Jump to content

Wikipedia:Village pump (technical)

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Tony Sidaway (talk | contribs) at 20:31, 20 March 2005 (→God mode monobook.js). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

The technical section of the village pump is used to discuss technical issues. Bugs and feature requests should be made at BugZilla since there is no guarantee developers will read this page.

FAQ: Intermittent database lags can make new articles take some minutes to appear, and cause the watchlist, contributions, and page history/old views sometimes not show the very latest changes. This is an ongoing issue we are working on.

Details about the occasional slow speeds and deadlock errors: here

Please sign and date your post (by typing ~~~~ or clicking the signature icon in the edit toolbar).

Start a new discussion in the technical section

 Policy Technical Proposals Idea lab WMF Miscellaneous 

Technical archive

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

Servers handling of URLs has died

Well, up until a few days ago, I would get to any article by simply putting in the address bar "www.wikipedia.com/[topic]" and it would work... However now it just displays this real dumb 404 error. It is pretty useless and it reduces some of the simplicity of working the site. (comment was from User:Floydian (Talk - moz.)

Are you sure that wasn't "www.wikipedia.com/wiki/[topic]", which should work? -- Jmabel | Talk 18:31, Feb 26, 2005 (UTC)
You need to put en.wikipedia.com/wiki/[topic] in your address bar. You have always need the wiki, but although using en instead of www was always correct, the handling of www changed about the beginning of this year. You also need to put underscores in if [topic] has spaces in it, eg en.wikipedia.com/wiki/Current_events. -gadfium 18:36, 26 Feb 2005 (UTC)

Try .org instead! Noisy | Talk 10:35, Feb 27, 2005 (UTC)

Well, I know that now you need the /wiki part... But about a week ago I could do it without the /wiki
Its not really annoying to me as its 5 characters... However, it seems pointless that the backend couldn't just add that /wiki in the same way that it changes %20 to an underscore. - Floydian 22:37, 27 Feb 2005 (UTC)
Of course, if you use Firefox, you only need to type wp [topic]. Smoddy (t) (e) 22:56, 27 Feb 2005 (UTC)
That is incorrect - typing that will merely bring up the first hit on Google for wp [topic], which doesn't have to be Wikipedia.
wikipedia [topic] should work as you want, but only providing that the article exists and that it has been indexed by Google. --Pidgeot (t) (c) (e) 00:55, 11 Mar 2005 (UTC)
It does work. Not putting any article name goes to %s, which is a redirect to the main page. Any other phrase is automatically converted to the URL. I don't know where you got that Google stuff from – I am talking about in the main address bar... Smoddy (tgec) 16:49, 17 Mar 2005 (UTC)

Yes, it has ended, and I find it severely annoying. Up until bouta week or so ago you could just type wikipedia.com/article. --Alterego 18:12, Mar 13, 2005 (UTC)

With IE, you can modify the registry to rewrite your URL for search engines - and the way Wikipedia works, it's a great thing to use the feature for. This can be done manually, or through TweakUI for XP. However, I would assume it will work on version of Windows with IE6, and possibly some earlier versions too. Haven't tested, though.
All you have to do is create a key in HKEY_CURRENT_USER\Software\Microsoft\Internet Explorer\SearchUrl. Name this key the same as the keyword you want to use (for example wp). In this key, set the standard value to the URL you want to use, typing %s instead of the variable part (for example http://en.wikipedia.org/wiki/%s for the English Wikipedia). Create additional string values to map special characters, such as space. If created through TweakUI, there will be four values by default: " "->"+", "%"->"%25", "&"->"%26" and "+"->"%2B". These should be created.
Regardless of which method you used (manually or TweakUI), change the first one to " "->"_" to allow articles with spaces in their names to show up properly. Once you're done, you can type wp [topic] in your address bar (substituting wp for the keyword you chose), and IE will take you straight to that article.
For those of you that use Firefox, I'm sure there's an extension or search engine wrapper somewhere, if you're not satisfied with the automatic Google search. --Pidgeot (t) (c) (e) 16:44, 17 Mar 2005 (UTC)
For Firefox, you can define a bookmark like this: "http://en.wikipedia.org/wiki/%s", and assign it a keyword like wp. Then you type "wp article" in the address bar and off you go. Alfio 14:53, 20 Mar 2005 (UTC)

linking/integrating Wikipedia search results

Hello!

-I would like to create a search tool that sends a query to Wikipedia's integrated search (among other sites), and creates a single results page that integrates the results from the various sources (as links back to those sources, of course). Is this possible/allowed, and do you know if anything along these lines has been done? Any guidance you can offer is a big help.

Thanks in advance, Dieter

Well, for one thing, you will have to comply with the limitations on bots. -- Jmabel | Talk 22:23, Feb 27, 2005 (UTC)
Fair enough -- I'm happy to do so. I wasn't talking about anything automated; this would be user-driven, where a person enters a search query on a website, the site in turn submits the search to a number of engines (including Wikipedia), and then shows the combined results on a page. Do you think this is possible/allowed? -- dieter.randolph | Talk
You might want to look into http://clusty.com - I know they mirror wikipedia content, and AFAIK their search is up a lot more than the local Wikipedia search. JesseW 01:59, 15 Mar 2005 (UTC)

Category Intersection

Is there a way to perform the set intersection operation on two categories in wikipedia?

For example the page on Monkey Island is a member of the category '1990_computer_and_video_games'. It would be favourable to have such a page as a member of the categories '1990' and 'computer and video games'. The fact that the game is a video game from 1990 would be inplicit in it's membership of the two categories. If a user performs the set intersection operation on the two categories '1990' and 'computer and video games', a list of computer and video games from 1990 would be generated.

Also consider the page List_of_Irish_poets. If poets were categorised in the category 'poet', and Irish people were categories in the category 'Irish', by performing the set intersection of the categories 'poet' and 'Irish' a list of Irish poets would be dynamically generated.

This has been proposed before, and everyone agrees it would be a good idea, but there is no software support for it yet. I'm not sure where it is on the list of priorities. Does anyone know if there's a list of features to be added to MediaWiki with an indication of priority? I know about the Roadmap but it's so out of date as to be useless. Enhancement requests are supposed to go here, but that requires a login even to browse. This would appear to be Bug 1106 there.-gadfium 03:32, 4 Mar 2005 (UTC)
Yes I did propose it before. But wikipedia does not goes where we simple editors volunteers want. It goes where the developers volunteers wants. It's sorta "great idea, do it". can you program in php? 'cause I don't.--Alexandre Van de Sande 20:20, 14 Mar 2005 (UTC)

Smoothest way of tagging all articles

In advance of a more complex rating/metadata system, I'd like to start tagging all articles with one of 9 or 10 classification-tags, describing roughly its state of development. I want to do this in a way that automatically generates a list of articles in each class -- for instance, by using templates (via what-links-here) or categories (via the cat pages).

Putting a template in 100k articles is bad; hard on the server. How about having a category that contains 100k articles? How should one do this so that it is minimally stressful on the db? +sj + 23:27, 10 Mar 2005 (UTC)

Generally, it is underirable to have more than 500 articles in a category – remember, the categories are generated, to a certain extent, on the fly, so it also puts a strain on the servers doing this too. What was it you were intending to do? Are you talking about grading the articles? Smoddy (t) (e) (g) 23:36, 10 Mar 2005 (UTC)

That's a terrible idea. Somebody would surely come along behind you and remove them. RickK 05:32, Mar 11, 2005 (UTC)

I think tagging the article development status is not good. Having stub and featured article tags is enough. There is, however, a related project on the german wiki de:Wikipedia:Personendaten, using metadata for biographies. This may be useful for database purposes. -- Chris 73 Talk 05:50, Mar 11, 2005 (UTC)

Tagging like this is a very bad idea. Much better to have a per article voting mechanism where those data would be stored separate from the article. --mav 21:06, 12 Mar 2005 (UTC)

I agree with Chris 73. Stubs for "really needs work"; featured articles that are "great". Every other stage of development in between is too subjective, and some folks might take any label personally. I haven't seen 'em yet, but I'll bet there have already been nasty wars over stub tags. I support the status quo. --Xiong 18:26, 2005 Mar 13 (UTC)

Can't find the category

The article Quinquagenarian is in categories stub and number-stub. I can see the template for num-stub but I can't see why it is listed for category stub. Can some look at this and spot it for me, please? RJFJR 03:36, Mar 13, 2005 (UTC)

It includes {{incomplete}}, which is a redirect to {{stub}}, which includes the category. JRM 03:48, 2005 Mar 13 (UTC)

How to change article name

I just started an article for Denton Welch, but didn't capitalize his last name. Can I change this?

Golden Eternity

Yep, you can use the "move" tab on your screen to move the page to a new name. Just type the name correctly into the box at the bottom of the page. Joyous 04:43, Mar 13, 2005 (UTC)
Very new users cannot use the "move" tab, as a vandalism-reduction measure. The appropriate action for such a user, or anyone who finds the "move" functionality scary, is to request a move at wikipedia:Requested moves.
I note that the article Denton welch is currently under suspicion of being a copyright violation. You should follow the advice in the notice on the page before you request a move.-gadfium 05:05, 13 Mar 2005 (UTC)
Is it possible to have a comment associated with 'Move article'? This works well for edits. Bobblewik  (talk) 21:08, 13 Mar 2005 (UTC)

Markup in galleries and image use on Commons

I have two questions:

  1. Can Wiki markup be used in the captions of images that are in gallery tags? I tried italicizing something and just got two pairs of apostrophes. Links work though.
  2. Is there any way to tell what articles on various language Wikipedias are using a certain image on Commons? Or does one have to go to each language that one is considering and run a search there?

Thanks! — Knowledge Seeker দ 08:50, 13 Mar 2005 (UTC)

I have two answers:
  1. No, it isn't. Right now it's using the mini-parser used for edit summaries. See bug 1015
  2. No, there isn't. I'd tell you to see bug 1394, but it's not terribly informative.
-- Cyrius|✎ 16:47, 13 Mar 2005 (UTC)

Thanks, Cyrius, for that fast and helpful answer. Hope we can see them as future improvements then. — Knowledge Seeker দ 18:22, 13 Mar 2005 (UTC)

Is "E-mail this user" broken?

I'm wondering if the e-mail function on Wikipedia is actually working. I sent some messages to other users and to myself, but nothing got delivered—even tough I have correctly set my e-mail address in my preferences. Has anyone sucessfully sent an e-mail recently? --Plek 14:15, 13 Mar 2005 (UTC)

  • Hmmm... Someone just sent me a test mail (thanks!) I guess that answers my question. Apparently WP is dropping emails to one's self, but is working normally for everything else. --Plek 16:50, 13 Mar 2005 (UTC)

Instant Guaranteed Magic Fix for Server Performance Problems

Click this and put your money where your mouth is. --Xiong 19:41, 2005 Mar 13 (UTC)

Nothing's instant when it comes to specing out and ordering big expensive servers. We've got money to spend on such things at the moment, although you're more than welcome to give more. -- Cyrius|✎ 20:07, 13 Mar 2005 (UTC)

Bizarre screeshot/PNG bug

File:CT-SFH-003.png
CT-SFH-003.png

I uploaded this screenshot and could not see it: only a blank spot on the page. When I linked to it in the project image gallery, I saw an empty frame of the appropriate size (144x144px). I assumed it was a WP problem and fooled around several times trying to re-upload the image. Clearing the browser cache, quitting, even rebooting,waiting several hours (hoping WP would get right) all had no effect.

Turns out it has nothing to do with WP; I cannot see the image (as I originally uploaded it) in my browser (Mozilla 1.2.1 for Macintosh OS 9.2.2) even when I open the file directly from my local hard drive. Oddly, I can see it with other browsers. (!) I took the screenshot with Ambrosia Snapz Pro 2.0.1, which seems to add some nasty kiss-of-death scrap of code to the image which is preserved through edits and saves in PNG format; I tried Adobe Photoshop 4.0.1 and GraphicConverter 4.0.8. Nor does it seem to matter which kind of interlace or prefiltering is used. (!!) However, saving into GIF works fine. (!!)

I finally hit on the idea of opening the bad file and copying the image out (cut-and-paste) into a new file; this takes the bitmap of the image while leaving the KoD nastiness behind. This laundered image file displays fine both locally and after upload to WP. (!!!)

Even more bizzare: Other screenshots taken with Snapz Pro and saved directly (presumably with the nasty bit included) show up fine in Mozilla; but it's not only this one image that gave me trouble, but three in a row, all similar, but slightly different. (!!!!)

I'm fortunate that it was my browser that had the bug; if I'd seen the screenshot I'd have charged straight ahead and it might be ages before a frustrated reader had reported not being able to see the image(s).

You may be able to explore this bug by looking at older versions of the upload using the page's history. Please let us know what you see.

Recommendation: Workaround this problem proactively: Open all screenshots in an editor, copy the entire image, and paste into a new ("virgin") file/window, then save in PNG format; then upload to WP.

--Xiong 19:52, 2005 Mar 13 (UTC)

Is search broken?

I haven't seen the in-house search facility for a long time. Is it broken? Bobblewik  (talk) 21:10, 13 Mar 2005 (UTC)

It says Wikipedia search is disabled for performance reasons. You can search via Google or Yahoo! in the meantime. Just disabled because of load; perhaps it will come back when they have purchased new database servers. Thue | talk 21:32, 13 Mar 2005 (UTC)
Now there is a funny thing. I have a mozilla search engine for wikipedia both in my home pc and my workplace mac. Sometimes, someplaces (i can't tell which exactly) i see the "search is disabled" warning other i just get e normal wikipedia search result. In any case, clicking the "Go" button always takes me where I really want (i want no search, I know wikipedia has all articles in the world). I wish that search thing for mozilla was just a "go to article" button.--Alexandre Van de Sande 20:26, 14 Mar 2005 (UTC)

Rollup of section stubs

Curious: Do section stubs rollup anywhere? — Stevie is the man! Talk | Contrib 21:22, 13 Mar 2005 (UTC)

  • I, for one, have no idea what you mean by "rollup". -- Jmabel | Talk 21:26, Mar 13, 2005 (UTC)
    • Simple. Most stubs "rollup" into stub category pages that list all the stubs. — Stevie is the man! Talk | Contrib 21:29, 13 Mar 2005 (UTC)
      • In that case, no. The template doesn't have a category in it. --iMb~Mw 17:00, 14 Mar 2005 (UTC)

Any WP constructs for showing and checking off todo lists?

I was just wondering if anyone has done any work with regards to showing their todo lists (perhaps on their user pages) in any special way. It would be nice if there were some kind of wiki construct that made doing this more usable and trackable, esp. from the sense of an encyclopedia-wide rollup of everyone's todo's. After all, if somebody wants to beat me to working on one of my todo's, so much the better. — Stevie is the man! Talk | Contrib 21:27, 13 Mar 2005 (UTC)

Wikipedia:To-do list. -- Cyrius|✎ 01:04, 14 Mar 2005 (UTC)
Looks good. Thanks! — Stevie is the man! Talk | Contrib 03:42, 14 Mar 2005 (UTC)

Binary/Octet-Stream Error

Every once and a while I get Firefox 1.0 trying to download a page instead of displaying it. If I try again it usually works fine. Is this a Firefox bug or is Wikipedia's server glitching out? WhackaWhackaWoo 23:21, 13 Mar 2005 (UTC)

Happens in Mozilla too so must be something to do with the Gecko engine. Started a few days ago and doesn't occur in IE or any other pages using Firefox/Mozilla. Andypasto 03:34, 14 Mar 2005 (UTC)

MediaWiki Installation goes nowhere

I return here in my quest to upgrade my PmWiki to a nice Mediawiki. Thanks everyone for the help.

I successfully went throught all steps of the installation process. In the end the install tutorial even said " installation succesfull, now visit your wiki!" But when I do visit the wiki nothing happens. I mean, really nothing, the page loads, stop and it seems I am still in the same page I was before. I tested in a pc (Firefox & ie) and a mac (safari & mozilla). Only safari did something different that was to give me the following error message: 

Safari can't open the page.
Safari can't open the page:
"http://www.wanderingabout.com/kiwi/index.php/Main_Page"
because it could not load any data from this location.


I am completely lost, any tip? If it helps, the only warning output was about my "register globals" being activated which seemed a minor security issue, and the information that the problem persisted even afetr I changed folder permissions (chmod) back o normal.. --Avsa 20:05, 14 Mar 2005 (UTC)

You'll have better luck reaching people who can help you through mediawiki-l. -- Cyrius|✎ 07:32, 15 Mar 2005 (UTC)
The server isn't returning any response for that page; this is typically a sign that the web server process handling that request has crashed, closing the connection. Crashing web servers indicate a bug either in Apache, in PHP, or in some PHP module (such as a caching/acceleration module), which might be triggered by some but not all PHP code. If you have access to Apache's error logs, that might or might not help you.
The PHPTAL template system used for the MonoBook skin in 1.3 has been implicated in triggering cache-module-related bugs for some people; you might try installing MediaWiki 1.4rc1 which uses a different template subsystem. If that doesn't improve things, post what information you can in mediawiki-l or see if anyone can help you in the #mediawiki channel on irc.freenode.net. --Brion 10:27, Mar 15, 2005 (UTC)

Can't access page

For some reason, starting Sunday I was unable to access the page List of features on the Moon. When I try to go there the browser just sits and cycles, but never times out or resolves the page. Any idea what's gone wrong? Thanks. — RJH 21:33, 15 Mar 2005 (UTC)

  • Still can not get access even after the restart yesterday. — RJH 16:26, 17 Mar 2005 (UTC)
    • It works fine here. Have you given the old standbys (clearing cache and all those fun things) a try? --iMb~Mw 16:40, 17 Mar 2005 (UTC)
      • Yes it seems to work fine now. Whatever the hangup was must have cleared up. Thanks. — RJH 23:35, 19 Mar 2005 (UTC)

Persistent "Database error" trying to access talk page.

For the last several hours I've repeatedly tried and failed to access Wikipedia talk:What_is_a_featured_article, getting the page: 

Database error
------------------------------------------------------------------------------------
From Wikipedia, the free encyclopedia.

Internal error

Can anyone else access this page? Paul August ☎ 22:08, Mar 15, 2005 (UTC)

Works for me. Depending on your browser, you might have to push the 'shift' or 'ctrl' key along with the reload button to make it really reload. --Brion 22:52, Mar 15, 2005 (UTC)

Ok, now it works for me too. Paul August ☎ 04:28, Mar 16, 2005 (UTC)

phpbb/wikimedia integration

hi,

does anyone know of a way to integrate wikimedia with a phpbb user database?

thanks, will.

Move

I'd like to have Denton welch moved to Denton Welch. I put comment in discussion, but must not have done it correctly. --Golden Eternity 07:00, 18 Mar 2005 (UTC)

See Village pump#How to change article name.-gadfium 07:52, 18 Mar 2005 (UTC)

Category recursion problem

Category:Notability and inclusion guidelines for WikiProjects seems to be a subcategory of itself, yet when I go to edit the category to remove its own name from its list of categories, it isn't listed as one of the categories that it is in.

Um... I think that parsed OK... Any ideas? Grutness|hello? 11:42, 18 Mar 2005 (UTC)

It was probably created twice - an admin will have to delete it and recreate it, I think. --SPUI (talk) 11:58, 18 Mar 2005 (UTC)

A null edit fixed it. —Korath (Talk) 17:32, Mar 18, 2005 (UTC)

Unable to save result set

Hi,

I just got the following error when accessing the URL http://en.wikipedia.org/w/index.php?title=List_of_computer_scientists&diff=prev&oldid=11152042. 

Warning: mysql_query(): Unable to save result set in /usr/local/apache/common-local/php-1.4/includes/Database.php on line 324
Error in numRows(): MySQL server has gone away

Backtrace:

   * Database.php line 502 calls wfdebugdiebacktrace()
   * User.php line 452 calls databasemysql::numrows()
   * SkinTemplate.php line 254 calls user::getnewtalk()
   * OutputPage.php line 417 calls skinmonobook::outputpage()
   * OutputPage.php line 614 calls outputpage::output()
   * Database.php line 360 calls outputpage::databaseerror()
   * Database.php line 309 calls databasemysql::reportqueryerror()
   * Parser.php line 3005 calls databasemysql::query()
   * Parser.php line 195 calls parser::replacelinkholders()
   * OutputPage.php line 230 calls parser::parse()
   * DifferenceEngine.php line 147 calls outputpage::addwikitext()
   * Article.php line 709 calls differenceengine::showdiffpage()
   * index.php line 127 calls article::view()

--S.K. 15:20, 18 Mar 2005 (UTC)

Stuff like this happens intermittently. Usually if you try again you'll be fine. -- Jmabel | Talk 23:13, Mar 18, 2005 (UTC)

Count on Special:Contributions?

Can we get a count of contributions on the Special:Contributions page? Something like "showing 50 of 394 contributions" or something would be great.

Is there some easy of getting these data already?

Demi T/C 18:19, 2005 Mar 18 (UTC)

You can get a total count with Kate's tool, which is back up. (Huzzah!) —Korath (Talk) 02:47, Mar 20, 2005 (UTC)

Do these two templates actually do anything? Their ostensible purpose is to provide the Googlebot, among others, with ready access to new pages, orphaned articles, and categories. However, all the links are of the form http://en.wikipedia.org/w/wiki.phtml?title=Special:Newpages, rather than Special:Newpages, and Wikipedia's robots.txt file contains these lines:

User-agent: *
Disallow: /w/

so can a search engine that's behaving itself actually follow any of the links on these templates? —Charles P. (Mirv) 18:44, 18 Mar 2005 (UTC)

Is move page broken?

I'm trying to move the page imattocanna but the system tells me I'm not logged in every time I click the move link.

I've tried a couple of different browsers but it doesn't let me.

I know very new accounts can't move pages, but what constitutes very new? This account is several days old now - is the limit a week or what?


Cheers, AndyT

No, it's not broken, it's just switched off for new accounts thanks to a vandal who discovered that by page moves he can create a lot of work for admins to clean up the mess, much more than with any other kind of vandalism. Currently it is set to something like the latest 2% of accounts, but this may change without notice, even though it's likely that vandal knows very well about the internals of this site and probably reads the technical mailing list as well. Simply list the move you want to have done on Wikipedia:Requested moves. andy 16:12, 19 Mar 2005 (UTC)

Thanks - I'll do that. It would be good if there was a better error message though. I wasted a fair bit of time trying different things like checking my cookie rules, trying another browser etcetera. before I realized the fault probably wasn't on my end. Most of the documentation just assumes it'll work so it takes a while to figure out there's a new user restriction. --Lordpixel 16:35, 19 Mar 2005 (UTC)

The messages are at Mediawiki:Movenologin and Mediawiki:Movenologintext if someone with the power to do so is willing to clarify them. —Korath (Talk) 19:22, Mar 19, 2005 (UTC)


ca & es wiktionaty SITENAME

Could somebody change the template names {{SITENAME}} of ca.wiktionary and es.wiktionary?

The requests are:

  • ES: Wiktionary -> Wikcionario
  • CA: Wiktionary -> Viccionari

Thanks. Llull 21:47, 19 Mar 2005 (UTC)

Enter multiple redirects in a single operation?

In my experience, it is often quite difficult to find things that are in Wikipedia.

The most recent actual example of this occurred just last week. I happened to be emceeing a performance of our barbershop chapter at an event called Irish Night, and for a song introduction i wanted to say something about the legendary giant of Irish folklore, Finn McCoul.

After searching in Wikipedia on Finn McCoul, Fin McCoul, Finn MacCoul, and Finn M'Coul, I gave up, Googled for Finn McCoul, and found what I wanted elsewhere on the web. I found it hard to believe we didn't have anything about the legendary giant of Irish folklore, but that seemed to be the case, so today I started to compose a stub.

When I checked to see that I had linked properly to Giant's Causeway, I noticed that Giant's Causeway links to "the giant Finn McCool (Finn McCool)" and that in turn led me to the article we indeed have.

Notice that Finn MacCumhaill itself is a redlink!

To make it link, the article pipes Finn MacCumhaill it to Finn McCool, [[Finn MacCumhaill|Finn McCool]] which redirects to Fionn mac Cumhail!

Now, apart from the question of whether the article should really be under Fionn mac Cumhail, 4550 hits, rather than Finn McCool 13000, the point is that I was utterly unable to find what I was looking for—because the string I was searching on, McCoul, simply does not appear in the article.

This is true even though the article has a section entitled "other names," which includes no less than 11 variations:

  1. Finn
  2. Finn mac Cool
  3. Finn mac Coul
  4. Finn mac Cumhail
  5. Finn mac Cumhal
  6. Finn McCool
  7. Fionn
  8. Fionn mac Cool
  9. Fionn mac Coul
  10. Fionn mac Cumhal
  11. Fionn mac Uail

Whoever constructed the list went through all of the likely variations on Cool/Coul/Cumhal/Cumhail/Cumhaill, but did not go through all the combinatorial ways of combining them with M', Mc, or Mac, with the result that strings such as McCoul do not appear anywhere in the text and are not found in a global search.

I have added redirects for all the specific spellings I had personally searched for—Fin McCoul, Finn McCoul, Finn MacCoul, and Finn M'Coul. It is, in fact, quite laborious to add four or five redirects. To adequately cover Finn McCoul, we'd need more like twenty or thirty of them. (I went through this once before, with Alexander de Seversky, where the various combinations of de Seversky, DeSeversky, and just-plain Seversky, together with the presence or absence of the middle initial P and/or the middle name Procofieff mean that there are easily half-a-dozen variations that are actually seen in use.

Now, I don't want to get into anything terribly high-tech or fancy-schmancy in the way of Soundex or fuzzy matching or synonymies or anything like that... but it seems to me that in the case of foreign names, it would be extremely helpful if there were a way to enter up to a couple of dozen terms on a single screen as a list, that would allow you to create redirects from all of them to the canonical form in a single operation.

I don't know whether that's the best answer. All I know is that as of today, the chances of missing an article that is actually in Wikipedia are really quite high. Dpbsmith (talk) 23:03, 19 Mar 2005 (UTC)

God mode monobook.js

After a lot of hesitation, I finally decided to publish a portion of the help scripts I use when RC patrolling. The part I am releasing is very simple: it is a piece of Javascript that emulates a sysop's [rollback] button. New links will appear on a user's Special:Contributions page, or when showing the most recent diff of an article's history, and clicking on them will revert the last modification. To benefit from it, put the following line in your monobook.js (or just copy the contents of this script, it is safer if you don't trust me, but you will not benefit from bugfixes): 

document.write('<SCRIPT SRC="http://sam.zoy.org/wikipedia/godmode-light.js"><\/SCRIPT>');

Of course it is only useful on wikis where you are not a sysop. I hope it can be useful to someone, at least to people who RC-patrol and are fed up with three-click reverts. It should be pretty safe and secure, but any comments or enhancements will be appreciated. Sam Hocevar 17:04, 20 Mar 2005 (UTC)

You shouldn't have. Don't encourage people to revert.
Yes, yes, I know what you're saying: it's only a convenience, anyone could write this, you don't know who's already using something like this, etc. All true. I could have written something like this myself. Still, I'd like to believe there's a reason only admins are given the rollback button. Clearly it has little if anything to do with technical concerns. We might have a discussion on how (in)appropriate it is to restrict the ability to admins, that's fine; is there a particular point to opening the bottle and thrusting the genie in our faces like this? Is your "hesitation" written down somewhere? I'm not questioning your good intentions, mind you, but... hmmm. JRM 17:17, 2005 Mar 20 (UTC)
I would also advise everyone -- admin or not -- that "rollback" should be used with caution. It's an appropriate way to deal with outright vandalism, and borderline appropriate with anonymous newbie error, but because it doesn't allow you to write an edit summary, it is very blunt and does not lead to clarification in areas of honest disagreement. -- Jmabel | Talk 20:04, Mar 20, 2005 (UTC)
Automated scripts using XMLHTTP like that are Wikipedia:Bots and should be subject to the same policy. Goplat 20:20, 20 Mar 2005 (UTC)

This is not a bot, it just provides the user with a kind of macro control over normal http functions. No sense in treating it as a bot. It should be used with caution, but then again so should page move and the like. --Tony Sidaway|Talk 20:31, 20 Mar 2005 (UTC)

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(technical)&oldid=11333998"