Jump to content

Wikipedia:Village pump (technical)/Archive O

From Wikipedia, the free encyclopedia

Glitch

The problem is when I'm logged in, on the upper maragine (correct me if I misspelled that, it looks like butter's sister) where it has "My Talk My Prefrences etc., etc." Well when I move my mouse on it, it shifts left far enough that "My Talk" is under the Wikipedia main page link. That would cause problems. The glitch starts when I go from Wikipedia to the Commons and back.--Calvinsupergenius 20:39, 27 February 2006 (UTC)

Ditto, when coming from other Wikiprojects. (Spelling fixed)--TJ 15:58, 2 March 2006 (UTC)

Bug 5116

I have requested a special kind of category piping at http://bugzilla.wikimedia.org/show_bug.cgi?id=5116. I will not go into specifics here; if you wish clarification, please leave them on the bugzilla page so I can make sure that the developers are also aware of my clarification. My proposal is mainly to deal with aircraft articles, though I'm sure that other topics have noted this problem as well. Ingoolemo talk 08:05, 27 February 2006 (UTC)

FYI, I think bug 491 is about the same thing. —Steve Summit (talk) 18:40, 2 March 2006 (UTC)

Revision navigation doesn't appear if a page is a redirect

This is copied from General complaints - I thought it was more likely to get a knowledgeable response here. Ta. --Cherry blossom tree 13:59, 23 February 2006 (UTC)

When searching through a page's History, if you choose a revision that is a redirect to another page, e.g. here and here, the revision navigation text, which reads like this:

From Wikipedia, the free encyclopedia
Revision as of 08:48, 11 October 2005; view current revision
← Older revision | Newer revision →

...will not appear. Can someone possibly fix this, please? -- RattleMan 01:36, 23 February 2006 (UTC)

Known, search BugZilla. Rob Church (talk) 22:42, 2 March 2006 (UTC)

e-mail subject as an argument to Special:Emailuser

How can I make a link to Special:Emailuser where a subject at my liking is included pr default in the Subject field of the email form? Like the subject variable in the mailto link in html. Shanes 20:38, 15 February 2006 (UTC)

Currently, all emails are sent with the title Wikipedia Email (or a similar variant). There doesn't seem to be any way to change it. There doesn't seem to be a bug for this, so I suggest you file one at Bugzilla as a feature request. — Ambush Commander(Talk) 21:20, 15 February 2006 (UTC)
I would caution though that e-mails sent through wikipedia should remain readilly identifiable as from wikipedia. So any user specified subject should probablly be appended to the standard subject, not replace it. Plugwash 22:15, 15 February 2006 (UTC)
Well, Wikipedia Email actually already is quite cryptic (it takes a while for me to figure out, "Oh, that's from Wikipedia!". It probably should be from wikipedia, with a reply-to set to the user who sent the email. Plus, there really should be a disclaimer appended to the bottom of the email. — Ambush Commander(Talk) 03:09, 16 February 2006 (UTC)
My main motivation for wanting this is to be able to automatically include the IP of blocked users in the mail-link on MediaWiki:Blockedtext. The page does note that users need to include the IP given in any mail, but I receive quite a few mails from auto-blocked users who ask to be unblocked but forget to include the IP. So I thought it could save some frustration and mailing back and forth if we just had it inserted in the subject field in the form. The subject field could also contain a string "Unblock Request" or some such that could up the importance level setting in any e-mail filters admins might have. Shanes 10:57, 16 February 2006 (UTC)
You're dealing with several different issues/feature requests at this point. — Ambush Commander(Talk) 23:51, 17 February 2006 (UTC)
No, I'm not. I don't understand why you think I am. Shanes 07:20, 23 February 2006 (UTC)

Okay... maybe not (after all, one of them getting implemented would probably be good enough). But, here's what we've covered:

  • Give users ability to edit a page like User:Example/Email which shows up on Special:Emailuser/Example, which allows users to put "reminders" to people emailing them.
  • Create a special type of email for complaining blocks, which automatically gives IP address (most people won't want to give their IP when emailing others for privacy concerns), or...
  • Allow a user to set-up multiple types of Wikipedia email forms accessible like Special:Emailuser/Example/Block with custom subjects/automatically appended things, etc.
  • Instead of using subject Wikipedia Email, allow users to specify a subject and simply prepend [Wikipedia Email] to it.

By the way, the subject variable in mailto: is not standard and while most mail clients honor it, it should not be used. — Ambush Commander(Talk) 21:16, 1 March 2006 (UTC)

We probably understand eachother, but to be clear: I just wanted a way to have a string inserted in the subject (or body, for that matter) field before the user edits and submits the mail form. The user can edit out any string inserted in any field before submitting it (just as in the html mailto: equivalent), which should take care of any privacy concernes. It's just that I get those mails from blocked users where they have simply forgotten to put the IP given to them on MediaWiki:Blockedtext ($3 to that page) in the actual form themselves making their mail pointless. I have to mail them back, wait for any respond, all of it just costing a lot of time and frustration on both parts. And I can see other uses than this special case for this option as well.
On your other note about how standard the subject argument to mailto: is, I do think it's a standard as much as RFC 2368 says:
The creator of a mailto URL cannot expect the resolver of a URL to understand more than the "subject" and "body" headers. Clients that resolve mailto URLs into mail messages SHOULD be able to correctly create RFC 822-compliant mail messages using the "subject" and "body" headers.
It doesn't say MUST, but SHOULD is pretty strong in RFC speak. But this is beside the point anyway, as I think it would be a nice option whether mailto: had it or not. Anyway, I understand now that it's not implemented in mediawiki and I should go beg on bugzilla. I probably will, next time I get frustrated by one of those IP-lacking unblock me mails. Shanes 23:22, 1 March 2006 (UTC)
Whoo, looks like I was mistaken (I knew that certain headers were not standard, here we go. Must brush up on my RFCs).
This doesn't look like it would be too hard to implement. Where would you like this string to be specified (in the link or in some user pref)? — Ambush Commander(Talk) 01:35, 2 March 2006 (UTC)

Explanation of AOL & autoblocker

Is there a project page somewhere that clearly explains the autoblocker and/or its impact on AOL users? The amount of email I get from AOL users confused about being blocked has gone up substantially lately and I want some uniform place to send them for answers. android79 22:17, 1 March 2006 (UTC)


Contents list

Any way to forcefully remove an auto-ToC from a userpage, not just for myself, but for other users who browse it? Sherurcij (talk) (Terrorist Wikiproject) 09:43, 1 March 2006 (UTC)

I think if you include __NOTOC__ somewhere on the page then that will do the trick.--Cherry blossom tree 11:26, 1 March 2006 (UTC)

Anyone know why wikilinks keep on changing from being underlined to not being underlined, and back again? Is this some edit war going on in the style formatting department? - MPF 12:17, 28 February 2006 (UTC)

It usually happens because your browser failed to load one of the stylesheets (most probably the one generated from your preferences). To fix, do a forced reload on your browser (Shift-Reload on Firefox). --cesarb 13:58, 28 February 2006 (UTC)
No, that doesn't make any difference, nor is there anywhere in my preferences where the underlining or not can be changed on its own without other much more obvious page layout changes which haven't happened (I've investigated both of these possibilities already!) - MPF 14:43, 28 February 2006 (UTC)
There is an option in my preferences to have links underlined or not. It's listed under 'Misc' and in a drop down box. --Cherry blossom tree 11:25, 1 March 2006 (UTC)
Thanks! Now how did I miss that one??? - MPF 13:41, 1 March 2006 (UTC)

How to Archive a Talk Page?

I can think of two or three ways I could archive a talk page, but I'd like to know what the official method is. For example, doing a Move would preserve all the edit history...and then I'd presumably make a new page out of the redirect that would create. But I could also just make a sub-page, then cut and paste most of the content from the original page, leaving the last few items in place. This would preserve continuity.

What's the officially sanctioned solution? --Kaz 19:08, 3 March 2006 (UTC)

Wikipedia:How to archive a talk page - Liberatore(T) 19:19, 3 March 2006 (UTC)
Both are usually used. Some people archive by cut-and-paste, some archive by moving; it's more of a personal preference. Avoid, however, the third method (archiving to history); it has significant drawbacks (the most important of them being making it much harder to search for older comments). --cesarb 21:12, 3 March 2006 (UTC)


Pressing enter in the subject line

I suddenly understand why forums that have lots of newbies, such as Wikipedia:Reference desk, end up with so many people adding vague headings with no accompanying text!

If you press enter after typing the subject on an "add comment" editing page, it acts as if you clicked "Save page". I imagine most people who do this aren't trying to be obtuse or unhelpful, or thinking that the box is a search engine -- they may have even read the directions. But pressing the enter key, which would almost certainly do the right thing in most e-mail clients, happens to do something completely unhelpful to everyone involved on Wikipedia. And then the newbie probably gives up and goes away and we grumble about the stupid incoherent newbies.

Would it be possible to change this to something reasonable? The best thing would be for the enter key to take you from the subject to the body, but that may not be possible with the limitations of forms. Having the enter key do nothing would still be better. rspeer / ɹəədsɹ 15:48, 3 March 2006 (UTC)

I believe Lupin committed a patch not 72 hours ago which should correct this. Rob Church (talk) 18:50, 3 March 2006 (UTC)

Can you reduce the space between the columns? - It's beyond me.

I need help with this one. There's a colum of white space down the middle (or almost middle) between the two columns. How can I reduce that so the gap is uniform with the gaps at the other edges of the same boxes? If you can do this, I will worship you as a markup GOD (GOD of Djinn).

The Thinker, a statue by Auguste Rodin, is often used to represent philosophy.

Philosophy ('love of wisdom' in Ancient Greek) is a systematic study of general and fundamental questions concerning topics like existence, reason, knowledge, value, mind, and language. It is a rational and critical inquiry that reflects on its own methods and assumptions.

Historically, many of the individual sciences, such as physics and psychology, formed part of philosophy. However, they are considered separate academic disciplines in the modern sense of the term. Influential traditions in the history of philosophy include Western, Arabic–Persian, Indian, and Chinese philosophy. Western philosophy originated in Ancient Greece and covers a wide area of philosophical subfields. A central topic in Arabic–Persian philosophy is the relation between reason and revelation. Indian philosophy combines the spiritual problem of how to reach enlightenment with the exploration of the nature of reality and the ways of arriving at knowledge. Chinese philosophy focuses principally on practical issues in relation to right social conduct, government, and self-cultivation.

Major branches of philosophy are epistemology, ethics, logic, and metaphysics. Epistemology studies what knowledge is and how to acquire it. Ethics investigates moral principles and what constitutes right conduct. Logic is the study of correct reasoning and explores how good arguments can be distinguished from bad ones. Metaphysics examines the most general features of reality, existence, objects, and properties. Other subfields are aesthetics, philosophy of language, philosophy of mind, philosophy of religion, philosophy of science, philosophy of mathematics, philosophy of history, and political philosophy. Within each branch, there are competing schools of philosophy that promote different principles, theories, or methods.

Philosophers use a great variety of methods to arrive at philosophical knowledge. They include conceptual analysis, reliance on common sense and intuitions, use of thought experiments, analysis of ordinary language, description of experience, and critical questioning. Philosophy is related to many other fields, including the sciences, mathematics, business, law, and journalism. It provides an interdisciplinary perspective and studies the scope and fundamental concepts of these fields. It also investigates their methods and ethical implications. (Full article...)


Philosophy ponders the most fundamental questions humankind has been able to ask. These are increasingly numerous and over time they have been arranged into the overlapping branches of the philosophy tree:
  • Aesthetics: What is art? What is beauty? Is there a standard of taste? Is art meaningful? If so, what does it mean? What is good art? Is art for the purpose of an end, or is "art for art's sake?" What connects us to art? How does art affect us? Is some art unethical? Can art corrupt or elevate societies?
  • Epistemology: What are the nature and limits of knowledge? What is more fundamental to human existence, knowing (epistemology) or being (ontology)? How do we come to know what we know? What are the limits and scope of knowledge? How can we know that there are other minds (if we can)? How can we know that there is an external world (if we can)? How can we prove our answers? What is a true statement?
  • Ethics: Is there a difference between ethically right and wrong actions (or values, or institutions)? If so, what is that difference? Which actions are right, and which wrong? Do divine commands make right acts right, or is their rightness based on something else? Are there standards of rightness that are absolute, or are all such standards relative to particular cultures? How should I live? What is happiness?
  • Logic: What makes a good argument? How can I think critically about complicated arguments? What makes for good thinking? When can I say that something just does not make sense? Where is the origin of logic?
  • Metaphysics: What sorts of things exist? What is the nature of those things? Do some things exist independently of our perception? What is the nature of space and time? What is the relationship of the mind to the body? What is it to be a person? What is it to be conscious? Do gods exist?
  • Political philosophy: Are political institutions and their exercise of power justified? What is justice? Is there a 'proper' role and scope of government? Is democracy the best form of governance? Is governance ethically justifiable? Should a state be allowed? Should a state be able to promote the norms and values of a certain moral or religious doctrine? Are states allowed to go to war? Do states have duties against inhabitants of other states?
Glossary · Outline  · A–Z index  · Featured content
Glossary · Outline  · A–Z index  · Featured content
Associated Wikimedia

The following Wikimedia Foundation sister projects provide more on this subject:

I've tried and tried, and just can't figure it out. I need the two columns squeezed together, with the same space between them as is between the boxes elsewhere on the page. The borders are fine, and work as predicted. But that big wide space down the center is driving me NUTZ. (And no, this is not for the philosophy portal. I chose it as a starting point because the chart code used is more stable than a div-based layout. By the way, I need the double outer border to stay. Thanks. --Go for it! 12:14, 3 March 2006 (UTC)

Not knowing exactly where you want to use this, I'm not sure exactly how to fix it, but I'd recommend some general cleanup and simplification regardless. Rather than clutter up this page, though, I'm going to play with a couple of versions on a sub-page. Let's take this question to User:Go for it!/table tests. Rossami (talk) 21:14, 3 March 2006 (UTC)
Is that better? After playing with the paddings and margins for a bit, I dug deeper, and couldn't figure out a reason for the table to have five columns with nested tables within. I simplified it to two columns, with the borders applied directly to table cells, rather than to one-cell tables. I think I've simplified the code as far as I can, removing all the excess paddingses and marginses, and probably over-commented it for someone who seems to understand CSS as well as you do. Still, I hope that's helpful for whoever might be maintaining it down the road. Does this do what you want? — Catherine\talk 22:08, 3 March 2006 (UTC)
Sorry, edit conflict with Rossami! I'll leave a note at Go for it!/table tests. — Catherine\talk 22:09, 3 March 2006 (UTC)

Image problem

Something odd has happened to the image at University of Cambridge - it's white with a vertical grey line in the centre Bluap 17:09, 2 March 2006 (UTC)

I got the box back to looking halfway decent by scaling the image down to 121px width (from 128px). Somehow, the thumbnailer only outputs a grey line if the width of the image is between 122 and 170 pixels (using 171 or more pixels it works again). Somthing strange is going on... I tried purging and shift-refreshing, and only got the line. -- grm_wnr Esc 20:57, 2 March 2006 (UTC)
Use ?action=purge on the image page to force it to regenerate any thumbnails in use for that image. --Brion 04:13, 3 March 2006 (UTC)

Category:Logos category page is unusable

Currently the image for every logo appears in the category page so when you go to the page it tries to load 200 images. On top of that there are so many logos that the first page on goes up to 9-11pdp and each letter has dozens of pages making the act of finding anything impossible. There are people that are trying to fix the problem by putting the logos into subcatagories but that doesn't solve the problem because the logo template puts the logos on the main page. Can anything be done to fix this problem? --JeffW 17:32, 28 February 2006 (UTC)

Sounds like you need to fix the logo template. -- Tim Starling 02:52, 1 March 2006 (UTC)
I suppose I might be able to do that if I was pointed to the right documentation, but I can't edit the template, it's Protected. --JeffW 14:22, 1 March 2006 (UTC)
I think changing
    • [[Category:Logos|<noinclude> </noinclude> {{PAGENAME}}]] to
    • [[Category:{{{1|Logos}}}|<noinclude> </noinclude> {{PAGENAME}}]]
would do the trick. Can someone make this change or give me access so I can make the change? --JeffW 18:14, 1 March 2006 (UTC)
Wouldn't it be better to make other templates for the other kinds of logos? --cesarb 22:07, 1 March 2006 (UTC)
So I should copy and paste the logo code to a new template and make the change there? Is there some sort of naming convention or should I just call it logo2? --JeffW 04:13, 3 March 2006 (UTC)
CesarB, that has been done, but several logos that only have a subcat tag still appear under "logos" as well as the resective subcat.--Esprit15d 19:22, 3 March 2006 (UTC)
Which ones? I just did a spot check of some of the logos on University logos and didn't find them on Logos. --JeffW 22:55, 3 March 2006 (UTC)

Collateral damage template suggestion

Given the ongoing problem of frequent collateral damage to innocent good-faith contributors due to autoblocking of dynamic IPs, I have been tinkering with a possible template that might make it easier to alert administrators of collateral damage problems. The underlying idea is that innocent bystanders who get caught in an autoblock would only have to fill in the relevant IP and their name in order to alert the relevant admin of the problem.

Proposed syntax of template:

{{Template:Collateral damage warning|IP=0.0.0.0|victim=John Smith}}

The template would look like this:

Dear Administrator:

The IP address 0.0.0.0 has been autoblocked under your signature.

Please note that this is a dynamic IP address that is used by many other contributors in good faith, and that the collateral damage your autoblocking is causing for these people is likely to outweigh any protective benefits of your autoblocking action.

This message has been sent to you by John Smith at this IP who would like to be able to contribute useful information to Wikipedia and therefore requests the block on this IP address to be removed.

Thank you for your help. John Smith 06:36, 28 February 2006 (UTC)

with the underlying syntax:

[[Image:Stop_hand.png|left|30px]] '''Dear Administrator''': The IP address '''{{{IP|(Sorry, user forgot to enter IP)}}}''' has been '''autoblocked''' under your signature. Please note that this is a dynamic IP address that is used by many other contributors in good faith, and that the collateral damage your autoblocking is causing for these people is likely to outweigh any protective benefits of your autoblocking action. This message has been sent to you by '''{{{victim|a good-faith user}}}''' at this IP who would like to be able to contribute useful information to Wikipedia and therefore requests the block on this IP address to be removed. '''''Thank you for your help''.''' {{{signed|~~~~}}}

with parameters in the above example set as IP = 0.0.0.0, victim = John Smith.

The idea behind the signed parameter is that it would be set to ~~~~ by default in order to give the administrator in question enough information about the affected user even if that user in their distraction forgets to give their name in the victim field. One of the reasons why I wanted to field this template here before adding it to the Template space is that I am not entirely sure whether the use of the four-tilde signature as the default value would be a syntax violation or not. Sandbox tests have been inconclusive. I am not a programmer, just a lowly Wikipedia contributor driven to boldness beyond my technical confidence out of sheer self-defense.

Comments, suggestions, and "watch it, you're doing this the wrong way" warnings would be greatly appreciated Vremya 07:28, 28 February 2006 (UTC)

I can't see most contributors knowing to use this template. I guess it's worth a try... Superm401 - Talk 00:44, 1 March 2006 (UTC)
The idea would be to add this template along with a brief how-to guide to the page that autoblocked users are redirected to when they try to edit. At the present time this page contains relatively vague instructions on how to contact the relevant administrator, whose main result seems to be that the collaterally damaged user fires off a snarky message along the lines of "Why the heavy expletive are you accusing me of vandalism?" to said admin. I am hoping that this template would serve the dual purpose of (1) not leaving innocent autoblocked users at a loss about what to do and (2) preventing admins from receiving the brunt of these users' frayed tempers.
Having said that, I would still appreciate any technical input, and particularly any advice on the wisdom or folly of incorporating the four-tilde shortcut into a template (and how to do it right if it is folly). As I said, sandbox tests haven't quite worked out Vremya 09:43, 1 March 2006 (UTC)
I can't help you with technical issues, and I don't do much blocking as an admin, but clearer instructions and a way to redirect frayed tempers sounds like a very good thing. I'd say go to it. JesseW, the juggling janitor 00:43, 3 March 2006 (UTC)

New magic words?

This is more or less a software feature request, but I wanted to ask for comments here first before submitting it at bugzilla.

I would find the following new magic words useful:

  • NEXTMONTH: To show the number of the next month
  • NEXTMONTHNAME: To show the name of next month
  • NEXTMONTHYEAR: To show the year that will be valid next month
  • NEXTWEEK: The number of next week
  • NEXTWEEKYEAR: The year belonging to next week
  • NEXTDAY: To show tomorrow's day of the month
  • NEXTDAYNAME: To show the name of tomorrow's weekday
  • NEXTDAYWEEK: To show the week number that belongs to tomorrow
  • NEXTDAYMONTH: To show the month that belongs to tomorrow
  • NEXTDAYMONTHNAME: Tomorrow's month name
  • NEXTDAYYEAR: The year that belongs to tomorrow's date

These magic words could help with testing. For example, on pages changing daily, like the main page, it would be possible to create a test version that will automagically show what tomorrow's main page will look like. Similarly, pages changing monthly could use this to preview next month's contents. This could be very useful to preview future content to see if the formatting of transcluded pages looks good. (I would love to have a test page that always shows tomorrow's version of Portal:Germany so I can make sure it will look alright). I am pretty sure some other pages could also benefit from "tomorrow" (or "yesterday") links that are made via magic words. Does this sound useful to anybody else or are there good reasons not to have such new magic words? Kusma (討論) 04:21, 27 February 2006 (UTC)

You can do it with templates already. Check the source (and hike up the template inclusion chain) for fr:Wikipédia:Le Bistro or the contents of the category fr:Catégorie:Modèle calculant une date. _R_ 18:33, 2 March 2006 (UTC)
I knew that was possible using several layers of templates, but that seems a somewhat hackish way to implement this. Are we allowed to use meta-meta-meta-templates for this kind of things now? Kusma (討論) 21:00, 3 March 2006 (UTC)

Search getting old results

Not sure if this is a known problem or not, couldn't see anything in the bugzilla. Anyway on 10 December 2005 I did a search for zeland (common mispelling of Zealand as in New Zealand ) and fixed most of them. However when I do a search for the world now like this it still shows the pages from which the word has been removed ( like Official Opposition (New Zealand) and Joan Hammond ) . I would guess there is some problem with the search database not updating. IS this a known bug? - SimonLyall 10:27, 23 February 2006 (UTC)

The search database is updated periodically. It probably just wasn't updated yet. --cesarb 13:56, 23 February 2006 (UTC)
It seems a long time, Also if I do a search for Sago Mine it shos up lots of results from early January. - SimonLyall 01:05, 25 February 2006 (UTC)
Looks like the database was recreaed a few days ago. I guess as long as it happens regularly (weekly perhaps?. It isn't a big problem. - SimonLyall 23:02, 3 March 2006 (UTC)

Viewing older entries in watchlist

Is there any way to view entries in my watchlist that are older than one month? I used to be able to view them all but now can't get past a month no matter what option I select. I've tried hacking the URL but that doesn't make any difference. Cjrother 23:42, 3 March 2006 (UTC)

That's currently not possible; the watchlist is drawn from the recentchanges database which contains only the last 30 days or so. --Brion 00:38, 4 March 2006 (UTC)


To create a plug-in for Wikipedia

Hii all ,

We have developed a web mapping application . And would like to integrate it with Wikipedia . Our mapping application uses MapServer,Php/Mapscript , as the scripting language and MySql as the database. Our application accepts user clicks on the image provided by us .And points are plotted .We have developed a Php file to accomplish this. Could anybody tell us how to create a plug-in for Wikipedia .

Thanks a lot for your help Munira

Do you mean Mediawiki, the software that runs Wikipedia, or the Wikipedia website itself? If its Mediawiki, see that article for links to the documentation for it. If it's Wikipedia the website, you'll have to present an extremely argument why we should run your software. -- Finlay McWalter | Talk 21:56, 3 March 2006 (UTC)

Hi , We intend to integrate our application with MediaWiki . Which documentation link are you talking about ? Thank You Munira

See m:MediaWiki extensions -- Finlay McWalter | Talk 12:53, 4 March 2006 (UTC)

For admitted copyvios- how about a new template?

It's evidently plenty worse when someone admits making a copyvio. Therefore, I thought why not have a new template about it?

Here it is: Template:Admittedcopyvio

If you want to see it right here:

{{Copyviocore}}

So, what do you think? Btw, would you care to tune it up further, make it more unique, and deal with its links? Thanks. --Shultz III 09:27, 26 February 2006 (UTC)

I don't think it's necessarily to have a whole new template for this - most everything is the same except for the author's claim that it's a copyvio. It doesn't affect process or anything. You can note this in the reasons on the copyvio listing page, and I think that's good enough. Deco 22:38, 26 February 2006 (UTC)
Unnecessarily complex. The worst part is the link to Wikipedia:Admitted_Copyvios, which I assume you're proposing we create; that would just serve to divide up copyright problems and make it harder to track and deal with them. I could see creating a speedy criteria for when the author states that something is a copyvio; but if we're going to send it through the copyvio-investigation process, we should plainly use the existing system and not create completely redundant second one. --Aquillion 08:05, 4 March 2006 (UTC)

New keyboard shortcut

I proposed a new keyboard shortcut for the admin [Rollback] button here - any thoughts would be appreciated. I'm just posting here to see if it is technically possible to create a shorcut, seeing as for example, the random article is always a set link (Special:Random), but the rollback button isn't - it contains letters and numbers and tokens etc. Thoughts? FireFoxT • 20:27, 5 March 2006


Category lists

See Wikipedia_talk:WikiProject_The_Beatles/Article_Classification#Building_from_the_category_dump Is there an easier way to generate a list of articles in a category that doesn't have any formatting, but is in alpha order? I can see how to do it from the category.sql dump but before I do that, thought I'd ask... rather not reinvent the wheel, being lazy and all that... Larger question, do other wikiprojects do article classifications and are there already tools to build the classification tables? (we built a couple by hand already) The Beatles have a LOT of articles in a LOT of categories. ++Lar: t/c 19:04, 5 March 2006 (UTC)

Strange background on image

The image used at {{Sockpuppet}} have a strange white background. There seems to be a bug in the imagemagick-interface. AzaToth 13:26, 5 March 2006 (UTC)

By "strange" do you mean "not transparent"? When I look at the same image in photoshop, it also has a white background, not a transparent one. -- Finlay McWalter | Talk 13:33, 5 March 2006 (UTC)
If you look at Image:Nuvola apps kdmconfig.png, you see that the image have an transparent background. AzaToth 14:13, 5 March 2006 (UTC)
Looking at its page on commons, it had a transparent background until yesterday, when someone changed it (with edit summary lossless compression). It now has a solid white background. -- Finlay McWalter | Talk 14:22, 5 March 2006 (UTC)
I looked at the image found at commons, and it has a transparent background. AzaToth 14:32, 5 March 2006 (UTC)
Firefox shows the old image as transparent and the new one as white. So does Photoshop. So does Paint Shop Pro. So does Inkscape. So does Opera. -- Finlay McWalter | Talk 15:05, 5 March 2006 (UTC)
I reverted the new uploaded image and all is well. The image had been, I believe, converted from 24-bit alpha to 8-bit palletized. You can't tell that it's not transparent on commons because of the white background used there (changing your skin to something else would probably make it more obvious). It seems the uploader was either running some kind of bot or had mass-downloaded a bunch of images, compressed them, and was reuploading them. It also seems he's aware of the problem, as a look at Special:Log/uploads on Commons shows he's been reverting a bunch of his uploads. (There's also a note on the users French Wikipedia talk page). —Locke Coletc 15:17, 5 March 2006 (UTC)
What I ment was I did download the image, and opened it in gimp, and there it indicated that the background was transparent. AzaToth 15:24, 5 March 2006 (UTC)
That's strange.. as with Finlay, I opened the image in Photoshop and it says it has a white background. It was also not dithered (hence the color banding) at all. The version I reverted to is transparent though (and also more colors, so no banding). How big is the file you opened in Gimp? —Locke Coletc 16:53, 5 March 2006 (UTC)
[0:0][azatoth@azabox img]$ file Nuvola_apps_kdmconfig.png
Nuvola_apps_kdmconfig.png: PNG image data, 128 x 128, 8-bit/color RGBA, non-interlaced
[0:0][azatoth@azabox img]$ ll Nuvola_apps_kdmconfig.png
-rw-r--r-- 1 azatoth azatoth 20195 2005-09-08 18:47 Nuvola_apps_kdmconfig.png
AzaToth 17:05, 5 March 2006 (UTC)
Strange, somehow you got the 24-bit alpha-blended version (20,195 bytes). Look at commons:Image:Nuvola apps kdmconfig.png, the image that was in use that was causing the white background on {{Sockpuppet}} is the image just prior to my reversion. (Here's a direct link to the broken file). The file size on the white-background image is 5110 bytes. —Locke Coletc 17:32, 5 March 2006 (UTC)

Primary Keys

Can someone point me to information about the wiki database structure? It looks like articles are associated with incremental primary keys, and I'm concerned that with high volume additions and edits, the key's index trees quickly grow out of balance and result in increasingly longer lookup times (until the db admin rebuilds the indexes). I can't be sure, though, without seeing an E/R diagram and key types. Rklawton 07:26, 5 March 2006 (UTC)

Well, there is a bit of documentation of the MediaWiki database architecture at MediaWiki architecture on Meta, but I believe that what you're looking for is the database layout tables, and more importantly, the annotated table database. Titoxd(?!? - help us) 07:31, 5 March 2006 (UTC)
Isn't a database's key index supposed to be auto-balancing? --cesarb 20:18, 5 March 2006 (UTC)
Autobalancing with sequential keys isn't free. It requires a lot of I/O and CPU cycles. Random keys with collision avoidance creates balanced trees by virtue of its randomness. Rklawton 23:17, 5 March 2006 (UTC)
It doesn't use a lot of CPU and I/O. It's designed to work well with any insertion order. Random keys, on the other hand, would use a lot more CPU and I/O (because it would have to check for collisions, which are not possible with sequential keys until the keyspace is exausted). --cesarb 23:32, 5 March 2006 (UTC)

My edits are messing up random sections of article

For some reason when I edit or revert on The Holocost article, it leaves behind many nonsense edits I didnt make. this hapened a few minuts ago [someone should please revert it since I cant fix it], as well as on the 26th of Febuary. also something like this hapend on the Chabad.org article on the 14th of Febuary (which also undid my own edits of that article on the 6th). Shlomke 03:36, 5 March 2006 (UTC)

It would appear that some software on your computer is removing any words which might be deemed offensive, for example "Hitler" or "Nazi" (if those words are removed by a subsequent edit, they are two words commonly used in reference to Germany shortly before WWII). Are you running some Net Nanny type program (some versions of Norton Internet Security have such a feature), or are you on a school or library computer which might have some such software installed? The only solutions, I think, is to get that software to turn off censoring of the Wikipedia.org domain, or to use a different computer for editing. Until you get this problem solved, please don't edit more articles, but do continue to edit here so we can sort the problem out.-gadfium 03:54, 5 March 2006 (UTC)

Yes, We run CYBERsitter on our computer. this makes sence for The Holocost article, though I dont see why it messed up the Chabad.org article. is it possible it was a double edit with someone else in there too? in any case, is there some way I can make test edits on The Holocost article to see if the problem is getting fixed? Shlomke 04:16, 5 March 2006 (UTC)

With Chabad.org, maybe you were looking at the history and edited an older version of the article by mistake. It can happen quite easily.
The easiest way to experiment with The Holocaust is to copy the wikitext of the article, and paste it into a subpage of your user space, e.g. User:Shlomke/sandbox. You can then edit that as much as you want, within reason.
I'm not familiar with CYBERsitter, but if you are the owner of the computer (eg the parent), and have the master password for the program, you may be able to configure it to ignore en.wikipedia.org, or at least you should be able to turn it off while editing Wikipedia. If you are a child whose parents have installed CYBERsitter, you may be able to persuade them to reconfigure it, but they also may not want you to have full access to all of Wikipedia, and it isn't our place to interfere with their decisions.-gadfium 04:50, 5 March 2006 (UTC)

Thanks, Shlomke 05:07, 5 March 2006 (UTC)

Template problem

I'm trying to edit a Korean Wikipedia template (ko:Template:llang) that uses three parameters, of which the third is optional. The template is to show what the name of something is in different languages. Parameter 1 specifies the language, and parameter 2 specifies the name in that language. The optional parameter 3 specifies how that name is written in the Korean alphabet. Currently, the operational part of the template looks like:

<small>{{llang/언어 고리|{{{1}}}}}: </small><span lang="{{{1}}}" xml:lang="{{{1}}}">{{{2}}}</span><small>{{{3|}}}</small>

So that if I enter "({{llang|en|English|잉글리시}})", then I get

(영어: English잉글리시)

with no whitespace between 'English' and the small '잉글리시'. But if I try to insert a whitespace in the template, for example as follows:

<small>{{llang/언어 고리|{{{1}}}}}: </small><span lang="{{{1}}}" xml:lang="{{{1}}}">{{{2}}}</span> <small>{{{3|}}}</small>

then I wind up with the problem that if I enter "({{llang|en|English}})" with only two parameters, I get

(영어: English )

with a whitespace at the end that I cannot get rid of (between 'English' and the closing parenthesis). Does anyone know how to fix this so that the whitespace is present when three parameters are entered but excluded when only the first two are entered? It would be really appreciated. --Iceager 01:16, 5 March 2006 (UTC)

Never mind, this problem has been fixed by a Korean Wikipedia user. The solution was to do the following:

<small>{{llang/언어 고리|{{{1}}}}}: </small><span lang="{{{1}}}" xml:lang="{{{1}}}">{{{2}}}</span>{{if|{{{3|}}}| <small>{{{3|}}}</small>|}}

Thanks anyway! --Iceager 01:19, 5 March 2006 (UTC)

Picture problem

Excuse me (and excuse me for talking horribly your language), I'm a french Wikipedia user.

For the french article about Rosin, I'm looking for pictures, and I've seen that Rosincake.jpg (you see, just there :) has got a PD license problem. Is it an error, what am I allowed to do, can I upload it on Commons... ?

I thank you a lot to have read my message, if it's possible, please answer me on my discussion page, it'll be more practical.

Again thanks, 82.125.185.57 21:12, 4 March 2006 (UTC) aka Pierre Vigué

This has been resolved. Makemi 06:19, 5 March 2006 (UTC)

E-mail confirmation not working

I've clicked Special:Confirmemail and received the e-mail, but when I click the link in the e-mail, the Wiki page it goes to says "Invalid confirmation code. The code may have expired." I've tried this a few times yesterday and today. What should I do? --Bruce1ee 08:04, 3 March 2006 (UTC)

Try again? I just confirmed both my Wikipedia and Commons email addresses, and it worked just fine. ~MDD4696 02:28, 4 March 2006 (UTC)
That happened to me; when I checked back I realised that not all the code was included in the hyperlink that appeared in the mail. The last ten letters or so appeared in the message but weren't part of the link, so clicking on the link sent me to a wrong address. Get around this by copying the full address from your message and pasting it into the address bar of your browser, Palmiro | Talk 20:05, 4 March 2006 (UTC)
Can you tell me what mail program you use? And if possible, would you mind forwarding me a copy of the mail to brion at wikimedia.org so I can check it? (Once used the code can't be used again.) --Brion 00:31, 5 March 2006 (UTC)
It worked. Thanks Palmiro for the advise – not all of the code was included in the hyperlink. BTW the mail program I'm using here is IE6 (it's a web-mail account). --Bruce1ee 12:46, 5 March 2006 (UTC)

Image:US Locator Blank.svg

The '''Image:US Locator Blank.svg''' map image used for Wikipedia:WikiProject Protected areas articles is blank. When one clicks on an article such as Arches National Park, the locator map is white and the red locator dot is all that can be seen. Is this a bug? If so, can someone direct me on how I get ahold of the developers.--MONGO 10:48, 28 February 2006 (UTC)

Probably something to do with the recent librsvg upgrade. I resaved and reuploaded the image, and it seems to render fine now. —Ilmari Karonen (talk) 11:28, 28 February 2006 (UTC)
Thank you for the very swift response...it is now working again...appreciate the effort!--MONGO 11:31, 28 February 2006 (UTC)
For a few hours, SVGs exported from Adobe Illustrator were rendering incorrectly (blank) due to a funky incompatibility. That's fixed; until the total regeneration is completed, you should now be able to manually flush those using action=purge on the image page. (You might have to force a reload of the images afterwards to actually see the new version.) Note that if it's hosted on Commons you may have to do the purge at Commons. --Brion 01:55, 1 March 2006 (UTC)
Just curious; was this a problem with Adobe Illustrator? If so, is there someplace this is documented? Thanks. =) —Locke Coletc 02:10, 1 March 2006 (UTC)
No, apparently a bug or limitation in libxml2. Illustrator's export (at least for some versions) uses inline DTD definitions to create named entities for the SVG and XLink namespace URLs, which apparently doesn't work right with libxml2 older than 2.6.22. If they rendered right before, it might have been due to older versions of librsvg not paying much attention to the XML namespaces... or something. :) --Brion 07:25, 1 March 2006 (UTC)

Unfortunately, the new rendering looks different - the 0px border no longer shows up on images like Image:Interstate 55.svg. Image:Interstate 99.svg shows how it used to look. --SPUI (talk - don't use sorted stub templates!) 23:16, 4 March 2006 (UTC)

Can you file this in bugzilla? I'll poke at it in a bit... --Brion 00:34, 5 March 2006 (UTC)
[1] - thanks. --SPUI (talk - don't use sorted stub templates!) 00:49, 5 March 2006 (UTC)

I need help with my disambiguation page

I just made this page:

http://en.wikipedia.org/wiki/AICE_(disambiguation)

because I am making a page that concerns an organization named AICE, yet there's another one named AICE w/ an article here already. How do I make it so that those who search for AICE go directly to the disambig. page?

If both AICE's are about the same in importance then there should be an article AICE that disambiguates to "AICE (context A)" and "AICE (context B)". This can be accomplished with a series of well planned article moves
  • Current "AICE" ->"AICE (context A)" (where context a is descriptive)
  • write your new article, create it named "AICE (context B)" (or move it to there from your sandbox, best to carry all of these moves out in a short period of time, see a couple problems down for what happens if you don't)
  • Update "AICE (disambiguation)" to point to both the above
  • move "AICE (disambiguation) -> "AICE"
then people entering AICE will get the disambig page.. However looking at the current AICE.. assuming it's not actually WP:AFD fodder, it's hard to imagine it's the more important or equally important org. In that case the NEW article, (the more important org) sholud just have a disambig back to the current AICE (after renaming to "AICE (competitive eating)" ... Hope that helps. ++Lar: t/c 22:47, 6 March 2006 (UTC)


Scapegoat

I suppose this must be a technical problem. People have left messages in my box, blaming me for "vandalism" on pages I KNOW I haven't touched. (I may be an ignornant newbie, but knowing that, I try to stay away from political discussions!) No one else uses my computer, and no one has broken into my house recently.

So either it's a technical issue or you people are all evil.

-The Incredibly Overbitten Newbie who only just discovered tonight that messages were capable of being left. Or would have anything (non-SPAM)interesting to say.

I'd be logged in if I had any idea when the system might get around to emailing me that pesky 'new password.' It's already been several minutes. — Preceding unsigned comment added by 207.200.116.203 (talkcontribs) 12:30, 6 March 2006

Last I checked, we're not ALL evil so it's probably a technical issue or a miscommunication. If they're leaving messages in your IP's talk page (that's what we call it here, not mailbox) it could well be because your IP address is shared with other users. Hope that helps... ++Lar: t/c 14:30, 6 March 2006 (UTC)
Your IP address points to AOL. This is a known problem with AOL: a user's IP address can change with every edit, and it's quite common to not only receive vandalism warnings meant for other people, but also to be accidentally blocked by a block meant for someone else. Some people say using another browser (for instance Mozilla Firefox) instead of AOL's own browser avoids part of the problem; however, this has not been confirmed. The correct technical fix depends on a change on AOL's server software. --cesarb 14:29, 6 March 2006 (UTC)

Suppressing categories for displayed example tempates

How can I prevent User talk:Danieljackson from showing up in cleanup categories such as Category:Disambiguation pages in need of cleanup? Melchoir 21:34, 6 March 2006 (UTC)

Can you "subst" all the templates, and then make sure they have the preceding colon before the category? Makemi 22:19, 6 March 2006 (UTC)
I could, but I was kind of hoping for a magical, non-invasive solution. Something like noinclude...? Melchoir 22:54, 6 March 2006 (UTC)

Well, it looks like BL Lacertae is less lazy than I. Thanks! Melchoir 01:15, 7 March 2006 (UTC)

A request for this problem has been posted to bugzilla. (see bugzilla:4407). Perhaps my bad English made the request to be forgotten. :( borgx (talk) 07:20, 7 March 2006 (UTC)


Something odd happened to Hungarian University of Applied Arts

There was an article about the above university. The name of the university changed on May 1st to Moholy-Nagy University of Art and Design. A user came along and copied the content of the page under the new name plus making a few edits, instead of moving the original page before editing (which was very bad, as he should have used the "move" button). I wanted to revert it by moving the old page to a new name, where I could insert the new content he created, but when I moved the page, it just lost all its history! [2] I wanted to reinforce the proper move to a new name, so as to save the history of the original page, and I lost it instead. Why? Could you please HELP me?

I mean:

  1. The history of the original page (Hungarian University of Applied Arts) should be undeleted and restored.
  2. The present, updated content of the page Moholy-Nagy University of Art and Design should be copied to the previous page.
  3. The page Hungarian University of Applied Arts (with the updated content) should be properly moved to the name "Moholy-Nagy University of Art and Design".

Thus, the result should be a page with the name "Moholy-Nagy University of Art and Design", which has the present (updated) content of this page and the full history of the page Hungarian University of Applied Arts, reflecting the implemented changes. It could perhaps involve a redirect from the latter page and all the other pages involved. I think an administrator is required for this task.

Thanks a lot in advance!

Adam78 21:13, 6 March 2006 (UTC)

It sounds like perhaps you will want to ask for help on Wikipedia:Requested_moves. Your writeup above seems good. Requested moves uses a very terse format, so the writeup probably should be placed on the talk page of the article that has the correct history and referenced from the move request. I'm not an admin so I may be wrong but that's what it sounds like to me. Good luck and happy editing! ++Lar: t/c 22:41, 6 March 2006 (UTC)
Thank you! I did what you suggested. The only problem is that presently there's no page which has the correct history. That's the history which has been lost, to my most surprise. Anyway, I hope that the people at the "Requested moves" will be able to help. Adam78 00:26, 7 March 2006 (UTC)
My suggestion, check every redirect you can find for the missing history. Is this it: http://en.wikipedia.org/w/index.php?title=Moholy-Nagy_University_of_Art_and_Design_%28Budapest%29&action=history ? Whenever you (or any user using the normal move page rather than admin tools) move a page, ***any page***, the redirect left behind should have the history... WP doesn't let you move a page without leaving a redirect behind. Unless you had an admin do a previous move, the history is out there on SOME page if you can just find it. Good luck... I'd advocate making sure no admin does any moves for you till you find it. (however take all that with a grain of salt, I may be confused!) Hope that helps, happy editing! ++Lar: t/c 01:36, 7 March 2006 (UTC)
Clarification: the place you move to gets the history, not the place you move from. The redirect left behind will have a single edit stating that it has been moved. -Splashtalk 01:41, 7 March 2006 (UTC)
Brain check, thanks Splash! But ya, I was nosing around and there's a tangle there all right. I stand by what I said, unless an admin was involved, some article has the history. Probably some history merging may be needed. Find all the redirects and leftbehinds and give the links to them all somewhere... ++Lar: t/c 02:00, 7 March 2006 (UTC)

Thank you and sorry for my inadvertence. Adam78 02:35, 7 March 2006 (UTC)

What's going on today?

Because I had a similar problem with KSTW ("this article doesn't exist, honest!"), this is related to the one above. Using IE 6, I'm getting a lot of "HTTP 500 Internal Server Error"s today, stating "There is a problem with the page you are trying to reach and it cannot be displayed." I get this mostly when trying to edit a page but I also had it happen on any attempt to go to any Wikipedia page for a brief time. What's going on? Morgan Wick 00:02, 6 March 2006 (UTC)

Technical problems. --cesarb 00:17, 6 March 2006 (UTC)

When I try to open the article Roland Fraïssé, it appears nonexistent, yet the history shows that it still exists. I don't know if this is related to the above problem or not. Spacepotato 02:08, 7 March 2006 (UTC)

?action=purge --Brion 08:11, 7 March 2006 (UTC)

Redirect type description with right-to left title

On this page: [3], the redirect description doesn't show up. However, it does show up on the corresponding diff: [4]. Is this a bug or a feature? Kusma (討論) 20:59, 4 March 2006 (UTC)

What's a "redirect description"? --Brion 00:29, 5 March 2006 (UTC)
It is a feature of certain templates, in this case, {{R from alternate language}}. Titoxd(?!? - help us) 00:32, 5 March 2006 (UTC)
Okay, so this has nothing to do with right-to left. The same thing happens with other templates on redirects too, see [5] and [6]. The description is not included in the redirect page, but is shown on the diff. Why is that a feature and not a bug? Kusma (討論) 17:37, 5 March 2006 (UTC)

The description is in "noinclude" so that it only shows up on edits.

--William Allen Simpson 10:36, 6 March 2006 (UTC)
Er, no. The noiclude in Template:R from misspelling only concerns a category, not the description. Also, I don't mean the "diff box" on the diff display. What is different is how the full article (in this case the redirect) is displayed below the diff box. Check the links and you should see what I mean. Kusma (討論) 12:29, 6 March 2006 (UTC)
Anyway, I found the bug report: this is bugzilla:927. Kusma (討論) 03:19, 7 March 2006 (UTC)

Cookies again (can't stay logged in)

Something similar came up a few weeks ago. I was abruptly logged out (no message or anything, just no longer logged in when I went to refresh my watchlist). When I tried to log in again, I get the message "Login error: Wikipedia uses cookies to log in users. You have cookies disabled. Please enable them and try again." However, I have cookies enabled (nothing changed there in a very long time) and I always log in with the "Remember me" box checked. I'm using Firefox. Any ideas? olderwiser 17:24, 4 March 2006 (UTC)

From what I've seen here earlier, my suggestion is to delete cookies and try to login again. Hope this helps, Shardsofmetal [ Talk | Contribs ] 19:25, 4 March 2006 (UTC)
Well, it is better now. I had cleared my cache, deleted cookies, rebooted -- all the usual Windows quick remedies. But for several hours on March 4 I could not log in. olderwiser 02:17, 7 March 2006 (UTC)

Suggestion: Name variations

It would be nice if the Wikipedia search could automatically try variations of names, such as Jimmy->James, Bob->Robert. There should be redirects for this, but there aren't always - a recent example I ran across was Cliff Stein for Clifford Stein. If there's support I could enter this at Mediawiki Bugzilla. Deco 02:05, 20 February 2006 (UTC)

It would be nice if Wikipedia's search could do a lot of things, but (as far as I can tell), it never will so I use Google site search. Superm401 - Talk 04:04, 20 February 2006 (UTC)
And this Google search box is WP specific with all of the nice stuff like spell check suggestions. hydnjo talk 18:38, 20 February 2006 (UTC)
That is a neat interface but you can also just type "site:en.wikipedia.org " before your query in any Google search box. Superm401 - Talk 06:11, 26 February 2006 (UTC)
It truly astounds me that there are people who think that typing such a string before a search query is an example of a good user interface. --Myk 03:11, 5 March 2006 (UTC)
I kinda like it actually, you do something through the advanced search page once or twice and then you notice the format of the search strings it generates. Why are so many people afraid of command based interfaces? Plugwash 22:13, 7 March 2006 (UTC)

Surely the lag can't be this long...

The article Valença turns up in Category:Portugal geography stubs, which it should, it's a stub created four day ago, according to the page history. Yet simply clicking on Valença tells me that the page doesn't exist. Four days seems a very long lag for the database to catch up... what gives? Grutness...wha? 11:11, 9 March 2006 (UTC)

Perhaps it's the bug Brion Vibber refers to above, in the section #Question about deleted pages. I have purged the cache for the article, and it's OK now. Eugene van der Pijll 21:43, 9 March 2006 (UTC)


Please look at the page cyclin-dependent kinase. A strange link was added named Media:example.ogg by an anonymous IP adress. It points to wikimedia but is unreadable. I don't know what to do with it. --User:AAM | Talk 08:40, 9 March 2006 (UTC)

Reverted the insertion. The user probably just clicked on a random button in the edit toolbar. Next time a link to such a file can be safely removed as harmless experimenting by newbies. — Kimchi.sg | Talk 09:49, 9 March 2006 (UTC)

Deleting own name w/o edits... possible?

The subject sums it up. I created a name that I now wish to change my name to. If I could delete my own account somehow I could then use the request to change name page. The problem with deleting an account is the GNU Free Documentation Licence correct? If there were no edits, could someone somehow delete their own account by submitting their only edit as a request to do so? Charangito 08:23, 9 March 2006 (UTC)

Just leave the old ones be, what's the big deal? --Brion 21:35, 9 March 2006 (UTC)

Wikilinking suffix text

When creating links, the system (as mentioned in the WP:Editing) is smart enough to extend the link to suffix text. For example [[Foo]]s extends the link to cover all of Foos: Foos. There appears to be one category of suffix text that this does not work for; namely, possessives—or any suffix starting with an apostrophe. As an example, [[Foo]]'s should extend to cover all of Foo's, but only links the text between the brackets: Foo's.

Semantically, I have no problem with this behavior, since the link should point only to the article "Foo". Graphically, however, this looks stupid in English. Foo's is one word, and should be linked as such. ... At least, in my opinion ;) Any thoughts on either why this isn't the default behavior or why it shouldn't be would be much appreciated. Seqsea (talk) 03:37, 9 March 2006 (UTC)

My guess (but this is uninformed armchair speculation) is that the definition of "suffix" is very simplemindedly "adjacent alphabetics", and that it deliberately does not cross punctuation, because in general it shouldn't, although you're right that there are cases in which you might want it to cross an apostrophe. I'm sure you knew this, but of course you can always get complete control over the link's appearance (at the cost of doing a bit more typing) by pipeifying the link: [[Foo|Foo's]] looks like Foo's and links to Foo. —Steve Summit (talk) 03:45, 9 March 2006 (UTC)

Suggestion: Better use for yellow "labels"!

If you place your curser on any internal Wikipedia link, a little yellow label appears, bearing the name of the page. For instance, the label for Eschatology reads "Eschatology". Try it yourself!

This is not very useful at all... You need to click and jump to a different article and wait in order to see what Eschatology is about...

However, introducing a slight change could really result in a miracle: if every editing page had a little field where you could write a short description, so that the label would be "Eschatology - The theological study of the end of the world", we would be saving thousands of unnecessary clicks a day!

The whole point of using new technology like computers is making things more efficient. Lets take another small step in that direction! - Jonathan Shafer, 14:38, 8 March 2006 (UTC)

Interesting idea! However some (a lot?) of the more seasoned editors use the popups tool, so may not be thinking much about this. With the popups tool installed in your monobook.js, when you hover over a link you get a big box of useful things to come up (see the screen shot at the link I gave) including the first few sentences, info on when and who edited it and lots of other great stuff. With that "crutch", maybe the impetus to address the underlying issue isn't as strong? But random users coming here won't be running it. Food for thought. ++Lar: t/c 15:04, 8 March 2006 (UTC)

Actually, the little yellow box shows you what page underlies the link. If I create this link --> what lies beneath, it's going to point to the page it's linked to, not to the name of the link. User:Zoe|(talk) 17:21, 8 March 2006 (UTC)

To make this (quite sensible) suggestion practical, we're difficient in an important piece of technology. Ideally the software would be able to extract a one or two line summary from the article; this summary could then be used for a variety of applications (your tooltip idea, generating summary lists, and via some RPC/SOAP type service for other applications, websites, etc.). A properly written introductory paragraph ("Joe Smith 1909-1924 was an important British film director, who made many movies with trains in them.") would serve this purpose well, but it's not practical for the software to extract that automatically from the wikitext (it's too hard to detect images, infoboxes, dablines, dispute boxes etc.). So we'd either need a seperate store in the database (and a separate little edit window at the top of the article) or some special markup denoting the start and end of the summary. With that done we'd also need a concerted effort to actually make sure intro paras really were summaries (and not the all too common "Joe Smith 1909-1924 was born on a pigfarm in County Antrim, where he developed a lifelong fear of frogs..."). -- Finlay McWalter | Talk 18:00, 8 March 2006 (UTC)

This suggestion is a good one, and it is already possible to do it, although the implementation is not very elegant. I've created an example for Eschatology. Try it. This is done by making a dummy redirect.

I also created a template a while back that can be used in a similar way to explain terms that might be unfamiliar. Here is an example for Eschatology  • . If this template were widely used, people would come to recognize that the blue dot will quickly explain an unfamiliar term. -- Samuel Wantman 01:45, 9 March 2006 (UTC)

I can't remember which one it is, but one of the 3rd party Wikipedia mirrors already offers these popup abstracts... Looks like it is thefreedictionary.com see for example their version of Eschatology -- Solipsist 21:12, 9 March 2006 (UTC)

I was going though the links in Special:Whatlinkshere/V10 to disambiguate them to the correct article (nearly all of them go to V10 engine). The only thing is, that any articles that use {{Piston engine configurations}} show up on the list even though that template was changed on Feb. 18 [7] to point to V10 engine. Is there some kind of cache that is keeping this from being correct? --rogerd 03:04, 8 March 2006 (UTC)

Those pages need a null edit (press edit then save) to refresh the template. If there are a lot of them, you can make a request at wikipedia:Bot requests.-gadfium 06:27, 8 March 2006 (UTC)
I made small edit [8] to trig the purge job and seems now the Special:Whatlinkshere/V10 is uptodate. Previous edit to {{Piston engine configurations}} was before the bugzilla:4857 fix I think. borgx (talk) 10:03, 8 March 2006 (UTC)
Thanks to both of you. Yes, it seems correct now. --rogerd 13:18, 8 March 2006 (UTC)

I read somewhere that...

I read somewhere that users could be given temporary SYSOP abilities in order to help clean up vandalism on a foreign Wikipedia. Apparently, this helps when, while a major act vandalism is in progress, most or all of its own admins are sleeping in their part of the world. However, how does this happen? Doesn't a native admin have to be present in order to give a foreign user admin abilities? Or how does this work? I never knew this could be done! --Shultz III 00:47, 8 March 2006 (UTC)

It generally happens on one of the very small wikipedias, where the adminship mechanisms don't exist yet (some of the smallest wikipedias have only a handful of regular users). In those cases someone will approach a Steward (although there may now be a page on meta for this kind of thing) and the whole thing is done in an informal, rather hand-wavey way. -- Finlay McWalter | Talk 00:55, 8 March 2006 (UTC)
Nice, nice. Is there a page where I can read more about it? Please show me the wikilink if you know. Thanks. --Shultz III 01:07, 8 March 2006 (UTC)
m:Meta:Requests_for_adminship is the meta page. xaosflux Talk/CVU 04:12, 8 March 2006 (UTC)
That page is for requesting admin access on Meta. For requesting adminship on other wikis that don't yet have a local bureaucrat, see m:Requests for permissions. Angela. 12:48, 8 March 2006 (UTC)

Where is?

Where is the raw CSS file for Wikipedia? I've been looking around MediaWiki:Common.css and MediaWiki:Monobook.css, but I could not find an entry for "usermessage" class and others. Thanks. – WB 00:09, 8 March 2006 (UTC)

Try looking around at Wikipedia:Catalogue of CSS classes. ~MDD4696 04:13, 8 March 2006 (UTC)

watchlist suggestion

I have a suggestion re Special:Watchlist. It'd be nice if it had something like "as of 07:25 December 5" (that is, "as of {{CURRENTTIME}} {{CURRENTMONTHNAME}} {{CURRENTDAY}}") at the top. —Steve Summit (talk) 15:12, 5 March 2006 (UTC)

Wow! That was quick. Thanks. (It would be even better in the "Below are the last %d changes in the last %d days" sentence instead of "You have %d pages on your watchlist", unless that's a harder part of the template to tweak.) —Steve Summit (talk) 22:42, 5 March 2006 (UTC)
The page that controls this text is MediaWiki:Watchdetails. The addition was reverted. I'm going to re-add it(in the different place you suggested), as I also find it quite useful. JesseW, the juggling janitor 10:03, 7 March 2006 (UTC)
Actually, the page with the bit you mentioned is Mediawiki:rcnote. I've now edited that page, with a pointer to this discussion. JesseW, the juggling janitor 10:08, 7 March 2006 (UTC)
And MediaWiki:wlnote, too. JesseW, the juggling janitor 10:13, 7 March 2006 (UTC)
Thanks! HappyCamper had made that change a couple days ago, as you saw, but reverted it because "there was a bit of resistance towards the implementation". Dunno what was up with that. See his comment on my talk page, under "Ah hah!". —Steve Summit (talk) 17:55, 8 March 2006 (UTC)

Styling

Can somebody dwelve and make sure the new NavFrame/NavHead format for footers has styles that gives it the same amount of whitespace as the current footers styled with class="toccolours"? While the new footers have several advantages (at least for the end users,if notfor wikipedians unused to this new syntax), they should stil look the same as the old footers. Circeus 20:19, 10 March 2006 (UTC)


text boxes

I somehow was able to mess up the textbox in the poppy artice. The picture somhow got remove. I need some help on fixing it. An explanation of how the syntax of these text bowes would also be nice.--BorisFromStockdale 06:48, 10 March 2006 (UTC)

I was able to fix it. Now my question is: how was it changed in the first place? The picture was fine, and then it disapperared. No change was recorded in the artice edit history? Any suggestions? --BorisFromStockdale 06:51, 10 March 2006 (UTC)
Never mind, I am on crack... I was looking at the wrong article... Sh*t; It's late, I better go to bed...

javascripts dont work

I added a couple of scripts to my monobook.js page that are supposed to put up extra functions in the tools section when in edit mode, but nothing happens. Everything seems to be in order, The scripts are from the Wikipedia scripts list, The cache is clear, I have javascript enabled in Firefox and ZoneAlarm. Any suggestions? Thanks! Herostratus 22:20, 9 March 2006 (UTC)

You have HTML tags in it, for one thing. Check Firefox's javascript console for errors if removing those doesn't fix it. — Omegatron 22:33, 9 March 2006 (UTC)
I do? Ooops, thx for checking, OK also did not know about javascript console, will find and use, thanks!!!Herostratus 13:38, 10 March 2006 (UTC)

If you set up a (MediaWiki 1.5) wiki, can it be set to get pictures from commons, the way Wikipedia does now? If so how? Gerard Foley 15:32, 9 March 2006 (UTC)

The functionality is there, the access is not. Your MediaWiki installation doesn't have access to Commons' MediaWiki installation's database, which is how the shared upload feature works. There's some discussion in place regarding an InstantCommons feature which would make this sort of thing available, but nothing concrete. Rob Church (talk) 07:41, 10 March 2006 (UTC)
I see. It would seem to be a logical thing to do, especially when (nearly) everything is under a free license. Gerard Foley 12:14, 10 March 2006 (UTC)

Question about deleted pages

I'm curious about a couple of deleted pages, International Male and Talk:Apple. Both articles display a "Wikipedia does not have an article with this exact name" notice, but the article history for both pages shows legitimate content and no indication of why the pages were deleted. Can someone explain what's going on with these articles? Why were they removed? --Muchness 15:25, 8 March 2006 (UTC)

They both look okay to me. User:Zoe|(talk) 17:22, 8 March 2006 (UTC)
They weren't showing up for me either, so I gave them both a null edit and now they seem fine. Perhaps a caching issue? --Bob Mellish 18:40, 8 March 2006 (UTC)
Odd. I tested on two different computers running different OSes and had the same problem on both. I added a couple of dummy edits to restore the pages. --Muchness 23:11, 8 March 2006 (UTC)
Not deleted, there was a cache bug a few days ago. Hit the purge link if you see it. --Brion 21:36, 9 March 2006 (UTC)
Thanks for the explanation. --Muchness 10:01, 10 March 2006 (UTC)

Page Moves and my watchlist

Why don't page moves appear as a recent change on "my watchlist"? -- Samuel Wantman 09:49, 7 March 2006 (UTC)

They do, unless there's a bug, which is possible. Be specific if possible. --Brion 21:38, 9 March 2006 (UTC)

this edit does not appear on my watchlist and it is the most recent change to the article. I looked at the changes for the last 7 days and did a text search using the old name and the new one. Both the old name for the article and the new one are being watched. Sorry I didn't ask a more specific question. -- Samuel Wantman 01:08, 10 March 2006 (UTC)

Image disappearing?

Hi all. Recently I uploaded Image:Actimagine logo.gif, and it displayed fine for a while, but all of a sudden it disappeared. As in, the image page was still there, but there was no image. I re-uploaded it and the image showed up again just fine. Unfortunately, now it's disappeared again. Is this just a server propagation problem, or might something be wrong with my image? ~MDD4696 04:27, 6 March 2006 (UTC)

Interesting... I cleared my cache and it showed up again. I'm still curious why that happened though? ~MDD4696 04:31, 6 March 2006 (UTC)

Try naming it something like actimagine0310226logo.gif That way the image has a unique date in it and the likelyhood of somebody overwriting it is slim. Miskatonic 00:55, 10 March 2006 (UTC)

The RfAr page seems to have disappeared. When I try to go to it from my watchlist, I am told that it does not exist, and that I can create it or search for it or request its creation. I don't want to create it, because if there is really a glitch, that might have odd effects. Robert McClenon 23:13, 5 March 2006 (UTC)

Seems to be there now. Rob Church (talk) 07:46, 10 March 2006 (UTC)

Where is this?

Class="MainPageBG"

I've looked in MediaWiki:Common.css, and it is not there.

Does anyone know where I can find the definition for this class?
--Go for it! 17:34, 2 March 2006 (UTC)

I'm also not finding it anywhere on Monobook. If you want to take a look, Wikipedia:Catalogue of CSS classes has a list of all the stylesheets used by all the skins. --cesarb 20:34, 2 March 2006 (UTC)

I haven't been able to find it with a quick grep through the skin directories in CVS. Where did you see this class referenced? Rob Church (talk) 19:48, 3 March 2006 (UTC)

The class is used on the Main Page, and the Main Page redesign draft. I'm beginning to think they are code remnants referring to a class that no longer exists. --Go for it! 05:04, 9 March 2006 (UTC)

That sounds like it's the case. Rob Church (talk) 07:51, 10 March 2006 (UTC)

Multilingualism

Isn't it possible to contribute in more than one language? I opened the main page in Zulu and it I suddenly wasn't logged in anymore, and then it said I didn't exist when I tried to log in! Is the English-language wikipedia completely seperate from the others? Do I need to create a new account for other language versions or am I just doing something wrong? Any help would be very appreciated! Thanks :) Joziboy 2 March 2006, 10:43

Unfortunately, yes, you do need to make a new account. Martin 10:45, 2 March 2006 (UTC)
Hopefully, this will be changing soon. It has been talked about for some time already. Rmhermen 14:41, 2 March 2006 (UTC)
One has to wonder how that will be implemented, since surely there are duplicate user IDs in other languages. If someone in the Greek Wikipedia, for example, had created a User:Zoe account, how would this proposed change make sure that they don't take over my English language account, or that I don't take over their Greek language account? User:Zoe|(talk) 17:41, 3 March 2006 (UTC)
Congratulations, you've hit the exact problem with single logon (well, the big non-technical one) ;-). The two suggestions floated, IIRC, were either to force everyone off the duplicate account - so all the different Zoes have to get new usernames - or to find some algorithm to determine which user gets it, whether that be by volume of contributions or date of registration or number of accounts held under that name or... something. As you can guess, neither of these are going to please everyone. Shimgray | talk | 21:23, 3 March 2006 (UTC)

Dunno where this is discussed, but it seems to me that the easiest display solution would be "User:en:Zoe" and "User:gr:Zoe" with a default as the local language. This might be the direction to go for removing all those once created, never contributed, user names (a master user number kept, but the name could be reused by assignment to a new User name).

--William Allen Simpson 10:20, 5 March 2006 (UTC)
Please do not start a whole new discussion about single sign-on here; see the referenced Meta page. Rob Church (talk) 07:52, 10 March 2006 (UTC)

Non-breaking space markup?

I like proper typography, but I also like simple wiki markup. It's nice that we can replace things like &mdash; with — in the markup now, but it doesn't work with everything. So I've been trying to think of wiki markup ideas for &nbsp;. What do you think about using three spaces in a row to signify a non-breaking space? I can think of some downsides myself, but I figured I'd throw it out there. — Omegatron 06:52, 11 March 2006 (UTC)


Saving a reverted image

I've spent 5 minutes searching for info & haven't found it by searching for "reverting an image" or "image revert" or by looking at image policy page or revert help pages... sooo...

There seems to be a rash of people uploading new photos over existing photo names. I can revert just fine to the old photo, and do so, and leave the user a note to reupload using a unique name. But is there a way to save both the old and new photo? I mean other than downloading a copy of the new one, then reverting, then uploading and labeling the new one myself (even if they have provided source/copyright info, which they don't always seem to)? Elf | Talk 02:43, 11 March 2006 (UTC)

As far as I know, no. How did you find out about these images? I'd be willing to help out. ~MDD4696 02:49, 11 March 2006 (UTC)

OK, I confess to a slight exaggeration: I've noticed at least 3, maybe 4 or 5, in the last week or two, and I don't know that I remember ever noticing one in the previous 2 yrs here. So a "rash" might be hype, but it does make me worry and wonder what else is going on out there. I have discovered them because: (a) one was a photo of mine that got overwritten and when I was looking at the index page of my uploaded photos, there was one I didn't recognize; (b) an image caption in an article on my watchlist was changed to say something completely different, and because I was familiar with the article, when I looked at the page, I knew that I didn't recognize the photo; (c) I was browsing an article in my watchlist for some other reason, and noticed that the photo didn't match the existing caption, and because I was familiar with the article, I realized that the photo had changed. In all cases, looking at the image's description page with its upload history at the bottom (as here), you can see that a different user has uploaded an image using the same name.

I'd also really like to know if there's some way to catch these--apparently an image being on one's watchlist does NOT reveal when it has been overwritten (as in case (a) above). Elf | Talk 03:49, 11 March 2006 (UTC)

Well, it's an interesting point, nonetheless. Maybe this deserves a mention in bugzilla? ~MDD4696 04:51, 11 March 2006 (UTC)

Turning off a specific edit button

Is there a way to turn off the [edit] button for a particular heading, while leaving all the other edit buttons on the page unaffected? --Go for it! 14:45, 3 March 2006 (UTC)

One method is to use <h2>This is a heading</h2> instead of ==This is a heading==. This used to cause problems when you tried to edit a later section on the page, as the correspondence between sections and headers was broken, but I think this was fixed quite some time ago. I haven't tried it, so use with caution.-gadfium 20:22, 3 March 2006 (UTC)
Take a look at my talk page, I think it wasn't fixed. —Nightstallion (?) 20:16, 10 March 2006 (UTC)
Obviously I'm wrong. It also doesn't suppress the edit button, so I'm doubly wrong.-gadfium 00:30, 11 March 2006 (UTC)

IE Error

In the last week or so, I've had some problems loading Wikipedia pages, mainly diffs and special pages. When I load them the page displays, then a system window pops up saying "Internet Explorer cannot open the Internet site <page address>. Operation aborted" and an OK button. I click the OK button and the page is replaced by a "cannot be displayed" message. Strangely, when I click the back button the page displays perfectly. This happens infrequently and to seemingly random pages, but never outside Wikipedia. If I go back and forth from a page where it has happened the error tends to occur again. It last happened on this user contributions page just a minute ago. Raven4x4x 10:37, 27 February 2006 (UTC)

Me too. --Quadalpha 03:07, 3 March 2006 (UTC)
Me three, for several weeks now - is this down to IE or is it a Wikipedia thing? Stephenb (Talk) 20:59, 7 March 2006 (UTC)
Me four! does anyone know why??!--Deglr6328 19:14, 11 March 2006 (UTC)
I used to have this problem, so I switched to Mozilla Firefox. --The1exile - Talk - Contribs - 19:50, 11 March 2006 (UTC)
I use both but when I can't use it the problem is really annoying.--Deglr6328 21:54, 11 March 2006 (UTC)

list alignment

I'd like to have two lists in a box. One list of text items should be aligned on the left, and the other text list gets aligned on the right (all the way to the right-hand side of the box).

How is this done?
--Go for it! 13:56, 12 March 2006 (UTC)

Sort of like one of the following? In the second form, carrying the numbering through to the right hand column is awkward. In general, I'd recommend focusing on content rather than presentation, so unless there's a really good reason it must look like this I'd suggest not doing either of these. -- Rick Block (talk) 15:28, 12 March 2006 (UTC)
item 1 left item 1 right
item 2 left item 2 right
item 3 left item 3 right
item 4 item 4


  1. item 1 left
  2. item 2 left
  3. item 3 left
  4. item 4

item 1 right
item 2 right
item 3 right
item 4


Mail server blacklisted by SpamCop

IF YOU ARE HERE BECAUSE OF THE MESSAGE ON YOUR WATCHLIST, OR FROM RECENT CHANGES, PLEASE READ BELOW FOR DETAILS, THE MESSAGE IS DISPLAYING FOR ALL USERS TO CONVEY THIS IMPORTANT MESSAGE

  • You only need to confirm your email if you:
  1. Have a wikipedia enabled email account and
  2. You have not yet confirmed your email.
  • To enable wikipedia email, or check your confirmation status go to Special:Preferences
  • The message on the Watchlist and Recent Changes will be removed in about 2 weeks.
  • For more information, see Help:Email confirmation
Thank you, xaosflux Talk/CVU 03:13, 4 March 2006 (UTC)

For some lovely reason our new mail server has been blacklisted by SpamCop, allegedly for sending mail to spamtrap addresses. (They provide no details by policy, of course, so there's no way to verify it.)

Since there's a tiny possibility that the user-to-user email feature actually could be abused, I've gone ahead and enabled the e-mail confirmation requirement for using email features. This is a bit annoying for the moment since you have to do it separately on each wiki.

I've disputed the listing, so hopefully we'll get it removed soonish and those who aren't getting email will, uh, start getting it again. --Brion 22:42, 1 March 2006 (UTC)

What this means for most users: You must go to Special:Confirmemail and tell it to email you a confirmation code. If you don't, you will not receive Wikipedia email. You have to do it with all your accounts, if you have more than one. --cesarb 04:11, 2 March 2006 (UTC) (link changed. Superm401 - Talk 05:46, 2 March 2006 (UTC))
Hmmm. All of a sudden, a mysterious message appears at the top of my Watchlist page:
Wikipedia e-mail confirmation has been enabled. To receive Wikipedia e-mail, you must go to Special:Confirmemail, request a code, and follow the link in the e-mail.
No explanation is given, nor is a link to a WP: policy page or a discussion page provided. No hint of this appears on the Community Portal. "Special:Confirmemail" also fails to explain what has changed, and why this process has been added. How am I supposed to know this isn't a very sophisticated prank? If we need this new process, we really ought to post a notice about it and include an explanatory link in the message that gets displayed everywhere. ~ Jeff Q (talk) 09:34, 2 March 2006 (UTC)
  1. If such a message showed up, it's because someone added it to the watchlist display in the hopes people would see it (eg, they "posted a notice about it and included an explanatory link")
  2. You don't need to confirm your address; only if you wish to receive email from other users through the wiki.
  3. If you do, it's through the preferences page. There's a nice big box showing your status, next to the box for opting in to allow receiving mail from other users.
  4. There is no WP: policy page or discussion page, as this is a technical decision I made based on the needs of the site.
  5. I've announced the reason for it here and on wikitech-l; as with all such announcements the information should be copied from these central places to whereever you guys would best see it on each wiki.
--Brion 10:16, 2 March 2006 (UTC)
It won't just be recieving email, you won't be able to email other users, which is the more relevant part for protecting from improper use. At least we'll know that everyone who can email users has entered a valid return address themselves. — Laura Scudder 15:00, 2 March 2006 (UTC)

Okay, so I confirmed my email... how do I get that annoying message off my watchlist? PaulC/T+ 16:17, 2 March 2006 (UTC)

Use #confirmemail { display: none }. --cesarb 16:54, 2 March 2006 (UTC)
Um, where and how do we use this? I think "Put #confirmemail { display: none } in page X" would be more helpful. --Yar Kramer 18:28, 2 March 2006 (UTC)
In your personal style sheet, which for you would be at User:Yar Kramer/monobook.css (assuming you're using the default MonoBook skin). —Ilmari Karonen (talk) 18:46, 2 March 2006 (UTC)
Does it require any leading spaces or anything else to escape the leading pound-sign? I've got it in my monobook.css, but the code keeps displaying on my watchlist, and the css page appears as a numbered list. —C.Fred (talk) 22:35, 4 March 2006 (UTC)
It does not need anything to escape the leading hash sign; the fact that it displays as a numbered list is a bug, and does not cause any problems. Check if you are editing on the correct place (monobook.css is only for Monobook; other skins use other names for the page), and clear your cache after adding it. --cesarb 16:34, 5 March 2006 (UTC)

Since the bare link to Special:Confirmemail has been confusing people (linking to the preferences would have been much better), I've edited its text (which is at MediaWiki:Confirmemail text) to add a link back to the preferences. It would be a good idea to change it to explain better the situation; I've only added a sentence, the rest is still the MediaWiki default. --cesarb 15:26, 2 March 2006 (UTC)

  • Would the SpamCop block have also affected the WikiEN-l mailing list traffic? I stopped receiving it yesterday. (But it started back up again today -- so maybe that means the SpamCop block is lifted?) —Steve Summit (talk) 15:48, 2 March 2006 (UTC)

I enabled my Wikipedia email function when I created my account. Do I need to follow this link and enter the confirmation code, or is this only for new users? Palmiro | Talk 16:50, 2 March 2006 (UTC)

Go to your user preferences page (Special:Preferences). It will tell you whether you need to confirm or not. —Steve Summit (talk) 17:05, 2 March 2006 (UTC)
Where do I insert the {display:none} line? I can't edit the watchlist page to insert text like that. Thanks. --Mmounties (Talk) 18:23, 2 March 2006 (UTC)
In your personal style sheet, which for you would be at User:Mmounties/monobook.css (assuming you're using the default MonoBook skin). —Ilmari Karonen (talk) 18:46, 2 March 2006 (UTC)
Should this same advisory appear in translation for all the other language wikis? ፈቃደ (ውይይት) 19:56, 2 March 2006 (UTC)
Yes, it should. The entry on the Server admin log quite clearly says it has been enabled for all public wikis. If you are an admin on other of the wikis, go ahead and do it; if you aren't, tell the admins there to do it. It's also particularly important for the admins to do the confirmation on their accounts as soon as possible, because it's one of the ways they can be contacted. --cesarb 20:23, 2 March 2006 (UTC)

As no one has bothered to explain why you need me to reconfirm my email address, I've disabled wikipedia e-mail for now. (I'm as paranoid about my security as you are about yours. :-P ) This is somewhat bothersome to me, as I'd like to be able to communicate with wikipedia e-mail. What gives? Kim Bruning 20:18, 2 March 2006 (UTC)

Let me explain then. Wikipedia has not asked for confirmation of your email address in the past; you could put any random email address in the email address field, and it would be used when someone tried to send email to you via Wikipedia. Now, however, the confirmation has been enabled; to be able to use the email features, your email must be confirmed. This confirmation is nothing more than proving that you are the legitimate owner of the email address. When you ask it to send you a confirmation email, it will send to the email registered in your preferences a short text message containing a link with a random key. You just have to open that link with your browser and your email has been confirmed as valid.
Here's a sample of how the message looks like (it's the one I received, sanitized to remove personal information):
Someone, probably you from IP address 127.0.0.1, has registered an
account "CesarB" with this e-mail address on Wikipedia.

To confirm that this account really does belong to you and activate
e-mail features on Wikipedia, please open this link in your browser:

http://en.wikipedia.org/wiki/Special:Confirmemail/deadbeefdeadbeefdeadbeefdeadbeef

If this is *not* you, please do not follow the link. This confirmation code
will expire at 04:07, 9 March 2006.
--cesarb 20:31, 2 March 2006 (UTC)
[That's the "how";] Here, I think, is the real reason [why]. As cesarb said, in the past, you could theoretically have put any random email address in the email address field. So that random person could have received Wikipedia email intended for you. So Wikipedia could be said to be spamming that person. In order to prove to concerned outside parties (e.g. spamcop) that it does not spam people, Wikipedia needs to be able to point to its policies which ensure that all the addresses it sends to are valid and requested.
In general, any site that sends mail (any site, for any reason) pretty much has to use a "double opt-in" scheme like this. Any other scheme eventually gets hijacked by spammers. I agree that double opt-in is a nuisance for all concerned, but it's unfortunately one of the prices we all have to pay in response to those pesky spammers. (I'm also not suggesting that Wikipedia email had actually been getting hijacked by any spammers yet, but I think it's clear that the "every site needs to use double opt-in, no matter what" policy does apply.) —Steve Summit (talk) 21:06, 2 March 2006 (UTC)
Isn't the correct name for it "confirmed opt-in"? "Double opt-in" sounds like you are doing something twice, which you are not (you only have to confirm once). --cesarb 22:01, 2 March 2006 (UTC)
"Double opt-in" is what a friend of mine calls it; I don't know which term is considered "correct". It's "double" because you have to explicitly do two things: ask for the email, and confirm that you asked. —Steve Summit (talk) 22:17, 2 March 2006 (UTC)
"Double opt-in" is the term we generally use in the consumer email business. You have to intentionally do TWO things (click a link and then submit a form, or check a box then submit). Confirmed would indicate that we sent you an email to say you have been signed up and you have acknowledged that yes, that was you, by responding. - PKM 19:26, 4 March 2006 (UTC)

RE:"How to get this off my watchlist notice" comments, you can bypass the box with the methods shown above, but the notice on the watchlist page is not going to be permenant, I've added it via MediaWiki:Watchdetails to get the word out to the thousands of editors who use wikipedia email, but don't read the mailing list or this page often. Comments regarding it's use can be made on MediaWiki talk:Watchdetails. I was bold in implementing this notice, even with it being in the mediawiki space, but don't intend for it to be up for more than 2 weeks. If confirmation are going to be left enabled, perhaps we could send out one last unconfirmed email to all wikimail holders, letting them know of the confirmation requirement? xaosflux Talk/CVU 01:37, 3 March 2006 (UTC)

There's a message at the top of my watchlist. I suppose I'm not the only one. :) It links here, but a link to an explaining page (what it's about) would make more sense. DirkvdM 06:32, 3 March 2006 (UTC)


I have got this "confirmmail" message 4 times now. Three times have I confirmed it this week. The IP number mentioned in the email is not mine tho. Does this matter? And how many times do I ave to confirm this?DanielDemaret 17:00, 3 March 2006 (UTC)

As long as the username on the confirmation email is correct, and you really requested the confirmation email to be sent, it's normal; the IP is probably a transparent proxy you are using (go to Special:Watchlist while not logged in; it should display the same IP address). You have to confirm only if Special:Preferences says you need to; you can see there when it was confirmed, and if it needs confirmation. --cesarb 17:09, 3 March 2006 (UTC)
I did not request the confirmation email to be sent.DanielDemaret 18:22, 3 March 2006 (UTC)

A banner ad. just appeared on my watchlist page, asking me to confirm Wikipedia e-mail. I have no (Wikipedia) e-mail address. How do I get rid of the banner advert? It takes up more than a quarter of my (small) screen!

Thanks -- quota 21:34, 3 March 2006 (UTC)

You have to use "#confirmemail { display: none }" on your monobook.css.
Well, i have been confirming my email on my several accounts, even on one i just made today. I think that if you have set the "Enable e-mail from other users" the confirmation should be sent for all accounts. Moreover, if you set an email on the registration process, the email confirmation should send automatically. Platonides 11:39, 4 March 2006 (UTC)
It is sent automatically when you register a new account. --Brion 12:14, 4 March 2006 (UTC)

This is a complete pain for contact with long term inactive users. E.g. someone has uploaded something and we want a confirmation of copyright status. Could there be a way to override this and send mail anyway for accounts older than a year. Mozzerati 22:54, 4 March 2006 (UTC)

Something that I think should be added to the Watchlist note is that this is a one-time deal and not an activation of e-mail receipts. I did't do it because I assumed it was a reference to the same sort of "e-mail receipts" the IMDb sends out. Every submission to IMDb generates an e-mail receipt which can quickly plug up your inbox if you're not careful (one of the reasons I don't contribute to the IMDb as much as I used to). And an IMDb sysop told me it cannot be turned off. Considering how many edits I do for Wikipedia, receiving a receipt for each one would have been disastrous. 23skidoo 20:22, 7 March 2006 (UTC)


How get rid of the banner ad?

Anyone know how to get rid of the banner advert at the top of watchlist? It goes on about how I 'must' do something to receive Wikipedia e-mail .. but why would I want that? I have a User and Talk page, and I would like to keep my e-mail private. I have tried following the links, and there is nothing there to say 'no thank you'. quota 21:00, 4 March 2006 (UTC)

Your answer would be here. It is only for users who have enabled the option to have users e-mail them via Wikipedia. If you haven't, you don't have to do anything. - File:Ottawa flag.png nathanrdotcom (TalkContribs) 21:06, 4 March 2006 (UTC)
Sorry I did not make myself clear .. I want to get rid of the banner advert, as it displaces all of the watchlist on my cellphone. How does one turn it off? Why is it there? Thanks! quota
More clarification: If it is only for users "who have enabled the option to have users e-mail them via Wikipedia" .. why is it displayed for users who have not enabled that? quota 21:33, 4 March 2006 (UTC)
Everyone sees that message, whether the option is enabled or not. You have the option of removing the message, though, with the instructions further down on this page. - File:Ottawa flag.png nathanrdotcom (TalkContribs) 21:39, 4 March 2006 (UTC)

E-Mail Confirmation

Anyone else seen this? Could this be a step towards something bigger and beautiful? — Ilyanep (Talk) 02:52, 3 March 2006 (UTC)

I've seen it but not quite sure I understand everything it is suppose to do and I can't find information about it. So what all is this suppose to be confirming? Ryokosha 03:30, 3 March 2006 (UTC)
Go to my preferences and make sure you have entered an e-mail address, and select the option there for getting a confirmation e-mail. You can also go to Special:Confirmemail to request the confirmation e-mail, so long as you have an e-mail address already set in my preferences. The confirmation is now required for e-mail functions to work. -Kmf164 (talk | contribs) 03:37, 3 March 2006 (UTC)
I've confirmed, so I don't know why the silly notice keeps appearing on every single page view. Make it go away! Michael Z. 2006-03-03 03:39 Z
Put the following in your User:Mzajac/monobook.css:
#confirmemail { display:none; }
This will remove the notice. -Kmf164 (talk | contribs) 03:51, 3 March 2006 (UTC)
But what the heck am I confiriming e-mail messages for? Just the standard to notify me of a reply to a watched page, updates to site and such? Ryokosha 05:39, 3 March 2006 (UTC)
No, actually for when someone clicks on the E-mail this user link on your talk page. You are not notified of changes to pages in your watchlist or changes in the site; it is exclusively for communication originated by other users. Titoxd(?!? - help us) 07:01, 3 March 2006 (UTC)
Anyone else seen this? Could this be a step towards something bigger and beautiful?

Meh, it's been a feature of 1.6 for a long time. :) Rob Church (talk) 18:51, 3 March 2006 (UTC)

Well, we do need confirmed email addresses for Stage I of something bigger... Titoxd(?!? - help us) 21:21, 3 March 2006 (UTC)
That's what I was looking at :P — Ilyanep (Talk) 19:33, 4 March 2006 (UTC)
So, is it? —Nightstallion (?) 16:25, 10 March 2006 (UTC)

See "Mail server blacklisted by SpamCop" up above for more information on this. —Steve Summit (talk) 02:19, 4 March 2006 (UTC)

I have a question: Is this email confirmation for all email addresses that we have. I have a Hotmail email address that has been giving me the s**** because of the spam that's inundated my account (from Ha'aretz to stuff from Ford) and I haven't checked my Gmail account in a while. Or is it just for the Wikipedia email? You've confused me. -Daniel Blanchette 17:21, 11 March 2006 (UTC)
Yes, what we're talking about here is Wikipedia attempting to confirm the (one) email address you've entered in your Wikipedia preferences. This has nothing to do with any of your other email addresses, or with use of your email addresses by any other sending sites. —Steve Summit (talk) 12:59, 12 March 2006 (UTC)

The table 'blobs' is full

Hopefully this edit won't destroy this page, but here's the deal: I was editing Wikipedia:Sandbox/Word Association, and when my edit saved, I was brought to the "this page doesn't yet exist" message. Knowing it to be in error, I cleared browser cache and reloaded the page. Then, I got "Error in numRows(): The table 'blobs' is full" with a bunch of gibberish. I tried going back and re-adding my edit, but still no dice. Then I tried to revert to an earlier version of the page, only to get more variants of the "table blobs" message. Anyone know what gives? — BrianSmithson 14:16, 13 March 2006 (UTC)

I managed to rv to 14:01 UTC and the to 14:02 UTC. This resulted in several blank edit boxes despite the edits themselves not showing as blanks. Sometimes, the diffs worked, sometimes they just showed a "no page, dude" message. Weird. -Splashtalk 14:36, 13 March 2006 (UTC)
Has this been resolved yet? It has affected a number of pages. ++Lar: t/c 17:43, 13 March 2006 (UTC)
This is in theory fixed now. The affected servers have been banged back into shape (mostly; some fixing still to do on srv33 but it should work transparently with automatic fallback). And I've fixed a bug in the fallback code which is likely the cause of the mysterious blank stuff.
There may still be some cached pages showing the 'no such page' message. Use the purge link to clear them if necessary... --Brion 23:04, 13 March 2006 (UTC)


Spell cheque

Is it possible to include a spell checking option for the preview of a page? No automatic corrections but it would be nice if wikipedia could highlight words that might be mis-spelled.--God Ω War 16:56, 12 March 2006 (UTC)

Realistically speaking, almost certainly not, however Firefox 2 (soon to be in alpha I belive) will include a spell checker. Martin 20:41, 12 March 2006 (UTC)
See Wikipedia:Village pump (perennial proposals)#Spell Checker at Edit Pages. Arbor 20:48, 13 March 2006 (UTC)

watchlist preferences - want it to default to 3 days

My watchlist is about 1400 pages long — it keeps on defaulting to a timespan of 12 hours so I have to continuously click it to "3 days" (to check if I missed out anything), which I find annoying. Is there anyway, in CSS, etc. for me to change this and set my default to three days, no matter the length of my watchlist? Is there a plan to implement this without all this tweaking in the future, too? Elle vécut heureuse à jamais (Be eudaimonic!) 01:41, 3 March 2006 (UTC)

Sounds like it could be an interesting feature request. Rob Church (talk) 18:52, 3 March 2006 (UTC)
I imagine a workaround solution would be some javascript that modifies the Watchlist link to include the 3 days parameter. — Ambush Commander(Talk) 22:09, 3 March 2006 (UTC)
My workaround has been to link directly to [9] and have that as a link on my browser's taskbar, for what it's worth. It's quicker to use than hitting the "my watchlist", anyway... Shimgray | talk | 12:25, 4 March 2006 (UTC)
Ah, that works for my home computer, but not for the public computers I use regularly. Elle vécut heureuse à jamais (Be eudaimonic!) 03:52, 7 March 2006 (UTC)
Test [{{fullurl:Special:Watchlist|days=3&hideOwn=1}} My watchlist]
Omniplex  03:44, 13 March 2006 (UTC)

Can these be used in the Wikipedia: namespace?

I'd like to make use of some of the following icons for a menu on a page in the Wikipedia namespace, but I need to track down information on the legalities of doing so before I proceed. What do you know of the licenses, copyrights, and trademark rights to these icons?

Image:Crystal Clear app mac.png - a link is included on this page to a crude translation of the author's permission. Though it's hard to tell, because the translation is pretty rough.

Image:Finder icon.png Image:Pages icon.png Image:Apple Dictionary Icon.png Image:ICal Icon.png Image:IPhoto Icon.png Image:Apple iDisk Icon.png Image:ISync icon.png

Though it appears that these icons are from a trademarked program. So I'm still not sure if they can be displayed as menu items for a page. Any help you could provide in tracking down the information clarifying to what uses we can legally put these icons to would be most appreciated. --Go for it! 13:28, 15 March 2006 (UTC)

No, since they are fair-use images. See both that link and WP:FUP#Policy, in particular item 9. They can be used solely in articles (meaning article space) that are about the thing they illustrate and nowhere else. Including here, so I de-inlined them. -Splashtalk 16:34, 15 March 2006 (UTC)

Browser font change error

I'm not sure what happend, I think I did something odd with my trackball. All of the sudden wikipedia and wiktionary are displaying in what looks like a small font (it doesn't show bolds and it's hard to read). Does anyone know what I did...and how to fix it? RJFJR 01:00, 15 March 2006 (UTC)

Seems better now. I don't know what happened or what fixed it. RJFJR 01:22, 15 March 2006 (UTC)
Most browsers have a view, text size, option, often also with keybaord and mouse shortcuts. xaosflux Talk/CVU 03:32, 15 March 2006 (UTC)

Does the article count include redirects?

Please provide the link to the documentation on this. Thank you. --Go for it! 04:02, 14 March 2006 (UTC)

Which article count? There have been many methods suggested for deciding what constitutes an "article", including "contains one wikilink", "longer than 200 bytes", "contains two newlines", and "contains one comma". --Carnildo 05:03, 14 March 2006 (UTC)
The one on the front page doesn't, though it does include disambiguation pages.--Cherry blossom tree 11:20, 14 March 2006 (UTC)

See Special:Statistics and Wikipedia:What is an article. —Steve Summit (talk) 14:24, 14 March 2006 (UTC)

No. "Articles" are defined as pages which are in the main namespace and which aren't redirects. They must also contain the string "]]". I don't remember offhand if there's a minimum size threshold, but I don't think so either. 86.134.49.239 22:10, 14 March 2006 (UTC)


Help - Portal code

I'm trying to understand how the portal template logic works because i want to copy it to my own mediawiki. I am running mediawiki 1.5.0. I'm trying to understand the logic in [10] This contains what looks like transcluding a template called {{fullurl}} . But no such template exists. I cannot figure out where "fullurl" is defined. What am I missing? Thanks. Nicholsr 22:11, 13 March 2006 (UTC)

{{fullurl}} is a built-in variable in newer versions of the MediaWiki software that gives you the URL of the page you're reading. I'm not completely sure if it appears on 1.5.0, but the latest stable version is 1.5.7, which contains a fair amount of security fixes. Titoxd(?!? - help us) 23:15, 13 March 2006 (UTC)
I believe the {{fullurl}} available only from 1.6 borgx (talk) 00:27, 14 March 2006 (UTC)

Please upgrade to MediaWiki 1.5.7 immediately. 1.5.0 contains a number of security vulnerabilites and major bug fixes.

That said, the fullurl variable is a feature of the parser in MediaWiki 1.6. 86.134.49.239 22:07, 14 March 2006 (UTC)

SUBPAGENAME etc...

Found a "new" magic-word in the code. This time it's SUBPAGENAME, it will return the name of the current sub-page. For example {{SUBPAGENAME}} gives here "Archive O" (here it is the same as pagename). if the page instead was Wikipedia:Foo/Bar, {{SUBPAGENAME}} will return "Bar".

Also I have found out what the GRAMMAR and PLURAL keywords are for.

GRAMMAR
does nothing on the english wikipedia, but the syntax are {{grammar:genitive|Foo}}, will render here as "Foo".
PLURAL
does simply return the first parameter here is the count are 1 (other wikipedias might also have a third parameter. Example: {{plural:count|wordform1|wordform2}} gives "wordform2". AzaToth 19:09, 12 March 2006 (UTC)

It is new, I added it some time last week in response to a request on wikitech-l. 86.134.49.239 22:05, 14 March 2006 (UTC)

Category Sort Blank bug

When there are a large number of items in category sort blank [[...| ]], the next 200 do not appear. Likewise, jumping to 0-9 and then trying previous 200, they do not page back to the beginning. So the problem is both directions. In Category:Redirects with possibilities, you can see the problem. It's existed this way for several months, so please don't fix this until the developers can look at it!

Yes, I know the problem is a bad category in Template:R to decade (edit | talk | history | links | watch | logs). But again, it's been that way for months, so leave it alone for testing purposes. I've already fixed (during the past two weekends) the problem in 4 templates that made this same massive 6,000+ entry bug at Category:Unprintworthy redirects. That's why I was looking for more examples, to determine whether it was a one time thing.

Anyway, just a heads up. I could not find a related bugzilla entry, and I'll post one there when I get home later today. --William Allen Simpson 17:46, 12 March 2006 (UTC)

Reported at bug 5241. --William Allen Simpson 12:52, 13 March 2006 (UTC)

Marked as a duplicate of bug #4912. 86.134.49.239 22:17, 14 March 2006 (UTC)

Main Page Draft

Hi. I was wondering how you got it so that the page title (Wikipedia:WikiProject Usability/Main Page) doesn't appear at the top. Thanks, Shardsofmetal [ Talk | Contribs ] 17:10, 11 March 2006 (UTC)

They used a DIV section placed over the title (view the source) I believe. That does raise a question though: it'd be nice if there were a command to suppress the title area. Something like __NOTITLE__? Or __NOBANNER__. (I guess the name's not important, but the functionality would be nice for the main page so they can avoid using the DIV trick to move stuff around). AFAIK this wouldn't be exploitable in any negative way (it wouldn't hide the buttons for discussion/editing/history/etc., and as a form of vandalism I'm sure it'd get reverted fairly quickly (bots could just look for insertions of __NOTITLE__; if that's possible)). —Locke Coletc 01:22, 12 March 2006 (UTC)
IIRC, it's actually in MediaWiki:monobook.js. æle 02:18, 12 March 2006 (UTC)

What's the point in including a tag to do something fairly useless and counterinformative? 86.134.49.239 22:04, 14 March 2006 (UTC)

Removing block log entries

I am an admin. When I went to block a user, I accidentally included a private email address. The entry is:

  1. 15:00, 7 March 2006 Woohookitty blocked "HeadleyDown (contribs)" with an expiry time of 90 minutes (block for violating civility rules on NLP. Katefan0 <(EMAIL REMOVED)>)

Could the email part of the entry be removed somehow? Or the entire entry since it was a mistake (and I redid the block) anyway? --Woohookitty(cat scratches) 17:34, 10 March 2006 (UTC)

Should you really have included the address here? ~MDD4696 22:59, 10 March 2006 (UTC)

You need to ask somewhere more prominent than this! 86.134.49.239 22:25, 14 March 2006 (UTC)

need developer to remove offending summaries

Could a developer please remove the offending summaries from the deleted history of Marian Rejewski and Marian Rejewski/cleanup? A vandal had used very offensive edit summaries, which could not be removed by those without developer access. Thanks. --Ixfd64 10:23, 10 March 2006 (UTC)

You need to ask somewhere more prominent than here. It does state at the top of the page and in the guidelines that developers aren't obliged nor guaranteed to read this. 86.134.49.239 22:00, 14 March 2006 (UTC)

Why is the search engine so crappy?

I type in 'bit torent', and get one result - Kratos Aurion. WTF? — Preceding unsigned comment added by 67.40.249.122 (talkcontribs)

Because torrent has two r's. Dragons flight 05:00, 2 March 2006 (UTC)

And 'Kratos Aurion' has only one? So instead of Bittorrent I get Kratos Aurion? Again, WTF? — Preceding unsigned comment added by 67.40.249.122 (talkcontribs)

Because that article contains both the words "bit" and "torent" (refering apprently to Torent Forest, a location in the game). It was apparently the only article using both words, as you spelled them. Dragons flight 05:18, 2 March 2006 (UTC)
Doesn't that majorly suck ass? I mean really, isn't there some way to tell it that there is at least a possibility that someone entering a bad spelling for bittorrent wants that instead of a completely unrelated computer game? — Preceding unsigned comment added by 67.40.249.122 (talkcontribs)
I'm sure the developers are working around the clock on the DWIM (Do What I Mean) Technology.boinger 14:53, 2 March 2006 (UTC)
Or, to use a slightly more technical term, fuzzy string searching. Weregerbil 15:14, 2 March 2006 (UTC)

Is anyone in charge of fixing the search engine? How do I help fix it? Is there a plan for rolling out a new version? — Preceding unsigned comment added by 67.40.249.122 (talkcontribs)

Not really. To do so would be a massive cost of time and effort. Unlike Google and Yahoo, we have only a few full-time developers, none of whose first concerns are the search algorithm. You're talking about something that would take years with the resources we have. It'd be nice, but there's not much we can do right now. Ral315 (talk) 06:23, 2 March 2006 (UTC)
What about making it an open source project? Surely, as popular as wp is, there would be volunteers to code it? I mean, there are os search algs that are better than that already. — Preceding unsigned comment added by 67.40.249.122 (talkcontribs)
The search engine is open source; dive in and Use the Source, Luke. --Brion 06:34, 2 March 2006 (UTC)
Is there a development road map? — Preceding unsigned comment added by 67.40.249.122 (talkcontribs)
Ahem, Google already fixed it for us, just type "bit torent site:en.wikipedia.org" in the google query box, you'll get what you want [11]. Isn't this great? Problem solved! Back to writing encyclopedia. --Vsion 14:22, 2 March 2006 (UTC)

A quick fix might be to add a link to the search results page: "You can search this term on Wikipedia using google", and that would do the site:en.wikipedia.org search. Just a random idea... Weregerbil 14:34, 2 March 2006 (UTC)

Alternatively, we could encourage our users to spell correctly... Anyway, whenever the MediaWiki search is unavailable, a Google search box pops up. (At least, it does for me) haz (user talk)e 15:39, 2 March 2006

Why is it ever switched on? There are no advantages, and lots of disadvantages to the MediaWiki search.
As a furriner I do appreciate sites that feature fuzzy searches, as correct spelling has an extra hurdle for us non-native speakers. Fuzzy searching is a long researched and published subject; ready code and almost ready pseudocode are available. Performance is a bit of a problem though. Exact matching is fast with a properly indexed database, whereas fuzzy algorithms generally require a heavier scan (though heuristics can be used to prune the search space). Weregerbil 16:01, 2 March 2006 (UTC)

In this specific case, there exists a redirect from "Bit torent" to "BitTorrent", but it does not show up in the search results. However, when one searches for "Bit torrent", the first two results are redirects to "BitTorrent". Now that doesn't make sense. --Someones life 15:59, 2 March 2006 (UTC)

The search index is badly out of date. That's probably the reason why bit torent isn't included. You're right, though, we do need a better search tool. —Ilmari Karonen (talk) 16:45, 2 March 2006 (UTC)
Thanks - the Google search works great! Why don't we switch over to that entirely? I don't want to be rude, but the built in search is an embarrasment. There are plenty of better ones out there, but it seems like the Google one is easiest. At the very least, put an option to use the google one wherever the other one is. In my case, someone on the phone asked me to check out bit torrent. So I type it in on Wikipedia, mis-spelling it (yes, I am, but then, so are a lot of people) and get absolute nonsense and no clue as to what to do next. Of course, in the real world, I type it into Google, realise I am mis-spelling it, and get the the main page for the software, but wouldn't it be nice if the seach on wp actually worked?
Google, as far as I know, still relies on web crawling to build their index. This would mean that Google searches with respect to Wikipedia will always be out of date and have the potential to overlook the obscure and little linked to topics. Hence we can't rely solely upon Google. Also, some people have issues with endorsing just one commercial search vendor. Ultimately search improvements are probably going to have come from inside. Providing intelligent fuzzy logic is a laudable goal, but dramatically increases the computational complexity of search. That doesn't make it impossible, but probably means that search will need to have some hardware dedicated to it if it is going to be made fancy without bogging down the rest of the site. Dragons flight 18:37, 2 March 2006 (UTC)


Well, it says just above that the search index is badly out of date, so Wikipedia search clearly isn't operating on live data. I think switching to Google is a good option, though perhaps there should be a customisation available to choose a different one?--Myk 06:36, 3 March 2006 (UTC)
It does work you just need to go back to school and quit complaining.. unless of course your willing to help fund development?--MatthewFenton 16:25, 2 March 2006 (UTC)

Well, it is broken, so I don't think that sort of rudeness is really necessary. I was just looking for Tariq Ramadan, after hearing about him on the radio, and typed Tarik Ramadan, not knowing how to spell it. I got only one result - Wilaya of Relizane massacres of 4 January 1998. How can you possibly say it works? For great justice. 16:37, 2 March 2006 (UTC)

Somehow I'm not surprised that Google, with an annual revenue of $6 billion and immense market capitalization, can afford all the fuzzylogic software it wants, complete with a hundred thousand servers (they really do have that many) to run it. Somehow I'm surprised that people expect free, non-profit Wikipedia to be able to manage it. In the meantime, until billionaires start to fund Wikipedia, we're just going to have to rely on accurate spelling or deal with the fact that typos will produce unwanted results. RGTraynor 17:03, 2 March 2006 (UTC)
So why not just switch off the one that doesn't work? For great justice. 17:34, 2 March 2006 (UTC)

The search engine is indeed crap, and I think the difficulty involved in improving it it is being far, far overstated. Yes, to make it perfect could take years, but to make it "good enough" should only take a few weeks for somebody who knows how to tweak regular expressions. (Don't ask me to quantify how I arrived at the figure of only a "few weeks", I pulled it from the same place you pulled the "years" figure). I've written similar searches myself. However, complaining here isn't the answer... there's got to be a place where we can request programmatic enhancements without getting the request shot down by people who know nothing about programming. The Crow 17:00, 2 March 2006 (UTC)

I agree that the obstacles seem to be being overstated. The heavy lift on doing a good enough job on this has already been done. For that matter, it's a project that might be given to students. Actually, yes, people do expect Wikipedia to manage it, in the same way that Mozilla puts out world class browsers and Open Office is expected to function. That's not being rude and ungrateful, just stating a fact. The original poster seemed to be asking how to help, and has not been given anywhere concrete to do that. For great justice. 17:34, 2 March 2006 (UTC)
The original poster actually used the words "How do I help fix it?" - pretty unambiguously offering to help, not just bitching. For what it's worth, I certainly don't have the coding skills to make this work, but I too would vote +1 for work on enabling fuzzy searching. I'd also vote +1 for people assuming good faith and pointing the original poster towards where he can help, rather than just slamming him (e.g. "It does work you just need to go back to school and quit complaining" - not very helpful). As has been pointed out, you may hear a word in conversation or on the radio and not know how to spell it, and for those without english as a first language I'm sure flexibility would sometimes help. Jamse 18:15, 2 March 2006 (UTC)
Not being a dictionary doesnt mean it doesnt work, Its not a spell checker nor should it be.. If you cant spell what your looking for then check it in a dictionary--MatthewFenton 17:58, 2 March 2006 (UTC)
Oh come on. You're saying it's feature, not a bug?! Of course it should be able to catch common mis-spellings and suggest likely matches. Especially with Arabic names, where translitteration is not always standard. Suggesting that I want Wilaya of Relizane massacres of 4 January 1998, when I asked for Tariq Ramadan is ridiculous. For great justice. 18:10, 2 March 2006 (UTC)

It can be hard to check somethng in a dictionary if you don't know how it's spelled :-) Going to school might not be a practical option to a large part of the world's population. Especially in the parts of the world where unbiased information is needed the most desperately. "Learn to spell English, lazy bones" is not the most constructive approach for helping them.

There are lots of concepts that are not in dictionaries (proper names come to mind). Also there are lots of words whose transliteration to English varies — witness the different spellings of Arabic words, as things like vowel transliteration are far from unambiguous.

Wikipedia has managed to get lots of volunteer help with content. From the comments in this thread I'm sensing that volunteer programmers have not been mobilized nearly to that level. Volunteer programmers certainly do exist; see the free software movement. Weregerbil 18:19, 2 March 2006 (UTC)

Fixing the search engine is a problem, but remember, there are many others, like monitoring and maintaining a fast, healthy server grid, finish the much requested article validation and single login features, etc. There's many requests and few coders, though. That said, if any of you has a clue as to how to fix it, you are welcome to submit a PHP patch to Bugzilla. Titoxd(?!? - help us) 18:25, 2 March 2006 (UTC)

COPY FROM MAIN VILLAGE PUMP: Does anyone realize that you can't get into the village pump by searching 'The Village Pump"? I typed it in the search and it came up with nothing, not even a relevancy match. I think it should be changed so that people can get in whether they searched it with or without 'the'. Pseudoanonymous 03:34, 28 February 2006 (UTC)
In fact, I've found that the search box is really awful at finding anything in Wikipedia: space. Two factors might exacerbate this: the checkboxes which you may need to check to get results there, and the fact that if you started the title with "Wikipedia:" it will look for a word starting with "Wikipedia:", including the colon, instead of taking the hint to search in the Wikipedia namespace. rspeer / ɹəədsɹ 18:33, 2 March 2006 (UTC)
In fact, I've found that the search box is really awful at finding anything. For great justice. 18:49, 2 March 2006 (UTC)

My guess would be that it's not so much that we don't have the development manpower, we don't have the servers. We don't even have the resources to update the literal full-text search indices now, from what I understand, so how are we supposed to improve them? (In my opinion, ads are the answer, but I realize that's an unpopular sentiment.) —Simetrical (talk • contribs) 01:39, 3 March 2006 (UTC)

I don't know about that - I think its less about the brute force of being able to search more, and more about searching smart. For example, right now, it seems that the search engine does a full text search of the article and title, looking for exact matches. I can't see there's much value in that. Fuzzy searching of titles would be faster, require less server time, and return more useful results. For great justice. 02:30, 3 March 2006 (UTC)
Actually, if we're worried about the servers, let's just turn off the search engine, and get free searching through Yahoo and Google. Their searches are better, free, and we use them anyway when the search engine is down. For great justice. 02:33, 3 March 2006 (UTC)

Share the hatred of ads. What about an 'ad mirror'? Loads of sites mirror with ads. Why not create a separate foundation whose purpose is to run an ad funded mirror (under the GFDL this is fine) and donating profits to WP? No one would fork, because it wouldn't be Wikipedia doing it. People who hate on it could hate on it. Who cares, it would be legal. And ... 3. Profit! For great justice. 18:38, 2 March 2006 (UTC)

Regarding Titoxd's comment above that you can just submit a patch to bugzilla: you wouldn't submit a patch to our bugzilla, you'd submit a patch to the Lucene issue tracker. The search backend is an external open source project, just like MySQL or PHP. There's not much point in complaining about it here. Lucene has been in active development for longer than Wikipedia has existed, I hardly think it's reasonable to expect us to have written our own vastly superior search engine in that time. -- Tim Starling 02:50, 3 March 2006 (UTC)

Thanks Tim - it's interesting how little is known about this. Seriously though - there is a vastly superior alternative available to us right now, for less work than we're doing now - why not just switch it off, and use Yahoo and Google? For great justice. 03:28, 3 March 2006 (UTC)
Yahoo and Google can't read our site structure (namespaces, etc), so can only provide a "dumb" word search. To have a more capable search which works with the wiki's structure (limiting common searches to the article space, etc), we have our own search engine. As noted above, it's based on Apache Lucene, a very capable and generalized open-source search backend library. If you're just going to whine about it, well, don't be surprised if no one pays much attention. :)
Additionally the foreign crawlers are limited to updating only after a long delay; at the moment this is true of our internal search as well but that's fixable if memory leak problems in the update daemon can be resolved. That's something Google or Yahoo can't provide.
A fuzzy title search à la "did you mean to type XYZ?" is possible and is in the code, but we have it disabled for now as it's slower to perform than the entire full-text search.
Kate's been working on improvements to the Lucene-based search for Wikia; if we can agree to get that work open-sourced we can move over to the updated code for Wikipedia as well. Otherwise it'll continue to sit around until someone else gets the inclination to put in some elbow grease. --Brion 04:11, 3 March 2006 (UTC)
I'm curious -- has anyone at the Foundation explored the Google Search Appliance, or is that not open-source enough for us? It looks hellishly expensive for the number of documents we have, but it might be worth it if it does the job we need it to do efficiently and well. See Google's page on it, and an external review. — Catherine\talk 04:43, 3 March 2006 (UTC)
We're aware it exists, but since it's totally proprietary it's not something we're considering. The Foundation's committed to running on a pretty much complete open-source software stack; part of what we're doing is making sure that everything we've got can be reproduced. --Brion 07:37, 3 March 2006 (UTC)
If the often-proposed machine interface to WP existed, Google could use that to search you, and (assuming it supported a suitably efficient "push") to keep their search up to date at all times. But there are always a zillion things to implement and I know you guys are under some tricky constraints. — ciphergoth 20:54, 5 March 2006 (UTC)
Thank you Brion - that's really helpful. If I understand you right, your saying that the search engine is better in theory than in practice, because WP had the fuzzy searching switched off to save server load? That improving the code wouldn't help, because the decent code is turned off? If that's the case, then it really is just a question of more money for more servers so that the decent search functions can be turned on? Thanks for an insightful answer. Just a comment from a user perspective, I have moved over to using google to search completely now, simply because it has the fuzzy title search. It would be great to see that in the main wp search, and would make it useable. Thanks again, For great justice. 17:44, 3 March 2006 (UTC)
Your sarcasm is appreciated; or are you really not aware that usability and functionality are not binary? --Brion 00:41, 4 March 2006 (UTC)
Are you sure that this user is being sarcastic? It's not obvious to me at all. — ciphergoth 20:54, 5 March 2006 (UTC)

Hmm.. i sense a bit too much whinging at each other, not enough point in the discussion. Please, WP:AGF, WP:NPA, WP:CIVIL, etcetera. Werdna648T/C\@ 05:48, 6 March 2006 (UTC)

??? no sarcasm intended, I was simply trying to understand the situation - so many contradictary explanations had been given above. I am not sure what you mean about usability and functionality. I am sorry if you thought I was 'whining' - I was trying to figure out where the problem is. If the problem is with the code, then the solution is better code, but if the code is already ok, but we don't have enough servers to run it, then that's a different problem. There seems to be a lot of 'don't criticise the search engine because it's unreasonable to expect us to have a good one and everyone is working hard'. I know they're working hard, and I'm not trying to suggest they're not, just trying to understand exactly why the engine is so bad for what I use it for, and, what I can do to contribute. If it's code, then not much. If it's money for new servers, then a little bit more. For great justice. 06:16, 6 March 2006 (UTC)
To claim that a fulltext search is "useless" without an additional feature for poor spellers is pretty silly. :) The search is useful. It could be even more useful, however. --Brion 08:06, 6 March 2006 (UTC)

I am not a programmer, so I guess I wouldn't know much about this; but at the risk of embarassing myself I would like to make a suggestion that could help with the problem temporarily. It seems to me the main problem with the search is that it does not check spelling. If that is so, then if it is feasible we could put a spellcheck on top or below the search box or program the spellcheck box to only appear if the search ends up with less than 10 results or something like that. We don't need to program the search to spellcheck, we only need a spellchecker. If thats too much can we make it so that the search only gets spellchecked when less than 10 results appear? I know typing stuff into a spellchecker is annoying but thats why it is a temporary solution and it is better then going to google for someone who has a slow internet. I know it might take just as much time but I am hoping it doesn't. Something does have to be done about this! A lot of people are using this website, I think its safe to say at least hundreds of thousands, if 1 out of 100 can't spell and have a slow internet then we would be inconveniencing thousands!!! Not to mention annoying tens of thousands who can't spell! I also think it is safe to say that only 5% if not less of us have never spell anything wrong in their entire life; so the suggestion that one should just learn how to spell is absurd. If a first timer comes to WikipediA searchs something and gets nothing, then we might not have that person as a second timer. Yes, we have other things to work on, but a good searchbox should be one of our top priorities; imagine Google not checking your spelling, then telling you, that you should just learn how to spell!(would we have heard of google if it did that?). The ads are not a bad suggestion, if it will make wikipedia better, then who cares whether you have a few ads on the side(do you divorce someone just because they told a few small lies?). Go to wookipedia or Memory Alpha the ads aren't bad. Don't say the seach box is useless or crappy a lot of people worked hard on that! :D Pseudoanonymous 05:38, 12 March 2006 (UTC)

Can someone explain why a search for "Duvall" doesn't find Duvall? Is it because the page was created on March 3rd? That was over 10 days ago. Thanks User:Amram99 4:33, 14 March 2006 (UTC)

As stated previously, the search engine currently does not immediately update. --Brion 07:19, 14 March 2006 (UTC)

Looking for a template

I am working on an article that received FA status a few days ago. In the past, although most of us working on this article had adopted the <ref></ref> reference system, and I had "translated" all previous source notes into this format, there were a couple of users who had refused to utilize the new system and had, instead, entered footnotes in whatever manner struck their fancy at a given moment, despite numerous requests from other editors to use the <ref></ref> system. The existence of the deprecated source notes inserted by the persons in question was noted in the FAC process and had to be corrected -- a time-consuming task that was carried out by the person who had nominated the article. My current concern is how, now that FA status has been achieved and all source notes are in the correct format, we can make it clear to those who wish to contribute to this article in the future that they must enter their source notes using the<ref></ref> system. Is there a template stipulating that usage of the <ref></ref> system is required that can placed on the Talk page? Thank you -- Polaris999 07:32, 16 March 2006 (UTC)

It is not, and cannot be required at the moment, partly because there are still uncorrected caveats to the Cite.php system. See Battle of Krasny Bor (refs in template breaks up)and Military hiatiry o France (Refs in Picture captions) for examples of articles that can't be converted. Circeus 11:45, 16 March 2006 (UTC)


Some characters disapear when pasting Wiki pages into MS-Word...

When I paste http://en.wikipedia.org/wiki/Russian_phonology into MS Word2000 many of the phonetic characters appear as boxes. What fonts do I need to add to Windows2000proSP4 for the chars to be preserved? THanks, MMCohen

Paste them into Notepad, rather than Word. Notepad is a full fledged Unicode editor, whilst your copy of Word may not be. ≈ jossi ≈ t@ 02:43, 16 March 2006 (UTC)

Image thumb layout unsatisfying

(This is a repost of a question submitted 2006-03-15 14:35:56 -0800, reverted (inadvertently?) at 2006-03-15 15:07:15 -0800)

In Washington Park, Portland#Children's Playground I'd like to display three thumbnail images across the page, preferably with left, center and right alignments. The current wikitext is

 [[image:Pdx_washpark_childrensplayground_elephanthouseleftside.jpg|thumb|center|Elephant house picnic area]]
 [[image:Pdx_washpark_childrensplayground_entrance.jpg|thumb|left|Children's Playground Entrance]]
 [[image:Pdx washpark childrensplayground back2front.jpg|thumb|Children's Playground Play area]]
 {{clr}}

But this displays the first image on a line by itself. The other two images are left/right aligned okay, but below the centered image, with plenty of space for the center image. It doesn't seem to be browser related: roughly the same appearance with IE 6.0, Netscape 8.0, Firefox 1.0 and 1.5.0.1 on Win2k and XP. The size of the thumbs has no effect as well. Putting the image references in left, center, right order got even stranger results: left appeared okay, but center was midway between center and right, then right was on a line by itself.

I tried <gallery> ... </nogallery>, but it doesn't make the image thumbnails satisfyingly large. Maybe there's a way to control that? (It does seem the proper way.) EncMstr 22:35, 15 March 2006 (UTC)

Someone changed the page to use a gallery, but without comment. Is this intended as an improvment?
I think it might be me, but the server had issues when I tried to save and I couldn't even see the final results. Circeus 02:12, 16 March 2006 (UTC)

Software solutions to copyvio

I've done a bit of research in the past into algorithms used by plagiarism detection systems. One of the simplest and most promising is described in the 2003 paper Winnowing: Local Algorithms for Document Fingerprinting, which I could implement a prototype of in PHP+SQL. All we would have to do is collect a corpus of text from a bunch of candidate sources, extract fingerprint data, and then we could have nearly instantaneous identification of any amount of copyrighted information anywhere in a submitted page. This is considerably more scalable than solutions based on the Google web service due to the web service's limitation on number of queries and query size. What do you all think? Deco 21:29, 15 March 2006 (UTC)

One point I didn't address above is what to do if a potential copyvio is detected; rejecting the document out of hand may be frustrating for users who are properly quoting material, or who accidentally conflict with it. I'd suggest instead that these changes are flagged on Recent Changes and the suspected copyvio portions highlighted in red in the diffs. This makes it all much more complicated, of course. Deco 22:26, 15 March 2006 (UTC)
The problem with this is the incredibly wide range of possible sources for copyvio material: articles on primary and secondary schools frequently contain material copy-and-pasted from the "about" or "mission" pages on the school's website, while biographies of minor pop-culture figures come from the figure's promotional home page. Copyvios from concentrated sources such as Britannica or Encarta don't happen very often. --Carnildo 23:27, 15 March 2006 (UTC)
I agree with Carnildo... I've never found a copyvio from what could be called a "standard" text; it's always been from a relatively small website directly related to the article's subject. This doesn't mean that there aren't copyvios from Britannica sneaking around, but in my experience they're much more rare. I didn't look at the paper, but since you talked about "collect[ing] a corpus of text," I'm wondering how you would address this problem of copyvios from smaller websites that don't really pop up until you google a specific sentence. Seqsea (talk) 00:03, 16 March 2006 (UTC)
I've once found a copyvio from Britannica, though it was only the first paragraph you can get for free off the website. I think 99.x% of our copyvios come from either the official website of the subject or another website dedicated to the subject. I do think copyvios from Britannica and the like are the most dangerous to us, though, since we're directly competing with them.--Cherry blossom tree 00:54, 16 March 2006 (UTC)
I wasn't actually considering copyvios from other encyclopedias, and I don't think I suggested anything to that effect. Most copyvios do come from the web. The corpus could be generated using a webcrawler, but because the web is so huge it would have to be targeted somehow to sites people are more likely to copy from. One way to do this would be to keep a list of websites known to be copyvio sources, but as Seqsea says, they're often small subject-specific sites. So this is tricky. Deco 01:47, 16 March 2006 (UTC)

Very recently, (as in today), I've noticed that Template:Half-Life series is no longer displayed in its entirety. Instead, we just have the title and a show/hide button. I checked the version histories of the template, and the change happened some time in January, however as I have only just noticed it, I'm guessing it's a recent change to the css instead.

This is useless for a template, yes the Half-Life template was too bulky maybe, but it was made for ease of navigation for newbies and other general users. A quick and easy way to jump between articles, however, now it's hidden. You're average user is not going to spot the "show" button, or even look at the navhead. Could someone change this back? There is some mention of this at Wikipedia_talk:WikiProject_Computer_and_video_games#What.27s_with_the_Show.2FHide.3F when pages with multiple navframes would have all but one hidden. Now they're all hidden, and it's rubbish. - Hahnchen 05:33, 15 March 2006 (UTC)

Templates - What the heck!?

Several templates( such as Template:PokémonGames and Template:Mario series) which previously displayed in their entirety are now hidden( collapsed) by default, and you have to click “[Show]” to see their contents. Since the templates haven’t been edited, I suppose it must be a change in the software? How can the templates be forced to display their contents by default? -- WikidSmaht (talk) 21:03, 15 March 2006 (UTC)

Not a software change, a stylesheet change. It looks like MediaWiki:Monobook.css was changed so that divs marked with the NavFrame class, as this one is, get the show/hide button. If you don't want it, you can change the div to some other class. --TreyHarris 21:31, 15 March 2006 (UTC)
The button was always there, but it used to show by default, now it hides by default. Why was this changed, who should I talk to about it? Looking at that stylesheet’s history, it hasn’t been edited since January. Are you sure that’s the source of the problem? -- WikidSmaht (talk) 22:29, 15 March 2006 (UTC)

Perhaps it is better if we can decide the default state for each template. Making separated div for example such as NavFrameShow for show by default and NavFrameHide for hide by default. borgx (talk) 00:47, 16 March 2006 (UTC)

Perhaps, but I’m still not clear on who changed what and why in the first place. And what( if anything) can be done in the meantime? -- WikidSmaht (talk) 02:13, 16 March 2006 (UTC)
I had a look at MediaWiki:Monobook.css round about the time I posted the original point above. But there hasn't been a change in that since January, so that wasn't the cause. I haven't found out yet what the cause actually is, but an awful lot of templates are doing an awful lot of no work by being hidden. Could someone revert this? - Hahnchen 02:30, 16 March 2006 (UTC)

These classes are from MediaWiki:Monobook.js, not from any of the CSS files. --cesarb 02:41, 16 March 2006 (UTC)

And, indeed, this is the change you all were looking for. --cesarb 02:42, 16 March 2006 (UTC)
Yes—sorry, my fingers typed ".css" when I was thinking ".js". Apologies for the confusion. --TreyHarris 02:45, 16 March 2006 (UTC)
Shameless plug: I'm making a Catalogue of CSS classes to make it easier to find where things such as these are defined and used. --cesarb 02:44, 16 March 2006 (UTC)

Textboxes get cut off

I'm using Mozilla Firefox 1.5.0.1 and I'm having this interesting problem: When I have a textbox that has a lot of text, most of the bottom of the text gets cut off and I can only see the top of the textbox. This is a problem if you are trying to edit wiki articles if half of the textbox is deleted and never shown in the textbox when the browser loads. (It's also difficult to copy and paste long HTML codes that people have put into textboxes especially templates for blogger sites.) Can someone help me? (I don't have this problem when I open the same page in MSIE.)

Wikipedia:Article size might have some useful information... in the mean time, try editing by sections. -- WikidSmaht (talk) 21:08, 15 March 2006 (UTC)
I'm not using a primitive browser, I shouldn't be having this problem. --Rachack 21:47, 15 March 2006 (UTC)
1) Can you provide a list of pages on which you encounter this problem?
2) Can you tell us what operating system and version you're running?
3) Are you running any kinds of proxy servers? (Ad-blockers, "sanitizers", JS-strippers, etc)
4) Are you using any Firefox extensions? Have you tried disabling or removing them?
--Brion 20:13, 16 March 2006 (UTC)

User Contributions

Is it possible to get a list of user contributions - just the same as the present list, but filtered to exclude all pages for which the user is (top)? (That is all edits to a page marked top, not just the most recent edit to it.) -- SGBailey 00:07, 18 March 2006 (UTC)


Wiki Molasses

You guys have a server down ? Wikipedia acts like Dial-up is faster, and I'm on DSL. I was directed here from WP:AN regarding this matter. Martial Law 21:27, 17 March 2006 (UTC) :(

Try 120 seconds, and a "Operation has Timed out." has appeared. Martial Law 21:43, 17 March 2006 (UTC) :o

'Edit' location screwed up

In Celtic knot the location of the 'edit' link for the History section is screwed up...somehow forced lower that the headline underline, possibly due to the images on the right. I can't seem to figure out a way to get this to sit normally, on the same line as the headline. Help? --Kickstart70 16:45, 17 March 2006 (UTC)

Page renders as I'd expect; section edit links will move about to accommodate block content. You could add a <br style="clear: both;" /> which might achieve the desired effect, but it doesn't make much difference. Rob Church 19:42, 17 March 2006 (UTC)
Sam Korn came and provided a fix, which is why it looks right now to you. It's a workable fix, but kind of a hack. Previous to this, the 'edit' link was rendered on top of other text within the History section. Ideally, this wouldn't be an issue, and I think that perhaps a containing box around both the included images would solve it more permanently. --Kickstart70 20:43, 17 March 2006 (UTC)

Help needed to fix a glitch

Hi, I've had a bit of trouble trying to format a number of tables in the past on this page. If anybody happens to know how to fix the problem of incomplete borders around these 3 tables without changing the arrangement of the contents inside, could you please so kindly fix them for me? Thanks heaps. mdmanser 14:08, 17 March 2006 (UTC)

Try removing the style bit about the borders. --Brion 22:39, 17 March 2006 (UTC)

WP Logo blurred in IE

File:Screenshot-logo-wiki-id.JPG
WP Logo blurred only on IE

Although this problem can easily get rid of by using Firefox, but I just want to ask if anyone have a workaround for solving this problem. The Wikipedia logo displayed blurred when using Microsoft Internet Explorer only (IE6 also confirmed). (see image) But seems that this problem is not consistent. I have found other PC with IE6 that work fine, but many disturbed with this condition. borgx (talk) 00:14, 17 March 2006 (UTC)

The reason is, that you are including your image by CSS (Monobook.css)! Maybe some versions of IE have problems doing so correctly ;-) You really should put up a request at m:Requests for logos (or better: Bugzilla!), because when Wiki.png is activated this effect will not be there any more. BTW: In my view the chief cause of this could be in ../skins-1.5/common/IEFixes.js at "Center image with hack for IE5.5", where the logo gets middled/centered. Maybe this causes that duplication (blurred logo)!? Try if the logo is shown correctly when you have JavaScript disabled, please. If yes, the following might work when added to Monobook.js:
if (isMSIE55 && !doneIEAlphaFix) {
 logospan.style.left = '50%';
 logospan.style.setExpression('marginLeft', '"-" + (this.offsetWidth / 0) + "px"');
}
If not (what I think; ->IE5.5), you could try to align the image (->#p-logo) 2px more to the left side in Monobook.css! --- Hopefully something of that will work; best regards, Melancholie 21:38, 17 March 2006 (UTC)
P.s.: Ah, there already is one: MediaZilla:5275 --Melancholie 21:50, 17 March 2006 (UTC)

Administrator needed! Please!

I did something silly and created a Wikipedia nonsense entry. Sorry. I regret it and wish to remove the entry's history. But I do not know how to do this. I suspect it requires someone with higher Wikipedia user rights than mine. My username is Marcusvox and you will see the silly entries I made on http://en.wikipedia.org/wiki/Sympatico

Can the history of my edits be deleted please, so this nonsense page will disappear for good? Once again, my apologies to the WIkipedia community for this lapse on my part. I do not make a habit of creating nonsense entries.--Marcusvox 23:34, 16 March 2006 (UTC)

That's very honest of you, but there's no need to delete them from the history since you've already reverted them out. -Splashtalk 00:28, 17 March 2006 (UTC)
Yes, there's no need to delete them, but there's also no need to keep them, so why not? Deleted. --cesarb 00:33, 17 March 2006 (UTC)
Thanks CesarB. I assume you are an admin. I will not be so foolish again. --Marcusvox 00:55, 17 March 2006 (UTC)
For one thing, they're no longer visible on the user contribs. Superm401 - Talk 02:01, 17 March 2006 (UTC)
But his request here to remove them is. --cesarb 16:06, 17 March 2006 (UTC)

Lost amethyst

The skin amethyst was apparently deleted, I had recommended it on Wikipedia:Customisation for legacy browsers without CSS. There should be a skin not depending on CSS magic doing its stuff the old way with pure XHTML, no WP:HIDE tricks, if possible no unnecessary translations of HTML 3.2 entities to UTF-8, and proper name= anchors for <references/>.

Willing to help test and maintain a legacy skin: Omniplex  12:41, 14 March 2006 (UTC)

The skin was removed because it had something wrong with it, if memory serves. I expect the exact reason is detailed in the release notes for 1.6 or the commit logs. 86.134.49.239 22:11, 14 March 2006 (UTC)
1) it looked terrible, 2) it was monumentally broken in every browser I tried it in, 3) the author didn't seem to have any interest in fixing or maintaining it. It would have been particularly ugly in browsers with poor CSS support; try the old Classic or Nostalgia skins, or the totally-blank-slate MySkin for that. --Brion 00:43, 15 March 2006 (UTC)
Apparently I was "moved" to Classic, no problems with that - I could do without the "insert UTF-8 char." JavaScript zoo on my legacy browser with (normally) disabled JavaScript 1.1, but nice to have for legacy browsers would be a waste of time. However a hidden reset-button would be fine to accelerate the insertion of action=raw results, poor man's variant of external editor. Omniplex  02:02, 17 March 2006 (UTC)

Hiding robot edits in watchlist

Buzilla: 4895

Please vote for this feature. My watchlist is unusable until this feature is activated. Of the 120 items on the first page of my watchlist, over 80 are from bots. It's as simple as setting $wgFilterRobotsWL = true;... we just have to wait for a dev to actually make this change... — 0918BRIAN • 2006-03-9 20:23

Please note that Brion Vibber has now marked this bug WONTFIX. Rob Church (talk) 07:40, 10 March 2006 (UTC)
And for follow-up, it's been REOPENED pending a fix to the feature Brian cited. Rob Church 07:44, 17 March 2006 (UTC)

Deleted Articles

I've noticed that it is some vandals favorite activity to create hoax articles. However, once these articles are deleted they disappear from the user's contribution list. Is there any way to see what articles a user has created, including deleted articles? It seems like it would be useful, since these articles take up more overall energy to fix than simple vandalism, and if it was known that a user tended to create hoax articles, then less time could be spent triple checking that it was indeed a hoax. Makemi 06:16, 5 March 2006 (UTC)

Not at this time. --Brion 06:57, 5 March 2006 (UTC)
But we could make note of it on their talk page esp. in the Edit Summary. It'll take an extra step to check out, but the edit summary would quickly bring this matter to attention, and it's not something the user can make go away. Rklawton 07:22, 5 March 2006 (UTC)
Is there an attempt to bring back deleted article list? --Masssiveego 03:40, 10 March 2006 (UTC)

Brion's working on changes to the way we mark specific data as deleted, which appears to include different levels of visibility. Looks like the old deleted revisions stuff will be reborn in shiny new packaging. Rob Church 07:49, 17 March 2006 (UTC)

Still trying to get rid of the banner ad in watchlist; help!

I have been trying for two weeks now to get rid of the banner ad in my Watchlist about 'e-mail must be confirmed' (I do not have an e-mail to confirm). Several people have made suggestions (thank you, all!) but none work. I have created a User:quota/monobook.css with the magic incantation. I have erased all files and temporary files. I have rebooted endless times. But I still get that banner which displaces all useful information off the bottom of the page.

Why isn't there a way to say 'yes I have seen this, thank you'? Following the links is hopeless .. one just goes around a loop.

Really would appreciate some help on this! quota 22:10, 9 March 2006 (UTC)

There's no "banner ad"; there's a helpful message someone put in about a change in the system. If you don't want to read it over and over, don't read it. --Brion 00:00, 10 March 2006 (UTC)
Well, with the huge space above and below and three lines in the box it takes about six lines of my screen, which means on the first screen ther are only 2 or 3 watchlist items. It's hard to see how that's 'helpful' :-) quota 09:15, 10 March 2006 (UTC)
Alright, the instructions for doing this are really simple, so read this several times if you need to:
  • Edit that file. Yes, it should be blank. Go to edit and it will bring up an edit screen.
  • Add this:

#confirmemail { display: none; }
...then and save the file.

  • You will then need to do a hard refresh of your browser. You will need to press Control-F5 if you're using Internet Explorer, or Control-Shift-R if you're using Mozilla/Safari/Konqueror or F5 if you're using Opera.
  • Go to your watchlist and the message should be gone.

- File:Ottawa flag.png nathanrdotcom (TCW) 00:15, 10 March 2006 (UTC)

Many thanks for the detailed instructions! They are pretty much what I'd garnered from other posts scattered about.
I've tried them again, following exactly that procedure, and there is absolutely no effect. (This on IE.) Any idea when the box will go away of its own accord? quota 09:15, 10 March 2006 (UTC)
Ah, a thought .. could it be that the reason it does not work is because I am using the 'Classic' Skin? If so, is there some other way to get rid of it? -- Thanks. quota 09:18, 10 March 2006 (UTC)
Use User:Quota/standard.css instead. Sam Korn (smoddy) 09:34, 10 March 2006 (UTC)
That seems to have worked. Thank you! 05:19, 17 March 2006 (UTC)
Maybe for these banner-boxes a checkbox 'please don't show me this again' would save many frustrating minutes and hours. I would much rather spend time editing than trying to work around this sort of thing. quota

infobox help

Can anyone tell me how to get tables into infoboxes ant how to create tables?

For a quick and dirty example see Template talk:-, for a more elaborated example see Template:Infobox Micronation(edit talk links history). Omniplex 14:05, 18 March 2006 (UTC)


Remove item from watchlist

There's an item which has been on my watchlist for a couple months now, and it's beginning to bug me. I've tried removing it a number of times, and I've cleared my browser cache, but every time it says "Couldn't remove item 'Wikipedia:BJAODN/This is your final warning, if you attempt another bad joke , you will be deleted nonsense'...done." Any ideas? Makemi 22:09, 17 March 2006 (UTC)

Is that the exact title of the actual page on your watchlist? Does clicking this link help? This is a bit difficult to debug, since no-one else can actually see your watchlist. Could you cut and paste the part of your complete watchlist containing the page you're trying to remove below? Or could you save the entire page to disk and e-mail it to me? —Ilmari Karonen (talk) 22:32, 17 March 2006 (UTC)

The link you gave resulted in the exact same error (yes, that is the exact name). In context it is

  • White flight
  • Wikipedia:BJAODN/This is your final warning, if you attempt another bad joke , you will be deleted nonsense
  • William Hunt
  • William Joyce (writer)
  • William Shakespeare
  • Wolfgang Amadeus Mozart
  • Women's rights
  • Zac Efron
  • Zaire
User
  • User:-TheRaven-
  • User:129.252.89.201
  • User:151.198.82.3
  • User:167.206.142.194
  • User:167.206.174.28

I guess it's weird because this was actually a vandal article, which is why it shows up in regular articles, rather than Wikipedia namespace articles. (eg. the title of the page was Wikipedia:etc., it wasn't truly part of the namespace). Perhaps that's why the software is confused. Thanks, Makemi 23:38, 17 March 2006 (UTC)

Now that's weird! But I think I've figured out a workaround: try this link. —Ilmari Karonen (talk) 23:49, 17 March 2006 (UTC)
Hey! You did it! Why did that work? Thanks, Makemi 00:02, 18 March 2006 (UTC)
When you watch or unwatch a page, the talk page is also watched or unwatched, and vice versa. The name of the page unwatched by the above link begins with "Talk:Wikipedia:", which makes it the talk page of a (supposedly impossible) main-namespace page with a name beginning with "Wikipedia:", which is what you somehow had on your watchlist. So unwatching it also unwatched the corresponding article page.
In fact, I'm willing to bet that it was the oddly named talk page that you'd watchlisted in the first place. Unfortunately Special:Watchlist/edit only lists the non-talk names, which means you can't unwatch such an oddly named talk page that way. Using the unwatch tab on the talk page itself would probably also have worked. This is, arguably, a bug. —Ilmari Karonen (talk) 00:30, 18 March 2006 (UTC)
Submitted as Bug 5280. —Ilmari Karonen (talk) 01:05, 18 March 2006 (UTC)
Thank you, that sort of makes sense, although generally when have I removed pages from my watchlist which have the same circumstance (I've only edited the deleted talk page, and both article and talk page are deleted) I don't have a problem, but perhaps the "Wikipedia"ness of it was what caused the bug. Thanks, Makemi 01:43, 18 March 2006 (UTC)

Images in articles give poorer google results

I discovered to my dismay that having images in an article produces poorer results from a google search (see [12]) than one without. A google hit on a wikipedia article with an image shows the caption of the image in the snippet rather than the first essential summary of the article. In the example search here for Nanortalik (search:[13]), the German wikipedia article gets #1 and with the essential information we would like to see displayed in the search result snippet. For the english Wikipedia (#10) that has an image at the start, the snippet does not encourage follow-through click due to its irrelevant snippet text. Any suggestions on how to prevent this, apart from having no images at the top of the page? Jens Nielsen 15:52, 17 March 2006 (UTC)

Wikipedia really doesn't have to worry about search engine optimization. We're the 18th most popular site on the internet, according to Alexa. If you build it, they will come. Superm401 - Talk 21:55, 18 March 2006 (UTC)

Replacing the image in the taxobox, and getting an image to sit where it belongs

In the article Hythe, Alberta, I have two problems. I want to replace the map in the taxobox with the map sitting above it, but I can't find where in the code the first map's name is. It must be in the taxobox code somewhere, because when I remove that, the map disappears too. Second, failing being able to replace the taxobox map, I want the other map to sit flush right. Using the code [[Image:dwalbertahythe.png|thumb|250px|right]] ought to do it, no? But it seems it doesn't quite. Can I please get a hand with this? Thanks! Denni 04:10, 17 March 2006 (UTC)

The box is created by {{Canadian Town}}, which has hardcoded into it the use of an image called "{{PAGENAME}} Location.png". The only way to get your image into the box, short of rewriting the template to allow an override of the image name, would be to re-upload your map over Image:Hythe, Alberta Location.png.
I'm not quite sure what your alternative request is. The dwalbertahythe.png is flush right for me (but I use the classic skin, which sometimes places images differently from monobook). I would suggest you put the new image below the box rather than beside it, which you can achieve just be shifting it down the wikicode a few paragraphs.-gadfium 05:24, 17 March 2006 (UTC)
That's just the default name for the image. It can be overriden with a template parameter. --cesarb 01:02, 18 March 2006 (UTC)

Need a bunch of users' opinions on sizes

At Template_talk:Current_U.S._Senators#Size poll, we're discussing whether or not the current version (D) is as good as the suggested one (M). Which version is smaller (which is the question of the moment). Many people are saying version (D), but it sure as heck doesn't look that way on mine. More pairs of eyes (and a larger variety of browsers) are needed, I think. Thanks! Matt Yeager (Talk?) 01:11, 17 March 2006 (UTC)

FWIW, I get "D" to be smaller - Firefox, Windows XP, 1024*768px. (I also can't see the colours, but that's another matter entirely...) Shimgray | talk | 01:41, 17 March 2006 (UTC)
D is much smaller (Firefox, winXP). I can barely see the colours. -- Finlay McWalter | Talk 01:49, 17 March 2006 (UTC)
D appears smaller to me as well. Also can barely distinguish the light colors. Thanks! Flcelloguy (A note?) 02:01, 17 March 2006 (UTC)
D is smaller (Mozilla SeaMonkey Suite); the other one doesn't even fit in my screen resolution, forcing a horizontal scrollbar. *Dan T.* 02:20, 17 March 2006 (UTC)
D is smaller, but not by much (Opera, large fonts). --Carnildo 02:52, 17 March 2006 (UTC)
D has the smaller type size, with a little bit of scrolling needed. M needs a lot of scrolling to see it all (on IE6 with a 1024 by 768 screen). Both with D and M the scrolling is irritating - Adrian Pingstone 16:26, 17 March 2006 (UTC)
D fits nicely on the screen, while M overflows and needs horizontal scrolling. Both versions are perfectly readable. Mozilla Firefox 1.5 on Debian stable, 1600x1200, 132 DPI. --cesarb 00:47, 18 March 2006 (UTC)
I've explained the reason for M being bigger than D at Template_talk:Current_U.S._Senators#Size poll. --cesarb 00:59, 18 March 2006 (UTC)

User-specified New Pages

Is ther a way to see all the "New pages" that an specific user has created, perhaps something similar to Special:Newpages but user specified? --MoRsΞ 23:08, 16 March 2006 (UTC)

Looking at a filtered Special:Newpages now; will be committing it to the codebase when I've done a bit more cleaning. :) Rob Church 20:21, 17 March 2006 (UTC)
That will be very useful. We already have Special:Log/upload so it's kind of odd that we don't have this. Thanks for working on it. Superm401 - Talk 21:50, 18 March 2006 (UTC)

Cite.php weirdness

The last footnote of Lack of outside support in the Warsaw Uprising unexplicably begins as bold, despite my having looked the entire paragraph for mismatched apostrophes... Can somebody gives it a deeper look? Circeus 02:13, 16 March 2006 (UTC)

It's caused by the single quote at the end of "told that the files had been inadvertently destroyed". I'm not sure how best to fix it, though; you might want to get rid of the italics there entirely. —Kirill Lokshin 02:21, 16 March 2006 (UTC)
(edit conflict) I fixed it. The closing single quote was joined with the two primes ending the italics. — Knowledge Seeker 02:29, 16 March 2006 (UTC)
m:Cite offers a list of known issues. Omniplex  00:06, 18 March 2006 (UTC)
This wasn't a problem with Cite.php at all, though. Superm401 - Talk 21:44, 18 March 2006 (UTC)

Watchlist by namespace?

I know it's probably been mentioned before, but there's no Talk:Watchlist so I'm bringing it here. What are the odds of being able to filter your watchlist by namespace, like you can with Contributions/Recent Changes? Sometimes I can't be buggered to deal with policy/RfA/*fD questions, but would still like to ensure that no one has vandalized Picts again.

I tried to split out the non-article-space pages I participate in and use Related Changes, but that's really hard to maintain and doesn't automagically update for things like Page Moves. And as much as I'd like to know the special syntax for showing ONLY WP-space watchlisted items, what I'm really concerned about is being able to only-show-articlespace on occasion. Thanks. -- nae'blis (talk) 16:47, 14 March 2006 (UTC)

You can ask about this on bugzilla:, the developers will probably see it faster there. æle 21:02, 14 March 2006 (UTC)
Thanks. See below for what I found. -- nae'blis (talk) 22:26, 14 March 2006 (UTC)

Sounds like something which would be useful to add. I'll get onto that if I have a spare moment, although I recommend filing a feature request if there isn't one.

If there is, please let me have the bug number so I know when to mark it fixed and can take other comments into consideration. 86.134.49.239 22:14, 14 March 2006 (UTC)

1862 seems to be an identical request, or nearly so, if I'm reading Bugzilla correctly. -- nae'blis (talk) 22:26, 14 March 2006 (UTC)

I've made a user script to do this: see Wikipedia:WikiProject User scripts/Scripts/Watchfilter. —Ilmari Karonen (talk) 17:32, 18 March 2006 (UTC)

Blank edit fields

Yesterday, between 14:03 UTC and 14:31 UTC, a problem occurred when editing a page where Wikipedia would return an empty edit box when editing a page. This presents a serious problem for bots: they might end up blindly blanking large numbers of pages. Was this problem a one-time thing, or are those of us who run bots going to have to code around it?

(For anyone worried, the only effect this had on my bot was that it blanked its own talk page repeatedly.) --Carnildo 05:01, 14 March 2006 (UTC)

Is it still happening? If so, we need much more information in order to reproduce, diagnose and fix it. 86.134.49.239 22:15, 14 March 2006 (UTC)
That's what I'm asking: Is it something that's likely to occur in the future (in which case I'll need to code around it), or was it a one-time thing? --Carnildo 05:23, 15 March 2006 (UTC)
Long-since fixed, yes. --Brion 20:23, 16 March 2006 (UTC)
Thanks. --Carnildo 21:18, 16 March 2006 (UTC)
Scratch that -- the bug (or a similar one) is still there. OrphanBot just blanked an article, and the logs indicate two possibilities:
  1. The page the server supplied contained an empty edit field, or
  2. The server supplied a partial page, cut off sometime after the "wpEdittime" field but before the end of the "wpTextbox1" textarea.
--Carnildo 21:45, 18 March 2006 (UTC)

<nopreview> and <previewonly>

Are there directives like those, similar to <noinclude> and <includeonly> ?

I have sections to edit in the middle of a large table. (take a look at Wikipedia:WikiProject The Beatles/Article Classification... each letter is a separate section) Because they are in the middle, when you edit one section, the text is not formatted in preview mode but is jumbled, making it harder to tell what you get when you save. I tried using includeonly which works if you edit an included subsection page (look at User:Lar/Sandbox and User:Lar/Sandbox/album Q, one of the sections) but not so well for what I want as when you click on the section edit tag it edits only a section of the inclusion.

Related question: how do you control the font size of the [edit] tag? It comes up in a very large font, I'd rather it was smaller than the section title. Or am I going about this the wrong way, using sections for editability? Should I be using something else? Forcing edits directly? ++Lar: t/c 23:39, 13 March 2006 (UTC)

No such directives exist because there's no need for them to. It sounds ridiculous. The styles applied to the section edit links come from the CSS files in the skins and can be overridden with the sitewide customisation or a per-user customisation. 86.134.49.239 22:09, 14 March 2006 (UTC)
I'm sorry you think "it sounds ridiculous". Frankly, I found that comment rather offputting. I'm not sure who you are so I'll try asking again. I think there is a need for these tags and I've given an example why. Did you take a look at the table I referenced? If you don't understand why, please suggest a different way to achieve it.
The section edit links in the example I gave (again, did you look at it?) are way larger than normal ones. I suspect because the section head is buried in a table cell which is changing what size they are. Changing the sitewide customisation for section heads is probably not the way to address the issue. ++Lar: t/c 23:07, 15 March 2006 (UTC)
I said it was ridiculous because it sounds like it to me. I don't see the need to hide things from preview, or only show them on preview, which is what such tags would be expected to do, no? As for the section edit links; well, providing context in the future would have helped that misunderstanding. No, changing the sitewide values isn't going to help. Rob Church 07:29, 17 March 2006 (UTC)
If you don't see the need to hide or show text or markup only during preview I suspect that perhaps you still haven't actually looked at the example I gave. Here is the problem again. Please take the time to look instead of spending time on writing snide commentary if you would be so kind (Please be civil, there is no context in which "ridiculous" is not snide or uncivil when applied to a suggestion, even if you think it actually is validly ridiculous, there are better ways of expressing your opinion). I have written it out again in more detail.
Go here: Wikipedia:WikiProject_The_Beatles/Article_Classification and scroll down to the tables... for example "Articles in Category:The Beatles albums. See how big that "edit" link is on any of the table sections? That's the first problem. Now click one, for example the "B" section: [14] What was a nicely formatted table section when looking at the whole thing is now a jumbled mess in preview, because it's a chunk of a table out of the middle. Editing it is difficult because you can't be sure you did it right, and you have no way to tell without saving (increasing server load and the number of edits to track). But if the putative previewonly tag existed, a pair of previewonly {| and |} bracketing the text would make it all highly legible.
Now... you could say, why do you want to be able to edit one section at a time instead of the whole table? Two reasons. the whole table is big, and you have to scroll around a lot to check your work, but worse, it's subject to edit conflict when multiple project members are busily classifying articles. Hopefully that explains why this is a nice to have. Or alternatively, perhaps there is another way of doing it? One way we thought of is having each table section be a separately included subpage (where using noinclude then gives you the same effect) but that creates hundreds of subpages to manage, which seems icky. ++Lar: t/c 11:48, 17 March 2006 (UTC)
That page gives me a headache just looking at the design. I see both points raised, yes. I still dispute that we should be allowing people to hide things from the preview; there isn't a massive lot of point in it for all our other pages, is there? Rob Church 20:27, 17 March 2006 (UTC)
Trust me, it gives me a headache too!... I mentioned nopreview for symmetry reasons but it's previewonly that would allow a fix, nopreview doesn't seem very useful. Do you have any suggestions for improvement of the page, keeping in mind it's relatively high traffic? We had thought (in addition to the separately included subpage mentioned above) of splitting the tables up at the sections, but that causes the cell widths to vary (even with forced percentages in the header row) from one section to the next. Thanks for any suggestions (or pointers on where else to ask for ideas). ++Lar: t/c 05:25, 18 March 2006 (UTC)

Moving category history

Can something like this: Category:Norma Elizabeth Boyd be fixed (moved to the main namespace) by a dev or something? Or should I just cut & paste move it and delete the category? --Sherool (talk) 18:36, 19 March 2006 (UTC)


Change Main Page

Anyway to override the new main page and view the old main page or am I asking for a hopeless cause. The new one honestly hurts my eyes! Mike (T C) 07:48, 19 March 2006 (UTC)

Main Page alternate (Classic 2006) (the Featured article, ITN, selected annivs, and DYK should still rotate properly) Raul654 07:53, 19 March 2006 (UTC)

Columns of content without yucky tables

I have been working to get rid of table-based visual layouts lately, and tried to tackle content in columns, like some people prefer with See also sections that have tons of links, etc. I don't really like them, but whatever; I might as well make them work well. Besides, with the CSS-based layout, people who really don't like the columns can turn them off in their monobook.css.  ;-) The original templates are at {{col-begin}}, and my new ones are at Template talk:Columns. I need feedback and it needs tweaking. Bypass your cache if you don't see the example. — Omegatron 06:14, 19 March 2006 (UTC)

Getting Username in Javascript?

I'm trying to store the username (of the currently logged-in user) into a string variable. So far I've tried

var myUserName = document.getElementById('contentSub').innerHTML.match(/'(.+)'/)[1];

but that fails because document.getElementById('contentSub') returns null. I also tried using MediaWiki:Yourname but I can't figure out the syntax for it. --Tifego 03:00, 19 March 2006 (UTC)

I'd suggest document.getElementById('pt-userpage').getElementsByTagName('a')[0].href and remove the "/wiki/User:" prefix. —Ilmari Karonen (talk) 03:19, 19 March 2006 (UTC)
Thanks. (But why not document.getElementById('pt-userpage').getElementsByTagName('a')[0].text instead?) --Tifego 04:03, 19 March 2006 (UTC)
You can do:
var myUserName = document.getElementById('pt-userpage').getElementsByTagName('a')[0].firstChild.nodeValue
"text" isn't a valid object. However, the elegant way to do it is read the cookie (ideally, you should use a cookie-reading function).
var cke = document.cookie;
var pos = cke.indexOf("enwikiUserName");
var start = cke.indexOf("=", pos)+1;
var end = cke.indexOf(";", pos);
if (end == -1) end = cke.length;
var myUserName = cke.substring(start, end)
-Superm401 - Talk 05:29, 19 March 2006 (UTC)
No, I think you're wrong about that being necesssary. Mainly because I'm already using the ".text" version and it works to get my username perfectly. --Tifego 05:33, 19 March 2006 (UTC)
To clarify: "text" has special meaning that makes it valid to use here although there's no node with that name. "text" is what's inside the node, i.e. <tag>text</tag>. --Tifego 05:43, 19 March 2006 (UTC)
You're right! Keep in mind, though, the DOM standard it to use textNode . Superm401 - Talk 05:47, 19 March 2006 (UTC)
Also, I still think it's cool to use a cookie. Superm401 - Talk 05:48, 19 March 2006 (UTC)

A better way to extract the cookie, using regular expressions, is:

var userName = decodeURIComponent(/((^|;)\s*enwikiUserName\s*=([^;]*)|$)/.exec(document.cookie)[3].replace(/\+/g, " "));

This also takes care of unescaping the username properly. —Ilmari Karonen (talk) 16:08, 19 March 2006 (UTC)

Spam filter problem

There's a problem with the spam filter, in which a person can make an edit like this one, but I'm not allowed to revert it. Can an admin revert that and someone please look into it? Thanks. tv316 22:59, 18 March 2006 (UTC)

Somebody reverted it already. No idea why you think that you can't do that, removing an unsigned Spanish text on a controversial talk page is IMHO no technical issue. Omniplex 07:57, 19 March 2006 (UTC)
It's not that I didn't think I could do it. It would not allow to me either rollback or manually revert it. Do you really think I would just post a link to the diff and say "Alright, guys. Have fun. Not my problem anymore!"? It becomes a technical issue the moment my revert gets denied due to the spam filter. tv316 22:53, 19 March 2006 (UTC)
The ON link in the "Al-Qaeda strategy" section triggers the spam filter (administrative rollbacks apparently work, though). I have fixed the problem by tweaking the link. Raul654 07:58, 19 March 2006 (UTC)