User talk:Simetrical

User talk:Simetrical/Archive

You asked me to unprotect the template so you could make some changes to it. Just reminding you that I did reduce it to semi-protection, so whenever you get around to making your changes... Later. — Jul. 1, '06 [13:58] <freak|talk>

I received a message from you that I edited the Colgate University page, and the edit was vandalism. I must let you know, however, that I was incorrectly identified through the IP address. All computers on the Colgate networked are masked under a small number (less than 10) IP addresses, so it could have been any other of the 3500 computer users on this campus. Not your fault, but I just wanted to let you know.

Hello, the law page looks a bit odd with a long blank space, because there's a short introduction and lots of content categories. Surely that counts as a good reason to deviate from the wikimanual you referred to when responding to the other person's complaint about this in August. I imagine wikimanuals are there for guidance on the more wild and undeveloped pages, where people fiddle with the style all the time and do little with substance. Best wishes. User:Wikidea

Hello, I've investigated a problem with the display of the [edit] links (next to the title of each paragraph) when there is more than one object (usually image) with "float" parameter in the preceding paragraph. In this case (using FF) the [edit] link does not appear on the same height as the title of the paragraph - but on the same height as the lowest floating preceding object - which apparently is the exact behavior specified in CSS 2.0. I think I have a solution, and I came to talk to you about it in advise of user:rotemliss on the Hebrew wikipedia - who said you take care of this aspect. The solution is pretty simple, instead of having the editsection CSS (on /skins-1.5/common/shared.css) defined as -

.editsection {
	float: right;
	margin-left: 5px;
}

set it to -

.editsection {
    display: block;
    text-align: right;
    position: relative;
}

("text-align: left;" on RTL projects). In this setup the [edit] link would always be in-line with the paragraph title, and lined to the top of the title - this works good on FF, and IE as far as I've checked, so if there are nothing I've missed I don't see a reason why not to replace it. תודה, Costello (talk) 20:27, 4 January 2008 (UTC)[reply]

This was tried, I reversed it. I observed distinctly different appearance in other browsers (e.g., old versions of IE). Specifically, it shifts up by somewhat different amounts in different browsers, whatever you shift it up by. (I assume you left out the "top: 1.2em" or similar statement.) For discussion of the issue, you can read bug 1629. I do plan to fix this, the question is only when. —Simetrical (talk • contribs) 04:27, 6 January 2008 (UTC)[reply]

...to the next New York City Meetup!

New York City Meetup Next: Sunday January 13th, Columbia University area Last: 11/3/2007 This box: view • talk • edit

In the morning, there are exciting plans for a behind-the-scenes guided tour of the American Museum of Natural History.

In the afternoon, we will hold a session dedicated to discussing meta:Wikimedia New York City issues (see the last meeting's minutes).

In the evening, we'll share dinner and chat at a local restaurant, and (weather permitting) hold a late-night astronomy event at Columbia's telescopes.

You can add or remove your name from the New York City Meetups invite list at Wikipedia:Meetup/NYC/Invite list.
This has been an automated delivery by BrownBot (talk) 01:29, 5 January 2008 (UTC)[reply]

Hi there. Hope the little debate we had yesterday wasn't too exhausting! :-) I was wondering if you would be able to point me towards the actual revision that implemented rollbacker? The impression I get from the "shell" stuff is that the switch is in some en-wiki specific thing, and that globally (for all the projects) the ability to turn on this new user right was available earlier. One of the points (I think) is that this software change wasn't available at the time of the earlier poll. I do remember seeing some talk somewhere about how some clunky patch that was used to change user rights had been made redundant by some change - was it this change that also allowed the rollbacker thing to go ahead? Any pointers would be much appreciated. Carcharoth (talk) 13:22, 10 January 2008 (UTC)[reply]

There are two separate concepts here that a lot of people get confused. One type of access you can have is to commit changes to the Subversion repository. This repository contains almost all the code used for running the site. Changes made here do not take effect immediately. Users with just commit access cannot do anything at all to the servers by themselves.

The other type of access you can have is shell access, i.e., more or less full access to the servers (possibly restricted somewhat so that you can't seriously disrupt the system). Shell users can run programs and store files and so on on the servers if they want, but the more relevant thing is that they are given the permission to change various configuration settings, they can update the code that's running on the servers to the latest version from Subversion (or any other version), and some (maybe all, I don't know, but at least all the roots) have direct access to the database. These things are not kept in Subversion, cannot be changed by people with only commit access, and are not publicly viewable. Therefore there is no revision to link you to.

I refer to developers and sysadmins separately, by the way, because the two groups are not only quite different in their rights, but they don't fully overlap. Sysadmins like JeLuF may have commit access, but in practice they never use it. They might not be familiar with the MediaWiki code, they might not even be familiar with programming at all beyond simple scripts. They therefore can't add new features to the software, and perhaps more importantly, they can't review things like new extensions. The two main sysadmins who are also developers are Brion Vibber and Tim Starling. (Other active or semi-active sysadmins can also write code, like Domas and River, but they don't presently tend to add new features to the software, or review them.)

As for the software change: up until a few months ago, there was no reasonable way for user-rights assignment to be given out on a modular basis. Either you were a steward and could change all rights however you wanted on Special:Userrights, or you could change nothing. When, in days of yore, it was perceived this was a little inflexible, someone coded up a totally different page, Special:Makesysop, and gave it a special right so that bureaucrats could use it. Later, stewards were having to field an excessive number of bot addition/removal requests, and there was quite a bit of lag involved (this was true until maybe a year ago, I think), and so Rob Church coded up yet another entirely different special page, Special:Makebot. What this meant is that if you wanted to create a rollbacker group, you would have to get a developer to write an entire extension for you, and it would have to be checked over and enabled by one of the two active developer/sysadmins I mentioned for reasonableness.

So several months ago, I changed Special:Userrights so any group could be easily configured to have limited access to it, for instance only to grant sysop, bureaucrat, and bot and to remove bot. None of the sysadmins used it that I know of, however, until a short while ago, when Werdna improved it further (by adjusting interwiki rights-granting, etc.). Now it is possible for any group to be given the right to add or remove any other group, and quite a few long-standing shell requests can be fulfilled (plus Makesysop and Makebot will be obsoleted once the Userrights interface is cleaned up a bit more). So when this was previously discussed, no, it could not have been implemented. It's only possible within the past few weeks. —Simetrical (talk • contribs) 19:06, 10 January 2008 (UTC)[reply]

Wow! Thanks for all that. I feel like I know how things work around here now! :-) I knew some of this already, but some of it I didn't know, or had forgotten. You may not be aware, but Doc Glasgow (following a suggestion by Jimbo) has filed an arbitration case to clear up the issue of consensus. I'm not sure how clued-up the Arbitration Committee are on what you just said above, so maybe you might want to post something over at WP:RFARB to provide the background? Just add a statement and type away (not sure if you know how that system works). Carcharoth (talk) 01:47, 11 January 2008 (UTC)[reply]

I have participated in ArbCom proceedings before on occasion. I imagine that there are people on the ArbCom who understand well enough how things work, and if not I'm sure they'll have the sense to ask if they're unsure. —Simetrical (talk • contribs) 17:59, 11 January 2008 (UTC)[reply]

As you made an edit to the incident listed in the Administrators notice board, it is requested that you confirm the details of the incident here (section 1.1.2)

This is as the incident is used as the basis of an argument and needs to be confirm by persons familar with the event

Regards --User:Mitrebox talk 2008-02-22 —Preceding unsigned comment added by 70.11.244.78 (talk) 07:57, 22 February 2008 (UTC)[reply]

There has been a mailing list created for Wikipedians in the New York metropolitan area (list: Wikimedia NYC). Please consider joining it! Cbrown1023 talk 21:39, 22 February 2008 (UTC)[reply]

New York City Meetup Next: Sunday March 16th, Columbia University area Last: 1/13/2008 This box: view • talk • edit

In the afternoon, we will hold a session dedicated to meta:Wikimedia New York City activities, and have salon-style group discussions on Wikipedia and the other Wikimedia projects (see the last meeting's minutes).

Well also make preparations for our exciting Wikipedia Takes Manhattan event, a free content photography contest for Columbia University students planned for Friday March 28 (about 2 weeks after our meeting).

In the evening, we'll share dinner and chat at a local restaurant, and (weather permitting) hold a late-night astronomy event at Columbia's telescopes.

You can add or remove your name from the New York City Meetups invite list at Wikipedia:Meetup/NYC/Invite list.

You're also invited to subscribe to the public Wikimedia New York City mailing list, which is a great way to receive timely updates.
This has been an automated delivery because you were on the invite list. BrownBot (talk) 03:32, 4 March 2008 (UTC)[reply]

The convention that the Chess Wikiproject uses is to put the "chess notation" tag at the top of the article. Bubba73 (talk), 15:32, 17 April 2008 (UTC)[reply]

The literal string "<ref>" could legitimately occur in a reference's text.^{[citation needed]} --Random832 (contribs) 03:37, 24 April 2008 (UTC)[reply]

I didn't say it was probable on the English Wikipedia. There's no intrinsic reason it's nonsensical, however. It might be worthwhile to assume it's not intended, however, since it seems to be much more likely in most circumstances that you left off the ending tag, I agree. —Simetrical (talk • contribs) 16:04, 24 April 2008 (UTC)[reply]

Well, the other thing is, there's a difference between <ref> showing up unescaped, and &lt;ref&gt; - and since you can't have references within references (arguably a bug), you should really be using the latter anyway. --Random832 (contribs) 16:20, 24 April 2008 (UTC)[reply]

Yes, that's true. If <nowiki>, <pre>, and entity escaping all worked to avoid the error message, it would probably be fine. I even once wrote a patch to that effect and attached it to a bug somewhere, before I got commit access. It might still work today (but probably not, with the addition of groups). —Simetrical (talk • contribs) 17:14, 24 April 2008 (UTC)[reply]

Template:FootnotesSmall has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for Deletion page. Thank you. — Rockfang (talk) 08:01, 15 May 2008 (UTC)[reply]

New York City Meetup Next: Sunday June 1st, Columbia University area Last: 3/16/2008 This box: view • talk • edit

In the afternoon, we will hold a session dedicated to meta:Wikimedia New York City activities, elect a board of directors, and hold salon-style group discussions on Wikipedia and the other Wikimedia projects (see the last meeting's minutes).

We'll also review our recent Wikipedia Takes Manhattan event, and make preparations for our exciting successor Wiki Week bonanza, being planned with Columbia University students for September or October.

In the evening, we'll share dinner and chat at a local restaurant, and (weather permitting) hold a late-night astronomy event at Columbia's telescopes.

You can add or remove your name from the New York City Meetups invite list at Wikipedia:Meetup/NYC/Invite list.

Also, check out our regional US Wikimedia chapters blog Wiki Northeast (and we're open to guest posts).
This has been an automated delivery by BrownBot (talk) 00:39, 20 May 2008 (UTC)[reply]

There are several proposals for changes to wikibits.js, here. Could you offer some guidance on how best to prepare the proposals for submission to bugzilla? Thanks! SharkD (talk) 21:48, 18 August 2008 (UTC)[reply]

File one bug per issue. Make sure patches are in unified diff format against the current version of wikibits.js. Ideally, prepare patches by checking out a copy of MediaWiki (or just phase3/skins/common/ if you like) and using the command "svn diff", or "Create patch" (something like that) in TortoiseSVN. Attach the patch to the bug report, and give an explanation of what it does and why in the report itself. If you want to commit performance improvements, make sure you know that they're actually a performance improvement. Set the "assignee" to Simetrical+wikibugs@gmail.com. —Simetrical (talk • contribs) 23:21, 18 August 2008 (UTC)[reply]

If I submit them all at today will it cause problems? I mean, will the software be able to apply several patches that all branch from the same, initial version? Or, should I wait until the first patch has been accepted before submitting the next? SharkD (talk) 00:59, 19 August 2008 (UTC)[reply]

As long as multiple patches don't change the same lines of code, or immediately adjacent ones, they should apply cleanly. If two changes do change the same lines of code, consider submitting them as one patch in one bug report, even if they're unrelated. Alternatively, just submit them separately, it shouldn't be hard to manually resolve any conflicts. —Simetrical (talk • contribs) 15:21, 19 August 2008 (UTC)[reply]

OK, thanks. SharkD (talk) 16:52, 20 August 2008 (UTC)[reply]

Hey, could you take a look at the discussion at Wikipedia:Village pump (technical)#nstab-main and tell us what you think? —Remember the dot ^(talk) 06:11, 7 September 2008 (UTC)[reply]

Alright, if you're not interested in the discussion that's fine. Would you consider taking a look at bugzilla:15507, at least? I made a patch to fix the issue, but no one has looked at it. —Remember the dot ^(talk) 20:55, 12 September 2008 (UTC)[reply]

Go to BetaWiki or someplace and talk to the Spanish and French translators. I don't know either language and am not going to commit changes to their localizations. —Simetrical (talk • contribs) 21:10, 12 September 2008 (UTC)[reply]

Thanks for pointing me in the right direction! I see you've beaten me to closing bug 15507, so thanks doubly. —Remember the dot ^(talk) 23:14, 14 September 2008 (UTC)[reply]

I didn't close it, Siebrand did. —Simetrical (talk • contribs) 00:15, 15 September 2008 (UTC)[reply]

Whoops. You're right. Thanks anyway. —Remember the dot ^(talk) 23:29, 15 September 2008 (UTC)[reply]

In re this, your patch r40499 is live now and works good. I was able to view my test page and watch the cat counts get corrected (with some judicious null-edits). At least, it sure seems that way!

Did you go any further with the idea of a job to recount all categories? That seemed like a strictly server-side function, which is certainly beyond my scope. Seems like a good idea though... Franamax (talk) 01:25, 24 September 2008 (UTC)[reply]

You shouldn't need null edits. Just viewing the category page should do it. As for the job, really all that would be needed is for the sysadmins to regularly run populateCategory.php with a very low value of maxlag and/or a high throttle (and with --force so it repopulates even though the table is populated already). The script's already there, but I can't make it run, since I only have commit access, not shell access. —Simetrical (talk • contribs) 14:15, 24 September 2008 (UTC)[reply]

Thanks! Viewing the cat page (I'm confident) updates the DB table. It seemed as though the null edits were needed to force an update on the page containing the PAGESINCATEGORY parser function. Would your count-update patch force updates onto the job queue for pages containing the parser function? I'm guessing no, since I don't see any way to obtain backlinks for parser functions - so it seems to me, the function-containing page will remain static. Though I'm now confused - if PIC is on a page, does that mean the page is dynamically regenerated every time it's accessed?

I did see a bit of tricky code UltraExactZZ uses to force an update on rendering his upage, see "fullurl:Category:Spam pages for speedy deletion|action=purge" therein - but I'm still confused on whether the page is dynamically rendered - does this mean UE's page never hits cache, by virtue of the PAGESINCAT function?

My confusion aside, who would be a good admin to push about setting up the job you describe? At least getting a one-time run? I see Brion has had only 17 or so server crashes to deal with in the last few days, would that be the first stop?

And I may try to update some doc's to reflect the latest changes - would you like to see diff's of same? Franamax (talk) 00:21, 25 September 2008 (UTC)[reply]

PAGESINCATEGORY just isn't updated until the page is reparsed. There are no backlinks, as you point out, so it's not possible to figure out which pages to regenerate on update; even if it were, invalidating the parser cache of bajillions of pages every time something is added to/removed from a category isn't really acceptable. The parser function is always going to be lagged, possibly by days, so it's not really useful for fine-grained tracking. For that purpose, visit the category page, or use a script.

You might try talking to Brion or Tim if you want the script run, I suppose. —Simetrical (talk • contribs) 01:09, 25 September 2008 (UTC)[reply]

Template:!vote has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for Deletion page. Thank you. Thinboy00 @076, i.e. 00:49, 3 December 2008 (UTC)[reply]

Regarding your recent edit to quaternion, I'm wondering whether I'm understanding it correctly; maybe it's just a terminology issue, but since octonions, quaternions, and complex numbers all have the reals as proper subrings, there should be three division rings (and not two, per your edit)? Thanks, Jens Koeplinger (talk) 02:30, 15 December 2008 (UTC)[reply]

Ah - just noticed that associativity of multiplication is typically part of a ring. Never mind. Thanks, Jens Koeplinger (talk) 02:38, 15 December 2008 (UTC)[reply]

I just asked "so what are the other two?" when I read that sentence, clicked through to Frobenius theorem (real division algebras), and added the one other nontrivial one listed there. Unless I'm grossly misunderstanding that article, if I'm wrong then it's wrong too. —Simetrical (talk • contribs) 02:44, 15 December 2008 (UTC)[reply]

Hi - the Frobenius theorem article is correct, it applies to associative division algebras. The octonions, in contrast, are a non-associative, finite-dimensional division algebra; and as I read here in Wikipedia, associativity of multiplication is typically included in the definition of a "ring", but it is also acknowledged here that alternate (though apparently less common) definitions exist, where associativity of multiplication in a ring is not required. That's what I had in mind. So, it's all good, no need to point this out in the quaternion article, I think. It was good for me to learn, though. Thanks, Jens Koeplinger (talk) 20:15, 15 December 2008 (UTC)[reply]

Hello, Simetrical. Based on the templates on your talk page, please consider joining the Article Rescue Squadron. Rescue Squadron members are focused on rescuing articles from deletion, that might otherwise be lost forever. I think you will find our project matches your vision of Wikipedia. You can join >> here <<. Ikip (talk) 15:34, 24 February 2009 (UTC)[reply]

Hi Simetrical, I just performed a massive update of {{pf}} that ended up deprecating {{cpf}}, so I redirected the latter. At some point after logging in, I intend to create a similar template for magic words (unless such a template already exists; I haven't searched for one yet), and I plan to tweak {{pf}} some more as well. Any thoughts? BTW, if you reply, could you please leave a {{tb}} on my talkpage? In all likelihood, I won't remember this when you do answer, or at least where it is. =P --Dinoguy1000 as 66.116.12.126 (talk) 08:32, 28 February 2009 (UTC)[reply]

I have no idea why you're saying this to me (I guess I contributed to {{pf}} at some point?). Do whatever you want. If you don't remember to check back here, that's just as well, so I haven't left a {{tb}}. —Simetrical (talk • contribs) 01:11, 1 March 2009 (UTC)[reply]

According to its history, you created it (albiet in 2006). I just thought you might appreciate the heads-up, even if only for curiosity's sake, and even after such a length of time. And, while I'm sure you didn't intend it, your answer sounded more than a bit irritable (not to mention the possible consequences of a developer telling an anonymous IP address to "do whatever you want" =) ). --Dinoguy1000 as 66.116.12.126 (talk) 05:24, 1 March 2009 (UTC)[reply]

Yes, sorry, I was a bit confused. I vaguely remembered having something to do with it, but I'm long past any desire to fiddle with Wikipedia's templates. Thanks for the heads-up anyway. —Simetrical (talk • contribs) 12:46, 1 March 2009 (UTC)[reply]

No problem. In retrospect, though, considering the age of the templates, I probably should have clarified why I was notifying you, specifically. And, if you don't mind my asking, what turned you off to working on Wikipedia templates? --Dinoguy1000 as 66.116.12.126 (talk) 03:22, 2 March 2009 (UTC)[reply]

I have better things to do. Like schoolwork, running a decent-sized website, and developing MediaWiki. —Simetrical (talk • contribs) 06:06, 2 March 2009 (UTC)[reply]

Aah, of course. =) Anything interesting or unusual happen on any of those fronts as of late? --Dinoguy1000 as 66.116.12.126 (talk) 08:01, 2 March 2009 (UTC)[reply]

You can check WP:SIGN for MediaWiki development news. —Simetrical (talk • contribs) 15:02, 2 March 2009 (UTC)[reply]

Hi, be careful with quoting claims that are apparently from Eben Moglen. Moglen frequently publishes claims in the semi-public that are contradicting his view as lawyer. The same happened with me for cdrtools. I have a private statement from Eben Moglen that there of course is no license problem in cdrtools. After Moglen send me this statement, he aborted his license review for Shuttleworth. Please correct the cdrtools page.... Schily (talk) 12:44, 11 March 2009 (UTC)[reply]

I provided what I believe to be a reliable source for the citation. My source is an official publication of the Ubuntu Technical Board that indicates that they believe they were acting under Eben Moglen's advice. If you have another reliable source that contradicts that assertion, please provide it. If you have unverifiable personal knowledge, that's not relevant to Wikipedia's discussion of the topic. Only reliable sources that can be verified by our readers are acceptable for consideration in writing Wikipedia articles. If you do indeed have such a communication from Eben Moglen, I suggest you ask him to make a public statement on the matter that can be cited in the article. —Simetrical (talk • contribs) 14:13, 11 March 2009 (UTC)[reply]

You do not have a reliable source but POV claims from some people only. It is simple for me to create a page that claims the opposite to what you found, the current claim is therefore not verifiable. POV statemens do not belong on WP, please remove it. Schily (talk) 14:23, 11 March 2009 (UTC)[reply]

Except that any page you create stating an opposite would fail WP:RS and thus would be unusable on Wikipedia. Simetrical's source is reliable because it's a statement (or rather, a meeting transcription) from Ubuntu concerning a stated fact about them - therefore, that fact is verified. There is no issue here, you appear to simply be trying to cause trouble. 「ダイノガイ千？！」^{(Dinoguy1000)} 18:45, 11 March 2009 (UTC)[reply]

WP:AGF, please. I'm sure he believes he's correct and is trying to improve the accuracy of the article. —Simetrical (talk • contribs) 22:30, 11 March 2009 (UTC)[reply]

My source clearly states the POV of a particular organization. "cdrtools cannot be legally distributed" would be a POV statement; "Ubuntu believes that cdrtools cannot be legally distributed" is NPOV. If you can find other notable parties that believe cdrtools can be legally distributed, you can feel free to add those to the article. I'm not aware of any significant ones other than Schilling himself. —Simetrical (talk • contribs) 22:30, 11 March 2009 (UTC)[reply]

All openminded parties believe that cdrtools can be legally distributed. This includes e.g. Sun (Sun legal did do an in depth legal review for cdrtools last Autumn and Sun lagel did give an OK) Slackware and others. I am not sure about the intention of Ubuntu as Mark Shuttleworth did break his promise about including cdrtools in Ubunto after Eben Moglen aborted his legal review for cdrtools. It is unlikely that Ubuntu does this because they believe, there is a legal problem; Ubuntu still happily distributes other software with definite legal problems. Cdrkit, GNUvcdimager and libcdio are all based on code from cdrtools and they are all in conflict with GPL and the Copyright and thus cannot be legally distributed. Ubuntu is informed about this problem but ignores it.

Besides Ubuntu, we curently mainly have two companies that only like to ship the undistributable "cdrkit": SuSe still "believes" that their customers need all the bugs in wodim that prevent it from being used and that their customers would not accept a working original cdrecord. RedHATs role is not clear (RedHAT maintainers told me that RedHAT managers forbid to include CDDL code into REdHAT but it seems that RedHAT recently upgraded star to a CDDL based version).

Eben Moglens legal opinion is that there is no license problem in the original cdrtools. He intentionally did not publish an "advise" not to publish cdrtools because this seems to be an opinion from someone else - guess who :-) If Moglen did publish information from private discussions I would publish our private mail discussion without asking him to prove his real opinion. Publishing leaked private discussions taken from the Ubuntu IRC is not a good idea as the related content is in conflict with the agreement of most parties that there is no legal problem with distributing the original software. Schily (talk) 16:31, 12 March 2009 (UTC)[reply]

If Sun and Slackware have officially stated that they believe cdrtools to be distributable, or if they actually distribute it, you should add that fact to the article, with citations. "All open-minded parties" is not a NPOV description and cannot be put in the article. —Simetrical (talk • contribs) 21:39, 12 March 2009 (UTC)[reply]

If POV cannot be in an article, why is there POV text in the article? All other facts I mentioned including the fact that Ubuntu does not care about legallity when publishing software are not POV and the latter fact would help to rate other claims. Schily (talk) 22:30, 12 March 2009 (UTC)[reply]

The statement "Ubuntu does not care about legality when publishing software" is not neutral. If you don't understand this, then I'm afraid there's not much point in discussing this further. —Simetrical (talk • contribs) 23:57, 12 March 2009 (UTC)[reply]

Your current text in the article is not neutral with respect to the cdrtools project - it rather reports about the POV from Mark Shuttleworth and I believe that POV statements do not belong into this WP article. You may add this to the article about Shuttleworth if you like to add things just for curiosity. Do you have a problem to understand that Ubuntu is currently in a self contradicting state (they claim that they "cannot" publish cdrtools because of supposed license problem but at the same time they happily publish software that definitely has license problems -> the fork, libcdio and vcdimager)? Schily (talk) 11:10, 13 March 2009 (UTC)[reply]

As I said, it is clear by now that there is (unfortunately) no point in continuing this discussion. —Simetrical (talk • contribs) 16:49, 13 March 2009 (UTC)[reply]

Hi Simetrical
Since I've read you'll be mentoring the GSC project I've got a question/suggestion.
I find our handling of JPEG images a bit wanting. I'd think that ideally, we should convert all our uploaded JPEG files into a lossless format and store them as PNG instead. The thumbnailing process should then decide whether an image is best delivered as a PNG or JPG, depending on the image content. Ideally, that determination needs only be made once, and stored as some meta information with the picture.
There are two reasons behind all of this. First, we have a sizeable amount of photographs stored in a lossless format which should really be served as JPGs, to save both the users and servers the time and bandwith. Second, we would reduce the problem of generation loss, so it wouldn't be always necessary to start from the original and attempt to redo all previous steps just to get the best quality. Currently, some editors are uploading JPGs with only little compression, which somewhat works around the problem but has a bandwith impact if the fullsize image is displayed somewhere on an article where the uncompressed image will be served, and not a compressed thumbnail.
I assume that thumbnail format and image format are already somewhat decoupled, judging by what I read about TIFF support. I'm unsure how hard it would be to do so down to the specific images, but I think it would be beneficial. I could think of both quick&easy and hard&slower algorithms to determine whether a picture is best served as PNG or JPEG, and as I said this only has to be done once per image. And if we someday want to support more complex formats, like XCF (which might be possible if the conversion happens on seperate servers) that decoupling would certainly be helpful, too.
Amalthea 11:09, 28 April 2009 (UTC)[reply]

To be honest, I don't know so much about our image thumbnailing right now. Hopefully this summer will be an opportunity to clean it up somewhat, and while we do that, we can look at our options for long-standing feature requests like this. There are clearly cases in which conversion from PNG to JPEG during thumbnailing would be useful, even if the storage format isn't consistently lossless. But in any event, none of this is related to the actual GSoC project, which is fairly narrow in scope and doesn't necessarily have to touch MediaWiki itself at all. —Simetrical (talk • contribs) 13:19, 28 April 2009 (UTC)[reply]

Ah, alright. Well, I'm happy if those possibilities are kept in mind (didn't know that was a "long-standing" request). I wasn't sure how integrated that dameon would be.
Thanks! Amalthea 13:34, 28 April 2009 (UTC)[reply]

Hi, I came accross {{PD-status}}, and was wondering if it is still useful/used in our current set of policies. If it's not, perhaps we should delete it ? —TheDJ (talk • contribs) 00:33, 4 May 2009 (UTC)[reply]

I have no idea. I haven't been involved with copyright in Wikipedia in a couple of years. —Simetrical (talk • contribs) 13:38, 4 May 2009 (UTC)[reply]

Hi, I believe you added the MediaWiki software for ALT text in images late last year. Thanks from everyone at WP:ACCESS! Several of us were wondering if you could adjust the onmouseover hovering tooltip to reflect the ALT text, if it's defined. It seems redundant to have it just repeat the caption, although I see the point for captionless images such as the thumbnails on the Main Page. If you could make the tooltip work for images of math-mode equations, we'd be especially grateful. Thanks in advance! Proteins (talk) 17:20, 5 May 2009 (UTC)[reply]

Why would you want the alt text to be reflected in a tooltip as a general rule? —Simetrical (talk • contribs) 20:30, 5 May 2009 (UTC)[reply]

Thanks for your reply! Here are a few reasons. First, if the caption is visible (as is customary on Wikipedia), then having it appear again in the tooltip is redundant. Second, as mentioned here, sighted editors would like to be able to read the ALT text more easily, both for checking it and for better understanding the image itself. Although the ALT text is intended mainly to depict the image for those who can't see it, sometimes even sighted people benefit from cues as to what they should be looking at. They see, but they don't see. I often had that experience myself when I was first learning architecture; I would see the building, but I didn't really take it in. Having ALT text easily available is another avenue by which we can inform our readers more effectively.

Conveniently available math-mode ALT text would help lay-readers understand the formulae better, I believe. Not everyone will understand mathematical equations written in symbols, but some will understand them if they're translated into English. Not many formulae have ALT text, it's true, but I intend to push the mathematicians to include more ALT text, and I aspire to write an automated ALT-text generator for LaTeX formulae this summer.

The usual method of finding the ALT text on the image "Properties" isn't convenient if the ALT text is long, which may often be the case; the ALT text scrolls off the edge of the popup window. You may wish to also speak to qp10qp, a prominent editor in literature who would like to foster more ALT text. Proteins (talk) 02:49, 6 May 2009 (UTC)[reply]

Alt text has the purpose of allowing people with images off (because they're blind, or otherwise) to understand the content of the picture. Its purpose is not to help people who can see the image understand what they're supposed to be looking at. The alt text should say no more or less than you would get from a brief inspection of the image. If it says less, the alt text should be improved. If it says more, the extra info should be moved to the caption, where everyone will be able to see it. The caption is the correct place to point out important details that the viewer might have missed, and that info will then be available to anyone, whether they're viewing the image or the alt text. Putting extra info in the tooltip is bad for accessibility, because most people won't think to hover to get the tooltip. Automatically mirroring the alt text in the tooltip will encourage people to provide bad alt text, namely alt text that's too verbose and says more than the image does.

As for helping editors check it: info that's useful to editors should be displayed on the edit screen. It should not be visible to ordinary users, most of whom are not editors. If editors want to easily check the accuracy of alt text, it would be easy to write a Gadget that would display it inline next to every image, which is much more useful than requiring you to hover over it.

I agree that it's pointless to have the tooltip reflect a visible caption. If a caption is visible, then the tooltip should be empty unless separately specified (not that there's any way to do that currently, I don't think).

Math is a whole separate issue. As a mathematician (entering a Ph.D. program this fall) who does quite a lot of chatting in plain text over IRC, I have to say LaTeX is actually pretty good for comprehensible plain-text math. Trying to write complicated equations out accurately in English is impossible. It would be like trying to follow a math class without looking at the board, which I know from experience just doesn't work. You get capitals confused with lowercase, parentheses are hard to track even if they're explicitly stated (which they usually aren't), etc. The best way of conveying equations in plain text (barring ASCII art, which doesn't make good alt text for multiple reasons) is basically just LaTeX, with some of the excess formatting removed. E.g., if we take this equation from Heat equation:

${\displaystyle \mathbf {q} =-k\nabla u}$

the alt text is:

\mathbf{q} = - k \nabla u

Good alt text would probably be "q = -k del u" or "q = -k gradient u". "q equals negative k times the gradient of u" would be very poor alt text, IMO. —Simetrical (talk • contribs) 14:26, 6 May 2009 (UTC)[reply]

The Gadget idea for registered editors sounds great to me, although others might not agree. For me, the hovering tooltip was only one option of several for helping editors to review the ALT text. The redundancy of the caption in the tooltip does seem pointless, as you say, but if it can't be removed easily, then it's probably not worth fixing. I also agree that the ALT text should not say more than the image, although IMO there's an art to crafting good ALT text that says no more than the image but says it well and usefully, even for sighted editors. But I also suspect that most editors won't take such care in crafting ALT text.

I haven't thought through the math-mode issues, but speaking as a physics professor who's used LaTeX continually for over 20 years, I'm not convinced that it's the best option for ALT text. However, a slightly parsed version of it might work well, and wouldn't require much tweaking to program. We should probably collect more data on what would be most useful for different categories of Wikipedia's readers. In the example you give above, it's obvious to us two that q is a vector field and u is a scalar field, but that won't be obvious to most readers. Speaking for myself, I'm not willing to sacrifice the understanding of all the others. A well-written exposition in the main article would (try to) make the equations intelligible to lay-readers, but as I'm sure you know, such exposition is still rare on Wikipedia. So, in my view, English versions of mathematical equations can serve a purpose on Wikipedia. Proteins (talk) 01:06, 7 May 2009 (UTC)[reply]

I agree that most editors won't take a lot of care in crafting alt text. But that's okay; it's a wiki, so the people who know and care about good alt text can write it.

As for LaTeX, extra info on the equations might be useful, depending on the audience. (Extra explanation makes the material more useful to lay readers, but less useful to knowledgeable readers. I've read popular articles on mathematical research where it took me halfway through the article to even figure out what field of mathematics they were talking about . . .) But if it is useful, it needs to be in the article text, not in the alt text. I'm not sure what sort of alt text you'd like for math images, but it should be as faithful and readable a translation as possible, not add extra explanatory info.

I do agree that LaTeX isn't at all the best alt text, but it's actually pretty decent: mostly too ugly and complicated, is all. Incomprehensible to lay readers, of course, but then, I don't expect that most of the equations in a typical math/physics article (e.g., Homology (mathematics)) will make any sense to non-experts anyway. If it could be simplified, that would be great, of course, as long as it doesn't become significantly more ambiguous.

Anyway, I guess the pending requests boil down to:

1. Don't duplicate caption in tooltip. There's probably not any need for tooltips at all on . . . well, basically any Wikipedia images. To stave off complaints, though, it might be best to add an extra tooltip= option for images in case people really want one. The code involving this needs to be refactored, it's kind of a mess . . .
2. Better alt text for math images. This is probably doable pretty easily, but most of the relevant code is in Objective Caml as far as I know, so probably not something for the faint-hearted.

If there aren't bugs open for these, you might want to file some. —Simetrical (talk • contribs) 21:17, 7 May 2009 (UTC)[reply]

3.5 years ago you pretty much built the Roman Censor article by adding the entry from William Smith's Dictionary of Greek and Roman Antiquities. The section on classes of punishable offenses described them as "threefold" but only discussed two. While wikifying the text you noticed this and added the comment "<!-- Huh? Only two listed -->"". Fortunately you did not change the word "threefold", so while reading the article today I noticed the discrepancy. I discovered that some online versions of Smith's Dictionary omit the second of three paragraphs, although the missing paragraph is present in many other online versions. I just finished adding it to the article. I'd be rather surprised if you remembered this ancient edit, but thought that you might be interested in hearing of a forgotten riddle's closure. -- Thinking of England (talk) 11:54, 21 June 2009 (UTC)[reply]

I took the liberty of making two small changes to your subpage User:Simetrical/WikipediaSister. The first change was the removal of an incorrect protection template, the second of a documentation page. Please see the edit summaries. I hope this is fine with you. If you have any questions, please write me on my talkpage. Debresser (talk) 14:15, 14 July 2009 (UTC)[reply]

See this discussion on the village pump. Graham87 10:48, 19 September 2009 (UTC)[reply]

FYI -- there is a discussion at [1] as to whether or not to allow the use of the all-numeric YYYY-MM-DD format in footnotes/references.

I'm mentioning it to you in the event that you would like to join in the discussion or follow it, as I recognize that this is an issue you have been interested in in the past. Thanks.--Epeefleche (talk) 08:24, 30 September 2009 (UTC)[reply]

You might've noticed already, but named entities aren't actually output from wikimarkup. On most pages, there seem to be just two problematic entities, one from MediaWiki:Copyright and one from MonoBook.php. —Ms2ger (talk) 19:25, 12 November 2009 (UTC)[reply]

No, I hadn't noticed that, actually. That's a very interesting point – it could be a much better solution than changing the doctype. Thanks for the pointer! —Simetrical (talk • contribs) 17:25, 13 November 2009 (UTC)[reply]

Just wanted to say thanks for correcting me after I removed a discussion that I'd started (and had been resolved) on a talk page. I'd seen other people remove discussions from talk pages after they'd gotten old and become irrelevant and thought that was appropriate. I appreciate you pointing me to the page on wikipedia that explains not to do that , and talks about archiving, etc. Whenever I think I'm starting to know almost everything about editing on wikipedia, I find out just how wrong I am - and I appreciate greatly the people who point it out to me when I screw up! So thanks! Spiral5800 (talk) 06:12, 12 January 2010 (UTC)[reply]

I removed you're changes in this article. read the comment and use the talkpage(of the article) please ;) mabdul 18:25, 10 February 2010 (UTC)[reply]

Thanks for adding that to the default of MediaWiki. You rock! :) Take Care...NeutralHomer • Talk • 22:50, 26 March 2010 (UTC)[reply]

Account Approval, ACCAPP-61

The ball is now, apparently, in your court - please give me a shout on my talk if you need anything from me. Many thanks,  Chzz  ►  21:21, 10 April 2010 (UTC)[reply]

Sorted, thank you.  Chzz  ►  18:29, 11 April 2010 (UTC)[reply]

Resolved

Hello. Your account has been granted the "reviewer" userright, allowing you to review other users' edits on certain flagged pages. Pending changes, also known as flagged protection, is currently undergoing a two-month trial scheduled to end 15 August 2010.

Reviewers can review edits made by users who are not autoconfirmed to articles placed under pending changes. Pending changes is applied to only a small number of articles, similarly to how semi-protection is applied but in a more controlled way for the trial. The list of articles with pending changes awaiting review is located at Special:OldReviewedPages.

When reviewing, edits should be accepted if they are not obvious vandalism or BLP violations, and not clearly problematic in light of the reason given for protection (see Wikipedia:Reviewing process). More detailed documentation and guidelines can be found here.

If you do not want this userright, you may ask any administrator to remove it for you at any time. Courcelles (talk) 18:03, 19 June 2010 (UTC) [reply]

class Inclupedia extends Wikipedia {
    public nonNotableArticle = array ( ... );
    ...
}