Misplaced Pages

10/12/2006 1:01AM: I managed to control how much text floats next to
the table, by using the <BR CLEAR=all> tag I read about here:
http://www.htmlhelp.com/reference/html40/tables/table.html
  The deprecated ALIGN attribute suggests the horizontal alignment of
  the table on visual browsers. Possible values are left, right, and
  center. Browsers generally present left- or right-aligned tables as
  floating tables, with the content following the TABLE flowing around
  it. To prevent content from flowing around the table, use <BR
  CLEAR=all> after the end of the TABLE.

Now that I know what to look for, I can find it with this search on Meta but not with this search on mediawiki.org. --Teratornis 03:37, 5 April 2007 (UTC)

tags using "header=1" are screwing up the edit section links

See User talk:Cthorntonjr for an example of what I mean. I used {{subst:nonsensepages|Bud Brothers|header=1}} there, and it leaves behind a section header like this: {{#if:1|=={{{header-text|]}}}==}} ... when clicking on the section edit, it won't take you to that section. I suspect Mediawiki doesn't parse the #if statement in a section header properly. This is a problem of all speedy notification tags that I've encountered so far that use the "header=1" parameter, and possibly others besides CSD notifications. — coelacan — 03:29, 2 April 2007 (UTC)

I believe this could be fixed by changing the innards of the templates to have a couple of <includeonly>subst:</includeonly> tags. I think this would work:

{{<includeonly>subst:</includeonly>#if:{{{header|}}}
| =={{<includeonly>subst:</includeonly>#if:{{{header-text|}}}|{{{header-text}}}|]}}==
}}

It's pretty ugly, but the result should be that only the wikitext for the header ends up in the page, without the parser functions. Mike Dillon 03:37, 2 April 2007 (UTC)

I see that the templates had these, but ais523 removed them. I'll ask that user to take a second look. — coelacan — 04:53, 2 April 2007 (UTC)

The problem is that with the old coding, the header appeared even if header=1 wasn't used. I'm not sure that I know a solution to this; ParserFunction/subst mixtures have always acted unusually in my experience. If you can make the change and it makes it possible to use the template without the header, please feel free to do so. (I was unaware that the new coding had a problem.) --ais523 13:47, 2 April 2007 (UTC)

Gallery display issues?

I haven't done any sort of bug reporting yet, mostly because I haven't heard of anyone else having the same problem. Many (not all, though close) of the pages I've gone to won't display images if they're part of a gallery. Sometimes a few thumbnails are shown, sometimes all but one or two, sometimes none, and the number in the gallery doesn't seem to have any effect on how many are(n't) shown. When the images are just regular thumbnails within the ] tags, they're ok (though I believe there've been one or two times when those won't display, either), as are images within infoboxes. I've tried purging and refreshing, but it doesn't help any. A few times it's even resulted in fewer pictures being displayed. Between all this, I'm seriously wondering if it's an issue with Wiki (in general, this issue isn't contained to Misplaced Pages), or with my computer (which is also why I haven't sent any bug reports). Anyone have any ideas what could be causing the problem, or how I can fix it? -Bbik 22:26, 14 April 2007 (UTC)

Which images are they? I believe this note from the top of this page applies:

If an image thumbnail is not showing, try purging its image description page (if the image is from Wikimedia Commons, you might have to purge there too). If it doesn't work, try again.

If you can point out an image you're having an issue with, we can see if the issue affects just you or others as well. Hope that helps. Mike Dillon 23:22, 14 April 2007 (UTC)

That line (among others) is exactly why I specified that I'd tried purging already. However, as for specific ones I'm having issues with, most recently it's this page, and the specific pictures are (working across, starting at the first one after the panoramas):

• Small Town: 1, 4, 5, 8, 10, 15
• Big Town: 5-9

I can find other pages I've had issues with too, if it'll help. If the numbering doesn't make sense, let me know and I'll copy the specific pages from the edit box. -Bbik 23:52, 14 April 2007 (UTC)

Just wondering if anyone has any ideas on this, since it's moved quite so far up the list by now, and no response to the listed pictures, either... -Bbik 07:54, 16 April 2007 (UTC)

So, should I assume no one has any idea about this, and I should just submit it as a bug, or? -Bbik 04:31, 20 April 2007 (UTC)

Sorry I lost track of this discussion. I'm not seeing any problems with the page you pointed to. Mike Dillon 04:52, 20 April 2007 (UTC)

Ok. I'll blame it on my computer for now and see if it's still a problem once I get it fixed (hopefully in the next few days), and then submit it as a bug if it is. Thanks anyhow. -Bbik 23:27, 20 April 2007 (UTC)

Question

Does anybody know why the indent wikimarkup in text ("::...:" or "**...*") doesn't work when there's a left-aligned pic? Is there a bug listed? NikoSilver 20:51, 15 April 2007 (UTC)

Can someone spare a few seconds to help me out with this please? NikoSilver 15:07, 16 April 2007 (UTC)

If you mean this:

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

It seems to be because the <dl> and <dd> don't clear the image div on either side, but the indentation is compromised only on the left side (probably to let it wrap naturally. Breaking it down and adding borders, we see:

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah

HTH. --Splarka (rant) 07:31, 17 April 2007 (UTC)

Thanks! I was supposing that this may be the case, but had no way of illustrating it as well as you did just now with the borders and all! It seems that the images displace bona-fide text, but they don't displace the indents as well, as evident if you continue your example sequence of indents (until they exceed the image width, which they do!):

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

Blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah.

So, shouldn't the wiki-markup (::: and ***) work around this html defect and force (somehow) the indents to be displaced as well (bypassing the html code in the resence of a pic)? I've worked on many articles with pics, and this is the most common problem that editors frequently face in format. To tell you the truth, I haven't noticed even one example, where the indent (if included) was actually intended to be omitted (why on earth would it be included in the first place then?). Shall we list a bug, or is there one already listed? NikoSilver 15:07, 17 April 2007 (UTC)

This is an interesting format issue here. I would like to ask again for input. Please don't be intimidated by the size of the section to participate, it's just examples of repetitive "blah"s to show the format issues. Should we list a bug? Is one already listed? NikoSilver 08:47, 19 April 2007 (UTC)

The problem is that (as the colored example above shows) floats do not, by default, displace the block boxes, only the inline boxes within them. Since the indents are nothing more than large margins on the block boxes, they are not affected by the floats. I knew I had seen something like that before, so I took about half an hour looking for it on my contributions (while waiting for the Kubuntu updates to download); it's here, and shows float-displace and/or -moz-float-edge are CSS properties which might help. In particular, float-displace: indent looks like it's exactly what you are looking for (and probably isn't implemented in any browser you might think of). --cesarb 03:19, 21 April 2007 (UTC)

Thank you very much for your time and for the explanation. This is too technical for me to understand, but it's nice to notice people have dealt with this before. I hope sometime a solution will be implemented. NikoSilver 19:42, 22 April 2007 (UTC)

Current recentchanges numbers only show net change?

For instance a green +1000 means that 1000 chars were added to the article, right? So let's take this, for example:

"I like pie."

If someone vandalized it into:

"I like egg."

It would be recorded as a zero change. Is there a way to change it so that it shows not just the net change, but the change components individually? That is, in the above example edit, it would show like this:

(diff) (hist) . . Pie‎; 03:15 . . (-3)(+3) . . Kirbytime (Talk | contribs)

Hmm?--Ķĩřβȳ♥♥♥ŤįɱéØ 03:16, 16 April 2007 (UTC)

Is this significant enough to create confusion the other relatively new number? Do vandals ever count the number of characters in an article, then perform ROT13 to keep the overall change at (0)? –Pomte 03:22, 16 April 2007 (UTC)

• I haven't seen it yet, but if that happens, we should just start looking at changes with net 0 character difference. - Mgm| 11:01, 17 April 2007 (UTC)

That would be really helpful. In RC patrol I often encounter 0 changes that are vandalism of the kind that is hard to spot. ≈ jossi ≈ (talk) 15:38, 17 April 2007 (UTC)

It's not just 0 change, it could be a major (±1000) change, but off by maybe ±20, so while over 1000 bits where changed, maybe 980 bits were added, so the change number won't show up as bolded in the recent changes.--Kirby♥time 06:01, 19 April 2007 (UTC)

IIRC, finding the minimal editing distance between any two strings (read articles) where adding, deleting, replacing, and moving are possible operations is very hard in terms of computational complexity. However, one could instead find approximate edit distances based on the "diffs" generated (which, btw, are not always optimal) between sequential edits and that might be useful enough. (I am not part of the technical crew of Misplaced Pages; I'm just making some comments.) Root(one) 20:05, 20 April 2007 (UTC)

Collapsing a lengthy caption?

Is it possible to use a script of some sort to hide a specific lengthy caption until the reader clicks "show caption"? This is suggested for the long caption at Atheism#Rationale. Something like what is used in {{hat}}, for example? Thanks! — BRIAN0918 • 2007-04-17 00:45Z

engineer

thank you for your time and answers, I may yet become an engineer.

SVG problems on my local copy of MediaWiki

File:Sunstar-wiki1.jpg

I got SVG working on my local copy of MediaWiki, but it doesn't quite render as well as I thought.

This is my template:

However, the SVG image of the stop hand has a white border, which it doesn't on Misplaced Pages - I'm using ImageMagick as the renderer. If I used Inkscape would that work better??

All advice is appreciated. Thanks, --SunStar Net 20:19, 18 April 2007 (UTC)

If it looks okay in Firefox but not MSIE, it might be a transparency issue. Apparently there's a javascript hack somewhere that's enabled on Misplaced Pages? --Interiot 06:38, 19 April 2007 (UTC)

• The SVG does not work in Firefox, the screenshot above is from Firefox. It's still rendering a white background with SVGs, doesn't do that with PNGs. Should I change from Imagemagick to Inkscape?? --SunStar Net 09:22, 19 April 2007 (UTC)

Misplaced Pages uses a modified version of librsvg. You might try running it instead of Imagemagick. I don't have any experience using Inkscape in batch mode, only as a vector image editor. Mike Dillon 02:53, 20 April 2007 (UTC)

• I'm using Windows XP to run my WAMP Apache server... how do I install RSVG on Windows XP to get it to render the SVG's correctly?? --SunStar Net 09:26, 20 April 2007 (UTC)

I have no idea. It looks like you can download it here, but I have no idea what's involved. You might find this page helpful as well. Mike Dillon 14:46, 20 April 2007 (UTC)

Recent changes to unwatched articles

Would it be possible to add an option to restrict Special:Recentchanges to articles that only a few (maybe less than twenty) people have on their watchlists? --Derlay 22:03, 18 April 2007 (UTC)

I don't believe so; the system doesn't store who has what watched like that (it's all tied through the account, not the article). EVula // talk // ☯ // 02:57, 19 April 2007 (UTC)

Admins have access to Special:Unwatchedpages for pages with 0 watchers, but a recent-changes for it would probably be too taxing on the database without some serious reprogramming. --ais523 15:13, 20 April 2007 (UTC)

How about this: every time a watchlist is viewed, update a timestamp on every page on the list, and filter Recentchanges by that timestamp. That should be feasible to implement, and actually better than a watchlist count, since watchlists of inactive users can't affect the result. --Derlay 23:45, 22 April 2007 (UTC)

That is a good idea Derlay. _InBC 00:13, 23 April 2007 (UTC)

The spell-checker SUCKS!

Why can't it be as good as Google?

There is a spell-checker in MediaWiki but it is disabled for performance reasons. x42bn6 Talk 01:17, 19 April 2007 (UTC)

No there isn't. --brion 14:43, 19 April 2007 (UTC)

You're probably seeing a spell-checker in your local browser. If so, you should be able to add new words to its dictionary when it's lacking. --Interiot 06:39, 19 April 2007 (UTC)

Because Google has billions of dollars of profits to spend on programmers to build a spell checker. Misplaced Pages has volunteers who do it in their spare time. Corvus cornix 18:50, 19 April 2007 (UTC)

Just to be clear: I'm talking about when you type a word in the search for an article and you misspell it--the results SUCK. Google picks up on your errors extremely well and usually knows what you meant.

Well, you can search Misplaced Pages with Google easily enough. --Teratornis 21:43, 24 April 2007 (UTC)

Also, you can create redirects for any terms you misspell; that helps out anyone else who searches for the same typo. EVula // talk // ☯ // 21:56, 24 April 2007 (UTC)

Software upgraded

After some delays, the database schema updates are complete and we've upgraded the software on the servers to the current MediaWiki code. There are a few cute little user-visible improvements, but mostly we've just got more bug fixes and things in. There might be a few new errors which crept in, too, so do give a shout about new oddities. --brion 13:49, 19 April 2007 (UTC)

Is there a way to remove the bytes from your visible edit summaries? Not really a fan of them. Quadzilla99 14:21, 19 April 2007 (UTC)

Having a quick peruse through the page source, I found the CSS class is history-size. Thus you could have something like

history-size{
   display: none;
}

I think I've got the right code. Harryboyles 14:29, 19 April 2007 (UTC)

The code should be .history-size{display: none;} with a dot. Alex Smotrov 15:42, 19 April 2007 (UTC)

• Sure is utterly annoying to have to go through an extra page to get user contributions. What was the point of that? Making vandal-fighting more tedious? --jpgordon 14:25, 19 April 2007 (UTC)
  • What do you mean? From what I can tell and from what it seems you're after, nothing's changed. Special:Contributions/Jpgordon still gets what you expect. Harryboyles 14:31, 19 April 2007 (UTC)
    • Yeah -- for a few clicks anyway, I'd get just the radio-button choices, without data; but it fixed itself somehow. Or maybe I was hallucinating, I dunno. (Still seems like unnecessary extra stuff; newbie=contribs-newbie should be its own page, and not clutter up the regular contributions page.) --jpgordon 14:35, 19 April 2007 (UTC)
      • It's in the page history now. This is what we're referring to (the bytes) see here: Quadzilla99 14:37, 19 April 2007 (UTC)
        • Wait, maybe I misunderstood you so. I just have to copy that to my css to turn off the bytes summary Harry? Am I understanding that correctly? Quadzilla99 14:39, 19 April 2007 (UTC)
          • Yeah copy it to your css file. Harryboyles 14:58, 19 April 2007 (UTC)
            • Ahhh, so much better. Thanks. Quadzilla99 02:02, 20 April 2007 (UTC)

Bug: WHOIS no longer works off the contributions page of an IP editor. CMummert · talk 14:36, 19 April 2007 (UTC)

Addition: It passes "Contributions" (the string) as the IP address to the whois pages. Harryboyles 15:01, 19 April 2007 (UTC)

This looks like a change in the behavior of {{PAGENAMEE}} on Special:Contributions. Apparently it used to contain the IP address or username. The IP address isn't passed to the wfMsg call, so it isn't possible to use "$1" or anything else in MediaWiki:Anontalkpagetext to fix this. Mike Dillon 15:10, 19 April 2007 (UTC)

WHOIS is still broken. The point is that all the query strings use 'Contributions' as the parameter. Shenme 18:36, 19 April 2007 (UTC)

Um. I don't think I contradicted that; I was just explaining why it is broken. All of the links in MediaWiki:Anontalkpagetext use {{PAGENAMEE}}. Mike Dillon 20:27, 19 April 2007 (UTC)

This appears to have been fixed. Maybe browser cache was to blame, who knows. Harryboyles 07:17, 21 April 2007 (UTC)

I'm interested in seeing two changes that I believe were announced but aren't in the deployed version:

1. Special:Contributions by IP subnet for /24 and /16 networks
2. Limiting of Special:Linksearch by namespace

Are there any plans to get these two things back into a future build? Mike Dillon 15:21, 19 April 2007 (UTC)

Brion ripped that out, seems it was too slow. Actually he took a lot of VoA's features out, including Special:Checkuser changes, this, ect. Prodego 20:32, 19 April 2007 (UTC)

Checkuser was not rolled-back. Only 1 function use was reverted (which was added to svn after the last site update), which was now re-added today. Voice-of-All 19:12, 21 April 2007 (UTC)

Is there some practical reason for lumping these two features together?

Not to mention it's completely redundant with Special:Contributions/newbies--VectorPotential 16:10, 19 April 2007 (UTC)

Agree. I don't see the point in this new feature either. --kingboyk 16:47, 19 April 2007 (UTC)

I presume it's useful for vandal fighting? - jc37 16:58, 19 April 2007 (UTC)

Special:Contributions/newbies is useful for vandal fighting, the above feature is just annoying and redundant--VectorPotential 17:00, 19 April 2007 (UTC)

And it isn't useful for vandal fighting, since new vandalisms are not showing up. Corvus cornix 18:39, 19 April 2007 (UTC)

It seems to have returned to normal ...for now--VectorPotential 20:13, 19 April 2007 (UTC)

And now it's back--VectorPotential 21:11, 22 April 2007 (UTC)

Not the end of the world but...

Made a typo fix on Self-injury and (i was logged on) it changed it all ok, but when i look at my watchlist i see a random ip number (which happens to be mine!)... i suppose it's a bug but dunno what happened, i made a edit to my user page, and my it appaeared as a contribution with my name...Would this have to do maybe with what the above user posted or? Ps is there any actual place to report bugs? Here i thought was most suitable? Later. petze 14:26, 19 April 2007 (UTC)

This might be fixable by just setting your browser to auto-log you in. I can't remember what's necessary, but I ran into the issue of me getting logged off when I had to reinstall Firefox. I fixed it, but I don't remember the fix except having Firefox remember my password and some other setting. Root(one) 20:12, 20 April 2007 (UTC)

Language problem

I'm Swedish and therefore, I use Swedish even on the English wiki (i.e. I don't write in Swedish but I've chosen that language under "My settings" for all the menus and stuff). At the top of Misplaced Pages, where we all have the links "My user page", "My discussion page", "My settings", "My watchlist", "My contributions" and "Log out", they are and have always been in Swedish for me, until now. They're all still in Swedish, except for "My watchlist" which is (starting today) for some odd reason in English. I have checked the "System messages" under "Special pages" and there, it still no longer says "Min övervakningslista" but "My watchlist" for me. Does anybody else, who doesn't have English in their language settings, exeperience the same problem and does anybody have a solution? /Ludde23 18:27, 19 April 2007 (UTC)

I only speak English, but after seeing this I got curious and switched my language setting to Swedish. It did the same thing to me. I don't know what is causing it, but it might be the recent upgrade. --LuigiManiac 18:33, 19 April 2007 (UTC)

This is indeed strange: in Italian it's "my watchlist", but in German it's not. Tizio 19:22, 19 April 2007 (UTC)

This is controlled by the MediaWiki:Watchlist system message. Since it has been customized on-wiki without customizing MediaWiki:Watchlist/sv, the Swedish version doesn't work. Since MediaWiki:Watchlist has the same value as the software default, deleting it should restore the cross-language functionality. I'm not sure about the difference between Italian and German, since neither one has been customized on-wiki. Mike Dillon 02:59, 20 April 2007 (UTC)

Hmm. It looks like MediaWiki:Mywatchlist may actually be the message that is used. From what I can tell, MediaWiki:Mywatchlist/de has a software-level default, but MediaWiki:Mywatchlist/sv and MediaWiki:Mywatchlist/it have the English phrase "My watchlist". I'd have to look at the source to see exactly what the logic is here, but it looks like it might be using MediaWiki:Mywatchlist and not MediaWiki:Watchlist. Mike Dillon 03:06, 20 April 2007 (UTC)

After looking at the source, it seems that it issue is that the language files define "watchlist" and "my-watchlist", but not "mywatchlist". I still don't exactly understand how this works, since SkinTemplate.php requests "my-watchlist", but I'm pretty sure that the missing keys for "mywatchlist" are the cause of the problem. The only language with definitions for "mywatchlsit" are zh_yue, bg, ar, zh_tw, de, en, fi, fr, he, and id. Mike Dillon 03:28, 20 April 2007 (UTC)

This was fixed by Brion in revision r21417, but it isn't live yet. Mike Dillon 15:17, 21 April 2007 (UTC)

Cascading protection not properly indicated

I seem to recall that when editing a page that was deleted protected via the cascading protection, there was some warning somewhere in the edit form about the page being protected (like the red message you got when editing a protected page), and telling which page caused the protection. Can someone confirm this behavior? I don't see that message any longer. Tizio 19:50, 19 April 2007 (UTC)

Non admins still get a message, but there is no longer a way for admins to tell if a deleted page is protected. Needs to be changed. Prodego 19:59, 19 April 2007 (UTC)

Yeah, that was weird. The "protect" tab doesn't even say "unprotect" and the only way to tell that something's protected via cascading protection is to go to the Protect Page to find out. Kind of annoying. howcheng {chat} 22:20, 19 April 2007 (UTC)

I just discovered this the hard way when I mistakenly believed that the main page's protection wasn't cascading. (I assumed that it had something to do with the incident earlier in the day.) I immediately manually protected all of the templates and images that weren't already protected. Then I removed and reinstated the cascade. I thought that this solved the problem, because the cascading protection message appeared when I went to edit the On this day section. (This was strange, because MediaWiki recently was changed to have the manual protection messages—including the semi-protection message, annoyingly enough—override the cascading protection message.) So I removed the manual protection, only to find that the message (and "unprotect" tab) had disappeared. I then logged out and found that despite the fact that the "edit this page" tab was present, the cascading protection actually was in place all along.

Has anyone filed a bug report yet? —David Levy 00:38, 20 April 2007 (UTC)

Yep, someone has. —David Levy 00:52, 20 April 2007 (UTC)

I just conversed with Voice of All via IRC. While he only cited a separate issue when closing the above bug report, it appears that most of the problems have been fixed in the most recent MediaWiki revision (coming soon).

It turns out that the "unprotect" and "view source" tabs have been intentionally disabled due to performance issues (the database queries are too stressful), but I suggested that the "unprotect" tab be replaced by a custom message generated via the addition of a database query performed only for sysop accounts (so it wouldn't be a significant server drain). He indicated that this should be possible, and he asked me to remind him about it after version 1.10 is released (because there's a code freeze until then). —David Levy 03:47, 24 April 2007 (UTC)

Bytes on history??

Anybody else notice that all recent edits in the history page now have the number of bytes? What's the use of this feature? -- _Hdt83 00:34, 20 April 2007 (UTC)

It's part of the most recent MediaWiki upgrade, and you can get rid of it by adding .history-size{display: none;} to your monobook.css--VectorPotential 00:41, 20 April 2007 (UTC)

Yes. And on some pages, all revisions have the number of bytes, and other only have a few or one. --TeckWiz is now R Contribs_@(Lets go Yankees!) 01:58, 20 April 2007 (UTC)

What is the meaning of this? Why did you have to add it? ↔ z (talk · contribs · autographs) 02:44:19, Friday, April 20, 2007 (UTC)

It looks like older edits made before the upgrade do not have the byte count. I think it was added for easier vandalism reverting. BANG! 02:57, 20 April 2007 (UTC)

This seems unnecessary on talk pages. Is it possible to hide it on discussion while still showing it for article histories, please? James S. 16:32, 20 April 2007 (UTC)

Yes. — Alex Smotrov 19:59, 20 April 2007 (UTC)

.ns-1 .history-size, .ns-3 .history-size, .ns-5 .history-size, .ns-7 .history-size,
.ns-9 .history-size, .ns-11 .history-size, .ns-13 .history-size, 
.ns-15 .history-size, .ns-101 .history-size {display: none}

Late to the party, but anyways... pretty unnecessary count, most revisions have virtually the same value. If only variations of +/-25% were shown, it would be better :-) -- ReyBrujo 22:00, 22 April 2007 (UTC)

Updates to Protected Pages

Thanks devs for implementing the update to: Special:Protectedpages, many sort options are now available. — xaosflux 02:17, 20 April 2007 (UTC)

Ooh. Nice pretty tools. Almost makes me want to... nah. Editing is still more fun. Carcharoth 03:04, 22 April 2007 (UTC)

Unused files / uncategorized images

When and how did Special:Unusedimages and Special:Uncategorizedimages get changed into galleries? Both are hideously bad as galleries and make patrolling a pain. Is this something that comes from the MediaWiki namespace or somewhere else? Thanks. --BigDT (416) 05:28, 20 April 2007 (UTC)

CSS bug in <code> boxes_boxes-2007-04-20T15:27:00.000Z">

<code> boxes have a line-height of 1.1em. This value is too small because with certain font sizes a little bit may be cut off from the lower part of characters. Underscores may disappear. Here is an example:

In my browser the lower part of the curly braces and the small g are cut off, the underscore disappeared.

I see this problem on Windows XP (both with ClearType turned on and off) with Firefox (fixed width font: Consolas, 15px), IE7 (default font settings) and Opera (default font settings).

I noticed this problem in Barton-Nackman trick where the underscores disappear from value_type in the first <code> box.

Changing line-height to 1.2 em solves the problem. BarroColorado 15:27, 20 April 2007 (UTC)_boxes"> _boxes">

That should be done site-wide, then, please. James S. 16:33, 20 April 2007 (UTC)

I hope this was the right place to report this ... I'm not sure if this should be considered a MediaWiki bug, or a Misplaced Pages bug (as the stylesheet is specific to this site). The culprit is in the declarations for pre in http://en.wikipedia.org/skins-1.5/monobook/main.css -- BarroColorado 18:13, 20 April 2007 (UTC)_boxes"> _boxes">

Attention administrators: The problem with 'g' is semi-bad, but changing '_' to ' ' in code is horrible -- please make this correction. Thank you. James S. 21:36, 20 April 2007 (UTC)

Just a clarification note: there isn't actually such a thing as a <code> "box", as <code> is an inline element. The box you are referring to is <pre>, which was generated in the above example because of the whitespace at the beginning of the lines (wikicode for pre). How do the following look to you?

Pre generated with whitespace:

{abcd_efgh}
xxxxxxxxxxx
{abcd_efgh}

Pre generated with <pre> :

{abcd_efgh}
xxxxxxxxxxx
{abcd_efgh}

Pure <code> with no pre:

{abcd_efgh}
xxxxxxxxxxx
{abcd_efgh}

--Splarka (rant) 07:56, 21 April 2007 (UTC)

Thanks for the reply! They look fine. I also removed the <code> tags from Barton-Nackman trick. -- BarroColorado 13:43, 21 April 2007 (UTC)

Graphic switching on Richmond County, Virginia

There is some rather odd behavior going on with the map of Richmond County, Virginia. Tested on Camino and Safari, both display a graphic with the state capital, Richmond City, highlighted rather than the county, which is northeast of the city. This graphic also appears on the page for Image:Map_of_Virginia_highlighting_Richmond_County.svg, which is referenced automatically via Template:Infobox U.S. County.

But when I downloaded the SVG to try and fix it, the downloaded SVG had Richmond County highlighted correctly. Seems like there's some weird substitution going on -- would a more advanced WikiGnome please check this out? Thanks. VT hawkeye_{talk to me} 17:13, 20 April 2007 (UTC)

Have you tried Shift-F5? Or WP:Purge? — RevRagnarok 20:09, 20 April 2007 (UTC)

I have verified that both versions look the same under Firefox (Windows). However, I'm a geography idiot and dunno if the image is the correct one - it's a fairly small red spot. I also have a image broken link for the seal. — RevRagnarok 20:12, 20 April 2007 (UTC)

OK, I reread the original post. I have the same problem. However, I asked it to purge the preview page and now it says the image is gone. And no record of it being erased in a deletion log. Yikes. — RevRagnarok 20:17, 20 April 2007 (UTC)

OK, I figured it out. SVGs are XML. There are two "paths" both named Richmond. One needs to be RichmondCo (the first). Then the fill at the end will highlight the proper one. (I did this locally in my text editor.) Bet the same problem happens with Roanoke. There are two of them too. I'd contact the original author and have him fix them all. The WikiMedia engine has a "last one" wins approach, the browser seems to do "first match." — RevRagnarok 20:22, 20 April 2007 (UTC)

I've fixed this for the following images:

• Image:Map_of_Virginia_highlighting_Bedford_County.svg
• Image:Map_of_Virginia_highlighting_Franklin_County.svg
• Image:Map_of_Virginia_highlighting_Richmond_County.svg
• Image:Map_of_Virginia_highlighting_Roanoke_County.svg

It had already been fixed for Image:Map_of_Virginia_highlighting_Fairfax_County.svg. The duplicate id issue is a problem for all of the other Virginia county maps as well, but they don't show any visible problems because there are no internal references to the duplicate ids. Since most of these county locator maps were generated using the same method by the same user, there are almost certainly maps in other states with the same problem. My fix was to append "_County" to the names of the paths for the counties whose names were the same as a city and update the xref. A safer way to deal with this would probably be to append "_County" to the paths for all counties. Mike Dillon 01:07, 21 April 2007 (UTC)

After looking at this closer, I don't think this will be a problem in states that don't have independent cities. The issue is that these maps were created from Census Bureau data and they treat independent cities and equivalent to counties for Census purposes. That's why the Virginia maps even have cities in them when other states only have counties. I took a quick look at the maps for the other counties listed in the Independent city article and their maps seem fine.

Also, I've uploaded Image:Map of Virginia counties and cities.svg to Commons with the ids for all of the counties changed to include "_County". Hopefully this can be a good basis for any future changes to Virginia county/city maps. Mike Dillon 01:38, 21 April 2007 (UTC)

Cascading notice in edit summaries

I've noticed that the cascading notice is not shown in edit histories (), but is shown in the logs. Is this something that can be done through the MediaWiki namespace, or does this need a bug report? 164.107.166.217 18:02, 20 April 2007 (UTC)

It's a bug, as far as I can tell. I have submitted bugzilla:9652 with my interpretation of why "" isn't included in the edit summary. Tizio 12:56, 21 April 2007 (UTC)

lowercase question

How do I properly use the lowercase parser function with a template variable? I have a switch statement and rather having something like |Test|test|TEST=output, I'd to just have one possible result which I could do if it was all lowercase. I've tried {{#switch:{{lc:{{{variable|}}}}} but that doesn't appear to be working. Any idea?↔NMajdan•talk 18:04, 20 April 2007 (UTC)

Nevermind, I think it is working.↔NMajdan•talk 18:13, 20 April 2007 (UTC)

Special:Filepath is inconsistent

Hi!

I'm trying to get the pathname for (arbitrary) images. Special:Filepath used to work(?) but now is inconsistent. For instance Special:Filepath/wiki.png (which resolves to http://upload.wikimedia.org/wikipedia/en/b/bc/Wiki.png) is found just fine, but Special:Filepath/Africa.png (which resolves to http://upload.wikimedia.org/wikipedia/commons/9/99/Africa.png) returns nothing.

Is it the "commons" thing? Any suggestions/workarounds? Thanks. --Saintrain 22:46, 20 April 2007 (UTC)

Wikimedia Commons is a different installation with a different db. Filepath seems to be restricted to local images on the same MediaWiki installation. Voice-of-All 23:20, 20 April 2007 (UTC)

Thanks VoA. Solution (at least for commons):

when Special:Filepath/Africa.png fails (404's), try commons:Special:Filepath/Africa.png .

but when both fail ???

New question:

1. Are there other DBs where WP images are stored?
2. What are they called?

Thanks. --Saintrain 20:43, 21 April 2007 (UTC)

No; only Commons (the global image repository) and the local repository. Titoxd 05:13, 22 April 2007 (UTC)

google error

A search for "wikipedia" on Google shows this : "two million articles and still growing." Did Google add up all of our WPs, or did they make an error?--Ed 00:14, 21 April 2007 (UTC)

The google hit you posted is for www.wikipedia.org which is for all wikipedia's, not just en.wiki. That's probably where they get 2 million from, although it's probably considerably more now. Ryan Postlethwaite 00:23, 21 April 2007 (UTC)

Possibly because that webpage appears to be static in some sense. x42bn6 Talk 02:15, 21 April 2007 (UTC)

Maybe GoogleBot may make a mistake. Jet123 (Talk) 02:54, 21 April 2007 (UTC)

Please tell me we won't celebrate reaching 2,000,000 articles? We still have massive backlogs on improving the first 1,000,000... :-( Is there any reliable way to measure whether the quality of the encyclopedia is improving? And anyway, the 2,000,000 figure might be wrong. Misplaced Pages:Misplaced Pages Signpost/2007-04-16/News and notes says: "All Wikipedias together have reached 7 million articles." Carcharoth 02:58, 22 April 2007 (UTC)

Not sure about the reliable part, but there's Misplaced Pages:Version 1.0 Editorial Team/Statistics. Titoxd 05:11, 22 April 2007 (UTC)

Ah, I think I see the problem. The "The biggest multilingual free-content encyclopedia on the Internet. Over two million articles and still growing." must be in some "alt text" thingy that the page instructs Google (and other search engines) to display. That needs to be updated to "over 7 million", though quite who does that, I wouldn't know. Carcharoth 03:01, 22 April 2007 (UTC)

m:Project portals. Titoxd 05:11, 22 April 2007 (UTC)

Thanks. No "over two million" thing there. I wonder where Google is getting that from? Carcharoth 00:12, 23 April 2007 (UTC)

From listing in Google Directory --h2g2bob 02:34, 23 April 2007 (UTC)

No error whatsoever. wikipedia.org does have over 2 million articles, hell, just de and en have that. -M 16:19, 23 April 2007 (UTC)

I've submitted a new version to DMoz, the ultimate source of the info --h2g2bob 04:09, 24 April 2007 (UTC)

Thanks. I've learnt quite a few things from this thread. Are lots of the Google blurbs taken from DMoz? Carcharoth 10:18, 24 April 2007 (UTC)

It seems to be that if it's listed there, Google will use the blurb. --h2g2bob 03:32, 25 April 2007 (UTC)

Special:PrefixIndex

I am confused, and wondering if I am misusing this or have found a gotcha. I was researching questionable user names and using Special:PrefixIndex. The username a concern had been raised about was User:The way, the truth, and the light which exists. But I can't find it using PrefixIndex and 'The w' (which does find User:The watershark for instance). What am I doing wrong? Can you help me see the light? Shenme 04:44, 21 April 2007 (UTC)

It looks like the user page for that user doesn't exist (nor any subpages), so it isn't found by that search. Mike Dillon 05:31, 21 April 2007 (UTC)

Prefix index is for pages. A user can exist but no have a page. Use Special:Listusers. --TeckWiz is now R Contribs_@(Lets go Yankees!) 05:33, 21 April 2007 (UTC)

Ah, very good. I had gotten a small clue when I came across a user whose sole presence was a /monobook.js subpage. Thanks! Shenme 07:53, 21 April 2007 (UTC)

http://en.wikipedia.org/Special:Listusers?username=The_way (you can pseudo-prefixindex link thusly). --Splarka (rant) 08:05, 21 April 2007 (UTC)

Article ~~~~ inaccessible but still in category

As you can see here, the article ~~~~ is still in Category:Protected redirects, but is no longer a valid title. Any idea for deleting it? It's a bit annoying because all other operations you do on the category listing fail beacuse of it (I have parsed it out, but it's an ugly kludge). Tizio 13:31, 21 April 2007 (UTC)

If it gives the "Bad title" error message, the only way to delete it is to wait for the next time when the developers run their bad title fixing script, which will rename the page directly in the database to a valid name prefixed with Broken/. After that's done, since the title is no longer invalid, it can be deleted normally. --cesarb 14:22, 21 April 2007 (UTC)

Great! Thanks. Tizio 14:25, 21 April 2007 (UTC)

Changing link trail

It has been suggested to change the link trail in MediaWiki:Linktrail from "/^(+)(.*)$/sD" to "/^(+)(.*)$/sD" to allow trailing apostrophes to be included in the link, as is desirable to prevent extra wikitext like here. Is there any reason why this has not been implemented? WARNING: If nobody replies within three days, I will implement this change. Please let me know if this breaks anything. Thanks. Samsara (talk • contribs) 14:32, 21 April 2007 (UTC)

Looks OK. Voice-of-All 17:23, 21 April 2007 (UTC)

I see no problem. –Pomte 19:10, 21 April 2007 (UTC)

Change implemented. Any problems, revert and discuss! Thanks, Samsara (talk • contribs) 22:59, 22 April 2007 (UTC)

Hmm... it doesn't seem to have any effect. Is there any reason why the effect might not be instantaneous? My only other guess is that the apostrophe needs to be escaped, so the variable would read "/^(+)(.*)$/sD". Samsara (talk • contribs) 23:16, 22 April 2007 (UTC)

' is not a special character as far as I know and doesn't need to be escaped. m:MediaWiki talk:Linktrail and m:Help:System message#Protection suggest that this page is ignored. –Pomte 03:00, 23 April 2007 (UTC)

In that case, I'm surprised that page hasn't been removed from the database, or rather, converted to a special page in case there's ever a need for someone other than devs to find out what this setting is. Presumably, deleting it will do no harm? I find it curious that the person who commented on the bug did not apparently know the setting was ignored. Samsara (talk • contribs) 09:09, 23 April 2007 (UTC)

Unicode issue (?)

What causes this kind of borkage? Is this a browser issue, an issue of someone copying and pasting the article into another program to edit, or a setting on the computer? I don't see it every day, but it can be messy to clean up. Thanks, Antandrus (talk) 16:20, 21 April 2007 (UTC)

Typically that is a browser issue, try downloading more langauge support. Voice-of-All 17:19, 21 April 2007 (UTC)

No, in this case Antandrus is correct. Whoever did the edit at that diff is using old non-Unicode-aware software (external editor? really ancient browser?) which replaced the legit unicode characters with old Windows-only gibberish characters. You see (if you don't use a MS operating system) this kind of mess all over the web on pages from the early 2000s and farther back. — SMcCandlish ‹(-¿-)› 03:17, 22 April 2007 (UTC)

They are not Windows-only gibberish characters, they are ordinary question marks, used for example by iconv to represent characters that aren't in the destination character set. According to the logs, the browser in question was Safari on Mac OS X 10.2.x (AppleWebKit/85.8.5). -- Tim Starling

problems editing page

For some reason I can no longer edit the Yue Fei page. I recently rewrote the entire page and have never had a problem with it before. Now when I try to edit any part of the page, it brings me here (It should say "Internet Explorer cannot display the webpage"). I even logged off and tried editing the page anonymously and it didn't work either. I can edit other pages, but not this one. There has to be a reason for it.

I just went to the help desk and some people reverted the page back to its original form. I noticed I can only perform one edit after another user does. After that, I can't touch the page. Why is this? (Ghostexorcist 17:39, 21 April 2007 (UTC))

I just switched to a different web browser and I still can't edit the page! (Ghostexorcist 18:35, 21 April 2007 (UTC))

Can you edit individual sections? x42bn6 Talk 18:44, 21 April 2007 (UTC)

I just tried and yes I can edit individual sections, just not the entire article. (Ghostexorcist 18:55, 21 April 2007 (UTC))

I think your browser might be timing out because it is taking too long to load the edit page or something - just a guess. Does it take a long time before timing out? x42bn6 Talk 23:48, 21 April 2007 (UTC)

Hierarchy of links pressed in the current session

In a given user initiated session, it would be helpful to have hierarchy of links that the user is clicking. This would help them being able to go back or move forward to link of interest in the given search. — Preceding unsigned comment added by 66.31.55.47 (talk • contribs)

That would be kinda cool, but I'm not entirely sure how feasible that would be. At the very least, there's always your browser's history. ;) EVula // talk // ☯ // 23:24, 21 April 2007 (UTC)

And why would anyone want to? We all have menu bar/Start bar clocks. This is 1995 silliness.  :-/ — SMcCandlish ‹(-¿-)› 03:18, 22 April 2007 (UTC)

National debt/Soc. Sec. clock

What exactly would the script that I would type into my User page to get the clock from http://zfacts.com/p/793.html onto the page? Alphabetagamma 00:48, 22 April 2007 (UTC)

It's possible to use this script if you want to see it yourself, but you can't make it work in such a way that other users will see it. Mike Dillon 01:46, 22 April 2007 (UTC)

Please fix the Category namespace edits problem!

I've noticed this before, but only just now realised what a problem it is. The essence of the problem is that when a category is renamed, it is done by deleting the old category and creating a new category under the new name. The edit history of the old category page is lost! This means that any edits anyone made to that category page are lost from their contributions list. This is not a problem when reviewing any categorisation you made of articles, as you do that by editing article namespace, and the edit history there is preserved when pages are moved/renamed. But categorisation of articles is only half the story. A lot of categorisation structural work is done by people categorising categories. Unfortunately, as category names get tweaked and changed, sometimes to suit changing fashions in category names, the edit history inevitably gets lost, and so does the history of who categorised the categories and edit summaries explaining the categorisations, and so on. Previously, I had only thought that edits to the "blurb" on the category page were lost (which was bad enough), but now I realise that the history of genuine categorisation edits are being lost. Is there no way that:(a) the edit history of category pages can be preserved, or even retrieved from deleted versions; and (b) a contributor can access their "deleted edits" to help review their genuine edits to deleted pages that, in the case of renamed category pages, technically still exist, but under a different name? Carcharoth 01:12, 22 April 2007 (UTC)

I have to concur. Categories should be movable just the same as articles. The amount of time wasted, in aggregrate, by Wikipedians having to go to CfD for the simplest things is really staggering. — SMcCandlish ‹(-¿-)› 03:20, 22 April 2007 (UTC)

See Bug 8685. Mr.Z-mantalk¢ 03:47, 22 April 2007 (UTC)

I understand that moving can't be done because of the way categories work, and so CfD is still needed, plus the running of a bot to do the rename, but what I want to see is a preservation of the edit history of the category pages. The bug report, and Titoxd's response, seems to say even that is not possible, but I don't quite understand why. Can anyone explain? Carcharoth 04:15, 22 April 2007 (UTC)

Ooh. Hang on. They are now saying it might be possible. This is terribly exciting!! <seriously, I'm not being sarcastic here - this is something I've wanted to see done for ages> I wonder if Cyde knows his bot is about to be made redundant... :-) Carcharoth 04:18, 22 April 2007 (UTC)

Well, not made redundant, but have its workload halved, or something. Carcharoth 04:22, 22 April 2007 (UTC)

It is not possible to update the category's contents automatically (I misread the question that was posted on the bug report). However, I'm not sure if there are any other obstacles to renaming a category page, besides leaving a lot of stale category links... Titoxd 04:24, 22 April 2007 (UTC)

Yes, a bot is still needed to do the grunt work of recategorising the articles from the old category to the new one. It seems the system would rely on people not to move category pages without a CfD being filed to ensure a bot did that grunt work. Maybe all Category pages could be move protected and then move unprotected and moved by the admin closing the CfD? That might be workable, and would still preserve the history. And correct me if I'm wrong, but couldn't this have been done all along? Can I sue someone for all my lost edits? :-) Carcharoth 04:28, 22 April 2007 (UTC)

Well, namespace-based default protection levels are possible (see Manual:$wgNamespaceProtection), so that can be made a default protection level (like editinterface is for the MediaWiki namespace). I am not sure whether that restriction is just for edits, but it is a start. It is still confusing, though... Titoxd 06:52, 22 April 2007 (UTC)

Could I possibly ask you (or someone) to make sure all the ideas in this thread end up in the bugzilla report? I'd really like to see a developer take a long, hard look at this, as losing edit history really shouldn't be acceptable, and if a way around it is possible, I think it should at least be clearly explained why it is not possible or won't be implemented. Carcharoth 20:58, 22 April 2007 (UTC)

As an aside, in that bug report, Titoxd said "categorylinks doesn't care if the category page doesn't actually *exist*" - a good demonstration is Category:Wikipedians who crack boiled eggs on the rounded end (must ensure the humour quotient stays high). Carcharoth 04:24, 22 April 2007 (UTC)

In case people are bemused by the red-link, click on it and visit the user pages of the people who categorised themselves this way. On second thoughts, don't! Carcharoth 04:28, 22 April 2007 (UTC)

I still don't understand why category redirects can't exist. Articles that are heavily linked to can be easily moved and the double redirects are quickly remedied, but articles that link to the old article name are left untouched. Why can't categories operate in the same way, with articles being left alone, but the category just redirecting to the current name? --MZMcBride 04:44, 22 April 2007 (UTC)

If you mean why can't a category page exist as a redirect, then the answer is that it should be able to. If the question is why can't one category act as a redirect and send its contents to another page, it is because the categorylinks table is not integrated with the page table, so the categorylinks table can point to non-existent pages, and ignore whether a page is a redirect or not. That could probably be remedied by adding a cl_to_id field, but that would be a massive change, not unlike the last database schema change. Titoxd 05:09, 22 April 2007 (UTC)

It should be fairly straightforward to join on the redirect table. Vote for bug 3311. -- Tim Starling 09:14, 23 April 2007 (UTC)

I've now read that bugzilla thread, but I'm afraid I didn't really understand it fully. Can anyone explain in simple terms what is being proposed there and how day-to-day categorisation and editing would change? Carcharoth 10:07, 23 April 2007 (UTC)

Okay, my input has been solicited on this, so here I am. There are an absolutely huge number of improvements that could be made to improve the categorization system. I'll start with the harder ones first. It would be really nice to have a way, within MediaWiki, to do true category deletion, merging, and renaming without having to go out and edit each individual article in the categories. But this could be far off in the future. The one problem that has been highlighted in this thread is the issue of category edit histories. As you know, there is no way to move one category page to another. It should be possible in just the same way that it is with article, but it is explicitly disabled, probably because most people would be confused and think that the "move" button should move all of the pages within the category when it was really just moving the text on the category page itself. As a result of this limitation, pyWikipediaBot compiles the names of the contributors to the old category page, copy-paste-moves the text over to the new page with the list of contributors in the edit summary, and then deletes the old page. This is clearly sub-optimal, and could be easily fixed by allowing category moves. --Cyde Weys 21:09, 22 April 2007 (UTC)

Thanks for the input. One specific question: instead of having category page moves explicitly disabled, is it not possible to have the namespace default "move protected", and then allow admins to remove the move protection, move the page, and then restore the move protection? (I'm not covering all details, but I think the general idea is clear). Carcharoth 21:26, 22 April 2007 (UTC)

Anything is possible, but I'd rather take some pride in the features I create and do the thing properly. I would propose to take the category links out of the article text on save, and put them back in at the end of the text on edit. I don't see any particular reason to preserve their precise location in the text. Then change the categorylinks schema to make cl_to be a page_id instead of a title, that way you get category moves for free. The fact that categories can be created without an associated page could be dropped, and the page could be created automatically when you first use a category. Versioning and review could be handled by storing a list of pages in a category in the text of the category page, and allowing editing of the category from either side. That would implicitly give you deletion and undeletion capabilities. Merging categories would become a simple cut and paste job, similar to merging articles. -- Tim Starling 09:25, 23 April 2007 (UTC)

That sounds good. I think. I'm not clear though, are you proposing that people would categorise by editing the category pages (and manually entering article names, like creating lists), rather than by editing the article pages as people do at the moment? That seems like a massive change to me. Or are you proposing both, the "from either side" bit? ie. You could categorise by adding category tag A to page X, or by adding page name X to category page A? Either way, would the edit history of both actions now be preserved, rather than partly lost as is the case at the moment when category pages are renamed (=deleted and recreated by cut and paste - losing the edit history of the category page, but not the edit history of the article being moved from one category to another)? It would be absolutely wonderful to see the category system upgraded to work more efficiently with redirects (and of course category intersections), but I don't want to lose sight of the original point I raised above, which was about preserving edit history. Carcharoth 10:07, 23 April 2007 (UTC)

Yes, I'm proposing that people should be able to edit categories either by editing the category pages or by editing the articles. Storing category membership in the category pages would allow for renaming, deleting, undeleting and merging in an efficient and intuitive manner. It would be difficult to preserve the ability to see category membership changes in article diffs, although not impossible. It may sound like a huge change, but the category code as a whole is relatively small and simple. -- Tim Starling 14:04, 23 April 2007 (UTC)

How would you deal with sortkeys that are generated from magicwords? Article deletion? Do you mean storing the category in the text of the category.. or only categorylinks? People also do care about the position within the text of categories and their order. I've seen people revert moving the categories above interwikis or someone alphabetically sorting categories whenever someone with AWB/pywikipedia/whatever comes by and messes with them. I believe this is because they want the categories that make sense to be the ones listed first instead of useless ones like 'Living people' or 'All articles with unsourced statements since August 2004'. Kotepho 14:34, 23 April 2007 (UTC)

Would you prefer it if it remained exactly the same as it is now? -- Tim Starling 15:48, 23 April 2007 (UTC)

I think it would improve things to be able to maintain categories more easily, but I think that it needs to be done carefully as to not make it harder to edit articles or for readers to use categories. Normalization is also a good thing, even if it is just a little. Slightly related to this is the same problem with images not being able to be moved. Kotepho 16:40, 23 April 2007 (UTC)

Another potential issue to consider here is categories that are added by templates, either with or without <includeonly>. Mike Dillon 15:56, 23 April 2007 (UTC)

As a frequent categorizer with no experience with bots, I would be overjoyed if Tim Starling's proposal to allow "editing the categories from either side." When I sort items into subcategories, for example, it is very tedious to go to each article indivisually and change "Category:Foo" to "Category: Foo in Canada" several dozen times. If I could re-categorize those articles directly in Categoryspace... my goodness, what a time saver! Fishal 04:22, 25 April 2007 (UTC)

Previous contents of a category

While reviewing my watchlist for red-linked categories that I had edited that had been deleted or renamed (I found one - Category:Texts by J. R. R. Tolkien), I found this deletion of a category. Not a very important one, and one that was empty at the time of deletion. My point is that it wasn't empty at one point, but I can't remember what was in it. Is there anyway to find out what used to be in that category? Carcharoth 01:19, 22 April 2007 (UTC)

It was moved in this discussion to Category:Works of J. R. R. Tolkien. It would have been nice of the deletion message has included a link to the discussion. Mike Dillon 02:15, 22 April 2007 (UTC)

The deletion message for the "Texts..." category did have a link to the discussion. Mike Dillon 02:20, 22 April 2007 (UTC)

Yes, I know. I followed that link as well. The empty category I am referring to was a different one, specifically this one. As I said, not very important. The "Texts..." one obviously is important! :-) Sorry, I shouldn't have mentioned two categories in the same post - it was bound to get confusing! The reference to the "Texts..." category was really an aside related to the previous section. Carcharoth 02:52, 22 April 2007 (UTC)

Undo CSS change to smallcaps

The CSS directive that controlled the size of text rendered in smallcaps was changed some while back. This change needs to be reversed or at least modified. It has resulted in text that is nearly illegible (and certainly crappy-looking) in Safari at default font and size (sample). At a blown-up font size, one can see that the smallcaps are much smaller than they should be, regardless what the scale of the base font size is (sample2). — SMcCandlish ‹(-¿-)› 03:10, 22 April 2007 (UTC)

redlinks to commons images

Images on commons redlink when you use the ] format, such as at , which actually exists . Worse, when you click such a link it brings up a page editor and provides no clue that the image actually does exist. Is there a work-around? Nardman1 05:00, 22 April 2007 (UTC)

Why is it a problem? -- Tim Starling 06:40, 22 April 2007 (UTC)

• Because I can't put text links to the images, at least using the inline system, making things like discussing whether the image belongs in an article impossible. Futhermore, Misplaced Pages has dozens of legacy discussions about images that were later moved to Commons, and the links to them don't work. Clicking the link leads one to a text editor without a proper preview, leading one to believe inaccurately that the image doesn't in fact exist. Nardman1 17:48, 22 April 2007 (UTC)

  Hm, let's see: Image:Cow-on pole, with horns.jpeg. Looks like it's already working. The deletion log correctly ignores the Commons image, but it doesn't mean the rest of the software does the same. --cesarb 19:53, 22 April 2007 (UTC)

  • D'oh. Nardman1 21:15, 22 April 2007 (UTC)

need help with {{archive}} template... possible broken link or incorrect usage?

Hello. I'm working on improving Misplaced Pages:Wikiquette alerts including the archive procedures.

We're using the {{archive}} template on the archived pages to ask users not to edit those pages.

But the link inside the template-generated page heading box directs comments to the (non-existant) archive talk page, rather than the main page at Misplaced Pages:Wikiquette alerts. Comments posted on the archive talk pages would not be seen by anyone, so this seems like a problem with the structure of the template - or maybe I'm missing an argument in the template.

An example of the result can be seen here: Misplaced Pages:Wikiquette alerts/Archive/2007

Is it possible this is a bug in the template, or are we not setting it up correctly?

Would someone please advise me on the proper way to set this up so the template does not create a redlink to the archive talk page and instead directs users to add new comments to the main page at Misplaced Pages:Wikiquette alerts... ?

Thanks! --Parzival418 06:38, 22 April 2007 (UTC)

The template is set up to use certain code that is getting confused by the slashes in the article title. I fixed the link provided by specifying which talk page is the current (code: {{archive|Misplaced Pages talk:Wikiquette alerts}}). Hope that helps. Cheers. --MZMcBride 06:56, 22 April 2007 (UTC)

{{archive|Misplaced Pages:Wikiquette alerts}} links to the desired project page. –Pomte 07:06, 22 April 2007 (UTC)

Thanks very much to both of you for the quick replies. I understand how to use the argument now and will fix the other archive pages.

I tried to find the syntax for this template before I posted my question here, but the template documentation page does not mention any arguments. I looked at the source of the page for the template but didn't know how to read the syntax. Now that I see how you handled it, I understand better what I read there, though just barely, and I still don't understand the "if" part of the template at the end.

Anyway... I thought you might want to know that the documentation page for this template may need to be updated to explain the use of the argument. I would update it myself but I don't want to take a chance on causing confusion if I don't get it right.

I appreciate your help, have a good evening. --Parzival418 07:22, 22 April 2007 (UTC)

I have added basic documentation. Feel free to improve on it by editing that sentence, and don't worry about changing the actual template as long as you're only editing text within <noinclude>. The parameter in question is {{{1}}} and the link is given with ], which means if the parameter isn't given, then link to the Talk: page above the current page. –Pomte 07:43, 22 April 2007 (UTC)

That's great, I get it now. Thanks very much! --Parzival418 07:49, 22 April 2007 (UTC)

Image question

This is probably not the place to ask this type of question, but how do you scan an image from TV show or movie? I'd like to scan a few fair use screenshots for some film/TB articles. Incidentally, I looked into the rationales needed and policies. Aaron Bowen 06:55, 22 April 2007 (UTC)

Your question probably would be better here. Cheers. --MZMcBride 07:04, 22 April 2007 (UTC)

Thanks, I'll ask it there too. Aaron Bowen 07:11, 22 April 2007 (UTC)

High Speed Bot Editing

A discussion that has involved in to having bot accounts edit at much increased rates (e.g. 3600 edits/hour) is taking place at Misplaced Pages:Bot_owners'_noticeboard#Cydebot_Block. Comment is welcome there. Thank you, — xaosflux 07:18, 22 April 2007 (UTC)

Adding classes to 'img' and 'href' elements

As part of adding microformats (see also WP:UF) to templates, I need to find a way to add classes to image tags (<img class="logo" >) and to external links (<href class="url" >). Is there a way to achieve this, with the the current software? Andy Mabbett 14:42, 22 April 2007 (UTC)

No. --cesarb 16:33, 22 April 2007 (UTC)

{{Rangevandal}}

Can anyone think of anything to add to this template, or should I call it done for now?--VectorPotential 23:34, 22 April 2007 (UTC)

I believe a WHOIS link still works, when asking about a range. That might be useful? – Luna Santin (talk) 01:14, 23 April 2007 (UTC)

Added--VectorPotential 01:28, 23 April 2007 (UTC)

Account creation weirdness

Is there any reason why a newly created account would not show up in the logs? An account request was made two days ago for Spray&Pray (talk · contribs · account creation), and the account did not exist when I checked it then. The username now appears in "listusers" but I cannot find a creation log for it. The requester reported receiving an e-mail with the account password (as is normal), but cannot login due to incorrect password. Any idea what is going on? jwillbur 00:14, 23 April 2007 (UTC)

• Could they have been created prior to the existence of the user creation log? or was it recently created?--VectorPotential 00:18, 23 April 2007 (UTC)

That was the first thing I thought of, but it was definitely created in the last two days because the e-mail that the requester received is automatically sent out when the account is created. jwillbur 00:23, 23 April 2007 (UTC)

We sometimes get unlogged operations (account creation, page protection, blocks, etc.) due to random deadlocks in an Apache signal handler. A deadlock causes the last transaction to be rolled back, and logging is often done in a separate transaction to the main operation to reduce lock contention. This will hopefully be fixed when we upgrade to Apache 2. -- Tim Starling 09:07, 23 April 2007 (UTC)

So, is there any way to fix this particular account? The messed up creation has caused the password to be unknown and no e-mail to be associated with it. Could the account be deleted to allow re-creation? The same thing seems to have happened to Greek0 (talk · contribs · account creation), by the way. jwillbur 17:22, 23 April 2007 (UTC)

Yes, done. -- Tim Starling 11:20, 24 April 2007 (UTC)

Thanks, Tim! Everything is straightened out now. jwillbur 16:56, 25 April 2007 (UTC)

'my watchlist' problem....

probably widely reported right under my nose, but i haven't seen it (sorry!) - but 'my watchlist' now has what seem to be markup characters either side of it - it says <my-watchlist> - any ideas why? - thanks, Petesmiles 11:24, 23 April 2007 (UTC)

I have the same problem. --TeckWiz is now R Contribs_@(Let's go Yankees!) 11:25, 23 April 2007 (UTC)

• For me it says "& lt;my-watchlist& gt;" - I was unable to find this on Special:Allmessages or Recentchanges for the MW namespace. >Radiant< 11:26, 23 April 2007 (UTC)

• just to clarify - that's exactly as it appears for me also - same as Radiant. Petesmiles 11:27, 23 April 2007 (UTC)

• As per Radiant. Carcharoth 11:28, 23 April 2007 (UTC)
  • Me too, except without the spaces. I think this is probably system-wide. Simply south 11:29, 23 April 2007 (UTC)
  • Same experience as Radiant, but it looks like it is gone now? Valentinian 11:32, 23 April 2007 (UTC)

yup - fixed now - thanks wonderful genies! - Petesmiles 11:39, 23 April 2007 (UTC)

Bot archiving for WikiProject talk pages

Is bot archiving allowed for WikiProject talk pages? It says here that use on article talk is not allowed, but it doesn't mention WikiProject talk. —AldeBaer 13:21, 23 April 2007 (UTC)

Yes, sure it is. User:Shadowbot3 does the archiving at Misplaced Pages talk:WikiProject Spam for instance. Graham87 13:49, 23 April 2007 (UTC)

Thanks. —AldeBaer 15:19, 23 April 2007 (UTC)

Google indexing

I've been asked whether google indexes the changing username archives. It's come up with regards to the right to vanish and people's privacy concerns.

We presume that the logs themselves are not indexed but is there a "no robots" type of thing to stop google indexing? If there isn't - could we do this to protect people who regret making their accounts with their real names? Thanks. Secretlondon 23:55, 23 April 2007 (UTC)

Logs aren't indexed, the archives (and the whole page) are. Robots.txt. Prodego 00:19, 24 April 2007 (UTC)

The developers have been known to add pages to robots.txt (i.e. prevent them being search-engine indexed) after a request at mediazilla:, as long as the change is obvious and/or you can point them to a discussion where there was a consensus for the change. --ais523 13:42, 24 April 2007 (UTC)

Case Sensitive

I have a question about the case sensitivity of Misplaced Pages. If anyone can help or offer input, that is greatly appreciated. Thanks in advance.

The problem began when I typed the link: Rebel without a Cause. However, the link did not work correctly, because the "w" in the word "without" was a lower case letter and not an upper case letter. I tried two other variants, and they both did work: Rebel Without a cause and Rebel Without a Cause. I tried a 4th variant, which did not work: Rebel without a cause. So, the question is: why do some of these variants work, and some do not? And, further, when I go to create a page, do I have to create "redirects" for every possible upper / lower case scenario? In this case, there are 16 possible configurations of upper/lower case letters for the 4 words. Cannot Misplaced Pages simply recognize that "rebel without a cause" (regardless of upper and lower case letters) should ALL direct to the same page? I am very confused and frustrated by this. Can someone please explain how all of this works? Furthermore, what if someone were to mistakenly type something like "Rebel WithouT a Cause" (note the typo of the upper case second "t" in "without")? If someone were to type that one letter accidentally as upper case, cannot Misplaced Pages nonetheless direct them to the correct article? If not, there are literally thousands of redirects that an editor would have to create -- no? Thanks for any input. The complexity of what should / could be so simple is quite frustrating. (JosephASpadaro 03:08, 24 April 2007 (UTC))

Also, what the search/go box will link to sometimes will not work with regular links. For example, I recently worked on the page Ibi. If I typed IBI into the search box and clicked Go, it would take me to Ibi. However, if I created a link IBI, it displayed as a redlink (until I made a redirect). Mr.Z-mantalk¢ 03:23, 24 April 2007 (UTC)

(edit conflict) See Misplaced Pages:Go button. As long as you have a redirect at Rebel without a cause or Rebel Without A Cause, all search variations end up there. If you're linking to the article from another page, the link has to be an existing article or redirect, but in most cases you would use the exact article name and not something odd like Rebel Without A cause. –Pomte 03:26, 24 April 2007 (UTC)

Thanks! (JosephASpadaro 03:40, 24 April 2007 (UTC))

This request is bug 453. -- Tim Starling 12:20, 24 April 2007 (UTC)

Bunched up edit links

Misplaced Pages:Meetup/London1 is vexing me. This is being done by the huge Wikimeetup template, not by images, so those fixes didn't help me. Is there any way to fix this? hbdragon88 06:28, 24 April 2007 (UTC)

I've a similar problem on Misplaced Pages:WikiProject Mathematics - I can't see any section edit links at all. --Salix alba (talk) 11:25, 26 April 2007 (UTC)

SVG images not working

Within the last few minutes, all SVG images have changed to bluelinks to their image description page and no longer show the image for me: for instance, {{tick}} displays "Y". Are other people getting this problem too? --ais523 14:24, 24 April 2007 (UTC)

They are not working when used in subst templates. -- FayssalF - 14:27, 24 April 2007 (UTC)

I cant actually view any of the images that now show up as blue links. I think it is also happening in not substed templates. -- Chrislk02 (Chris Kreider) 14:29, 24 April 2007 (UTC)

I was just about to comment on them not working, svg images that have been cached seem to work. But resize isn't working Richard Thompson (Talk! | Contribs) 14:31, 24 April 2007 (UTC)

I as just about to say something, they don't show in templates or the image pages themselves... Must be a problem with the renderer. _InBC 14:32, 24 April 2007 (UTC)

Ok, not just me. -- Chrislk02 (Chris Kreider) 14:33, 24 April 2007 (UTC)

Seems to be enwiki-only; for instance de:Vorlage:Begriffsklärung is showing an SVG image fine at the moment. --ais523 14:34, 24 April 2007 (UTC)

Is the issue connected with the thread below? – Riana 14:36, 24 April 2007 (UTC)

That is a distinct possibility. _InBC 14:37, 24 April 2007 (UTC)

I think everyone's having this problem. I noticed a pattern. Someone vandalized the image and then reverted by a Wikipedian but the image doesn't come back. Most national flags can't display edit history and when I click on file history nothing happens. OhanaUnited 14:42, 24 April 2007 (UTC)

Looks like they are back Richard Thompson (Talk! | Contribs) 14:45, 24 April 2007 (UTC)

Yep, seems to be working fine now. – Riana 14:55, 24 April 2007 (UTC)

See also bugzilla:9677. --ais523 16:06, 24 April 2007 (UTC)

Imagemap issues

Imagemap isn't working, just showing up as a lot of code and <imagemap>: image is invalid or non-existent. Any reason? Anything to do with the above thread? – Riana 14:36, 24 April 2007 (UTC)

Check out the top right of my userpage for an example. – Riana 14:37, 24 April 2007 (UTC)

I dont know enough about it, although if there is a problem with the images, then I would think the problem causing all the svg's to show as blue links might also be triggering the imagemap problem, possibly because it cannot find the images. Again, this is not an area of my expertise. -- Chrislk02 (Chris Kreider) 14:41, 24 April 2007 (UTC)

Whoa... I cannot even view your user page right now. It gives "database error" and "from within function "Image::upgradeRow". MySQL returned error "1223: Can't execute the query because you have a conflicting read lock (10.0.0.231)"." This was the same message I got trying to view a SVG image itself minutes ago. —Resurgent insurgent 2007-04-24 14:37Z

Database error? :S – Riana 14:54, 24 April 2007 (UTC)

(edit) And the blue question mark SVG at the top of this page is back. Still get the error for your user page. —Resurgent insurgent 2007-04-24 14:38Z

at last check I could view them on the image page, however they are not showing upo in userboxes. -- Chrislk02 (Chris Kreider) 14:46, 24 April 2007 (UTC)

Is this fixed now? -- Tim Starling 08:09, 25 April 2007 (UTC)

According to brion, yes (bugzilla:9677) (and it appears to be working on pages which contain the code in question, such as User:Riana). --ais523 13:54, 25 April 2007 (UTC)

Link to part of a page

How do you link to part of a page. I want to do it like #part of Page but I don't want it limited to the headings in the page. 86.41.131.174 14:33, 24 April 2007 (UTC)

You can put in a new 'anchor' in a page (that you can link to using #) by typing <span id="anchor name"></span>. --ais523 14:35, 24 April 2007 (UTC)

Why do I have the option of editting other people's comments?

Or deleting the entire "Talk" page, for that matter. Doesn't this take "anyone can edit" a little to far? — Preceding unsigned comment added by Lumarine (talk • contribs) 16:16, April 24, 2007 (UTC)

Nope, it works great. If a user abuses the privilege then they can be blocked, and any damages easily undone. _InBC 16:17, 24 April 2007 (UTC)

Not to mention, there's pair of bots that automatically revert page blankings of any kind--VectorPotential 16:18, 24 April 2007 (UTC)

No, this is stupid. Comments belong to a historical record which should never be edited, only appended to. The Misplaced Pages system is looking at everything as a nail because you have a hammer. --Ideogram 21:00, 26 April 2007 (UTC)

The issue comes when someone posts "OMG U GAYWAD!" There's no reason why we should have to go through a long post deletion process to remove that. -Amarkov moo! 21:04, 26 April 2007 (UTC)

Let admins delete entire posts just as they can delete entire articles now. There is nothing most people should be able to do to another person's post; and only uninvolved admins should be able to delete it if it is abusive. --Ideogram 21:17, 26 April 2007 (UTC)

Actually, one thing lost on modern[REDACTED] is the will to wiki:ReFactor (for instance from wiki:ThreadMode to Wiki:DocumentMode). In one sense, the ability is vestigial. In another sense, it's sad that this has been lost --Kim Bruning 21:05, 26 April 2007 (UTC)

What we really need is dynamic views on the underlying data where you can choose to view the unfiltered historical record, or a "cleaned-up" version approved by (someone). --Ideogram 21:22, 26 April 2007 (UTC)

I am so going to look into wiki interface design, sometime in my copious free time. --Kim Bruning 21:24, 26 April 2007 (UTC)

All comments are kept forever in the history tab. _InBC 03:40, 27 April 2007 (UTC)

Sure, but finding them back can be fiendishly tricky. --Kim Bruning 04:02, 27 April 2007 (UTC)

Archive

I would like to know how I can archive my talk page. How do I create the page and how do I make the archive box on my current talk page? Thanks in advance! Felixboy 17:48, 24 April 2007 (UTC)

You can do it yourself or have a bot do it periodically. See WP:ARCHIVE. –Pomte 18:02, 24 April 2007 (UTC)

Ok, thanks alot! Felixboy 18:20, 24 April 2007 (UTC)

Image:Rickey henderson.jpg at Rickey Henderson

Hello, cannot get Image:Rickey henderson.jpg to show in the template at Rickey Henderson. Probably something simple, but is has beat me, and at least one other person. --Ezeu 18:21, 24 April 2007 (UTC)

• It seems, that every instance of {{Infobox MLB player}} is in fact currently broken, no few pages using this template seems to contain a correct image link, see Special:Whatlinkshere/Template:Infobox_MLB_player--VectorPotential 18:26, 24 April 2007 (UTC)

• Correction, it seems that a number of articles have this template improperly applied, such as Barry Bonds, Ozzie Smith, etc.. Unfortunately I don't have time to go through all 2500 articles that use this template--VectorPotential 18:30, 24 April 2007 (UTC)

• Correction you seem to have figured it out the same time as me, it's the missing width parameter that's doing it--VectorPotential 18:33, 24 April 2007 (UTC)

• Yep. Thats it. Thanks. The template is probably better off with a default width instead. --Ezeu 18:34, 24 April 2007 (UTC)

• Done, set at {{{width|200}}}--VectorPotential 18:39, 24 April 2007 (UTC)

Section edit on right click

Using MSIE7 in WinXP, Whilst reading List of emotions, I right clicked on the link Lojban to open it in another window. Instead of the article, I got a section edit window of the current page section. This also happens with artificial language in the same heading. Why does this happen? How can it be stopped? -- SGBailey 20:46, 24 April 2007 (UTC)

Special:Preferences → Editing. Matthew 20:49, 24 April 2007 (UTC)

Thanks -- SGBailey 21:08, 24 April 2007 (UTC)

List in infobox breaks the infobox

The templates {{Infobox Ship Career}} ans {{Infobox Ship Characteristics}} are used like this

{|  <!-- infobox table start stuff -->
{{Infobox Ship Career
|arg=Career value
...
}}
{{Infobox Ship Characteristics
|arg=Characteristics value
...
}}
|}  <!-- table end -->

Under some circumstances, if one of the Career value entries is a LIST (* or #) the infobox breaks (a </TABLE><TABLE> is inserted between the template outputs). Note that all the TABLE code is outside the templates. Nor is there any LIST code in the templates.

Some of the code that breaks the infobox is outlined in the talk page. Because the break happens between the templates suggests a template problem, but because of the wierd circumstances (please see the talk page) it "feels" like a rendering bug.

It's a real puzzler. Thanks. Saintrain 22:06, 24 April 2007 (UTC)

I think I've fixed it. See reply here. –Pomte 01:45, 26 April 2007 (UTC)

Proposal: Stabler versions

Idea: There can be one more tab at the top of article pages, marked "Stabler version". Clicking on this would display a version of the article selected by an algorithm such that it's usually a version that seems to have been approved of by three editors.

Usually, it would be the version left by the 3rd last editor to edit the page. But if reverts have been happening, it would be the version people keep reverting to (if any). The algorithm could be simple or more complex to handle more complex situations such as partial reverts and overlapping edits.

If editor A changes one part of a page and then editor B changes another part of the page, implicitly editor B has indicated approval of editor A's changes. If, on the other hand, B reverts A's changes, that would indicate disapproval. The algorithm would try not to return a version which any later editor had reverted, although in complex situations of overlapping edits it might not always be able to avoid it.

In cases where the last few edits have been vandalism and reverting of it, the algorithm would return the version that people keep reverting back to (provided at least 3 editors, including the one who created it, have left it at that version.) --Coppertwig 23:07, 24 April 2007 (UTC)

You may want to check out stablepedia.org then. Voice-of-All 00:22, 25 April 2007 (UTC)

We are neither affiliated with, nor endorsers of that site officially by the way. But yes, that is what you are looking for. That site might be a bit mean to our servers though, but compared to other stuff, it is nothing. Prodego 02:34, 25 April 2007 (UTC)

What does Ken Stabler have to do with anything? Perhaps you mean "more stable version"? Corvus cornix 18:12, 26 April 2007 (UTC)

Isn't Misplaced Pages workong on a method for "grading" certain versions of articles? I thought I remembered hearing something like this several months ago.↔NMajdan•talk 18:25, 26 April 2007 (UTC)

Could it be WP:PR? --saxsux 20:11, 26 April 2007 (UTC)

You might want to take a look at Misplaced Pages:Why stable versions, m:Reviewed article version, and mw:Extension:FlaggedRevs. -- John Broughton (♫♫) 20:34, 26 April 2007 (UTC)

Edittools problems on personal wiki

I have copied part of MediaWiki:Edittools for use on my personal wiki (thanks for making that box, by the way -- it's really handy) and installed the CharInsert extension. However, some of the characters that look just fine in Misplaced Pages turn up as diamond-enclosed question marks on my wiki. I have scoured the CSS on this site as much as I can, but couldn't find anything that fixed the problem. How can I make the characters work? Thanks in advance to anyone who can help. — Tuvok 03:57, 25 April 2007 (UTC)

It sounds like a character encoding problem related to copying and pasting. Is you wiki public? Mike Dillon 03:59, 25 April 2007 (UTC)

Unfortunately not. It's actually only local as of yet; I'm still looking for hosting. I can post the code from any pages I have in user subpages if it'll help (like LocalSettings.php or MediaWiki: namespace pages). Thanks for the quick response! Too bad I wasn't online :).

I actually tried copying and pasting a couple problem characters directly from Character Map, but it didn't fix the issue, at least with the character or two I tried it on. — Tuvok 21:22, 25 April 2007 (UTC)

You might try copying and pasting the contents with a different browser to see if you have the same problems. Mike Dillon 01:00, 26 April 2007 (UTC)

And here I thought IE would be good for something... Wish I could say it worked. Thanks for the tip, though. I noticed that my edits copying from Character Map weren't actually saved. Since the characters didn't change (according to MW, at least), it didn't save the edit. I'll have to keep working on this... — Tuvok 03:40, 27 April 2007 (UTC)

Help with a thumbnail

I tried purging the article, the wikimedia description page, and the[REDACTED] description page, and tried adding ?! to the article's URL, but no luck. Is there some way to have this image appear here? There are two versions in the wikimedia history. The larger one is what I am trying to see in place. Thank you.-Susanlesch 09:24, 25 April 2007 (UTC)

I can currently see the image on the image description page and as a thumbnail. --Iamunknown 09:29, 25 April 2007 (UTC)

Thank you for the magic touch. I see it now too! -Susanlesch 09:33, 25 April 2007 (UTC)

Glad I could help. :-) --Iamunknown 09:40, 25 April 2007 (UTC)

Request for edit history change

These days, vandal reverting is so fast, that sometimes vandals can be reverted two or three times in one minute.

But edit histories only show time in hours and minutes.

Perhaps the edit histories should be changed to hours, minutes, and seconds, so that you won't confuse all the edits that happened at 8:55 or whatever.

Also, occasionally, if two edits happen one right after the other, those two edits (just those two) will be backward on the edit history. Adding a seconds feature would help, unless the problem only happens when edits are within a second of each other.

Virtual Cowboy 12:00, 25 April 2007 (UTC)

Anyone may currently select something quite similar to your suggestions and to ISO 8601#Combined representations. Go to Special:Preferences, click the "Date and time" tab (if you are in the monobook skin) and click the radio bullet next to "2001-01-15T16:12:34". --Iamunknown 12:04, 25 April 2007 (UTC)

OK, I just did that, but it looks really awkward (all smooshed together). Is it possible to have the old format and just add seconds? Virtual Cowboy 12:08, 25 April 2007 (UTC)

I agree that it is difficult to look at. The time functions in PHP, as in MediaWiki ParserFunctions, are rather versatile and could definitely append a colon and two digits for seconds to a more familiar timestamp. It was be relatively easy (I would think) to implement, if you filed a bug report for more options with time stamp formatting and provided an attachment, I don't it would be excluded.. --Iamunknown 12:16, 25 April 2007 (UTC)

Moving an image?

For examle, I am creating a PNG of File:TalkTalk.gif, and therefore would find it much easier to rename the image extension from .gif to .png, so I can edit it from there. Is there any way to do this? Otherwise, is there a template to show that it has been superseded? δσώпҹ 14:28, 25 April 2007 (UTC)

No, a new image must be uploaded in the PNG format; you can't just rename the image or move the image description page. You can move the talk page over to the new image's talk page. Mark the old GIF copy with Template:PNG version available or Template:obsolete. —Remember the dot 16:48, 25 April 2007 (UTC)

How can I find my AMA request accoun?

I seem to have lost the bookmark for it. None of their pages seem to direct me to it. How is this achieved? HeadlessJeff 18:46, 25 April 2007 (UTC)

What is an AMA request account? Is it a alternate account of yours? John Reaves (talk) 19:03, 25 April 2007 (UTC)

Help with load time, resource use

Hi. Minneapolis, Minnesota was really slow to load, so I cut some text and templates. Now I am wondering if there was something else I ought to have done instead. Why is it so much bigger in bytes than these comparable articles? Are there ways to use bytes better? Thanks. -Susanlesch 19:13, 25 April 2007 (UTC)

• Looks to me that Minneapolis is just a larger article with more pictures; compared to say the Ann Arbor example you give. If there is twice as much stuff, it will take twice the time to load; can be a little less than double (there is a fixed cost in loading anything) but also can be a little more (there is a quadratic effect, a longer load is more likely to collide with other network traffic or system usage, and face longer delays). Pete St.John 20:43, 25 April 2007 (UTC)

• Thank you for the reassuring opinion. Yes, "just larger" makes sense in the case of Ann Arbor. But I am not so sure, looking down the list. Could 32 citations make that much difference between, for example, Minneapolis and San Francisco -- which has a larger "Edit this page" at 92K, but smaller "Pre-expand" etc.? While I am asking, what is the relationship between the "Edit this page" byte count and the others (850946, 317371 and 245963 in the case of Minneapolis)? It may help to note Minneapolis has three templates above the footer (which explains the high "Template argument" maybe). -Susanlesch 23:31, 25 April 2007 (UTC)

Templates normally count much more towards the pre-expand include limit than non-template text (they don't count any more towards the post-expand include limit unless they expand to a lot of text, but it's the pre-expand limit that's hard on the servers). See Misplaced Pages:Template limits; yes, it is all the cite templates that are hard on the servers, as they contain a lot of ParserFunctions and similar complicated code that take time to parse. --ais523 07:54, 26 April 2007 (UTC)

• Many thanks for the link (it answered a couple of questions I had). Would omitting the approx. 130 cite templates help speed up the page? It would only take a short time to rewrite them to look like the same format (and might even save a few more bytes in the page in the process). -Susanlesch 19:41, 26 April 2007 (UTC)

• The content is just about the same as before, but I removed the cite template markup in Minneapolis, Minnesota to see what would happen, see table below for before and after. Is this removal of the 'cite's going to speed up loading? Save server time? Or am I better off reverting to the cite templates that other editors are accustomed to? (Regarding editor preferences, I added about 120 of these in the first place so can comfortably go either way I think. I'll check to see what style of citation markup was used last year. Trimming 6k seems worthwhile to me.) -Susanlesch 06:20, 27 April 2007 (UTC)

Getting a partial source, with action=raw?

Is there a way to get the source (ie. raw wikitext) of a section of an article? I tried adding &section=x to the URL (as per how section editing is done) but it still pulls the whole article. If there is no current way to do this via a URL, is there a good way to parse the whole wikitext to split into sections (ie. a regex to split on)? —Daniel Vandersluis 21:11, 25 April 2007 (UTC)

Here's the MediaWiki parser regex for making headings (converted to JavaScript):

       for ( i = 6; i >= 1; --i ) {
           var h = "";
           var r = i;
           while (--r >= 0) h += "=";
           re = new RegExp("^"+ h +"(.+)"+ h +"(\\s*)$", "gm");
           text = text.replace(re, "<h"+ i +">$1</h"+ i +">$2");
       }

You could a split/explode just before the regexp, and you'd have an array of sections. As far as I know, there's no current way to do this via URL. Datrio 07:37, 26 April 2007 (UTC)

Thanks! —Daniel Vandersluis 13:49, 26 April 2007 (UTC)

Arabic text and numerals

Is there a known problem with numerals following arabic text? See the first line of Sirr Al-Khatim Al-Khalifa, for an example. If I change that capital i at the start of the date to a figure 1 the whole date jumps to before the arabic text. It's obviously something to do with arabic reading from right to left. Sandbox experiments confirm the problem and it happens here whether I'm logged in or not, on XP using Firefox or IE. If I replace the space following the arabic with &nbsp; it renders OK in the editor, but still jumps on the page: viz.

سر الختم الخليفة 1234 456 789 test

I can fix it by entering some plain text (e.g. born) before the numeral, but that's not the point! I've tried searching for references to this, but the nearest I can find are this and this (search for "arabic text"). I'd appreciate if someone could confirm this is a problem, and is not just something wonky here. Is there a fix? - a <span> perhaps? Smalljim 21:53, 25 April 2007 (UTC)

Yes, simply use {{bdo}} (a template replacement for the unsupported <bdo> tag):

سر الختم الخليفة 1234 456 789 test

As to the rendering order on the editor, I have #editform textarea { direction: ltr; unicode-bidi: bidi-override } on my user stylesheet; it's the exact same CSS the template above uses, but being applied to the edit area. I don't know if it still works; I added it a long time ago, and the skin might have changed in the meantime. It's worth a try. --cesarb 23:59, 25 April 2007 (UTC)

That CSS will cause the entire edit box to render left-to-right no matter what's typed in it (including Arabic); I suspect that wouldn't be an appropriate fix for someone who reads or writes Arabic, as it would be a similar effect to all the words appearing backwards in English (which would be hard to read!). --ais523 07:51, 26 April 2007 (UTC)

Thanks for pointing me to the {{bdo}} template, cesarb. That's allowed me to fix the article. I'm not too bothered about rendering order in the editor: I don't expect to come across this often, and the addition of the letters in the template cures it. But it is odd that it only happens with numerals following Arabic. How do I find out if it's a known bug, or something I ought to report somewhere for the good of the community? Smalljim 09:53, 26 April 2007 (UTC) Oops! edit collision - thanks for the explanation below, Tim. Smalljim 09:58, 26 April 2007 (UTC)

The problem is that the HTML specification and all the browsers use the Unicode Bidirectional Algorithm, which although it may be quite versatile, does need tweaking at times to get things to display correctly. The problem here is that weak characters such as numerals adopt the type of strong characters logically before them, which in this case is the arabic text, not the latin text following. I wouldn't recommend using the override characters (LRO/RLO or HTML's BDO), since their purpose is to make text go backwards, e.g. {{bdo|rtl|unicode}} -> unicode. Rather I'd suggest separating the numbers and the arabic text with an LRM character. LRM is a non-displaying character with strong left-to-right type, so it makes weak characters after it become left-to-right. Using a pasted LRM character works both in the edit box and on display:

سر الختم الخليفة‎1234 456 789 test

Using HTML character entities works on display:

سر الختم الخليفة‎1234 456 789 test

--- Tim Starling 09:44, 26 April 2007 (UTC)

Stop people from editing a page

How do i stop other people from editting a page i made? — Preceding unsigned comment added by Meanmachine4242 (talk • contribs) 22:00, 25 April 2007 (UTC)

You can't. This is the encyclopedia where anyone can contribute, after all. x42bn6 Talk 01:46, 26 April 2007 (UTC)

Appendix namespace

• m:Appendix namespace

In case it hasn't been noted here already: The Spanish Wikipedians are proposing that all Wikipedias in all languages gain an "Appendix:" namespace (like the ones in the Spanish Misplaced Pages and the English Wiktionary) and that all Wikipedias have an Encyclopedic Support Contents Policy. Uncle G 23:21, 25 April 2007 (UTC)

• Technically doable, but please, let's not. Basically, it boils down to some people thinking that tables, reference lists, and lists shouldn't be in the main namespace, but others disagreeing. Titoxd 23:26, 25 April 2007 (UTC)
  • Agree with Titoxd. No more red tape. Please! Valentinian 23:35, 25 April 2007 (UTC)

The place for discussion is the talk page of the Meta page linked to above. The people making the proposal won't see know about your objections if you write them here. Uncle G 15:20, 26 April 2007 (UTC)

• • • Well, I'm talking about the English Misplaced Pages here. Other wikis may find it useful, but not here. And they do know about me not liking it, I participated in the process to enact that thing... Titoxd 04:04, 27 April 2007 (UTC)

How to prevent my pages from being deleted

I have been notified that User:Jefferson Anderson is trying to delete my personal pages. How can I prevent this from happening? Sincerely, --Mattisse 00:22, 26 April 2007 (UTC)

I suppose you mean that he is going to either nominate or speedy delete certain pages under your user space. First of all, check whether those pages fit the User page guidelines, otherwise there is little to do. Second, if the articles are nominated for deletion at Miscellany for deletion, discuss there why you think those pages are relevant for the encyclopedia. Finally, if the discussion ends with the deletion of the pages, or if the pages are speedy deleted (review whether a Criteria for speedy deletion really fits the case), you can always request a Deletion review. Note that you can't stop the page from being speedy deleted if it qualifies for it, nor you can stop anyone from nominating a page for deletion if seen unnecessary or provoking. -- ReyBrujo 00:36, 26 April 2007 (UTC)

In other words, there is nothing I can do. Thanks, --Mattisse 01:07, 26 April 2007 (UTC)

Be sure to sign your posts with ~~~~. John Reaves (talk) 01:18, 26 April 2007 (UTC)

How can I get back material that has been removed? How can I retrieve lost ANI posts? Please? Mattisse 02:00, 26 April 2007 (UTC)

I did sign with Mattisse 02:02, 26 April 2007 (UTC). Now I will sign by clicking the toolbar.--Mattisse 02:02, 26 April 2007 (UTC)

Can I get my pages protected so I can get some rest? Please? --Mattisse 02:05, 26 April 2007 (UTC)

: I cannot see how such a protection would be in line with the Misplaced Pages:Protection policy. Jefferson Anderon is not "deleting" your personal pages; as far as I can see, he removed some content from them which, incidentally, anyone including me can view simply by going into the revision history of your user page. I don't know the history between you two, but I venture to say that this thread will not work to improve relations. May I suggest you blank this thread after reading my message? Regards, Iamunknown 02:13, 26 April 2007 (UTC) Responded less civilly on ANI and, after realizing what I'd done, hopefully more civilly on editor's talk page. Suggestion to close thread? --Iamunknown 03:34, 26 April 2007 (UTC)

How do I close a page that I made from being editted?

Sincerly Meanmachine4242

• You don't. This is the encyclopedia anyone can edit, it's a collaborative process--VectorPotential 00:34, 26 April 2007 (UTC)
  • This is not entirely true. They can request for the page to be protected. I'm not sure the requirements, but you could look into this. --AjaaniSherisu 11:14, 26 April 2007 (UTC)
    • The most common reasons to protect an article would be a high amount of vandalism from several different users, and an edit war. See Misplaced Pages:Protection policy. --ais523 15:21, 26 April 2007 (UTC)

The repeated blanking of Monk Bonasorte (edit | talk | history | protect | delete | links | watch | logs | views) by Meanmachine4242 (talk · contribs) is probably relevant here. Uncle G 15:16, 26 April 2007 (UTC)

• Closed, user blocked indefinitely. Tizio 16:57, 26 April 2007 (UTC)
  • Please remember that if an editor is blanking a biographical article, xe might be ineptly attempting to remove negative biographical information. That is not necessarily vandalism, and your immediate reaction should not be to block the editor. In this instance it was negative biographical information sourced to an article in the St. Petersburg Times. Uncle G 12:13, 27 April 2007 (UTC)

Moving content and images to Wiki's in other languages

Are images and text moved automatically from the English[REDACTED] to other language WikiPedia's or vice versa? Or does all of the moving between WikiPedia's in different languages have to be done manually? If moving images is done automatically, is there something that checks to see if an image with the same name exists on the other WikiPedia?

I am trying to figure this out so that it might help in another Wiki where we want to have one version of the pages for internal users and another version for external users in the organization that runs the wiki. Talk to GM 00:37, 26 April 2007 (UTC)

The different language editions of Misplaced Pages are completely separate - articles do not correspond word for word, but editors are encouraged to import content through manual translation, assuming competence. For your wiki, it sounds like you might want to run a bot to manually copy pages, but I don't know as much about that. Nihiltres 00:51, 26 April 2007 (UTC)

Images, if they have the right license, go to Wikimedia Commons, an image and media repository that can be accessed by all of the Wikimedia projects (all Wikipedias, Wikibooks, Meta, etc). They get uploaded manually, but should have on the image description page the details from the page on the original wiki. It is possible for developers to move pages along with their history (e.g. from Meta to the MediaWiki wiki), but generally we just copy it. Harryboyles 03:30, 26 April 2007 (UTC)

Don't run a bot - it is not necessary. You can just export and import pages with Special:Export, or even better, with Special:Import. It requires a bit of work on the server side (to set up interwiki sources), but it is overall more efficient. Titoxd 04:20, 26 April 2007 (UTC)

Monobook sidebar

Could you please explain how to position the sidebar (the left side of your screen) to the right? ↔ z (talk · contribs · autographs) 03:32:29, Thursday, 26 April 2007 (UTC)

I think this does it:

#column-content {
    float: left;
    margin: 0 0.6em 0.6em -12.7em;
}
#content {
    border-color: #AAA rgb(170, 170, 170) rgb(170, 170, 170) transparent;
    border-style: solid solid solid none;
    border-width: 1px 1px 1px medium;
}
#column-one {
    padding-top: 1.5em;
}
#p-cactions {
    left: -1.5em;
}
#p-logo {
    overflow: hidden;
    position: relative;
}

This seems to work alright in Firefox at least. Mike Dillon 04:16, 26 April 2007 (UTC)

Color of internal links

In the default monobook setting, assuming no custom CSS is used, what six-digit hexadecimal value describes the default color of unclicked internal links, such as this one (before clicking on it)? Pyrospirit 03:46, 26 April 2007 (UTC)

rgb(0,43,184) or #002BB8. Does this fool you? x42bn6 Talk 03:48, 26 April 2007 (UTC)

No, but this does! Gracenotes § 13:39, 26 April 2007 (UTC)

Thanks. Time for a totally real link to the Misplaced Pages article on fish! Pyrospirit 23:05, 26 April 2007 (UTC)

Revert to specific revision sometimes fails when vandal actively at work?

I had one unhappy reader inadvertently point out to me that a reversion using Lupin's popups didn't quite work correctly. Pointing at the correct revert point, I got a mix of that version and the most recent vandal edit. Thinking all's good I wandered off having left vandalized text in the article!

Now today I see that this happened to someone else, under pretty much the same circumstances, but using another tool Twinkle. So I'm thinking this is a bit more a problem than what tool is used.

Peek at this history listing starting with the revert. Mentality has pointed to the last good version "revision 125237527 by JoanneB" and started the revert. And nothing complains. But, the second bad edit by 67.128.112.59 sneaks in and remains after the revert is finished. A whole section of text got deleted without anyone the wiser.

In my case, in this history listing you can see the last one of three consecutive bad edits by 88.110.112.166 snuck in before my revert and stayed after the revert was done!

So... how do you know a revert hasn't worked, in that yet another vandal edit has snuck in? Do you have to do a "Compare selected versions" with the revert point and the result, just to see they aren't the same again/anymore? Shenme 05:05, 26 April 2007 (UTC)

It happens when the vandal is editing a section that isn't being changed by your revert, so usually that means they're editing a section they haven't edited before. I usually check the user's contributions after doing the revert just in case. I revert manually and it's happened to me as well - see my edit and this edit to Australian Security Intelligence Organisation which removed some vandalism but also added some in a different section of the article. Graham87 07:11, 26 April 2007 (UTC)

Hmm, but I thought Twinkle checked for reverts made during the revert process? In other words, even if they make a change whilst you're in the process of reverting their vandalism, it reverts both of them? And why does the wiki history display their vandal-edit before the anti-vandalism revert? I don't know I've just woken up :( Mentality 08:20, 26 April 2007 (UTC)

What happens is that the vandal makes a change to a section that isn't vandalised during the revert process, and the vandal's edit saves first. Then, when your revert saves, the server notices the edit conflict but thinks 'those edits are to different sections, I can avoid the conflict by just changing both sections the way the users wanted' and silently merges the edits, so the vandalism done while you're reverting isn't reverted. (This relies a lot on the exact timing.) Twinkle doesn't notice either because the edit-conflict warning never came up from its point of view. I hope that explains what's happening; I'm not sure how to avoid it (it's a case of two simultaneous changes which have to be placed in some order in the history!). --ais523 08:25, 26 April 2007 (UTC)

Ah, I get it now, thank you :-) Mentality 08:52, 26 April 2007 (UTC)

Indeed, pointing out that different sections are involved makes sense. Drat, more work. Thanks! Shenme 19:22, 26 April 2007 (UTC)

I have just done some further reading and experimentation and I have discovered that the above is either completely incorrect or there has been a change in MediaWiki of which I am not aware. A more likely explanation of the problem is server lag. help:revert#Reverts do not cause edit conflicts explains that reverts never cause edit conflicts. I have checked all the examples above plus three more I know of using the ISO 8601 date format (which is the only direct way of getting at the times of edits to the nearest second that I know of); it can be accessed in the date and time section of the preferences. Looking at the following diffs plus the ones provided above in ISO 8601 format should make things clear: , and . In all of these examples and the ones given above, the maximum time between edits is 6 seconds, which is most likely not enough time for all the servers to be alerted that an edit has taken place. Therefore, the problem is related to server lag, and has nothing to do with edit conflicts. Graham87 14:01, 27 April 2007 (UTC)

how do you delete and article you made?

meanmachine4242

If nobody else has edited the article since, log in to the account with which you made the article and place {{db-author}} on it, to request an admin to delete it. Otherwise, use WP:PROD or WP:AFD as usual. --ais523 11:03, 26 April 2007 (UTC)

User contributions

Any ideas why the "user contributions" link has disappeared when I visit people's user/talk pages? --Dweller 11:10, 26 April 2007 (UTC)

It's still there for me, and there's nothing obvious in your monobook.css or monobook.js that could be causing your problem. --ais523 11:16, 26 April 2007 (UTC)

Odd. It seems a problem for me with some users only. --Dweller 12:26, 26 April 2007 (UTC)

I take it that they are real users, rather than orphaned userpages? --ais523 12:29, 26 April 2007 (UTC)

Maybe it's alreay obvious, but the link is not shown un subpages. Tizio 12:34, 26 April 2007 (UTC)

User:Pupster21/monobook.js

Hello, it's me again again requesting a blanking of my monobook.js. However, this time I am requesting that an admin give me rc patrol admin-like revert script because I'm not sure I'm doing it right. --Pupster210 15:56, 26 April 2007 (UTC)

You seem to be attracting a stroke of bad luck, Pupster. I'll make a post on WP:AN/I. Valentinian 16:24, 26 April 2007 (UTC)

I've just blanked it and I'm giving you WP:TW. Cheers, alphachimp 16:30, 26 April 2007 (UTC)

I don't know how to use TW. --Pupster21 Talk To Me 17:21, 26 April 2007 (UTC)

Can you teach me or give me a simpler one? --Pupster21 Talk To Me 17:23, 26 April 2007 (UTC)

• WP:POP is pretty straightforward--VectorPotential 17:25, 26 April 2007 (UTC)

Add please. --Pupster21 Talk To Me 17:26, 26 April 2007 (UTC)

I'm not an admin, I can't edit your monobook (: --VectorPotential 17:27, 26 April 2007 (UTC)

... --Pupster21 Talk To Me 17:27, 26 April 2007 (UTC)

What you need to do is to go to your monobook and add the following code:

// ]
importScript('User:Lupin/popups.js');

You might wish to open the edit window for this section and copy-paste it directly. Alternatively, you can open your monobook and type:

{{subst:navpop}}

That should do the trick. Valentinian 19:04, 26 April 2007 (UTC)

If it is broken as in "it doesn't look right", you need to purge the page after you change it. Click here once you make a change to fix that. Even if you don't the code will still work as long as you clear your cache. Prodego 00:15, 27 April 2007 (UTC)

Security

I'm getting a lot of alerts that my information is being sent unencrypted, including from the Sign in / create account page. Someone could easily intercept my password when I log in, especially if I run an unsecured wireless network, which most people do. This needs to be fixed ASAP. --Ideogram 17:42, 26 April 2007 (UTC)

If you want secure login, you can use the secure Misplaced Pages interface. Login is unencrypted by default. (I'm not sure why the situation is as it is.) --ais523 17:47, 26 April 2007 (UTC)

Oh my god, that is so wrong. Ideogram 18:21, 26 April 2007 (UTC)

I use my insecure password for wikipedia. Misplaced Pages accounts were never thought to be exceedingly important. --Kim Bruning 18:28, 26 April 2007 (UTC)

So creating socks is no big deal? Ideogram 18:39, 26 April 2007 (UTC)

You can make non-abusive sock puppets already. If your password was stolen, someone would be able to use your account. For most people, this isn't worth doing - anyone can edit Misplaced Pages already! Admins, etc. should be a bit more careful. --h2g2bob 18:49, 26 April 2007 (UTC)

Why are group and role accounts discouraged? Why couldn't I "accidentally" lose my password? Ideogram 19:05, 26 April 2007 (UTC)

Because it's a violation of WP:GFDL and therefore a copyright violation. If more than one person uses the same account, Misplaced Pages cannot attribute the edits to one particular person. Corvus cornix 20:13, 26 April 2007 (UTC)

I think there's an inconsistency in your policies here. --Ideogram 20:27, 26 April 2007 (UTC)

In what respect? Corvus cornix 20:32, 26 April 2007 (UTC)

citation coding

Hello, I have added a source citation to the RNC 2004 article, but I can't figure out how to code it correctly. Currently both citation links are pulling up only one of the actual citations (you click the little super text number and it is supposed to take you to the full information for that citation, but its taking you to the same citation information for both instead of taking you to the correct information about each separate citation).

I hope that makes sense.

Here's what I'm talking about at the bottom of the Police Tactics section on the following page... http://en.wikipedia.org/2004_Republican_National_Convention_protest_activity

You'll notice that the 44 citation is listed twice. I couldn't figure out how to make the second citation appear as 45. If you look at the code for the page all the info both cited articles is listed there, but the second citation says 44 not 45. Also, how do you make that appear below the article in the list of citations.

I honestly did look around a lot on the help pages, but I didn't find anything that tells exactly how to do the coding to make a new citation appear correctly with the article and also at the bottom in th list of citations.

I'd appreciate any help on this. It would be nice to be able to learn how to do it, rather than just some one fixing it for me. But, I'll take what I can get I guess.

Thanks!— Preceding unsigned comment added by Misterman8 (talk • contribs)

Man, the citations on that page are a mess. I think I fixed it though. Try looking through WP:CITE, WP:FOOT and WP:CITET.↔NMajdan•talk 19:54, 26 April 2007 (UTC)

Location map dot varies location between Firefox and IE; one correct, one not.

{{Infobox City}} has a provision using the template {{superimpose}} to manually superimpose a dot atop of a blank map. When I first tested it, it seemed to work great. However, now on certain pages there seems to be a glitch when viewed using Internet Explorer (WinXP), but Firefox (MacOS 10.4 and WinXP) is fine. Here is an example—Airdrie, Alberta. I thought that maybe it was the map itself (Image:Division No. 6, Alberta Location.png) but Didsbury, Alberta uses both this map and Infobox City (aka Infobox Settlement) and there doesn't seem to be a problem or a noticeable difference in dot placement. As a control, Langdon, Alberta uses this map and the {{superimpose}} template but doesn't use infobox city and it seems to be fine. Can anyone explain this and is there a solution for this? —MJCdetroit 19:49, 26 April 2007 (UTC)

Looks fine in IE6, but not in Firefox 2 on WinXP. I'm pretty sure it's because the image is centered in the table cell. See this table with exaggerated width:

We need a way to make the surrounding <div> wrap around the image only, and not the entire table cell. –Pomte 01:19, 27 April 2007 (UTC)

In your preview, when viewed in FF 1.5.0.11 and Safari Version 2.0.4 for Mac OS, the dot is about 2½ inches left of the shaded area. I'll preview it tommorrow in WinXP and IE & FF when I get to work.

It looks like the author of Template:Superimpose may have had a problem with IE and png images before. I would have asked him if this is the problem here, but Papayoung is currently inactive. —MJCdetroit 02:31, 27 April 2007 (UTC)

At least in this example, the issue is not with the type of image but with CSS layering. One way to solve it is by specifying the width of the <div>:

I've added a removable red border to make it obvious what has changed. Someone with more knowledge of CSS may have a better idea. –Pomte 03:23, 27 April 2007 (UTC)

Some localised infoboxes use {{Location map}}, which does not contain superimpose. I haven't seen any faults when viewing that with Firefox of IE. Maybe the difference between the two template codes could reveal something. - 52 Pickup 13:11, 27 April 2007 (UTC)

Thumbnail problem

I have uploaded better (cropped) versions of some pictures to Commons recently. The problem is that the relevant Misplaced Pages article (Tadley) is displaying thumbnails of the earlier versions, though these are in the aspect ratio of the latest versions. Clicking on the thumbnails does correctly display the latest versions. Purging the image description page cache (both WP and Commons) has no effect. Alan Pascoe 20:29, 26 April 2007 (UTC)

Have you tried purging the page itself? Also, the job queue length is currently at over a million, so some image tasks may be delayed. Nihiltres 00:12, 27 April 2007 (UTC)

Password restrictions changed

Due to an attacker mass-abusing accounts with weak passwords, passwords that are the same as the username can no longer be used. Affected accounts can reset their password by e-mail to something more secure. --brion 21:55, 26 April 2007 (UTC)

I've tested r21640 locally. Such accounts are in fact blocked from logging in, however if the attacker has obtained a session cookie for any remaining user accounts with eponymous passwords (obtained prior to the software change), he could still edit with those accounts by providing old (yet non-expired) cookies. —freak(talk) 23:17, Apr. 26, 2007 (UTC)

What about users that didn't specify an e-mail address? --24.178.78.17 01:28, 27 April 2007 (UTC) (was User:All in )

Can you not log in at all? --Deskana (fry that thing!) 01:30, 27 April 2007 (UTC)

No, I get the error message "Login error: Incorrect password entered. Please try again." when attempting to log in. --24.178.78.17 01:35, 27 April 2007 (UTC)

I think you need to register a new account. I was going to suggest usurpation, since the legal issues don't exist if you're the account holder, but you don't really have any way of proving that you are. --Deskana (fry that thing!) 01:36, 27 April 2007 (UTC)

Annoying image preview issue

It used to be the case that you could only click on an image on its image page if it had a higher-resolution version than that shown in the preview. If the preview (the image embedded in the image page) was at the maximum size, you couldn't click on it, which gave the user a clue (without having to read the full and preview dimensions) that you could view a bigger version. Now the image is always linked, so you often end up clicking on it to see a bigger version, but it's just the same. The new behavior is annoying. Is it a bug or a feature? Example of new behavior: Image:Raunkiaer.jpg; it used to behave like this: . --TotoBaggins 05:14, 27 April 2007 (UTC)

I think the caption makes it pretty clear:

Size of this preview: 600 × 406 pixels

Full resolution (600 × 406 pixel

In Special:Preferences -> Files, you have a preference for the maximum image size shown. If an image is smaller than that limit, then that's the original size. –Pomte 06:59, 27 April 2007 (UTC)

Image:Cricket World Cup trophy.jpg

I cropped this image from the original because it was off-centre. Now it seems to be displaying a stretched and distorted version of the original image when it displays at 100 and 200px. But at 80 and 120px it displays correctly. See the image page for sample displays. Is this a server purging issue, and if so why won't it fix? I'm seeing the same thing from IE as well as FF and a good 10 minutes has elapsed, so I presume its not in my browser cache. Any help appreciated. —Moondyne 06:34, 27 April 2007 (UTC)

The original uploader has now reverted the image back to the off-centre version (which is fine), but then the 80 and 120px thumbnails changed to centred versions with distortion. And now when I look, all thumbnails have changed to off-centred with no distortion. I'd love to know the answer to this, please. —Moondyne 07:28, 27 April 2007 (UTC)

Sometimes when you upload an image, it loads the new image but not the dimensions of the new image, causing the new image to be displayed exactly the same size as the old image. I think this is more of an issue with your browser cache than it is with MediaWiki software. A forced refresh normally fixes the problem, which probably explains why when you came back to it later, it showed it correctly. --Deskana (fry that thing!) 12:28, 27 April 2007 (UTC)

Reducing image file size

Hello. I have been cleaning up some images. In some cases I find images that I can reduce the file size for but with little visual change. For example, optimising a PNG image with optiPNG. I have been replacing images if the file size is approximately halved. This will conserve bandwidth, and make the image more accessible to people on slow connections, but fill up the server with potentially unnecessary files.

Is it worth optimising the image and uploading a new version? If so how much difference makes it worthwhile?

I think in most cases the server reprocesses the image before sending it to the browser. This may make optimization irrelevant, but I wouldn't like to say for sure. Notinasnaid 07:18, 27 April 2007 (UTC)

Yes, images that are resized lose the advantage of optimisation. In some cases a smaller resized version may have a bigger file size than the large original! So optimisation may only be worthwhile for viewing large originals, but is it even worth it for them?Iain 07:33, 27 April 2007 (UTC)

Odd internal error message

I just tried to delete an image and got this message:

There seems to be a problem with your login session; this action has been canceled as a precaution against session hijacking. Please hit "back" and reload the page you came from, then try again.

This is the first time I've got this - is it happening to anyone else? If not, what's the reason it might have happened to me? – Riana _ऋ 12:15, 27 April 2007 (UTC)

Have you ever had the "loss of session data" error when trying to edit? It may be related to that, but that's juts a hunch. Are you still getting the error? --Deskana (fry that thing!) 12:21, 27 April 2007 (UTC)

I do get "loss of session data" every so often, but that seems to fairly common - I've never had this one before! And I've got it twice since. – Riana _ऋ 12:27, 27 April 2007 (UTC)

Either way, I've changed my password just in case. – Riana _ऋ 12:32, 27 April 2007 (UTC)

I wouldn't think changing your password is necessary, this seems to be more of a software problem than anything else. Your account is as likely to be secure as mine is, I think. --Deskana (fry that thing!) 12:34, 27 April 2007 (UTC)

Well, it sure didn't help - I just got it again. Grrr – Riana _ऋ 12:37, 27 April 2007 (UTC)

Presumably what's happened is that it's got confused about the cookies and thinks you're trying to hack your own account? (By the way, I don't think the password's got anything to do with the seession data.) --ais523 12:49, 27 April 2007 (UTC)

That's strange. I got logged out a few times too. – Riana _ऋ 12:55, 27 April 2007 (UTC)

Edit links bunched, floating doesn't help

See Template:Infobox Skyscraper and Circle on Cavill. I changed the template to add float: right; clear: right;, but it's not working. I purged both pages several times as well. I also checked out Misplaced Pages:How to fix bunched-up edit links, but it just said to do float: right; clear: right; which I had already done, and putting a section clear I don't think will work, and the div solution doesn't seem needed here, as it's an infobox and this problem doesn't exist elsewhere. Any help will be appreciated. MECU≈talk 12:54, 27 April 2007 (UTC)

Nevermind, it was the image. MECU≈talk 12:57, 27 April 2007 (UTC)

A little help perhaps?

I apologize for making a personal request, but… I just would like to request a little help. IF you are not busy doing something constructive, and you know a bit about wikicoding, could you please take a look at my user page, and tell me why my stupid menu bar is refusing to stay up at the top, and continues to hide at the bottom where no one can see it? Thanks! N i g h t F a l c o n 9 0 9 0 9' 13:21, 27 April 2007 (UTC)

Silly question but I need answer

Yesterday I could not find the Tecnical Pump!

I need to go on a Wikibreak as quickly as possible and want to know how to do it -- where to get the banner, etc. I also need to have my pages protected while I am gone, as there isa person who blanks parts of them. If I am not around to guard my pages, I cannot protect them myself. Is this possible? Thank you? --Mattisse 13:23, 27 April 2007 (UTC)

The fastest way to find this Village Pump is to type WP:VPT into the search box and press Return. To 'go on wikibreak', type {{wikibreak}} at the top of your talkpage (and userpage as well if you wish); this will add the banner. You can request an administrator to protect your userpage/talkpage at WP:RFPP, but normally this wouldn't be done (normally RC patrollers would revert vandalism to people's userpage in their absence); in particular, usertalk pages are rarely protected because people might need to send you messages (which you would then be able to read when you came back from Wikibreak). Hope that helps! --ais523 13:29, 27 April 2007 (UTC)

That you so much for you answer. I do not need my talk page protected. It is my user page and other person pages. A person has been blanking selected parts of them. Vandal patrols have not noticed nor have they reverted them. I have had to go through each page, look at the history and try to restore the blanked sections. Do you have a suggestion? (I have considered saving all such pages in a text file on my desk and then just replacing my user pages with the text file when I return. That is all I can think of to do.) Sincerely, --Mattisse 13:56, 27 April 2007 (UTC)

If a particular user is targeting you for vandalism, you could let administrators know at WP:AN/I; you could request for your pages to be protected or the user in question to be blocked there, if it's a big enough problem. (If other people don't edit the pages in question, you could just revert them to the pre-wikibreak version when you come back from a wikibreak; Misplaced Pages saves the old version for you so you don't have to save it yourself.) Hope that helps! --ais523 13:59, 27 April 2007 (UTC)