Jump to content

Wikipedia:Village pump (technical)/Archive

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Crypticbot (talk | contribs) at 00:11, 8 June 2006 (Automated archival of 8 sections older than 7 days from Wikipedia:Village pump (technical) and removal of 18 sections older than 14 days). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Village Pump - Archive

DO NOT EDIT OR POST REPLIES TO THIS PAGE. THIS PAGE IS AN ARCHIVE.

Discussions older than 7 days (date of last made comment) are moved here. These discussions will be kept archived for 7 more days. During this period the discussion can be moved to a relevant talk page if appropriate. After 7 days the discussion will be permanently removed.

Post replies at Wikipedia:Village pump (technical), copying or summarizing the section you are replying to if necessary.

Note: Please add new material at the bottom of the page. Preceded by the following: =Sections archived on ~~~~~ =



Sections archived on 00:18, 2 June 2006 (UTC)

Fonts and Unicode support in SVGs

This was originally posted at User talk:Mysid.

Is there a list somewhere of the fonts that are installed on the Wikimedia servers, so that I know what fonts I can work with in SVGs? It seems that at least Helvetica is installed (although it might go by some other name). A list of supported fonts would be very useful to SVG authors.

In a related vein, it would be nice to know which Unicode characters can be put in text labels in SVGs. For example, I want to include an H with a cedilla in my text labels (as in AL BAḨR AL AḨMAR), but I've tried both U+1E28, which doesn't show up at all on Wikipedia, and an H followed by U+0327, which just shows up as a normal H (see Image:Bkell-test.svg). I've thought about converting the text to a path, but that destroys the editability of the label. Suggestions are most welcome. —Bkell 19:18, 24 May 2006 (UTC)[reply]

SVGs are currently rendered with rsvg. As for what fonts are available, that's a good question and we don't have the answer ready. --Brion 22:43, 24 May 2006 (UTC)[reply]
I have created Image:Fonts.svg which uses a number of fonts (mostly those that with my Gentoo Linux installation and were not installed later). Some others are emulated (e.g. Courier New becomes Courier) but I think these are all that actually look like they're supposed to:
  • Bitstream Charter, Bitstream Vera, Bitstream Vera Sans, Bitstream Vera Sans Mono, Bitstream Vera Serif
  • Courier
  • Luxi Mono, Luxi Sans, Luxi Serif
This is probably incomplete, and I can't identify which fonts are actually used for the generic Sans and Serif fonts, but it's a start. —da Pete (ばか) 14:57, 25 May 2006 (UTC)[reply]
It looks as though Bitstream Vera Sans Mono, Courier, Courier New, and Luxi Mono all refer to the same font, which appears to be Courier. (It's certainly not Bitstream Vera Sans Mono, since it's a serif font.) The same goes for Bitstream Vera Serif, Luxi Serif, Serif, and Times New Roman. Bitstream Charter appears to be a different font, though. So it appears that the only distinct fonts that actually exist are Bitstream Charter, Courier, Helvetica, and Times, possibly under different names. —Bkell (talk) 17:31, 25 May 2006 (UTC)[reply]

I created a javascript link button for Wikipedia and would like to donate it to the cause. Where would I put it?

You can see the link button on the following page: http://www.geocities.com/boltbaits/magic/MagicCard.html

It is at the bottom of the orange box. Instructions for installation and use are below that.

--BoltBait 01:43, 25 May 2006 (UTC)[reply]

Wikipedia favicon in search bar changed?

Today I noticed that the Wikipedia icon in my browser's search bar changed from the stylized "W" to a tiny version of the Wikipedia globe logo. (This is the search bar in Firefox which allows you to pull down a menu of search engines; my default is Google, but I also have a number of others enabled including Wikipedia.)

Given the size of the image I assumed they were favicons; however, the favicon for Wikipedia in the URL field of the browser toolbar remains the stylized "W". Did something change to cause the search bar icon for Wikipedia to change? (Did the main Wikipedia favicon change, but I still have the cached old one?) Is there a way I can get the "W" back for my search bar? (I much prefer the "W" since it's distinctive enough to cue me that I'm searching Wikipedia and not Google.)

This is using Firefox 1.5.0.3 on MacOS 10.4.5. Thanks, MCB 01:44, 25 May 2006 (UTC)[reply]

That search plugin is maintained by a third party, who presumably changed the icon used. --Brion 02:16, 25 May 2006 (UTC)[reply]
Thanks. I just wanted to make sure something had not changed at Wikipedia. MCB 18:52, 25 May 2006 (UTC)[reply]

How do I find the list that is at the bottom of a page?

danct 01:51, 25 May 2006 (UTC)

I have editied this page:

http://en.wikipedia.org/wiki/The_Last_Camel_Died_at_Noon

and I borrowed the format from someone else becasue it looked great. But there is a box at the bottom with a list of the stories, and I want to edit that list, and I want to creat other lists for other topics I contribute to.

I see

at the bottom of the edit page, and it is the thing I want to edit, but when I click on the hyperlink, it just takes me to her page.

I don't even know what to call things so I can't look them up.

Thanks for any help

--danct 01:51, 25 May 2006 (UTC)

{{Amelia Peabody}} is a Template, which means that it resides in template namespace (Template:, much like the Village Pump is in Wikipedia: namespace), so to edit it, you could type "Template:AmeliaPeabody" in the search box and click "Go". This should take you to Template:Amelia Peabody, where you can edit it like you would any other page. Jude (talk) 01:56, 25 May 2006 (UTC)[reply]
This is caused by a feature called transclusion, which basically allows you to add the content of one page into another without copying the code. The form {{Templatename}} is used to translude a page. In this case the page Template:Amelia Peabody is the page you want to edit. It is also possible to transclude user pages, talk pages, and even articles, using the same format. Normally if you do not include a prefix, it will use the template namespace, i.e. {{Pagetotransclude}} would put the text of Template:Pagetotransclude on the page. To use other namespaces you need to include the prefix (like {{User:Userpagetotransclude}}). Hope this helps, Prodego talk 01:58, 25 May 2006 (UTC)[reply]
Just to add to the above. The templates used in a page can be seen listed as links at the bottom of the page (below the editable window) when you've clicked "edit this page". Shanes 03:47, 25 May 2006 (UTC)[reply]

Suppressing images

Is it possible to suppress images, and only show the text that is associated with them? Sometimes my connection is slow, and I'd like to be able to load up the pages better. Is this possible? --HappyCamper 04:24, 25 May 2006 (UTC)[reply]

You should be able to tell your Web browser not to download images. Look in the preferences somewhere. —Bkell (talk) 04:25, 25 May 2006 (UTC)[reply]
Ah, is that the trick? Thanks! --HappyCamper 04:28, 25 May 2006 (UTC)[reply]

Translation of some symbols into squares.

My computer doesn't interpret correctly some symbols. For example "-1" (written in the original format as "& minus ;1" is shown as a square followed by 1. Can this problem be solved? Thanks a lot. Paolo de Magistris, Rome, Italy. demaag@simail.it

This is almost certainly a font issue, see m:Help:Special characters. -- Rick Block (talk) 14:02, 25 May 2006 (UTC)[reply]

I'm sure it's been discussed before, but could someone point me to a previous discussion (or bug#) regarding whether things like FiNaL FaNtAsY could be automatically hooked up by mediawiki software, instead of redlinked? A new user asked me about why the links don't go to the right place, or at least go to Final fantasy, since that's what happens when you hit "Go" in the search box, and I didn't have an completely adequate answer. --Interiot 20:26, 25 May 2006 (UTC)[reply]

I suppose it's because capitalization is important in article titles (Aid and AID are distinct articles, for example). So if I wanted to start an article about some FiNaL FaNtAsY that's actually capitalized that way, it would be more difficult to do so if I could never get a redlink to appear. —Bkell (talk) 20:37, 25 May 2006 (UTC)[reply]

Redirection is a good workaround. --DLL 21:32, 25 May 2006 (UTC)[reply]

How to get an entry based on value or null (Revised question)

This is a revised entry of the original question so I can make it clearer.

I want to create a template which is used to get either a specific entry or all of them depending on the parameter.

Let's say I have a list like this

  • Item 01 - May
  • Item 02 - Critical
  • Item 03 - Exact

And I want to put it into a template, let's call it FUZZ so that if I simply use a macro call of {{FUZZ}} I get

  • Item 01 - May
  • Item 02 - Critical
  • Item 03 - Exact

But if I use {{FUZZ|01}} I get

  • Item 01 - May

But if I use {{FUZZ|02}} I get

  • Item 02 - Critical

I know in some way this requires {{#if:}} or some macro, but I'm not sure how.

Is there a way I can do this? Second, is there a way I can have the macro generate text only if the parameter is null? So that I could put a header if the parameter is null but no header if the value of {{{1}}} is not null.

Paul Robinson (Rfc1394) 22:53, 23 May 2006 (UTC)[reply]
I'm not sure if this is exactly, what you want, but perhaps something like this?
{{#switch: {{{1|}}}
 | 01 = *Item 01 - May
 | 02 = *Item 02 - Critical
 | 03 = *Item 03 - Exact
 | #default = 
  *Item 01 - May
  *Item 02 - Critical
  *Item 03 - Exact
}}
You might want to check out m:ParserFunctions for a better explanation of things. In regards to null-parameters, {{#if: {{{1|}}}|do this if it is set|do this if it is unset (null)}}, is that what you're looking for? Hope that helps! Jude (talk) 02:46, 24 May 2006 (UTC)[reply]
Thank you for your response. I'm already doing something like that now, what I think I want is something like this:
{{#switch: {{{1|}}}
 | 01,#default = *Item 01 - May
 | 02,#default = *Item 02 - Critical
 | 03,#default = *Item 03 - Exact
}}

or

{{#switch: {{{1|}}}
 | 01
 | #default = *Item 01 - May
 | 02
 | #default = *Item 02 - Critical
 | 03
 | #default = *Item 03 - Exact
}}
So that I only have to include each of the items once. For default, it shows all of them, for any one of the entries, it only shows that one. So that for each entry, if it is that value OR null, then it shows the item. This way I can get both with only one entry for each item, as opposed to having to do them twice. Paul Robinson (Rfc1394) 22:45, 25 May 2006 (UTC)[reply]
There is one thing you might be able to do that, and have only one line per option, though it is a bit dirty:
 {{#switch: {{{1|}}} | 01 | = <br />Item 01 - May}}
 {{#switch: {{{1|}}} | 02 | = <br />Item 02 - Critical}}
 {{#switch: {{{1|}}} | 03 | = <br />Item 03 - Exact}}
This puts a switch on each option that only shows that option if the first parameter is either blank or matches the number of that parameter. This means of course you can only call one: {{FUZZ|##}} or all: {{FUZZ}}. There is probably a better way to do it though. --Splarka (rant) 23:09, 25 May 2006 (UTC)[reply]
Streamlined the code a bit. | 01 | case with no = in it passes to the next argument, | = with nothing before the = assumes empty string, so | 01 | = basically means "case '01' OR ' '". And the default is not needed. The only problem I see with this now, is a line return added for the omitted lines. *shrug* --Splarka (rant) 00:05, 26 May 2006 (UTC)[reply]

Sections archived on 00:11, 3 June 2006 (UTC)

Is it possible a sysadmin could install the new <sort> extension?

Bug 4870: <sort> feature: automatically sort lists, e.g. on lists, dab pages has been kindly resolved by Rob Church, but when I emailed him to ask, he said it was not installed on Wikipedia. Is anyone here a system administrator? Could you get it installed? Thanks, --unforgettableid | talk to me 22:30, 24 May 2006 (UTC)[reply]

I think it needs a little more work: the ordering is not ideal at the moment, and it should probably be profiled for performance on large lists before we consider it for Wikipedia. --Brion 22:59, 24 May 2006 (UTC)[reply]

Brion and I had a brief discussion about it on IRC. I'll get some profiling sorted and improve the sort method a little better. Since it was a quick hack for a friend which also happened to fulfil a request, it wasn't supposed to be brilliant in the first place. robchurch | talk 00:33, 25 May 2006 (UTC)[reply]

Just curious: how do you profile PHP code? Cheers, --unforgettableid | talk to me 03:00, 26 May 2006 (UTC)[reply]
There are various ways, but one which is quite nice is to use the Xdebug extension to output profiling data and load that up in KCachegrind for visualization and analysis. --Brion 22:11, 26 May 2006 (UTC)[reply]

Occasionally I'll take some image in Wikipedia that looks bad, bring it into Photoshop, and clean it up. Then I'll upload it under a slightly different name, so as not to destroy the original, and change the link to the image.

The article looks fine, but the original image is now orphaned. Even if I link to the old image from the page of the new image, "What links here" doesn't pick it up. So now the original image can be deleted, which isn't what I intended. See Image:Yellowstar.jpg for an example. Is there some way to avoid losing the original? --John Nagle 03:35, 25 May 2006 (UTC)[reply]

If the new image is a replacement for the old one, then just use the same filename. The old version will still be available in the history. —Bkell (talk) 06:31, 25 May 2006 (UTC)[reply]
I don't think simply being orphaned is enough reason to delete an image; it has to have some other problem, such as being unencyclopedic or not having a proper license. But if an orphaned image is licensed under a free-use license, and you want to prevent it from being deleted for being an orphan, you can always upload it to the Commons. —Bkell (talk) 17:37, 25 May 2006 (UTC)[reply]
The really obvious solution to this problem is to add text to the old image's page linking to the new image. I like to put something like: "Do not use this image. It has been replaced by Image:XXXX." This is only really appropriate though for format changes - usually you can just upload the new version over it. Warning, though: if you're editing a JPEG, make sure you upload the new version at high quality, or you will risk eventual generational deterioration. Deco 09:07, 26 May 2006 (UTC)[reply]

Word wrap around images

Typically with this software, words will wrap around left-aligned images but not right-aligned images. This always causes annoying blank spaces below pictures on the mainpage. Is there any way to go around this default setting?--Pharos 06:08, 25 May 2006 (UTC)[reply]

Examples might help, but I think the wrapping works on both sides essentially the same (assuming images are floated either right or left). With a left floated image, text that starts a new line is obviously just underneath the image. With a right floated image, a short sentence that doesn't extend to the end of the line (or a series of these) may make it look like the text is not wrapping even though it is. Do the examples at Wikipedia:Picture tutorial exhibit the problem you're seeing? -- Rick Block (talk) 14:12, 25 May 2006 (UTC)[reply]
Just look at the mainpage nearly any day on In the News or On this day or occasionally Did you know... for examples of the problem. I've been thinking of the exact way to describe it, and you're right that it really isn't a right-left issue. The problem is simply that one sentence, if it touches an image on its first margin-kiss, is also restricted to that width on its second and third lines. I think it's just more of a problem for these boxes because of the specific format of bulleted sentences.--Pharos 19:38, 25 May 2006 (UTC)[reply]
I see what you mean; look at User:Bkell/Sandbox for example. —Bkell (talk) 19:48, 25 May 2006 (UTC)[reply]
One possible solution is to put the floating image inside the bullet list. which will sort-of wrap multiple bullet items around it (this doesn't work good for left-aligned images however). See the section section in User:Bkell/Sandbox. --Splarka (rant) 02:29, 26 May 2006 (UTC)[reply]
I've been considering it, and I think the solution would just be to graphically reproduce the standard Wikipedia bullet symbol through Unicode without the text of the following sentence actually being formatted as a bullet. This shouldn't be too hard to do, I think, for someone who is has just a little experience with Unicode.--Pharos 22:46, 25 May 2006 (UTC)[reply]
The bullet image is here: http://en.wikipedia.org/skins-1.5/monobook/bullet.gif (defined in .css via list-style-image: url(bullet.gif);). It is 5x13 pixels with a 5x7 area at the top and 5x1 area at the bottom set as transparent (white). The visible area is a 5x5 pixel box RGB: 99, 140, 156 (#638c9c). So it could be emulated with a crude div:

Test

The margins and padding could be adjusted as needed (like, in a template with editable defaults, eg: padding-right:{{{pr|5}}}px. Be good as Template:Bullet I guess. --Splarka (rant) 02:44, 26 May 2006 (UTC)[reply]
This seems like the most viable short-term solution so far.--Pharos 03:37, 26 May 2006 (UTC)[reply]
Splarka and Pharos, what browsers are you using? —Bkell (talk) 03:56, 26 May 2006 (UTC)[reply]
Firefox (the latest update), but I also have access to IE and I see your point about the differences.--Pharos 04:30, 26 May 2006 (UTC)[reply]
Firefox (a very embarrasingly old version). --Splarka (rant) 06:10, 26 May 2006 (UTC)[reply]
Do you mean the blank line before the bullet list starts? I don't know of a way to suppress this, if this is what you're after. -- Rick Block (talk) 00:57, 26 May 2006 (UTC)[reply]
Hmm, upon further investigation this appears to be a Firefox-only issue (maybe Mozilla too). It looks fine in Internet Explorer and Opera. —Bkell (talk) 02:35, 26 May 2006 (UTC)[reply]
Splarka, your solution where the floating image is inside the bulleted list doesn't look good in IE (it indents the bullet and the first line a little bit for some reason) or in Opera (the bullet appears next to the image, not next to the text). My guess is that a real solution will require some CSS wrangling, or maybe a Firefox bug fix. (I should also point out that I'm still using Firefox 1.0PR on Windows XP, not anywhere near the latest version.) —Bkell (talk) 02:39, 26 May 2006 (UTC)[reply]
Yah, seems to be a problem in the different way browsers handle <img src> inside of <li>. --Splarka (rant) 06:10, 26 May 2006 (UTC)[reply]

It might be more direct to enter this as a bug report on Firefox. -- Rick Block (talk) 15:01, 26 May 2006 (UTC)[reply]

I found the reason for the non-wrapping behaviour on the first example at User:Bkell/Sandbox; it's the -moz-float-edge: margin-box on resource://gre/res/html.css (the default rules for HTML on Gecko). Changing it to -moz-float-edge: content-box fixed the issue (see the third example I created at User:Bkell/Sandbox). Why is it set to margin-box, and what does it actually mean, I don't know. Since it's a Gecko-specific property, it will be ignored by other browsers, so it should be safe. --cesarb 21:21, 26 May 2006 (UTC)[reply]

Taking a look at CSS3, looks like -moz-float-edge is somewhat similar to float-displace. That, however, still does not explain why it's the default on li elements. As an aside, it looks like -moz-float-edge is a cute way to avoid the problem where the horizontal line below a section heading "cuts through" a transparent float: just make the section headings (of which the horizontal line is the lower border) -moz-float-edge: margin-box, and they'll stop just before/after the float (since the headings are supposed to be a single line, wrapping isn't an issue). --cesarb 21:41, 26 May 2006 (UTC)[reply]
Found a testcase at http://www.brunildo.org/test/floatFlow.html. As you can see there, using content-box with a left float can make the bullet be placed to the left of the image, which is probably why margin-box is the default. That shouldn't be a problem with a right float, unless you start worrying about things like bidirectionality. It also has a link to https://bugzilla.mozilla.org/show_bug.cgi?id=163110. --cesarb 21:53, 26 May 2006 (UTC)[reply]

"Pagename" magic word

Does anybody know what code to write in order to have the pagename magic word writing "page 1"'s name in it and then, if I put in, say, may talk page {{PAGE 1}} not rendering "User:Argantino/talk" but "page 1"? —Argentino (talk/cont.) 14:28, 25 May 2006 (UTC)[reply]

There could be a pagenumber template, to use instead of page name, which gives what it says.
However, pagination on the web is slightly different from that on a text processor. There are no page numbers that I could imagine. --DLL 21:29, 25 May 2006 (UTC)[reply]
Do you mean {{SUBPAGENAME}}? On a page like User:Argantino/talk it would return "talk", or User:Argantino/page_1 it would just return "page 1" ({{SUBPAGENAMEE}} would return "page_1"). Or, are you trying to change the firstHeading on a page? Sorry, question is confusing. Rar. --Splarka (rant) 21:58, 25 May 2006 (UTC)[reply]
Sorry, i mean: if i create put { Pagename } to an article X what do I have to do when i put in my userpage { X } to call the article put in the place where i had put {pagename } to get "X" and not to get "Argentino?. —Argentino (talk/cont.) 17:16, 26 May 2006 (UTC)[reply]
Why do you write {{PAGENAME}} in article X instead of just writing "X"? —Bkell (talk) 17:21, 26 May 2006 (UTC)[reply]
I just wondered, because i'm organizing the anniversaries for each day in Argentina for the Portal:Argentina and I think that writing { Subpagename } in every article is better than going to Portal:Argentina/Anniversaries and start making a header/day. I see i'll have to write down each day name, so it will take much more time (i cant copy & paste). Anyway, thank you —Argentino (talk/cont.) 17:39, 26 May 2006 (UTC)[reply]

Complicated Image Request

I love the colored maps I find when searching different nations and states on Wikipedia I was interested to know if anyone can create or find a way to bring the State of Alaska and the Territory of Yukon together on one map. Kind of a supranational map for topics relating to North America's first-order administrative units. Any help I could receive on the matter would be great or a direction or a wikipedia image specialist I could consult. Thank you. 209.193.56.228 01:02, 26 May 2006 (UTC)[reply]

One option is to look for an online atlas. :-) Cheers, --unforgettableid | talk to me 01:11, 26 May 2006 (UTC)[reply]
What exactly are you looking for? A map of Alaska and the Yukon, and nothing else? —Bkell (talk) 02:03, 26 May 2006 (UTC)[reply]

I got troubles reading characters correctly, although I...

Hello: I got in internet options enabled for East-European languages to make readable,I also downloaded Lithuanian fonts in my C:\WINDOWS\FONTS map /folder. Nevertheless, when not on original Lithuanian websites, I get always letters / signs not readable or not according the language. I tried all possible encodings of East-European fonts and the UHF-8, the problem is persisting.Do you or one of your colleagues know what I must do inorder to make it working 100% properly ? My e-mail address is: Kris_aus_Antwerpen@hotmail.de Thank you in advance, Chris.

Sorry, Wikipedia is not a tech support forum. You can get help at www.googlegroups.com. Cheers, --unforgettableid | talk to me 02:58, 26 May 2006 (UTC)[reply]
This is a known problem in Internet Explorer. Use FireFox. —Ruud 22:24, 26 May 2006 (UTC)[reply]

"/ad/" in URL's of images is blocked by default in AD blocking plug-ins

Some URL's of images in wikipedia has the term "/ad/" in them (example: http://upload.wikimedia.org/wikipedia/en/a/ad/MSdying.jpg ). the term "/ad/" is blocked by default in most commonly used AD Blocking plug-ins and programs thus blocking many images in wikipedia articles. It would be wise to stop using the term "/ad/" in the URL's to avoid that from happening. (btw, is this the right place to report this?) --TheYmode 06:48, 26 May 2006 (UTC)[reply]

This is a known issue. At some point in the future we're going to change the way images are stored and referenced, and when we make those changes we will keep this in mind. In the meantime, you can fix your ad blocker's configuration. --Brion 07:06, 26 May 2006 (UTC)[reply]

JPEGs and generational decay: a solution proposal

As more and more users join Wikipedia, our forces of technically adept image editors is also growing. Wikimedia Commons is a thriving project. This creates the problem that soon, many of our images will be iteratively refined and edited over time, just as our articles are. For bitmap images, SVGs go a long way towards making this more convenient for many types of images.

For JPEGs, however, the problem of generational decay is inevitable: images repeatedly saved at 70% or 80% will acquire visible artifacts and all their edits will have to be repeated from the original version. Some images may have had older versions stripped for space under the mistaken impression that nothing of value was lost. We need to develop a solution now and implement it before this happens.

One partial solution, recommended by many users, is to just use very high quality JPEGs, such as 95% quality. Because thumbnails are rendered at medium quality, there's no problem with download times for readers. This staves off generational decay for quite a bit longer while keeping filesizes somewhat reasonable, but it does not eliminate decay. This is what I'd recommend in the absence of any software changes.

A bad solution: Use JPEG's lossless mode. This isn't a really good idea, as in Huffman mode JPEG lossless has unimpressive compression rates of about 49% on average, while the arithmetic coding mode is encumbered by patents. It also has poor tool support.

A better solution: Upload PNGs and render their thumbnails as JPEGs (ideally with the option to choose the thumbnail quality level). PNG delta compression does pretty well with continuous tone images, exhibiting an average compression rate of about 56.5% ([ref table in JPEG-LS paper), which isn't much worse than the best lossless formats (although this is after a rather slow pngcrush run). This would only require small software modifications to enable the selection of format for thumbnails, and would dovetail nicely with another important feature, specifying color depth of PNG thumbnails. This also has the advantage of extremely wide tool support.

The ideal solution: Use a specialized lossless compression format for photos such as JPEG LS or JPEG 2000 lossless mode. Average compression rate in the JPEG-LS paper is about 60%, and another study shows average compression rates of 66%. JPEG-LS outdoes PNG by about 10% on typical photos[1]. Unfortunately, tool support is currently extremely limited, with only an extremely alpha open source impl, and a single mature Photoshop plug-in distributed by HP Labs. The effort required for Mediawiki just to decode such files for re-rendering is prohibitive. These formats are also encumbered by a variety of patents (albeit freely licensed ones) and are relatively slow to decode.

So, in light of the above, I propose the middle solution. Let's add support to the parser and renderer for specifying the format of a thumbnail. We know convert will happily do all the hard work for us. Sound good? Deco 10:04, 26 May 2006 (UTC)[reply]

We also could do with divorceing the image description page name from the file format so that new versions can be uploaded in png format when the original is a jpeg. Plugwash 13:47, 26 May 2006 (UTC)[reply]
See also bugzilla:4421. —Bkell (talk) 17:50, 26 May 2006 (UTC)[reply]

When do new articles appear in search index?

Having created a couple of new articles a few days ago, they still do not appear in the search index and you can only find them if you enter the direct name of the article or link to it from another article that links to them?

How often does Wikipedia update its search index to pick up new articles?

--Tompage 16:54, 26 May 2006 (UTC)[reply]

Scroll to the top of this page and read the "Frequently Asked Questions" section. ~MDD4696 17:04, 26 May 2006 (UTC)[reply]

Sound files

How do you add sound files to an article? I'd like to know how to create a sound file as well as how to wikify it. In order to avoid copyright issues, I'm thinking about recording some plainchant myself for the chant articles. (The music is centuries old, but the commercially available recordings are copyrighted by their performers.) What kind of copyright release or permission would I need to give, and how would I give it?

I haven't been able to find info in the help section. Thanks! --Peirigill 21:20, 26 May 2006 (UTC)[reply]

Wikipedia:Sound_and_Wikipedia#Audio. --Splarka (rant) 22:32, 26 May 2006 (UTC)[reply]
Oh, and to embed it: See {{Listen}}, {{Audio}}, etc. --Splarka (rant) 22:36, 26 May 2006 (UTC)[reply]

Sections archived on 00:10, 4 June 2006 (UTC)

Is there a way to make it so that a page does not show up in another page's "What links here"? I wonder, because some people link various things on their userpages, which results in a mix of namespaces in the What links here, which in my view makes it a less useful tool. In addition, many are flooded with recent changes from date links included only for preferences, rather than being contextually relevant. I've tried prefixing links like this with a colon, not sure why that would help, but I tried it. Does anyone know if there's a wiki-fix? Mak (talk) 18:26, 22 May 2006 (UTC)[reply]

If a page links to another page, then it's counted in Special:Whatlinkshere for that page. Because, y'know, we want to keep track of what links where, etc. So no, we won't be allowing you to suppress that information from our link tables, because we need them for various important functions. 86.138.46.182 20:22, 22 May 2006 (UTC)[reply]
Example link: Oroshigane. But that breaks all sorts of normal conventions, so it should be used sparingly. Alternatively, vote for Bug #4624, and then it won't be a big deal if people spam whatlinkshere. --Interiot 23:33, 22 May 2006 (UTC)[reply]
Ah! That bug was really more what I was thinking of. I'm sorry, robchurch (if it's you), I really don't intend to break your wiki. Please accept my pleas of ignorance. Mak (talk) 23:38, 22 May 2006 (UTC)[reply]
It's extremely useful having all pages that link from something show up in whatlinkshere, irrespective of namespace, so I'd be against this. it would be nice, though, if there was some way to sort whatlinkshere in the same way as sorting user contributions (i.e., have a toggle at the top that allows you to pick a namespace). Grutness...wha? 03:08, 23 May 2006 (UTC)[reply]
Side note: WP:AWB can create its list of pages from the What links here of a page and filter out non-main pages. The list can be stored as a text file (and thus edited and read back). --Ligulem 07:42, 24 May 2006 (UTC)[reply]

Links inside <gallery> tag descriptions are not included in Whatlinkshere, but that's just a bug. Also, going through the interwiki map with the "w:" prefix supresses it: w:stuff, but that's also most probably not a cool thing to do... -- grm_wnr Esc 13:46, 23 May 2006 (UTC)[reply]

That bug should have been fixed some time last week. It won't magically fix existing pages with galleries until they've been null edited (since the link tables won't be updated) though. Rob Church (talk) 19:23, 23 May 2006 (UTC)[reply]
I like the current "What links here" as it is. PS If you can add a count to each page to show how many people have it on their watchlist, you'll be very popular with me. This is quite a good rule-of-thumb guide to the accuracy of an article. Stephen B Streater 19:43, 23 May 2006 (UTC)[reply]
That seems like a security risk, but something admin-only like Special:Wathclistcount/Pagename or Special:Mostwatchedpages would be partially useful (like Special:Unwatchedpages ... TINC!). --Splarka (rant) 07:25, 24 May 2006 (UTC)[reply]
I slightly see your point. Well a point anyway - that it could be used by vandals. Perhaps I'll become an Admin just in time to use the Admin-only version ;-) Stephen B Streater 12:07, 27 May 2006 (UTC)[reply]

"Protected deleted pages" not ignored with Special:Random

Protected deleted pages come up with the random article button. This is rather irritating, and probably doesn't send a good message about the quality of Wikipedia to the users who encounter them. Sarge Baldy 21:32, 24 May 2006 (UTC)[reply]

Well, MediaWiki has no concept of a protected deleted page. As far as it can see, there's a page with content on it, which is fair game for Special:Random. robchurch | talk 21:51, 24 May 2006 (UTC)[reply]
This is a wider problem for other uses, because MediaWiki can't distinguish between normal articles and stubs or disambiguation either. I would like to take this opportunity to beg the devs to find a way to develop such a capacity—though they probably are anyway, so I also offer thanks.
I don't think the 'deleted page' is such a major issue, however, since there are probably very few in existence. (Don't quote me on that.) Ingoolemo talk 22:00, 24 May 2006 (UTC)[reply]

MediaWiki does have a concept of stubs, but so few users have discovered the user preference required to make it recognise them, that few people know it. It also knows that any page containing Template:Disambig can be referred to as a "disambiguation" page.

One solution for the protection issue here might be to split protection data off into a new table, using namespaces and titles to keep tabs on it. This would allow administrators to protect pages that didn't exist. robchurch | talk 22:08, 24 May 2006 (UTC)[reply]

What we do at Uncyclopedia, to prevent them showing up in Special:Randompage, is make the pages a protected redirect to a Project: namespace page with a message about why the page was locked. This would solve the problem without changing the code. (see Uncyclopedia:Uncyclopedia:CVP). --Splarka (rant) 07:10, 25 May 2006 (UTC)[reply]
PS: It would also fix the pollution of Special:Shortpages discussed here. --Splarka (rant) 07:53, 25 May 2006 (UTC)[reply]

Splarka's suggestion could do with presenting on the Village Pump for Proposals. 86.133.53.232 03:43, 27 May 2006 (UTC)[reply]

Redlinked images

When I click on a redlinked article that's been deleted, the creation page that opens up includes a link to the deleted edits and deletion log at the top ('cause I'm an admin). When I click on a redlinked image, however, I just get the image upload page with that image name in the first field. Is there any way to have the deleted edits link show up on that page? The only way to get it seems to be to copy the name of the image into the search box. Thanks. Chick Bowen 05:21, 25 May 2006 (UTC)[reply]

Quite possible. Open a feature request for it and I'll poke it if I get a moment or two. 86.133.53.232 03:20, 27 May 2006 (UTC)[reply]

Wikipedia Zeitgeist

I posted the following on the proposals page I may have been looking in the wrong place but I couldn't find anything which lists which are the most searched articles. I'm thinking along the lines of Google Zeitgeist or eBay Pulse., and it was suggested I post here. Would this be feasible? I'm sure there'd be plenty of interest in such a feature. (Yorkshiresky 19:50, 25 May 2006 (UTC))[reply]

That question recurs periodically. --DLL 21:30, 25 May 2006 (UTC)[reply]
Theoretically possible, but not set up at the moment. --Brion 00:31, 26 May 2006 (UTC)[reply]
Special:Mostrevisions is a slightly different take on that (it's cumulative, not yearly or daily, and it counts writers, not readers), but with Wikipedia's exponential growth, it's weighted towards more current events. --Interiot 00:43, 26 May 2006 (UTC)[reply]
A while ago, I was writing an informal article about Wikipedia and wanted to illustrate it with examples of the most-read articles, and was surprised that this statistic was not kept. Oddly, there are obsessively exhaustive statistics about all sorts of things at WP:STAT and stats.wikipedia.org, but this is not among them, presumably for performance reasons. But given the frequency with which this is requested, I wonder if it might be a better priority for the limited CPU cycles on our servers that can be devoted to stats than some of the rather esoteric stuff that is kept. Since I'm not a developer, what would be a good way of communicating that? --MCB 00:11, 28 May 2006 (UTC)[reply]

Thumbnail rendering issues

I've been tracking down problems with rendering of thumbnails and SVG images; we're now logging all failures, and various sources of trouble are being poked at.

Temporary rendering errors should now be handled more gracefully as well; instead of embedding an ugly error message into the page, the error is logged in the background and the page will optimistically assume the image will be fixed in the future (which, likely, it will be when the page gets loaded). This should cut down on annoying "Image thumbnailing failed!"-type errors getting stuck in cached pages. --Brion 01:19, 27 May 2006 (UTC)[reply]

Wiki Monitor

Can users ping Wikipedia, or is this just an admin function? Wallie

What does this mean? You can do a normal ping on en.wikipedia.org in the standard manner, of course. Your question is far too ambiguous to expect a targeted answer. robchurch | talk 11:59, 27 May 2006 (UTC)[reply]

Vertical space problem

Why is there a lot of vertical space at the top of this article? I thought it was the template being included there, but after removing several newlines in it, and trying both null edits and real edits on the article itself, I can't get the space to go away. - dcljr (talk) 19:09, 27 May 2006 (UTC)[reply]

It is the tempalte. I'm trying to fix it but I'm causing new problems. The problem was far too many table rows in the template. Jeltz talk 19:55, 27 May 2006 (UTC)[reply]
I think that it is fixed now. What the fix was was removing all the "|-" in the table since there already were html table rows. I guess this made either the browser or Wikipedia go nuts. Jeltz talk 20:07, 27 May 2006 (UTC)[reply]

Sections archived on 00:09, 5 June 2006 (UTC)

Line Spacing After Super/Sub script

Using subscript or superscript on wikipedia introduces increased line spacing which doesn't look so nice. Why can't this be fixed? It doesn't do it in word processors. You don't find it in books. So why here? --Username132 (talk) 20:32, 23 May 2006 (UTC)[reply]

Welcome to the web. --Brion 00:08, 24 May 2006 (UTC)[reply]
Okay, thanks, but I've been using the web for eight years. It doesn't explain the wikipedia subscript/superscript issue. --Username132 (talk) 18:31, 24 May 2006 (UTC)[reply]
Web browsers are not word processors, and in fact they sometimes behave differently. You are reading Wikipedia with a web browser. The way that web browsers render text with subscripts and superscripts matches the description you give. --Brion 22:48, 24 May 2006 (UTC)[reply]
Thank you. I'll go and hassle Opera. --Username132 (talk) 23:05, 24 May 2006 (UTC)[reply]
Given that web browsers are not word processors, Wikipedia should avoid the use of subscripts and superscripts unless absolutely necessary. The ref tags being added to so many articles make these articles unreadable. The template or style should be changed so that footnotes are not superscripted. -- FRCP11 14:37, 25 May 2006 (UTC)[reply]
It might be possible to fix this with CSS, using something like SUB, SUP { line-height: 1%; }. Test case:
This is a narrow paragraph containing absolutely no superscripts or subscripts at all. It should therefore, presumably, have a constant line spacing.
This is a narrow paragraph containing superscripts(like this) and subscripts(like this). Hopefully it will nonetheless have a constant line spacing.
This is a narrow paragraph containing superscripts(like this) and subscripts(like this). It will therefore probably not have a constant line spacing.
While this does work in my browser (Firefox 1.5), the difference it makes is a whopping one pixel. However, if other browsers (as it would seem) render the second and third examples above differently, the CSS fix may be worth considering. —Ilmari Karonen (talk) 12:51, 28 May 2006 (UTC)[reply]

Linking to an edit page, but with the some content in the URL

I want to be able to like to a page that doesn't exist, and have it fill the edit box with a basic template. I'm sure I've seen it, but I can't remember where, so I can't see how it was done. So, how do I do that? --Lorian 22:15, 23 May 2006 (UTC)[reply]

m:Help:InputboxSimetrical (talk • contribs) 05:19, 24 May 2006 (UTC)[reply]
Ok well I have got the code I need to make an input box, but how would I make this: somerandompagethatdoesntexist preload another page, if it doesn't exist? --Lorian 07:31, 24 May 2006 (UTC)[reply]
Like this.Ilmari Karonen (talk) 12:53, 28 May 2006 (UTC)[reply]

ref/note referencing system problems

Hi, I have a problem when using the ref/note system for citing references when several points in the article text point to the same source. I have been using ref label|XXX|No\a etc fine for a while on the Chew Valley page without any problem. I was then advised to rearrange the section & renumbered the refs to take account of this. Now, if you look at the geology section you will see several links to No [7] in the text which point to different resources in the list at the end. I (& several others) have fiddled with this but are unable to get this to work properly. Any help would be appreciated. Rod 19:58, 24 May 2006 (UTC) Possibly resolved by using User:Cyde/Ref converter Rod 20:11, 24 May 2006 (UTC)[reply]

It looks fine to me. Can you describe the problem? --Brion

I finally managed to fix this by using User:Cyde/Ref converter Rod 08:27, 28 May 2006 (UTC)[reply]

Special:Contributions/newbieIP?

Special:Contributions/newbies shows contributions by individuals whose accounts are new. Is there an equivalent for IP addresses, i.e., people posting from IP addresses that don't already have a posting history to Wikipedia? — WCityMike (talk • contribs) 16:30, 26 May 2006 (UTC)[reply]

Recent Changes has a checkbox to show only non-logged-in contributions. — Saxifrage 01:18, 27 May 2006 (UTC)[reply]
Appreciate the answer, but it's not what I'm looking for. Recent changes by non-logged-in individuals might be longstanding editors on Wikipedia who've just chosen not to register. Special:Contributions/newbies reflects new users' edits -- basically, the system recognizes that the users are new and puts them in this category. Does it do the same for new IP addresses -- recognizing, say, that the IP address ##.###.##.## hasn't ever logged in here before and assigning those edits to a particular log? Newbie IPs, in other words? — WCityMike (talk • contribs) 01:30, 27 May 2006 (UTC)[reply]
No, there's no functionality similar to that which you describe at this time. "Newbies" is calculated based on the age of the user account...to look at the history of IP addresses, and grab every "new" IP address, we'd have to scan the entire revision table every time the page was viewed, which is an absolute nightmare from a performance perspective. robchurch | talk 12:00, 27 May 2006 (UTC)[reply]
Okay. Thanks.  :) — WCityMike (talk • contribs) 15:32, 27 May 2006 (UTC)[reply]
Also, I would caution against drawing too many conclusions about whether an IP address is "new" or not, since many IPs are shared (proxy) or dynamic. --MCB 00:15, 28 May 2006 (UTC)[reply]

To avoid all kinds of scary breakage (and potential security issues with broken HTML), <nowiki> and friends are now explicitly excluded from use in URL and wiki links.

It seems a few people have been using this little hack with <nowiki> to get around the way template parameters containing "=" get split up; alas this now will break your links. As a replacement, I've fixed it so that = (&#61;) and other standard character entities are properly decoded/normalized in URL links. (This had been special-cased for & previously, while any others remained broken.)

So you can write http://example.com/?foo=bar with &#61; and it should pass through a template parameter without breaking the parameter. --Brion 01:13, 27 May 2006 (UTC)[reply]

I had fixed things so that {{urlencode:}} and friends worked with message parameters correctly, but had to revert it due to some weird side effect which produced massive breakage on the Chinese-language site. Will investigate more later... --Brion 09:33, 28 May 2006 (UTC)[reply]

Printing underlined links.

None of the skins print them, and my attempts at editing User:Jeandré/monobook.css didn't work. -- Jeandré, 2006-05-27t14:15z

You have an error in your syntax. It should be:

@media print {
a:link { text-decoration: underline !important; }
}

. Notice the location of the semicolon. I haven't tried it so it might not solve your problem but that is the way the syntax should be. Jeltz talk 15:31, 27 May 2006 (UTC)[reply]

(edit conflict) Try removing the semicolon before the !important. Also take a look at common/commonPrint.css to see how they are removed by default. --cesarb 15:33, 27 May 2006 (UTC)[reply]

It doesn't seem to removed by commonPrint.css. At least not for "blue" internal links. Redlinks and external links could have their underlines removed by commonPrint.css, but I'm not sure. Jeltz talk 15:39, 27 May 2006 (UTC)[reply]
/* MSIE/Win doesn't understand 'inherit' */
a, a.external, a.new, a.stub {
	color: black ! important;
	text-decoration: none ! important;
}

/* Continue ... */
a, a.external, a.new, a.stub {
	color: inherit ! important;
	text-decoration: inherit ! important;
}

Looks like removing the underlines for all links (including internal links) to me. --cesarb 16:15, 27 May 2006 (UTC)[reply]

a:link, a:visited {
    color: #520;
    background: transparent;
    text-decoration: underline;
}

I'm referring to this. This is before the part you pasted here but I think that this one might have precendence. Not sure though about the rules of precendence in CSS. Jeltz talk 16:53, 27 May 2006 (UTC)[reply]

That does not have !important. Rules with !important always have precedence over ones that don't, irregardless of any other precedence rules. --cesarb 03:45, 28 May 2006 (UTC)[reply]

Thanks. The following works in Ubuntu/Firefox:

@media print {
a, a.external, a.new, a.stub, a:link, a:visited { text-decoration: underline !important; }
}

Should it be added to the meta user style page, and if so can it be safely shortened? -- Jeandré, 2006-05-27t18:46z

What is going on with this image?

Why does this image [2] display all screwy in the articles its in? can someone just delete the old one for me?--Deglr6328 06:27, 28 May 2006 (UTC)[reply]

The 250px thumbnail for that image is persistantly cached a bit, it looks like. "Fixed" (in quotes) by resizing slightly. --Splarka (rant) 07:15, 28 May 2006 (UTC)[reply]
thanks!--Deglr6328 08:47, 28 May 2006 (UTC)[reply]

Semantic Wikipedia

This week I attended a day of the WWW2006 conference in Edinburgh. Although I didn't attaend the particular paper another delegate pointed it out to me after I talked about WP in the context of peer review processes. The paper by Max Völkel, Markus Krötzsch, Denny Vrandecic, Heiko Haller and Rudi Studer entitled Semantic Wikipedia talks about "an extension to be integrated in Wikipedia, that allows the typing of links between articles and the specification of typed data inside the articles in an easy-to-use manner", as part of becoming "a resource of semantic statements, hitherto unknown regarding size, scope, openness, and internationalisation. These semantic enhancements bring to Wikipedia benefits of today's semantic technologies: more specific ways of searching and browsing. Also, the RDF export, that gives direct access to the formalised knowledge, opens Wikipedia up to a wide range of external applications, that will be able to use it as a background knowledge base."

Does anyone know if these proposals have been considered by Wikipedia & whether there are any plans or timescales for implementation? Rod 08:34, 28 May 2006 (UTC)[reply]

Maybe someday. There are no specific plans. There are no timetables. --Brion 09:31, 28 May 2006 (UTC)[reply]

For about five pages I've tested {{PAGENAME}}/, the subpage / (slash). It usually works, but something (my browser or the server) sometimes redirects the base page to a new slash-subpage (or vice versa) for a day. It's probably a bad idea here - for PURL partial redirects it's a feature.

How about automagically redirecting all namespace:basepagename/ slash-subpages to [[Special:Prefixindex/namespace:basepagename/]]? The special page doesn't depend on the "real" subpage-feature, it should work in all existing "real" (non-negative) namespaces. A few legacy slash-subpages might be a problem.

The proposed feature could also replace Template:Subpages(edit talk links history). Example, it would redirect Wikipedia:Wikipedia Signpost/ to Special:Prefixindex/Wikipedia:Wikipedia Signpost/. -- Omniplex 15:48, 28 May 2006 (UTC)[reply]

Whitespace handling

I think that one aspect of how Mediawiki handles whitespace here at Wikipedia should be changed. One linbreak is ignored and two linebreaks means a new paragraph, but what bothers me is that addtional whitespace you get if you have more than two newlines. I have often seen it breaking the nice flow of articles with too much space between paragraphs. I propose chaning this bahviour to just ignore additional linebreaks after the first two (like LaTeX does). This would make life easier for the editors. An additional benefit is that this removes one of the possible issues with skins with indentation instead of margin between paragraphs. I see no major con with this change since it is rare to want to force longer distances between paragraphs and in those few cases I think <br/> could be used, I can't really think of anythin useful that this would break. It might be quite major change but I still think it would be a great improvement.

Discussion at WP:VPR#Whitespace handling Jeltz talk 18:54, 28 May 2006 (UTC)[reply]

Then don't put extra whitespace. --Brion 22:27, 28 May 2006 (UTC)[reply]
I wrote this because I don't want people to have to care about whitespace. I personally am careful not to leave unnecessary ugly whitespace in an article that I have edited and I have removed whitespace from a lot of articles. I have grown tired of copyediting away whitespace (but not that much, I like copyediting :)). But should we really demand that all users have to care about something like that with little benefit (as far as I can see)? And why not try to make the syntax better and easier to use for everyone? Comment tags and templates actually make this a bit tricky sometimes.
My point is that I think the way Wikipedia handles whitespace currently does more harm than good. What I would like to hear is other peoples oppinions, and also the feasilbility of changing this behaviour. Is this something the people on Wikipedia desire? All who answered at WP:VPR were possitive, but they were quite few. I could understand that this could be a lot of work and might break something, and maybe not an important feature to most people. But I don't think that "don't put extra whitespace" is an answer because I think that that is the problem. Another thing that people have to think of, and for little benefit as far as I can see. Jeltz talk 23:58, 28 May 2006 (UTC)[reply]

Title-changing template: Followup

Okay, I established that the bit of code -{T|realtitle}- which changes an article's displayed title is implemented in meta:Automatic conversion between simplified and traditional Chinese. What I haven't figured out is, why isn't this used to fix the capitalization problem? For example, we could put -{T|iPod}- at the top of IPod and it would show up correctly. This is not a CSS hack and it does not move the page. Ashibaka tock 21:28, 28 May 2006 (UTC)[reply]

Odd line-break in References section

See Nokia Eseries and look at the References section. Why does entry #1 have that odd line-break? I must have made a syntax-error, but I can't find it. --Notmicro 22:27, 28 May 2006 (UTC)[reply]

Because of the break between the ref tag and the {{cite web}}. Ashibaka tock 22:34, 28 May 2006 (UTC)[reply]
Thanks for the help, but why doesn't the same thing happen with the 2nd reference? --Notmicro 22:49, 28 May 2006 (UTC)[reply]
There's a real mystery. It probably has something to do with the cite.php code. Ashibaka tock 23:50, 28 May 2006 (UTC)[reply]

Sections archived on 00:14, 6 June 2006 (UTC)

How to get an entry based on value or null (Revised question)

I want to create a template which is used to get either a specific entry or all of them depending on the parameter.

Let's say I have a list like this

  • Item 01 - May
  • Item 02 - Critical
  • Item 03 - Exact

And I want to put it into a template, let's call it FUZZ so that if I simply use a macro call of {{FUZZ}} I get

  • Item 01 - May
  • Item 02 - Critical
  • Item 03 - Exact

But if I use {{FUZZ|01}} I get

  • Item 01 - May

But if I use {{FUZZ|02}} I get

  • Item 02 - Critical

I know in some way this requires {{#if:}} or some macro, but I'm not sure how.

Is there a way I can do this? Second, is there a way I can have the macro generate text only if the parameter is null? So that I could put a header if the parameter is null but no header if the value of {{{1}}} is not null.

Paul Robinson (Rfc1394) 22:53, 23 May 2006 (UTC)[reply]
I'm not sure if this is exactly, what you want, but perhaps something like this?
{{#switch: {{{1|}}}
 | 01 = *Item 01 - May
 | 02 = *Item 02 - Critical
 | 03 = *Item 03 - Exact
 | #default = 
  *Item 01 - May
  *Item 02 - Critical
  *Item 03 - Exact
}}
You might want to check out m:ParserFunctions for a better explanation of things. In regards to null-parameters, {{#if: {{{1|}}}|do this if it is set|do this if it is unset (null)}}, is that what you're looking for? Hope that helps! Jude (talk) 02:46, 24 May 2006 (UTC)[reply]
I'm already doing something like that now, what I think I want is something like this:
{{#switch: {{{1|}}}
 | 01,#default = *Item 01 - May
 | 02,#default = *Item 02 - Critical
 | 03,#default = *Item 03 - Exact
}}

or

I'm already doing something like that now, what I think I want is something like this:
{{#switch: {{{1|}}}
 | 01
 | #default = *Item 01 - May
 | 02
 | #default = *Item 02 - Critical
 | 03
 | #default = *Item 03 - Exact
}}

So that I only have to include each of the items once. For default, it shows all of them, for any one of the entries, it only shows that one. Paul Robinson (Rfc1394) 22:40, 25 May 2006 (UTC)[reply]

You could do it using multiple #switches:
{{#switch:{{{1|}}}|01|= *Item 01 - May}}
{{#switch:{{{1|}}}|02|= *Item 02 - Critical}}
{{#switch:{{{1|}}}|03|= *Item 03 - Exact}}
Notice that each #switch has two equivalent cases: the item number and the empty string. This solution does, however, have the problem that entering an invalid (but non-empty) item number causes no items to be shown. —Ilmari Karonen (talk) 12:33, 28 May 2006 (UTC)[reply]
Rfc1394 re-posted the question below, and I proposed almost the exact same solution (but with <br /> instead of *. GMTA! --Splarka (rant) 02:23, 29 May 2006 (UTC)[reply]

I want to keep a list of wikipedia links for reference, but I dont want it in my whatchlist and see it popup every time there is a change or edit. Is there a way to do this and keep it private like my watchlist? --Shlomke 20:56, 28 May 2006 (UTC)[reply]

There's a good chance your web browser has included this advanced functionality for over a decade. You'll probably find it hidden under such names as "Bookmarks" or "Favorites". --Brion 22:26, 28 May 2006 (UTC)[reply]
You don't have to make it obvious you are 'watching' the pages, Shlomke. Create a sockpuppet or ask another user to create a user page for you with the list of articles wikilinked (it can be anyone, anywhere on en.wp, like for example User:Splarka/secretshlomkepage). Privacy through obscurity. Then, you just bookmark the 'related changes' link (eg Special:Recentchangeslinked&target=User%3ASplarka/secretshlomkepage) which gives you a list of all edits to the pages listed there over the last several days. Make sure to only edit it as another user or as an IP. No way for anyone to trace it back to you, except via this conversation ^_^. --Splarka (rant) 02:32, 29 May 2006 (UTC)[reply]
Theres nothing I'm trying to hide here. What I'm trying to say is, I have a list of links on wikipedia just for reference and I don't want to see them every time I check my watchlist. I know I can make a subpage on my username and put them all there. However I enjoy the privacy of my watchlist (hence keep it private), and was wondering if such an option exists already on Wikipedia (i.e. a "Bookmarklist"). This is realy a matter of conveniecne (over "Favorites" or something in my browser) and I'm not trying to "hide" my edits. But it doesn't look like this exists yet, but thanks anyway. --Shlomke 03:33, 29 May 2006 (UTC)[reply]
I think that is probably much easier to do inside your browser. That's what I do, using Firefox. I have a list of about 40 Wikipedia links (to tools, lists of templates and categories, pages like WP:AFD and WP:AIV, etc.) and keep them in my browser toolbar so that they're in an always-accessible pull-down menu. Much more convenient than anything server-based like a watchlist or private links page. -- MCB 18:24, 29 May 2006 (UTC)[reply]
Thanks, I'll try that. --Shlomke 20:03, 29 May 2006 (UTC)[reply]

Seeing edits to deleted articles?

Is there any way to see what edits a user has made to deleted articles? It would appear that once an article is deleted, if you look at a user's contributions, their edits to those articles are no longer listed. It's occasionally useful to see these, when trying to decide if a user is a habitual vandal or not. -- RoySmith (talk) 00:07, 29 May 2006 (UTC)[reply]

Try [3], but keep in mind it won't be up to date for en.wikipedia (at least, it's not showing my latest deleted edits). --james(lets talk) 00:44, 29 May 2006 (UTC)[reply]

Is anybody else having the problem that redlinks suddenly include a question mark at the end? I know there was an option in preferences to either have red links, or regular links with a question mark, but now both seem to have the question mark. Was this something that was changed and I missed the announcement, or is somthing going wonky? Essjay (TalkConnect) 02:51, 29 May 2006 (UTC)[reply]

As I can't replicate the issue in my browser, your cache probably went bananas. Doing a hard reload should do the job. Titoxd(?!? - help us) 03:18, 29 May 2006 (UTC) Titoxd(?!? - help us) 03:18, 29 May 2006 (UTC)[reply]

It seems to have fixed itself, but I couldn't get anything to work for about an hour...I closed out, I cleared caches, I did just about everything I could think of. Even the box in preferences that says "Format links like this" was wonky. Anywho, it's fixed itself now. Essjay (TalkConnect) 03:35, 29 May 2006 (UTC)[reply]

This is a known cache buglet. Hopefully we'll have it sorted out soonish. --Brion 05:08, 29 May 2006 (UTC)[reply]

Template:Cite_web date-rendering

I'm puzzled by the behavior I'm just now seeing in, for example, Template "Cite web" and Template "Cite press release". When I view the results, the date defined by "accessdate =" renders following my Preferences setting for "Date and time / Date format"; however the date defined by "date =" does not conform to my Preferences setting, and appears to render "raw". See for example the External Links section in DisplayPort. Is this behavior correct, or have I missed something? --Notmicro 07:37, 29 May 2006 (UTC)[reply]

See Template talk:Cite web#Removing_links_from_dates. Someone thought that the formatting was unnecessary. Go figure. --iMb~Meow 11:47, 29 May 2006 (UTC)[reply]

"Wikipedia" search result...

Hi there.

I have no clue whether or not this is the right place to post this, but could someone look into the "http://en.wikipedia.org/wiki/Wikipedia"..?

Don't know if it's supposed to look like this, but I guess not...

Thanks,

Troels.

The page was vandalized and quickly reverted (less than a minute later). If you see vandalism on any page in Wikipedia, you're encouraged to directly fix it. Please see Wikipedia:Vandalism#Dealing with vandalism and Help:Reverting. -- Rick Block (talk) 17:29, 29 May 2006 (UTC)[reply]

Comments requested. -- Zanimum 13:43, 29 May 2006 (UTC)[reply]

Not sure I agree with the principle, but it should be noted that the improvements I hope to make to the blocking mechanism during this summer will allow for the technical part required. robchurch | talk 13:59, 29 May 2006 (UTC)[reply]

nowiki and ref weirdness

Recently (Possibly related to #nowiki_in_links) ref tags that contain something wrapped in nowiki have started to produce odd output (GUID?).
Example: Test[1] (<ref><nowiki>stuff in nowiki</ref></nowiki)

Result:

  1. ^ stuff in nowiki

(^ UNIQ71e533182945ac5c-nowikib61493f7bc55c1200000001 )

One simple solution would obviously be not to do it, but there is a weird set of circumstances on Daniel Brandt that people wish to cite a site that is in the spam blacklist by listing the url and not having it linked. Another possible solution is to use <ref>http://url</ref> but that might trip the spam blacklist were it to start working again. Kotepho 15:22, 29 May 2006 (UTC)[reply]

I'm investigating this. --Brion 21:43, 29 May 2006 (UTC)[reply]
Aaaand it's fixed. --Brion 23:08, 29 May 2006 (UTC)[reply]
Seems quite similar to bugzilla:5384; it's probably the same cause (failing to expand the placeholders). --cesarb 22:55, 29 May 2006 (UTC)[reply]

User with some admin capabilities

I was just wondering if it was technically possible for a user to have only some admin capabilities. For example, the only admin tool Im really interested in is the delete button. Is it possible to have only that. --Osbus 15:24, 29 May 2006 (UTC)[reply]

No, currently only admins have access to admin capabilities, and all of them at that. While technically it could be implemented (with some work), I don't believe there is any consensus to do so. Wikipedia:Requests for rollback privileges was proposed a while ago, but it didn't get anywhere. ~MDD4696 15:37, 29 May 2006 (UTC)[reply]

It is technically possible to grant individual permissions separately. We don't have any interfaces to make it possible on Wikimedia sites, at present. For instance, bot status has to be granted via MakeBot; it adds a user to the bot group, which has bot permissions. Sysops have rollback, delete, protect and block permissions, etc. robchurch | talk 22:32, 29 May 2006 (UTC)[reply]

Watchlist

Is there a way to import my watchlist from one account to another? --GeorgeMoney T·C 20:32, 29 May 2006 (UTC)[reply]

No. robchurch | talk 20:50, 29 May 2006 (UTC)[reply]

Sections archived on 00:16, 7 June 2006 (UTC)

Linebreaks ignored in blockquotes?

Look at the wikicode for this.

foo

bar

baz

All linebreaks are ignored. To get one to work, you have to manually insert <br /> or <p>. Why? —Simetrical (talk • contribs) 05:43, 29 May 2006 (UTC)[reply]

Since this issue has been raised before, allow me to not insult your intelligence by failing to explain how to use a search engine. robchurch | talk 13:54, 29 May 2006 (UTC)[reply]
I'm probably using the wrong keywords, but I searched the VPT, its archives, and Bugzilla for "blockquote", "line break", and "<br" and found no relevant results anywhere. —Simetrical (talk • contribs) 06:14, 30 May 2006 (UTC)[reply]

List items

I have recently noticed the the indention of list items is different for the ordered and unordered lists.

Example:

  • This is a level 1 unordered list item
    • This is a level 2 unordered list item
      • This is a level 3 unordered list item
        • This is a level 4 unordered list item
  1. This is a level 1 ordered list item (note this lines up with a level 2 unordered list item)
    1. This is a level 2 ordered list item (note this lines up with a level 4 unordered list item)
      1. This is a level 3 ordered list item (note this probably lines up with a level 6 unordered list item)
        1. This is a level 4 ordered list item (note this probably lines up with a level 8 unordered list item)

The indention of them is quite different. Could this be modified so that the levels of both ordered and unordered lists are the same at each level?
-- Lady Aleena talk/contribs 11:03, 29 May 2006 (UTC)[reply]

It looks like it's just a css difference. Try looking at Special:mypage/monobook.css, esp. any ul or ol entries. But to get this changed on a wide scale would probably be something difficult to gain consensus for. – Xolatron 15:45, 29 May 2006 (UTC)[reply]

From http://en.wikipedia.org/skins-1.5/monobook/main.css :

ul {
	line-height: 1.5em;
	list-style-type: square;
	margin: .3em 0 0 1.5em;
	padding: 0;
	list-style-image: url(bullet.gif);
}
ol {
	line-height: 1.5em;
	margin: .3em 0 0 3.2em;
	padding: 0;
	list-style-image: none;
}

You can add

ol {
margin-left: 1.5em;
}

to Special:mypage/monobook.css to make the ol indentation the same as ul. — Ian Moody (talk) 16:22, 29 May 2006 (UTC)[reply]

You mean to tell me that there might be resistance to this tiny change? It boggles the mind.
-- Lady Aleena talk/contribs 17:21, 29 May 2006 (UTC)[reply]

I changed the Monobook skin once. It looked nicer, IMO, but I had to revert it 'cause I kept getting screamed at. Never underestimate humans' desire to whine about issues that are, in fact, of bugger-all importance. robchurch | talk 22:34, 29 May 2006 (UTC)[reply]

How does one get the monobook.css changed without some going up in flames? Is there a process akin to the Categories for deletion process or is this it?
-- Lady Aleena talk/contribs 06:06, 30 May 2006 (UTC)[reply]
Ask at MediaWiki talk:Monobook.css. Any admin can do it. —Simetrical (talk • contribs) 06:16, 30 May 2006 (UTC)[reply]
You can change your own monobook.css without any problem, but any changes to the meta one will likely cause much more trouble than it's worth. Unfortunate, but any change that effects enough people is always difficult to pass, because at least a small percentage will oppose it, but this could become a large number of people. – Xolatron 19:59, 30 May 2006 (UTC)[reply]

RSS feed?

Did I miss some kind of announcement? I just noticed that article histories have an RSS feed! That seriously rules. :) Is this just a test or will this be permanently? --Conti| 01:34, 30 May 2006 (UTC)[reply]

I might change the number of items, but yes it's here to stay unless something goes horribly awry. --Brion 03:01, 30 May 2006 (UTC)[reply]


why does happen?

I attempted to post a few times, but the resulting page looks messed up?

example

\\\'\\\'\\\'Charon\\\'\\\'\\\'

Because you are using a misconfigured web proxy. Such proxies are not allowed and this IP has now been indefinitely blocked. More at User talk:216.66.21.176. Dragons flight 04:42, 30 May 2006 (UTC)[reply]

Is there an error in the Norwegian (native speaker) babel-box entry?

I noticed that all the users listed as native norwegian speakers, i.e. {{User no}}, were presented with the following text: "Denne redaktøren har norsk som morsmål." A word-by-word translation to English is "This editor has Norwegian as [his/her] native language". In Norwegian, "Redaktør" implies administrative privileges, and is much closer to "Administrator" than to "User". Being a native Norwegian speaker myself, I was confused, first thiking: user:gisle is an administrator? Well, that makes sense because he is a well known and respected computer specialist. But when I browsed the other users, I realised that this was probably a mistake. Could somebody please have a look into this, and consider changing the text to "Denne brukeren har norsk som morsmål."? --vibo56 08:52, 30 May 2006 (UTC)[reply]

Why don't you change it yourself? It's a wiki, after all. Or you can at least propose the change on the talk page. Template:User no. pfctdayelise (translate?) 09:17, 30 May 2006 (UTC)[reply]
Done, and thanks for the tip. I didn't know how to get there. --vibo56 09:37, 30 May 2006 (UTC)[reply]

Change Namespaces

I'm the sole administrator (and user, at the moment) of the Latin Wikibooks. One of the namespaces, however, has been left as 'Wikibooks' when it should be 'Vicilibri'. I've had a look round the MediaWiki namespace, but it appears to be buried deeper than that. How do I change the namespace? Daniel () 12:25, 30 May 2006 (UTC)[reply]

File a request on BugZilla asking to have the site name set to the correct translation. robchurch | talk 15:08, 30 May 2006 (UTC)[reply]

SVG purge and recreate

...doesn't seem to be working, or I'm doing it wrong. See Illinois Route 255.

The image linked to is: http://upload.wikimedia.org/wikipedia/commons/thumb/9/90/I-255.svg/25px-I-255.svg.png

and the link to purge it is: http://upload.wikimedia.org/wikipedia/commons/thumb/9/90/I-255.svg/25px-I-255.svg.png?action=purge

but I still haven't been able to get the image to be recreated. I've purged in Camino and checked in Safari, shift and option-reloaded, etc. but it doesn't seem to be working. Is this the right procedure? —Rob (talk) 17:54, 30 May 2006 (UTC)[reply]

The purge link for that image is http://commons.wikimedia.org/wiki/Image:Illinois_255.svg?action=purge . --Brion 19:14, 30 May 2006 (UTC)[reply]

Sections archived on 00:11, 8 June 2006 (UTC)

Image tagging

Got a question on image tagging that wasn't made clear by the copyright tag page. If I find something like a photo taken by a user that doesn't make me question if it's a copyvio, but it's orphaned, should I mark it with source and license tags as I have here, or something else? And, same question, if it seems to be an obvious copyvio problem, is this correct? Thanks. Tijuana BrassE@ 04:21, 24 May 2006 (UTC)[reply]

Another one: how do I treat images that are only used for user pages, i.e. Image:Brady.jpg ? Tijuana BrassE@ 04:27, 24 May 2006 (UTC)[reply]
  1. All images must have source and license information, whether or not they're orphaned.
  2. Obvious copyvio problems should be {{imagevio}}d and PUId, as you did.
  3. All images must have source and license information. Images used outside of articles must have a free (not fair use) license.
  4. Ask questions like this at Wikipedia talk:Copyrights rather than here. That way you're more likely to get answers from people who know about copyright. —Simetrical (talk • contribs) 05:23, 24 May 2006 (UTC)[reply]
Not copyright related, but I'd also like to add that user mugshots and other photos of "random" non-notable people should just be put straight on IFD if they are not actualy used on a userpage or something (these ones are used so it doesn't relate to them spesificaly). If the uploader can't be bothered to use them (on Wikipedia at least, soemtimes they use Wikipedia for file storage and use the images elsewhere) we should just get rid of them regardles of license, since all they do is clutter up the place and waste space. --Sherool (talk) 13:06, 30 May 2006 (UTC)[reply]
Well, unless they're conceivably useful for something. Then they should be Commonsed. —Simetrical (talk • contribs) 07:52, 31 May 2006 (UTC)[reply]

User:120

hello, I'm a French contributor registered on French wiki as fr:Utilisateur:120 and I would like to use "120" as username to register on English wiki. I can't create my account even if it seems there is no User:120 already registered. What should I do ? (apart from choosing another nick...). Thanks, (you can either answer here or on French wiki)

I would've said that 120 is already taken since there is a "User contributions" link, but the user creation log doesn't list its creation. How come there's a contributions link for User:120 but not User:Ksbndslkcjn? ~MDD4696 15:00, 29 May 2006 (UTC)[reply]
Perhaps 120 trying to create his account as 120 caused a partial creation, resulting in a contributions link but no creation log instance. 120, you should be able to create an account; see User:123 - he exists. If you'd like, I can try to create an acocunt User:120 with a generic password and tell you it here, or you could leave your e-mail and I could e-mail you a difficult to guess numbers and letters combination, which you could change after logging in. – Xolatron 15:39, 29 May 2006 (UTC)[reply]
Yes, I think you're right. Maybe it's necessary to delete the partial creation. Anyway, if you can create the account, my e-mail is <vincent point mourre et wanadoo point fr> Thanks, fr:Utilisateur:120
Didn't work. It says the account already exists on my computer as well. Have you tried logging in with the password you originally chose? If that doesn't work, you might need an admin or bureaucrat to delete the account. Otherwise you could choose a name such as "l20" (lowecase L then 20) or I20 (uppercase i then 20). If you'd like to remove your e-mail and this comment, please do so. Sorry. – Xolatron 16:45, 29 May 2006 (UTC)[reply]
User accounts cannot be deleted. robchurch | talk 15:22, 30 May 2006 (UTC)[reply]
Ach! No, bad idea. Misleading usernames are bad. ~MDD4696 20:21, 29 May 2006 (UTC)[reply]
Um, Special:Listusers says that User:120 exists and has registered. The new user log wasn't around until recently, so it most likely did not log the creation of the user. What you need to do is to check if the user has any contributions, and if it doesn't, try to plead your case at WP:CHU. Titoxd(?!? - help us) 20:04, 29 May 2006 (UTC)[reply]
not even any deleted contributions: [4] Agathoclea 20:08, 29 May 2006 (UTC)[reply]
The account does exist, as indicated by Special:Listusers; as Titoxd has stated, the creation log does not list all accounts, only those created within the last few months. If the account existed prior to that time (there is no way for us to know) then it won't be logged. As for WP:CHU, we cannot rename an account to a name that already exists; the interface literally will not perform the request. Policy does not allow us to forcfully rename accounts, and without a policy allowing usurption, we are powerless to perform such requests. The best advice I can offer is to support the usurption proposal at Wikipedia:Usurpation. Essjay (TalkConnect) 12:37, 31 May 2006 (UTC)[reply]
My suggestion: this name is so simple that it's like to be in use at many projects. Rather than have a zillion variations, you're better off with just two. For example, I use "Deco" and "Dcoetzee" where that is not available. Deco 23:48, 31 May 2006 (UTC)[reply]

How can I get subject stats?

Is there any way - other than counting manually - of finding out how many articles there are in a higher level subject category? My interest is Category:Opera. Thanks. - Kleinzach 22:39, 29 May 2006 (UTC)[reply]

try CatScan. pfctdayelise (translate?) 09:45, 30 May 2006 (UTC)[reply]
Thanks. It wasn't really successful, but perhaps the English Wikipedia is impossible to count at the moment because of database corruption? Regards. Kleinzach 22:39, 30 May 2006 (UTC)[reply]
Oh yeah, good point. Sorry, I forgot about that. pfctdayelise (translate?) 02:38, 31 May 2006 (UTC)[reply]

help!

I wasn't sure where I should post this question, but...I can't log back in using my username PlainJane. I asked for a new password to be emailed, but it didn't ever arrive. What can I do? 220.253.35.37 01:08, 30 May 2006 (UTC)[reply]

Make a new username?...and if it concerns you, post a little thing on your userpage informing others of your old username. --Osbus 01:51, 31 May 2006 (UTC)[reply]

How can I find out when a project last underwent a SW upgrade?

Special:Version doesn't say when the version last changed. I guess this information is available somewhere, but where?

In not-unrelated news, commons:Special:Upload is missing the license selector box for languages other than English. bugzilla:6126 would like some eyes, thanks --pfctdayelise (translate?) 04:58, 30 May 2006 (UTC)[reply]

Several times a day, and all our sites run from the same software so no project is upgraded at a different time (with some rare exceptions over the last several years). --Brion 06:35, 30 May 2006 (UTC)[reply]
OK, so... has bugzilla:6126 been looked at, at least? Perhaps I am the only one who considers this reasonably important? If it's difficult to fix and will take some time that's fine, I'd just like to know that it has at least been looked at. pfctdayelise (translate?) 08:12, 30 May 2006 (UTC)[reply]

That bug has now been fixed. robchurch | talk 13:49, 31 May 2006 (UTC)[reply]

Reconciling Mozilla and Explorer formats

I have a problem with a format for a template. Referring to Template:AcademyAwardBestPicture_1961-1980, in which I used <br> to ensure that the movie title won't be separated from the year it won, the appearance looks alright when viewed from Explorer, but when using Mozilla, it looks like this. Can someone help us out? Copysan and Joey80 08:23, 30 May 2006 (UTC) (Please direct suggestions to Copysan since I don't have Mozilla in order to determine whether the suggestion is working properly, thanks!)[reply]

You've lost me. In that image, no movie title is separated from the year in which it won. For gluing them together, I don't see why you'd use <br> (or, since this is XHTML <br />). Instead, 1964:&nbsp;''[[Revenge of the Killer Vampires|Revenge&nbsp;of&nbsp;the&nbsp;Killer&nbsp;Vampires]]'', no? -- Hoary 08:57, 30 May 2006 (UTC)[reply]
The issue seems to have been resolved by User:Ian Moody. Kudos to him! It looks fine on my Firefox now. I'll ask User:Joey80 what he thinks. Copysan 20:29, 30 May 2006 (UTC)[reply]
You can also use {{nobr}} instead of &nbsp;. --cesarb 05:22, 31 May 2006 (UTC)[reply]

I had tried to add these shorts Kit_shorts_Barcelona to FC Barcelona, but i wasn't able to do it. Could somebody help me? Spurs229 15:21, 30 May 2006 (UTC)[reply]

There's apparently no way to use an image for the shorts of a football kit, only a color. I've restored the color that was there previously. —Bkell (talk) 15:29, 30 May 2006 (UTC)[reply]
that's too bad... thanx anyway Spurs229 16:17, 30 May 2006 (UTC)[reply]
Actually, it is possible to use an image. But which is more encyclopedic? The flat color makes it easier to tell what color the shorts are... using an image with texture makes that more difficult to determine. ~MDD4696 18:56, 31 May 2006 (UTC)[reply]

How do I revert a page that's been vandalized?

Is there some special way to do it? Or do you just go through the article and try to restore it? KarenAnn 00:44, 31 May 2006 (UTC)[reply]

See Help:Revert. --PS2pcGAMER (talk) 00:49, 31 May 2006 (UTC)[reply]