right. There is something called local knowledge, which I have occasionally cited! It seems to silence some of the more aggressive copy-ed bots. ClemRutter (talk) 20:51, 3 January 2008 (UTC)

Is the Main Page counted as an article in the total number of articles?

It's a page within the main namespace. The MediaWiki software includes a Main Page by default, but I'm curious if anyone knows.--h i s s p a c e r e s e a r c h 08:16, 3 January 2008 (UTC)

According to Tim Starling, yes, the Main Page on the English Wikipedia is counted as an article in the total number of articles. However, there are times and settings in MediaWiki that can affect this, but it is counted on the English Wikipedia. -- Cobi(t|c|b) 08:42, 3 January 2008 (UTC)


Changing the recent changes text

I'm posing this here because i'm hoping a developer may be around who can do this. There has been a discussion here about adding links to recent changes for featured and good articles. The suggestion is to add them in to the utilities section, however i suggested we instead replace the "changes in last 1 | 3 | 7 | 14 | 30 days!" part on the grouns that it's been obsolete for years. We can't do that ourselves. I don't know if this a two minute job for a developer or if it would take much too long to be worthwhile. Any coments? Theresa Knott | The otter sank 18:40, 3 January 2008 (UTC)

Scrap that. Someone suggested a reason why we should do that anyway. I'll leave the note here in case anyone's interested in commenting on the discussion. Theresa Knott | The otter sank 01:08, 4 January 2008 (UTC)

Medlibrary using Wikipedia content directly?

There is something curious about medlibrary.org. Apparently, they have a pretty up-to-date version of Wikipedia articles, but they aren't directly leeching from us. For example, they have the article about the list of best-selling video game franchises, indicating that "This page was last modified 11:33, 2 January 2008." However, I continued working on the article, yet they aren't mirroring the modifications. I believe they leech articles upon request, and cache them in their site. Anyone can confirm this? -- ReyBrujo (talk) 16:09, 2 January 2008 (UTC)

They're listed as a mirror here: Wikipedia:Mirrors and forks/Mno. -- John Broughton (♫♫) 23:41, 3 January 2008 (UTC)
Yeah, I was just wondering how is it possible they can mirror content that is not in the dumps (especially since, well, we don't have current dumps as far as I know). -- ReyBrujo (talk) 01:34, 4 January 2008 (UTC)
Sorry, misunderstood the question. I'm pretty sure that all the mirror sites use their own web crawler to update articles, either sweeping through the entire mainspace, periodically (say, once per day) or perhaps updating more frequently for more-read articles (though I don't think they particularly care about being absolutely current). It wouldn't be particularly efficient (I think) to do a retrieval from Wikipedia when a read request is received - first, Wikipedia is often slow to respond, and second, the mirror site is adding advertisements and doing some formatting changes to the Wikipedia text, which means (I'd guess) that using a cached version is absolutely critical.
You could, of course, check the article of interest (say) once per day, at this mirror site, and see, over the course of a week, how often it's updated. -- John Broughton (♫♫) 13:24, 4 January 2008 (UTC)

Userpage Code Problem (BoxBox Method)

Could someone please look at my page code and see if anything is wrong with it? On my computer (Mac OS X 10.4.1 and Safari 3.0.4) it refuses to render properly. If it is just Safari acting up, is there a work-around? Thanks. Dictator57 (talk) 05:21, 4 January 2008 (UTC)

  Resolved
--TheDJ (talkcontribs) 12:42, 4 January 2008 (UTC)

Unusual problem downloading article

Whenever I try to download the article Attachment therapy, I get a pane that tells me "File Download — Security Warning". It asks me whether I want to open or save the file.

This does not happen with any other article on WP, and I'm at a loss to understand why it does with this one. It must be some kind of technical problem, either on WP, or perhaps this proxy that I must use.

I must edit WP through a proxy because I currently live in China. Could it be fouling things up for me? I still wouldn't understand, though, why it would happen with only this article. Kelisi (talk) 10:29, 4 January 2008 (UTC)

When you save "download", I'm assuming that you're just doing File/Save from your browser. I just tried this (Firefox 2, Windows XP), didn't get any warning pane. Looking at the article, it's a bit unusual in that the wikitext begins with a hidden comment; perhaps your browser is interpreting that as meaning that the file is some sort of executable, though I'm inclined to doubt that - the HTML source, which is what I think you'd actually be downloading - doesn't even include that comment, and the actual article text is way down on the (source HTML text) page. -- John Broughton (♫♫) 13:16, 4 January 2008 (UTC)
Wild thought: Some programs have special reactions to certain file names. Could "Attachment" in the name set off an alarm bell in your browser? Does it happen with other articles in Special:Prefixindex/Attachment (or with that link itself)? Which browser do you have? And do you mean download as in save permanently on your hard disk, or just as in click the article link to view it? PrimeHunter (talk) 22:35, 4 January 2008 (UTC)

Is dynamic content possible (e.g. active chessboards)?

I've been editing Chess-related articles and been unimpressed with how chess games are presented (see for example Evergreen game). The presentation on most games sites is better because it uses dynamic Web content (Java and / JavaScript) to show just one board and to highlight moves in the scoresheet alongside the board -see for example Richard Reti vs Savielly Tartakower 1910.

Is it possible to include this type of facility in Wikipedia?

Its uses might be wider than Chess-related articles, for example any 2-player turn-based game (Draughts, Shogi, Go (board game), Reversi, Stratego etc. I appreciate that in practice a different implementation may be need for each species of game, since they all have different notations for describing moves and start positions (if e.g. one wants to show only the finish or other crucial stage). Ideally one would also want a facility to synchronise annotations (e.g. alternative ways of playing) with the moves, but that's asking an awful lot. Philcha (talk) 13:56, 4 January 2008 (UTC)

You can't use Javascript on wikipedia, nor a Java Applet, so it's no possible to do anything like the example you gave, because of the risk certainty that people would use such scripts for exploits, or other stupid vandalism. You can use animated GIFs but they aren't dynamic in the sense that there can be no user input, to control the animation. You could create a template to show static chessboard layouts using transparent PNGs (or GIFs) though. I don't know if you've ever followed a link to one of those sites that messes up the whole browser by constantly resizing and moving the window, and attempting to launch a series of software, connect you to irc etc. That's what would happen to wikipedia if you could use Javascript on pages, I think that's why we can't use it. Jackaranga (talk) 15:17, 4 January 2008 (UTC)
I think at this point in time, just give external links to sites that do use the dynamic content. But, of course, there is no guarantee that their sites won't go offline after a while, but that's the best you can do. — Frecklefσσt | Talk 15:26, 4 January 2008 (UTC)
I actually wrote some JavaScript do to this via MediaWiki for a personal wiki, but it needs a change in the sitewide code, and my version wasn't particularly polished and likely had bugs. It would be best to do it as a software extension rather than a very large piece of JavaScript anyway. --ais523 20:34, 4 January 2008 (UTC)
Wikipedia articles should ideally be printable. Can you suggest a way to mark up chess games that allows presentation either for a browser or for a printer? -- Tim Starling (talk) 01:11, 5 January 2008 (UTC)
I'm thinking that the best way would be some sort of extension markup to write down a game, and have it appear as a static image in a page, and (if available) use JavaScript or some similar language to make the display interactive, just sticking with the static version if JavaScript isn't available. Printing the page would print whatever was currently on the screen. The actual markup could be something like PGN (for chess) inside a pair of extension tags, and maybe an option to state what position should be shown by default, which would display as a list of moves and as a position, all made interactive. This is rather blue-skies thinking, though, I don't think there's any existing code for anything that advanced. --ais523 21:44, 6 January 2008 (UTC)

Multi-Column References

Is there any way to override the multi-column references in articles and force it to display in a single column format? May be some CSS for my monobook? --soum talk 12:39, 4 January 2008 (UTC)

Try this, it seems to work for me in Firefox (2.0.0.11) in a quick test:
div.references-small {
    -moz-column-count:1 !important;
    -webkit-column-count:1 !important;
    column-count:1 !important;
}
A page could get around it by coding the DIV directly and omitting the class instead of using {{reflist}}. You could apply it to all DIVs or even "*" if you really want to make sure it overrides. Anomie 13:48, 4 January 2008 (UTC)
My understanding (not from personal experience) is that IE7 doesn't handle multi-column reference sections correctly (CSS 3.0 standards?). If so, the above css (if it works in IE7) would be nice to make accessible via the Gadgets tab, so that editors could fix this problem for themselves with a couple of clicks. -- John Broughton (♫♫) 23:01, 5 January 2008 (UTC)
IE doesn't do multi-column references at all: it displays them in a single column always. The above CSS does nothing to change that, and in fact is intended to make browsers that do support the multi-column references display them in a single column. Anomie 00:28, 6 January 2008 (UTC)

Excessive list bullets

Would someone like to tell me where the six extra empty bullet points in template:db-t3 are coming from? Happymelon 21:04, 4 January 2008 (UTC)

The only thing that I can think of is that it's to do with the interaction with other code in db-meta that needs to use the reason for other purposes, such as a parameter to other templates. Try removing the bullet points if that doesn't make the template too ugly; otherwise, use Special:ExpandTemplates to expand the db-meta and fix it by hand. --ais523 21:38, 4 January 2008 (UTC)
{{db-meta}} hides the reason a second time, and the list bullets don't get hidden as the <span> tags can't go around them. I've replaced them with •. –Pomte 03:20, 5 January 2008 (UTC)
They come from {{Hidden-delete-reason}}. As said, a list within span tags with display=none is converted to a list with display=none for each list element. Therefore perhaps span should be replaced by div in this template.--Patrick (talk) 09:10, 5 January 2008 (UTC)

Magic words -- overconfident!

re: Mess in http://en.wikipedia.org/wiki/1632_writers#Breivik.2C_Aamund Thought I'd 'ease' cites (note the <xref> in said mess...) with a typing aid...

The nesting of the magic words seems to be malfunctioning in an unanticipated way... I'm getting a subst: when I thought I'd see a subst'd value, and I currently need to boogie down the road on an errand. Unfortunately, in my overconfidence, I already advertised the template in two posts.

Can someone please take a peek and see if you can fixup the nesting? I suspect it has to do with the overides using {{{1}}}, {{{2}}}, and/or {{{3}}}. Adding if logic will mess up the whole idea, since parserfunctions get subst'd too. I've already deinstalled one pair of included only subst: commands (for the padding) in trying to see what was happening. (Disabled the usage on the template page too! See "--x>" below! <g>)
  • The template has an {{lts|Adt}} edit link currently there in that article's section, and should someone fix it up, I'd appreciate "obvious tidy up edit" to the debugging there, including the 'x' in <xref>'s check AND a note to CBD that this is done here. (I suspect he's sleeping)
  • The usage on the template page can be enabled by removing the 'x' in --x> circa the noinclude start.
  • This is a similar technique in long use now in my {{DATE}} template... sigh. // FrankB 18:11, 5 January 2008 (UTC)

Why isn't there a creation log?

To go along with the deletion, protection, and move logs? I think it could have some potential value...--h i s s p a c e r e s e a r c h 07:59, 6 January 2008 (UTC)

What would be the advantage, when we already have Special:Newpages?-gadfium 08:04, 6 January 2008 (UTC)
Well, newpages moves past quickly, whereas with a creation log it would be recorded in the log for that particular page indefinitely. Say a page was created on March 16, 2006, deleted on August 15, 2007, then recreated as a new page but not deleted on November 3, 2007. You wouldn't have known how long the original page lasted for.--h i s s p a c e r e s e a r c h 08:14, 6 January 2008 (UTC)

Template indicate with AfD closed?

Is it possible to do something with AfD closures so that it indicates in the AfD TOC when the AfD is closed, or does the fact that AfD uses transclusion prevent that? I know it's considerably simpler at the BLPN noticeboard, which I frequent, given that the tickets are on the actual page, but I don't want to presume it's impossible to do with transclusion without asking. People who do technological stuff (I do not) are constantly impressing me with their ingenuity. :) It would certainly make it easier for admins to close if we could simply look at the TOC and see what's outstanding. --Moonriddengirl (talk) 16:43, 6 January 2008 (UTC)

Just in case there are interested (and technologically clueful) editors there, I'm duplicating this question at the talk page for AfD. Magic wands appreciated. Kind reminders that magic wands do not exist also appreciated. :) --Moonriddengirl (talk) 20:16, 6 January 2008 (UTC)

Template:Edit0

What do you thing about the Template:Edit0? I thing it would be very useful in every article. --Janezdrilc (talk) 16:17, 5 January 2008 (UTC)

At one time, we did have the ability to edit just the first section of an article. But it was done away with (not sure) because the edit links were unsightly. Now I just edit the second section and then replace the "=1" in the URL with "=0" to edit the first section. So, in short, I think, in practice, every article used to have the effect of that template, but it was done away with. — Frecklefσσt | Talk 17:51, 5 January 2008 (UTC)
You can enable an edit link to appear for you for section 0 of pages by enabling the option in the Gadgets section of your preferences. mattbr 20:54, 6 January 2008 (UTC)
It's already a user preference (using JavaScript) — there's no need for a template. æ²  2008‑01‑06t21:13z —Preceding comment was added at 21:13, 6 January 2008 (UTC)
And there also is {{Edit-top-section}}, which is a bit more flexible. EdokterTalk 21:31, 6 January 2008 (UTC)

Need help

Hi does anyone here know how to help me figure out what are the time codes for different time zones like (EST for eastern standard time) and the daylight times during daylight savings time. I need the following to complete my interchangeabe timezone clock

UTC-0 daylight savings time (for daylight savings time I want the DST that has the time difference specified ex. UTC-5 regular is EST DST is CDT not EDT)

UTC-1 daylight and standard (standard meens not DST)

UTC-3 standard

UTC-10 daylight

UTC-11 standard

UTC-12 daylight

UTC+1 standard

UTC+2, UTC+3, UTC+4, UTC+5, UTC+6, UTC+7, UTC+8, UTC+9, UTC+10, UTC+11 daylight and standard

If there are other UTC+# or UTC-# feel free to list it here with the above but please keep it to whole hours (no minutes). Thanks in advance for the help Alexfusco5 22:07, 6 January 2008 (UTC)

Try List of time zones. --ais523 22:09, 6 January 2008 (UTC)
Thanks, I'm an idiot Alexfusco5 22:38, 6 January 2008 (UTC)

ArticleViewHeader

The ArticleViewHeader is installed here. If I'm not mistaken, it gets a page from MediaWiki:Foo for example, and on Foo, it puts that at the top of the page. I was thinking we do something like this for the sandbox. See also: [[1]] Soxπed Ninety Three | tcdb 04:42, 7 January 2008 (UTC)

Template bug on Barack Obama

Can someone have a look at the bottom of the article (the related article boxes)? Apparently there's a glitch somewhere making the parser functions visible, but I can't find it at all. I presume there's an other template somewhere in the page that is breaking it (the templates by themselves are working). Thanks! -- lucasbfr talk 10:17, 7 January 2008 (UTC)

Too many templates on the page, I believe. Converting a few citation templates to manual formatting. –Pomte 10:24, 7 January 2008 (UTC)
The article includes too many (or too large) templates. Click "View page source" in your browser (ctrl-U in Firefox) and you'll see the error messages about the pre-expand include size near the bottom. A possible fix is to use less templates or to shorten the transcluded templates (e.g. by using /doc subpages or by using simpler templates instead of complicated if constructs). Kusma (talk) 10:25, 7 January 2008 (UTC)
Crap the idea did not even cross my mind. I assume the error message is Pre-expand include size: 2032717/2048000 bytes? (just for next time). -- lucasbfr talk 10:32, 7 January 2008 (UTC)
That's it. The counter stops counting when it's about to reach the limit, but anything near the limit is suspicious. (The message itself appears on every page.) --ais523 10:33, 7 January 2008 (UTC)

Linking to a category

Am I missing something obvious? I tried to create a link to a category in a talk page with this

 blah blah [[CategoryInks]] blah blah

When rendered, the link is rendered as nothing. For instance, here it is again:[[Category:Inks]] (placed in nowiki tags by PrimeHunter). Note that a colon is missing from between "Category" and "Inks" in the above block. If the colon is replaced, the rendering there too is null. Frotz (talk) 12:58, 7 January 2008 (UTC)

To link to a category type [[:Category:categoryname]] The extra colon before Category turns it into a link, instead of placing the page into the category. Hope this helps. DuncanHill (talk) 13:01, 7 January 2008 (UTC)

Yes, I have edited the post to remove this page fromCategory:Inks. PrimeHunter (talk) 14:14, 7 January 2008 (UTC)