This is an archive of past discussions with User:Happy-melon. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
This seems to have stalled recently. Mind having a look to see if we can go with the simplest thing which works, and then jettison the old code? Just looking at some examples where the current code doesn't drop in nicely, Apollo 11 has some stacking / float issues. Chris Cunningham (not at work) - talk21:55, 23 November 2008 (UTC)
Awesome work on those, by the way - the current layout is clean and stylish. Have you a full list of what this has deprecated, or what work still needs to be done to get all our audio clips using this? Chris Cunningham (not at work) - talk15:23, 25 November 2008 (UTC)
The most important task is actually to clean out trancslusions of {{Sound sample box align left}} and {{Sound sample box align right}}. These were used as 'wrappers' to hold other templates; with the new code for {{listen}} this was causing very ugly formatting when a listen template was inside a sound sample box. I solved that issue by blanking those templates, but this inevitably broke any instances where the contents weren't listen templates. So the priority should be to replace all transclusions of those two with appropriate listen templates. Then we of course need to go through all transclusions of {{listen}} itself and make sure that they are now appropriate, especially since the default position has changed from left to right. Once that's done, I can see from a quick glance that {{listen}} deprecates the Template:Multi-listen start/item/end system as well; all the other templates seem to be for inline audio. Which is not to say that we can't have a go at clearing them up later... :DHappy‑melon14:57, 26 November 2008 (UTC)
I see that you are still interested in Flagged Revisions. I actually played sometime ago in LabWiki with sighting/unsighting too (without sysop access). So I am interested of your experience. Ruslik (talk) 14:34, 26 November 2008 (UTC)
The sound templates
By removing the formatting for what you refer to as deprecated templates, you just broke a LOT of articles. The very definition of deprecated is that you should slowly lean away from using it. Erzsébet Báthory(talk|contr.)12:02, 25 November 2008 (UTC)
Hmm... I did considerable investigation and concluded that this was the best way to fix a lot of articles :D The audio templates are enough of a mess as to preclude us doing pretty much anything to fix them without breaking some in the process; make omlette, break eggs, sort of thing. I agree it's unfortunate so I'll be working to fix those articles that are broken as quickly as possible. Happy‑melon14:36, 25 November 2008 (UTC)
Just here to tell you I hate you. You broke a lot of articles, without any indication in those articles why they had to be broken. it took me several hours to find out who was the pseudonymous vandal. I assume I am not the only victim of your vandalism. Erik Warmelink (talk) 10:15, 26 November 2008 (UTC)
Well, no. I should thank you for your "improvements" to the layout of wikipedia. Unfortunately, someone (who doesn't even use colour in the signature) didn't understand your art and pages like the Fall (band) were reverted to their dull, boring appearance. Erik Warmelink (talk) 21:46, 26 November 2008 (UTC)
Hmmn, maybe. It is raining at the moment :D. If I don't get around to doing this within a couple of days, I'll do a find-and-replace to change the mfd tags to a bespoke notice. Hopefully it won't drag that much. Happy‑melon16:48, 25 November 2008 (UTC)
Hi Happy-melon. Just wanted to mention that I have responded to the message you left on my talk page. But I copied it to Template talk:Fmbox#Merging "tmbox-small" and "ombox-small" and responded there, since it was a continuation of the discussion there.
Hey, sorry to keep dragging you back to this (from the looks of your talk page you've had a lot of it recently), but I was wondering if you wouldn't mind checking over some proposals I've made for further changes to the Listen template. I would also like to say well done for getting a template updated when it needed it. There are too many templates that need changing which are left in disrepair, and it is worth "breaking" a few articles to fix them once and for all. Thank you – Ikaratalk →04:40, 28 November 2008 (UTC)
That usually happens to me when I'm wikistalking a vandal and stumble across a comment page (in their edit history) for a user, article, or template that I've never encountered before. :-) — 72.75.110.31 (talk) 00:31, 1 December 2008 (UTC)
Hi there. As you have recently edited Template:SharedIPEDU I thought you would be the best person to approach. Would it be possible to place a soft block on User:89.207.208.9. This IP address is one of several proxy servers for approximately 30 primary and secondary schools in the local area. Because the proxy covers so many users it is not possible for abuse to be reported or acted upon in any useful fashion. It would also assist teachers in some fashion if anonymous editing is disabled, by removing the temptation for kids to vandalise! Thanks in advance carelesshxtalk09:56, 2 December 2008 (UTC)
Hi there! While I can understand your concern, I don't think a block is warranted in this situation. The edit rate from the IP is extremely low, and not all of its edits have been unconstructive; this indicates that there are some bona fide editors from the IP who could be inconvenienced by a block. More importantly, however, blocks are almost never instituted pre-emptively; there must be a demonstrable pattern of vandalism or unconstructive editing such that it is self-evident that instituting a block would benefit the encyclopedia. Blocks have as their first and only purpose the protection of the encyclopedia; the convenience of the institution responsible for managing the IP is primarily the responsibility of their own IT staff; if they want to prevent editing, they can simply blacklist http://en.wikipedia.org/w/index.php?action=edit from their end. As there is no evident benefit to the encyclopedia from blocking this IP, I'm afraid I must decline. Happy‑melon14:45, 2 December 2008 (UTC)
Hi again! I have done some checking and have noticed that there has been vandalism from several of our IP addresses (currently 89.207.208.2-11), and that at least two of these addresses have been soft-blocked already. For various reasons, these IP addresses have only been in use in some schools for less than six months, so it is likely that there would have been less vandalism prior to this (I believe the standard duration of a block is 6 months?). Given this, I think rather than being pre-emptive, a block would be useful simply to catch the IP addresses that were missed before. In any case, I have no particular axe to grind, I just thought it would be useful to point this out. carelesshxtalk15:28, 2 December 2008 (UTC)
Yup - I was missing something (like the fact that you were just fixing the redirect and a typo slipped in). Cluebat has been self-administered. -- MarcoTolo (talk) 18:26, 3 December 2008 (UTC)
Huh? No one was "test[ing] changes". We were simply tweaking a color to appear as intended (purple). Now you've changed it back to pink on many displays. —David Levy18:54, 3 December 2008 (UTC)
Indeed, because you're doing one thing, someone else is doing another, and the result is a flip-flop of styles on a huge number of pages. Is it a crippling error that's rendering the site unusable for anyone? No? Then we can damn well take the time to get one edit right rather than five edits to faf our way to a solution. WP:BRD etc etc, you must know the routine. I don't even fully understand what you think is wrong. Happy‑melon18:59, 3 December 2008 (UTC)
Incidentally, my original version of that edit summary was "don't sandbox changes..." before I decided that "sandbox" is not really a verb... that might better indicate what I was trying to convey. Happy‑melon19:00, 3 December 2008 (UTC)
1. The changes were minor and had no effect on the actual site's interface.
2. The color would only "flip-flop" for someone whose cache was purged between edits. For most users (excepting those who just reached the end of their thirty-day cache), it didn't even change once.
3. I explained what I think is wrong: the color is supposed to be purple (and is labeled as such), but it looks pink on many displays.
4. "Testing" or "sandboxing" implies that the edits were performed without prior examination of their effects. This is not the case. —David Levy19:16, 3 December 2008 (UTC)
I'm sure you know just as well as I do that your comments 1 and 2 are both entirely correct and totally irrelevant. Of all the possible places to not make sequential 'corrective' edits, the interface is the most absolute. If you do something and it doesn't work exactly as it's supposed to, we revert and discuss. Period. I recognise that there is a problem, indeed it looks slightly wierd on my display too. I want this fixed just as much as you. But there's no reason whatsoever why we can't have a discussion that's more thorough than a set of sequential edit summaries to make sure that what we come up with works properly before we implement it. Happy‑melon19:23, 3 December 2008 (UTC)
Well, no, I don't see those comments as irrelevant, as I was responding to your claim that "the result is a flip-flop of styles on a huge number of pages."
The edits did work exactly as they were supposed to. Again, they did not affect the site's interface. They slightly modified the background shade used in some non-mainspace templates.
To be clear, I have no problem with someone reverting because he or she dislikes the change (or conveying such a complaint, at which point I would have immediately reverted myself), nor do I mind discussing the matter. I only object to the notion that it was wrong of Edokter and me to collaboratively perform very minor edits without prior discussion. —David Levy20:08, 3 December 2008 (UTC)
They are irrelevant because the principle that great care should be exercised in editing the interface and that you should revert and discuss at the first sign of any trouble applies no matter how inconsequential you think the edit is or how effective you expect it to be. You are aware that it is possible to configure browsers such that they do always use the latest stylesheets? As I said on MediaWiki talk:Common.css, I ahve no problem whatsoever in you attempting to fix this problem, indeed I applaud you for it. But I don't see that sequence of edits as "collaboratively perform[ing] minor edits"; I see an edit, and objection, and a second attempt. The objection is a clear sign that this modification is not, in fact, as trivial and uncontroversial as it first appeared, and that's the time to ask whether there are other factors to consider. Emphasis, of course, on the "ask". Happy‑melon20:17, 3 December 2008 (UTC)
1. Why do you keep referring to "the interface"? Again, this change did not affect the site's interface.
2. If Edokter had reverted, I wouldn't have performed a second edit without discussion. But he didn't revert; he worked to improve upon my edit. And then I didn't revert either; I worked to improve upon his edit (and he's expressed the opinion that I succeeded). That's collaboration.
The only person to revert was you. And if you're genuinely dissatisfied with the changes, I have no problem with that. —David Levy20:36, 3 December 2008 (UTC)
Over at MediaWiki talk:Common.css, I had the idea to adjust the "notice" color, and Edokter came up with a shade of blue that I think is a substantial improvement (regardless of which shade of purple we use as the "move" color). I'm hoping that this would alleviate your concern. Here's a comparison:
Would you mind explaining how exactly these boxes are message boxes, intended to be used in other namespaces? Thanks. —Ms2ger (talk) 12:28, 5 December 2008 (UTC)
They're not, and so using the "ombox-small" class is indeed a misnomer. That's why we currently have a generic "mbox-small" decaching in Common.css; after January 1 2009 we can change {{sister}} to use the new class. See this discussion for more details. Happy‑melon14:38, 13 December 2008 (UTC)
Okay, I missed that (wondering why that ended up at Template talk:Fmbox). I'm not sure they're really message boxes at all, though. Wouldn't that mean they should really look like amboxes, with the thick left border? (And btw, you missed {{Wiktionary}}.) Thanks. —Ms2ger (talk) 20:45, 13 December 2008 (UTC)
Trust Wiki
It has occoured to me that many persons do not trust Wikipedia due to the fact that anyone can edit it. I think we should advertise that we constantly chech sources and cites. Many people think that we do not have the capeablilty to check all edits. We should allow the general public to know of our technologies to prevent and reverse vandalisim and the punishments of offenders. We are not trusted by all but I believe in Wikipedia to make it happen. Someone could start a project to do these things.
P.S. Pass this message to every single wikipedia user starting with the project council. I have chosen you for the "H" section. Please pass it on to all members under "H" on the project council. And reply on my talk page.
Please revert/fix your changes at Template:WPBannerMeta
Hi Happy-melon ! I do not know what you have done today at Template:WPBannerMeta, but the result I see is that some template have images that do not scale correctly now. For example see Talk:Grand chess: the photo is much bigger than what it should be. Could you please fix that, or revert your changes ? SyG (talk) 18:11, 14 December 2008 (UTC)
{{Chess-Wikiproject}} does not specify a size for the left image, meaning it uses the default sizes which, as you note, have been recently changed. I personally feel that the larger image is nicer and more useful, but if the chess wikiproject feels otherwise the size of the image can be easily set using the |IMAGE_LEFT_LARGE= and |IMAGE_LEFT_SMALL= parameters, to set the size of the image in large and small formats respectively. Have a read of the documentation and let me know if there's anything you still don't understand. Happy‑melon18:20, 14 December 2008 (UTC)
All right, thanks for this explanation, I have added the parameters you indicate and it works fine. There is still a problem with the Portal image, though, as its size in the template seems to have increased as well (?) and I do not find how to fix that. SyG (talk) 19:39, 14 December 2008 (UTC)
None of the changes made affected the size of the image used in the portal box, which is not and has never been configurable. Happy‑melon19:42, 14 December 2008 (UTC)
You need to be more specific than that. Where and when does it not nest, what OS/browser, and can I have an example? :DHappy‑melon13:28, 15 December 2008 (UTC)
Protection categorization discussion
Hi, there's a discussion going about changing the way the protection templates categorize the pages as protected. Since this would affect {{pp-meta}} and the other templates and you've been involved there in the past, I thought you might appreciate a heads-up as part of our (you, me, David Göthberg) little "template triumvirate". :p {{Nihiltres|talk|log}} 03:06, 15 December 2008 (UTC)
Deletion of empty staple category
I suggest that Category:Disambig-Class Anthroponymy pages be undeleted. This is a category that should be populated, and I'm sufficiently familiar with the subject to know that it will be. WikiProject assessment categories really should never be deleted even if they are empty, as they are often created simultaneously, en bloc, at the start of the project's assessment process. __meco (talk) 11:45, 15 December 2008 (UTC)
Because they are clearly not being used. The template is transcluded on only half a dozen pages, and not one of them has a defined value for |class=. The assessment system depends on ten categories, none of which exist, and there is none of the necessary preparation to ensure that the assessments are collected by the 1.0 bot. As such, when none of the necessary infrastructure exists, allowing the template to make quality assessments is actually doing more harm than good. Happy‑melon13:27, 15 December 2008 (UTC)
Okay, I think I understand. The thing is though, I'm only just now (like yesterday/today) working towards re-initializing the project to which that template is attached. In the next few days, I was going to apply the template across the board, build up the Project's SOP, etc. Now, being that I haven't already, it's no big deal. However, please don't take any affront when I revert that edit in a few days after filling the categories to which it points. Sound okay? — pd_THOR|=/\= | 21:25, 15 December 2008 (UTC)
Certainly not, I will crack out a bottle of champagne to see another WikiProject getting back off the ground. Well maybe not a whole bottle... :DHappy‑melon21:39, 15 December 2008 (UTC)
Hi, I have a quick question for you. It's probably a silly question. When you substitute a template you normally want it to contain as few bytes as possible. When the template contains parser functions it would be sensible for the result of these to be substituted in, not the functions themselves. But this is not what happens, is it? For example, when I use Template:Ifu talk I get a whole load of stuff which I don't need. For example I will paste it below (look at the code to see what I mean).
There are two customary approaches, depending on whether the template is supposed to be always substituted, or only sometimes substituted. If the template should only be substituted, you can put <includeonly>subst:</includeonly> immediately after the opening braces of every parser function and magic word, which will cause them all to be substituted. Note, however, that this will cause corrupt output when the template is not substituted. If the template should only sometimes be substituted, you can put {{{subst|}}} in the same place, then by calling the template {{subst:foo|subst=subst:}}, you can achieve the same effect. They really ought to be documented somewhere, but I'm not sure where. Hope this helps. Happy‑melon15:54, 16 December 2008 (UTC)
It is indeed helpful! Next question: is there a template which will take a fullpagename and give you its talk page? Obviously you can just add Talk: if it's a page in mainspace, but that won't work with other namespaces. (I want to use it on a page other than the page in question, so namespace detection won't work.) Thanks again, Martin16:18, 16 December 2008 (UTC)
No, and this is a longstanding issue that has resulted in some very messy workarounds. It's not really possible to do reliably without either mw:Extension:StringFunctions, or some other form of regular expression testing, both of which have been requested for years with no development. Happy‑melon17:13, 16 December 2008 (UTC)
What does it take to get a particular extension installed? Surely we just get consensus and ask for it? Extension:CreateArticle is on my wishlist. Martin08:17, 17 December 2008 (UTC)
templatepage
Hi H-M, I've been hacking at the templatepage template. Hope you don't mind! I'm ready for the parameters to be passed from the core, when you get a chance. Thanks! Martin14:22, 18 December 2008 (UTC)
I've cheated. What I did was to vary the subpage called by the main WPBannerMeta template, so it calls /templatepage directly when it's on a template page. That means you get all the parameters that exist, after normalisation, and I can also remove the /templatepage call in /core, which saves quite a bit of code. That does mean, however, that the example banner no longer appears at the top of the template page, which I'm working to correct at the moment. I'll make sure FQS is passed through for you. Happy‑melon14:52, 18 December 2008 (UTC)
I think you broke it. Could you look at the coding for when a yyyy-mm-dd format is used. It is only outputting a year, and the wrong year at that. MrKIA11 (talk) 17:40, 18 December 2008 (UTC)
My apologies, I clearly didn't read your statement carefully enough, and also managed to avoid edit conflicts throughout that edit set! Sorry for the misinterpretation. Happy‑melon21:06, 18 December 2008 (UTC)
On the contrary, it does precisely that: the parameter is defined with a null ('') value, which is probably the source of your problems. Happy‑melon15:12, 21 December 2008 (UTC)
Hmmm, this is confusing. So how can you test for that? Does {{#if:{{{parameter|}}}|yes|no}} give yes or no if the parameter is null? By the way I have made some changes to templatepage. I haven't completed solved the problem of creating the categories, but now you only have to type one thing which can then be pasted onto all of them. See if you can work out what I mean, although the wording is not very clear yet! Happy‑melon22:49, 21 December 2008 (UTC)
That expression will return false ("no"), as #if: tests for non null input. What you'd need to do would be to set a default value of something like our beloved mu character, and use #ifeq: between that and mu, which is what WPBM does all over the place for the category optouts. I'll have a look at your new /templatepage code, I'm interested to see what you've managed to come up with! Happy‑melon22:52, 21 December 2008 (UTC)
search query
Hi Happy Melon, I saw your proposal at the village pump and thought you would be just the chap to ask about a search query. Every month or so I search through article space for "posses " and usually find a bunch of typos plus about a hundred plurals of posse. Is there any way to order the search by most recently edited? That way I wouldn't need to look at anything older than my last patrol.
Also is there a way to screen out a particular combination, so I could search for poop but exclude "poop deck"? ϢereSpielChequers14:31, 19 December 2008 (UTC)
An inquiry
Hello Happy-melon, recently at the Village Pump we engaged in some conversation about citing refs multidimensionally. You had mentioned that the feature has been requested for some years, and I wondering if you could provide links to any of the prior discussion that you could dig up, so that I might make a case for its necessity (and demand) now. Many thanks, ←Spidern→21:24, 19 December 2008 (UTC)
Hmn... I can't find any off the top of my head; I know there were several discussions on one of the pumps, but I can't remember which one! Try the archives of VPT and VPR; sorry I can't be any more specific. Happy‑melon18:49, 21 December 2008 (UTC)
Recat?
Why are you recategorizing and eliminating the future categories in articles?--SRX 18:23, 21 December 2008 (UTC)
I'm mainly a developer on {{WPBannerMeta}}, the meta-template used by some seven hundred WikiProject banners including {{Pro-wrestling}}. Following recent changes there support for some of the more obscure 'extended 1.0 assessment' classes are going to be dropped, including Future-Class. While it is possible to use a more advanced feature of the banner to continue support for individual projects, we're keen to only implement that where it is genuinely important; the whole point of the change is to spare the 99% of projects the burden of having to maintain a Future-Class system just because the other 1% use it. So I've been working through tracking categories chasing down exactly which projects have implemented a Future-Class system and then taking a best-guess as to whether they're actually using it, or have just created it because the banner previously told them to :D. In WPWrestling's case, I deduced from the substantial overlap with WPFilm, which does actively use the Future-Class, that most of the dozen-or-so articles in Category:Future-Class Professional wrestling articles were there because they were tagged the same as another banner (usually {{film}}) by passing taggers, not because they were carefully selected to be so. I'm sorry if I've made a mistake in deducing that WPWrestling is like the majority of other projects and has just been caught up in the 'flow'. As I said, there is a workaround to continue supporting Future-Class for this project banner, but I'm keen to only implement it where it is genuinely required. So, the real question is, is this assessment genuinely useful to this WikiProject? I've asked this question on a numberofotherprojects with mixed reactions. Happy‑melon18:40, 21 December 2008 (UTC)
Bloody hell you archived that quickly! I went back to your page and couldn't find anything of mine, so went off and assumed. Ok, workaround it is! Happy‑melon18:52, 21 December 2008 (UTC)
Well it has served our project well. Without the future class we were creating more and more stub articles when creating PPV events. We couldn't get under 600 until we implemented the Future-class. It has been very helpful since. There isn't alot of articles tagged for future but in the next coming month about 10 or so will be tagged with as future. It isn't a must have for the project but it is a useful class.--WillC18:53, 21 December 2008 (UTC)
Recently you and I participated in the discussion on WP:VPP, which was about granting 'editprotected' right to some users. Authors proposed to begin "splitting" the sysop package. However the most amusing discovery (for me at least, after I studied Localsettings.php) was that there is nothing to split! Contrary to what this template shows sysops do not have 'editprotected' right (and no account has). They have only 'protect' right that allows them both to change protection levels and edit protected pages. So the template contains a serious error that needs to be fixed. Since you created the template I want to ask your opinion on the matter before I make any changes. Ruslik (talk) 19:46, 24 December 2008 (UTC)
I see what you mean, a cursory examination of Special:ListGroupRights and the API confirms that you're right: the editprotected permission is not in fact being assigned. Certainly go ahead and correct the template, no need to ask! Happy‑melon21:20, 24 December 2008 (UTC)
Mr Melon, your remark that Integer tincidunt. Cras dapibus. suggested that you were a sterling chap, and your recent bout of template-smithing beyond the call of duty clinches it. Sir Claude and I are delighted to present you with the Order of the Waxed Moustache for services to "HOP".-- Hoary (talk) 14:28, 25 December 2008 (UTC)
Hi. I notice that you gave WOSlinker, a user I'd never heard of till a few minutes ago, a barmy star For tireless work in converting WikiProject banners to use {{WPBannerMeta}}, a largely thankless but utterly necessary task.
I hope that my recent edit of Template:WikiProject HOP, in which I undid WOSlinker's unannounced and unexplained work of a few days earlier, didn't screw anything up. If it did, let's discuss the matter, on that template's talk page. -- Hoary (talk) 08:40, 23 December 2008 (UTC)
While I take strong exception to your positions and poor choice to reignite the withdrawn Template:Video TfD, it's somewhat less personal to you than it appears. You seem ok in general, and I recall sharing your position in a previous venue.
The basic problem is your taking up the bad karma TfD project of that loose cannon editor Chris. (And I can prove it, so that's not considered a personal attack - read WP:NPA) This means TfD projectwise, you are now Chris, or at least are temporarily personifying Chris' bad reputation.
He did one outrageous thing after another to me: deleted at least 15 of my talk page posts for example. He made transparent excuses, tried to snow me on template performance and programming issues, walked right up to the edge of an ANI vandalism report – until finally I did an investigation to find out why he was wasting so much of my time. In swaths of campaigning across WP, not only did he make himself unpopular with heavy-handed and combative article editing (followed by a predictable rejection for adminship), but he actually managed to get himself censured by Jimbo.
What I've told you must be of substantial concern to you, or at least it would be to me if our roles were reversed. Beware of further Escalation of commitment. (Please reply here if desired)Milo12:13, 28 December 2008 (UTC)
Hi, and thanks for the comment. I am aware of the circumstances surrounding the template and the previous TfD, which I did read before renominating. I agree with your overall assessment of the nomination: it was underhanded and entirely inappropriate. The TfD should have been speedy-kept. However, to extend from there to say that anyone who nominates {{video}} for TfD is "now Chris" is quite simply an expression of bad faith. I am not Chris, I am not "taking up" the 'quest'. I am re-raising the serious questions that still remain over the usefulness of this template, which were buried under legitimate outrage at the nom's actions in the previous TfD, and I am hopefully doing so in a much more transparent fashion. There has been no underhanded blanking - I removed the template from only two pages before nomination, one of which I had added it to just a few minutes earlier - no canvassing, no skullduggery as far as I'm aware. The actions surrounding the previous TfD were deplorable, but have no relevance to today, where we must consider the template on its current merits. In my opinion, those merits do not compensate for the additional hassle of maintaining and using the template, and as such it should be deleted. If you have arguments on the merits of the template, they are urgently needed at the TfD. If you have arguments on the merits of the nominator, please think twice about what you're saying and why. If you think I have abused admin tools, you want WP:AN; otherwise you're looking for WP:RFC. Nice working with you anyway. Happy‑melon13:23, 28 December 2008 (UTC)
The summary was being parsed by the preprocessor, which means wikilinks became html links, wikimarkup became xml markup, etc. That's no good if we wanted that summary to be available for preloading into deletion summaries. I've just rewritten the script at sysop.js to pick up the "delete-criterion" value for CSDs use that to preload the deletion summary; I can't see any reason why that can't be extended to picking up a "delete-reason" id from other deletion templates (like XfD and PROD) and prefilling those deletion summaries as well. To do that, of course, we have to make sure the wikimarkup survives all the way to the deletion summary box, so urlencoding is a necessity so that it's translated back into raw brackets and braces when it fills the summary box. Hope this clarifies. Happy‑melon10:32, 29 December 2008 (UTC)
Okay, it would be nice if in the future, you used some more descriptive summaries or made a note on the talk page about this. Changing something like this that's only used by bots/scripts without warning will most likely break any existing usage (like the script I'm working on). Mr.Z-man17:07, 29 December 2008 (UTC)
Ooh, I'm sorry :S I had a check through various script files and couldn't find any uses, but of course I can't check bot code. Sorry if I've caused you inconvenience; it is (as usual) something we should have thought of at the time rather than realised later. Happy‑melon11:29, 30 December 2008 (UTC)
Forwarding banners
Hi Happy, I am trying to work out how to get a banner to 'forward' to another banner. For example, I would like to replace Template:User TFBH2 by the text
Probably the former; it ensures that the dependent banner is also properly categorised, and it would display the category warnings if there were any. Mu change is in the works, I'm utterly fed up with it; I'm just trying to work out how to not break 700,000 pages in the process... Happy‑melon22:26, 30 December 2008 (UTC)
Hi Happy, thank you for restoring the automatic deletion summaries. However, your edits to MediaWiki:Sysop.js had an unintended (minor) side effect. The dropdown menu options used to be rendered in plain text, for example:
G4: Recreation of a page which was deleted per a deletion discussion
The options now appear like this, with all of the wiki-formatting displayed:
[[WP:CSD#G4|G4]]: Recreation of a page which was [[WP:DEL|deleted]] per a [[WP:XFD|deletion discussion]]
This makes the deletion menu toolbar unnecessarily wide, and it requires a bit of extra scrolling around to delete a page that isn't already tagged. Would it be possible to render the dropdown menu as it was before, without losing the autofill feature? Thanks, take care. --Bongwarrior (talk) 07:34, 29 December 2008 (UTC)
P.S. - I should have mentioned that this is only a problem for admins with lower screen resolutions (the computer I user most frequently is 800x600, for example). The deletion dropdown menu should fit just fine as it is for most other users. --Bongwarrior (talk) 09:27, 29 December 2008 (UTC)
Hi, and thanks for bringing this up. However, I don't believe that anything has actually changed. My edits to sysop.js do not affect the dropdown menu, in fact they don't affect the delete confirmation screen at all! What they do is change the target of the "delete" tab's link to add a preload parameter, so that clicking on a delete tab when there is a CSD template present will take you to a delete confirmation screen with the summary already filled in. The dropdown menu, which I have also been playing with, is at Mediawiki:Deletereason-dropdown, however I don't think anything has substantially changed there; all we've done is play around with what's included in which namespaces. Are you sure that it was definitely like that before?
If this is causing a problem for you, you can add this line:
Yes, Deletereason-dropdown is exactly the same as it was before, which initially confused me pretty good. The deletion toolbar got wider yesterday, but Deletereason-dropdown hasn't been edited in any way that should have caused that. I saw that MediaWiki:Sysop.js was changed recently, but I wasn't sure if that was it or not, so I did a temporary revert of MediaWiki:Sysop.js and tried to delete something in order to test my theory. The toolbar was back as it was before, so that was definitely it. --Bongwarrior (talk) 14:50, 29 December 2008 (UTC)
Ok. I still have no idea what prompted that transition. I have added a nicer rule to MediaWiki:Sysop.css which forces a maximum width for that box, so hopefully it no longer falls off the edge of your screen. Happy‑melon11:43, 31 December 2008 (UTC)
Hi! I took your advice and substituted the parser functions on Template:Afc talk. But for some reason the test for whether it was an IP talk page (in which case lc=uc) stopped working - it always found that lc didn't equal uc. Can you explain this for me? Thanks, Martin23:28, 2 January 2009 (UTC)
The functions lc: and uc: need to be substituted themselves, as do all the variables; otherwise they are treated as text strings: "{{lc:Foo}} == {{uc:Foo}}", which is of course false. If in doubt, subst everything :DHappy‑melon23:33, 2 January 2009 (UTC)
Yes, I think so; the CSS was added to common.css on December 1 2008, so we're well past the decaching time. Why, is it not working somewhere? Happy‑melon20:39, 3 January 2009 (UTC)
An editor has asked for a deletion review of DTFD and DTFDB. Since you closed the deletion discussion for this page, speedy-deleted it, or otherwise were interested in the page, you might want to participate in the deletion review. Ipatrol (talk) 22:43, 3 January 2009 (UTC)
Hi there. Since you seem to have been involved in the development of the "class" parameter of WPBannerMeta template used across many Wikiprojects, I thought you may have some insight on a problem I'm running into. I'd like to create categories that filter pages by their status and quality assessments (e.g. a list of pages from WP:MCB that are in both the categories "Top-importance" and "Start-Class"). I'm able to create these lists fairly easily in AWB.
However, when I add the category "Top-importance and Start-Class MCB articles" to the talk page of articles that fit those criteria (e.g. Talk:Active_transport), the resulting category page does not alphabetically sort the articles by their mainspace title. Instead, all of the articles are listed under 'T'; the sort is being done on the title of page that the category is being added to (i.e. "Talk"). See Category:Top-importance and Start-Class MCB articles for an example of my problem.
Do you have any idea how I could get these categories to sort properly? The reason I assume you may be a good person to be asking is that the articles sort correctly based on the value of their template's class parameter (e.g. Category:Start-Class MCB articles). If you don't know what could be causing this or are too busy to look into it, perhaps you could point me to someone who could better help? Thank you! Emw2012 (talk) 23:56, 3 January 2009 (UTC)
Hi there! I can see what you're trying to do, and I recognise the usefulness of it. Before going any further, however, you should probably be aware that the system you're trying to create will eventually be deprecated by the 'second generation' 1.0 bot, the code for which is currently in development. The new bot (coming soon to a blue moon near you :D) will do category intersections automatically. Have a read of the collaboration page, and decide whether it's worth still doing the manual intersections you want, or if it would be easier just to wait for the new bot.
If you do decide to proceed, the reason the pages categorise to "T" by default is because you have not set a category sortkey, so they sort automatically by namespace then by pagename. Again you have two choices: firstly, you could wait until T18552 is fixed, which will change the default sort behavior to be what you want. Or you could manually specify a sortkey for each page. See help:Category#Sort key. I hope this is helpful... Happy‑melon10:12, 4 January 2009 (UTC)
FlaggedRevs
Hi, sorry for trolling to your page... Is FlaggedRevisions intending to be a fundamental change to Wikipedia? It is not described very clearly anywhere. It always says stuff like "Article validation is..." but what is "FlaggedRevisions is..."? Just curious having seen your entry on Jimbo Wales talk page. Could you describe what I would see different if anything? ~ R.T.G00:43, 3 January 2009 (UTC)
Hi, thanks for explaining it. I guess it doesnt make me feel threatened but is the queue of seers going to go through it all? I guess it could be a lot more interesting than admin vandal crap. Anyway, excuse the French please. Again many thank yous for your time. ~ R.T.G03:56, 5 January 2009 (UTC)
Do not make untested changes to the MediaWiki interface
Do not make any more changes to the MediaWiki interface unless you've thoroughly tested them first. This change causes the deletion drop-down box to become huge in Webkit-based browsers. krimpet✽00:53, 5 January 2009 (UTC)
Another Flagged-Revision Proposal
I started a flagged revision proposal which hopefully can be used as a starting point for the flagged revision trial, I thought you might like to take a look and feel-free to add to it as well. Y. Ichiro (talk) 04:12, 5 January 2009 (UTC)
Hi! I was away for the past week. I see some activity on the implementation of FlaggedRevs, including a straw poll. Unfortunately I'm still clueless. I think it would be a good idea for a summary of the pros and cons of FR be placed above the poll. I'm interested in the Flag revs bit, but certainly do not have time to go through the discussion. Thanks! =Nichalp«Talk»=06:55, 4 January 2009 (UTC)
I'm sorry this FlaggedRevisions thing is not gaining consensus. I can't for the life of me see why people are so opposed to trying something new, because without a trial we can't even make an informed decision. Martin13:55, 4 January 2009 (UTC)
I'm sorry, but I have changed my mind. This edit from Ruzlik declares
that sighting is to be used only against obvious vandalism, libel, and copyvio.
Infers that sighting will be only so used, because it ought to be.
Your draft says nothing about (1), and (2) is the idea that Wikipedians will do what should be done without an enforcement mechanism. This is utopian anarchy. I oppose this draft (since it cannot now be changed) unconditionally, and shall oppose all future drafts which share these defects. SeptentrionalisPMAnderson22:58, 4 January 2009 (UTC)
I don't know if if helps any that I don't agree with Ruslik's stance on that. The draft says nothing about that because the point, my point anyway, is that the "when should edits be sighted" issue is yet another thing that we should be experimenting with. I don't think that it's helpful to declare that that is the only acceptable way to use FlaggedRevisions, that's far from the case. In my opinion, the fact that this proposal says nothing about this issue is a strength, not a weakness, because it avoids just such sweeping assertions that disenfranchise and discourage editors such as youself. Formally, Ruslik's assertion is not supported by the trial proposal, which deliberately avoids prescribing anything on this issue to leave it open to further consensus-building discussion. I hope this goes some way to ameliorating your concerns. Happy‑melon23:06, 4 January 2009 (UTC)
My judgment is that this is more of the vagueness Ozob objects to. However, if you are going to try another draft with this approach, do consider a list of points it is not discussing, and why. SeptentrionalisPMAnderson23:12, 4 January 2009 (UTC)
I answered in this way because I thought that you was speaking about trials, not about general proposal, which is just a technical implementation of FR. The first four trials are clearly about obvious vandalism, libel, and copyvio. Ruslik (talk) 16:55, 5 January 2009 (UTC)
I suspect the answer to this question will be the same, but is there a function such as {{SUBCATEGORIESINCATEGORY}} (or similar) that will only count subcategories? PC78 (talk) 17:57, 6 January 2009 (UTC)
Thanks for all your hard work on the Flagged Revisions debate
Hey there, just wanted to pop in and say that even though we currently sit on different sides of the Flagged Revisions debate, that I wanted to recognize all of the conscientious effort that you are putting into this difficult issue and all of your diligence in responding to objections and keeping the discussion moving. Keep it up!
Thankyou for the comment, it's much appreciated. My only regret is that the steady stream of !votes in the attention-grabbing "support"/"oppose" sections often display a complete ignorance of the discussions and debates going on just a few screens below. If people were inclined to show a bit more interest in reading round the subject before making their mark, it would be a much improved process for all concerned. Thanks anyway for being one of those who do actually participate in the most important part of any straw poll: the surrounding discussion. Happy‑melon22:00, 8 January 2009 (UTC)
A lot of the headers in Category:Wikipedia header templates should be omboxes, and are not, which creates some annoying situations when adding them to new pages (using old shortcut conventions, etc). I have tried converting some, but the complex variable functions are beyond me. Can you help? MBisanztalk04:56, 11 January 2009 (UTC)
How does the |category=no function work in the meta (i.e. to stop the banner from categorizing a page where it isn't desirable to do so)? I was thinking it might be worth adding such a function to {{Film}}. Might be able to figure it out myself, but I'm not sure which subtemplate of the meta I need to be looking at. PC78 (talk) 20:48, 11 January 2009 (UTC)
Short answer, it's everywhere :D. Everywhere that a category is added by the banner, it performs this test:
{{#ifeq:{{{category|¬}}}|¬|[[Category:Foo]]}}
And everywhere in the 'chain' of templates, the parameter is passed like this:
|category={{{category|¬}}}
So, if at the end of the chain (or anywhere in the middle, for that matter) the category parameter is not specified, then a default of the "not character" ¬ is set. That default propagates all the way through to the conditional, which evaluates to true and so the category is added. However, if the |category= parameter is explicitly set to anything else anywhere in the chain, then what reaches the conditional is not the default, so the category is suppressed.
Hope this helps your understanding, but why don't you convert {{film}} to use WPBM? I'm now increasingly prepared to bet money that there are few if any banners out there that can't be converted with no loss of functionality. Film certainly seems to be a 'textbook case', that could be fairly easily converted. Why don't you give it a try? Happy‑melon20:54, 11 January 2009 (UTC)
Ah, so I would only need that first bit (the test), and you can use any word (not just "no") to get the desired result? As for conversion I dunno, the |futyear= and |futmonth= parameters look a bit scary. PC78 (talk) 21:25, 11 January 2009 (UTC)
It's not too scary, and don't forget that you can take that nest of catgories and just drop them onto a new banner; because it doesn't produce any visible output, it doesn't matter what it is or how it's coded. It can certainly be simplified, but that's besides the point. The rest of the banner is fairly vanilla. Happy‑melon22:08, 11 January 2009 (UTC)
Meta templates
You seem to fairly involved with meta templates, so I was wondering if you knew anything about {{Userbox}}. It doesn't look like the sort of thing that should be subst'd, but the documentation suggests otherwise. PC78 (talk) 22:13, 11 January 2009 (UTC)
It's not something I've ever seen before, or something I'm particularly involved with. My immediate reaction to anyone saying the word "subst" is to shudder anyway; templates should only be substituted if they're supposed to look like text - substituting anything that's obviously a template is almost always unnecessary, and just results in it looking like a bad template in a few months/years time. Unless the benefit of having the content unchanging - warts and all - is absolutely irreplaceable, you will never see me support template substitution. But no, I haven't ever seen that before. Happy‑melon22:17, 11 January 2009 (UTC)
Oh gosh, long story :D. Short version is that there's a PS2 game called mashed that I used to play quite a lot with friends; it's a great icebreaker to play a car racing game where being in the lead just makes you a target for the guided missiles on the cars behind :D. One of the AI drivers is called 'melon', and the taunt-line that always plays when his car wins is "I am now a happy melon". Since my friend who always played melon was by far the best driver amongst us, we tended to hear that a lot. Years ago when I came to choose an anonymous e-mail address, it seemed like a natural choice; now it's grown into a whole collection of accounts on various websites. It still makes me smile occasionally... Happy‑melon13:33, 12 January 2009 (UTC)
then I have a job for you ;) We were discussing changing the colour of Project-Class and noticed that several changes are necessary for this. It would seem natural that the contents of Template:Project-Class col were used wherever the colour is used, but that doesn't seem to be the case. For example neither Template:Cat class nor Template:Project-Class use it, so I'm not sure how it all works. You probably have a better idea and a lot of these are fully protected anyway. Thanks, Martin15:48, 12 January 2009 (UTC)
There are about three sets of templates here, none of which actually interact in any way. We have {{FA-Class}}, {{FA-Class td}} for an HTML version, and {{FA-Class col}} which just adds the colour. I might try and make WPBM independent of the system entirely... what do you think? I might also have another crack at implementing the colours through CSS. Happy‑melon18:30, 12 January 2009 (UTC)
Hmm, what would be the advantage of going independent? Then there would be yet another system. Wouldn't it make more sense to rationalise them and make them work better together? Martin19:27, 12 January 2009 (UTC)
Don't forget that whatever WPBM uses will eventually become the de facto standard. I tried putting these into CSS some time ago, but reverted because it broke horribly; I think I now know why, and I'm tempted to have another go. If we can get it working using pure CSS, it will mean all sorts of things, from only having one template to call for each of importance and quality, to being able to eliminate the last {{!}} calls in WPBM. Happy‑melon19:32, 12 January 2009 (UTC)
Hope you don't mind me adding a comment on this too. I think it would be better to have just one Class template rather than a separate one for just WPBannerMeta as then templates such as Cat class could use it as well. Template:Class would be a good name and then pass through the parameters to that one template, e.g. {{Class|class=A|type=htmltable|category=category}} with type being either htmltable, wikitable or color/colour. You would just need to use the {{#switch: {{lc:{{{class}}}}} code like you've done in WPBannerMeta/class. One advantage of having all the Class code in one place is that the What links here would actually work against the template better as if you try to find things in the template-space using A-Class or A-Class td for example, there doesn't seem to be that many because they are using code such as {{{{{class|}}}-Class}}. -- WOSlinker (talk) 19:42, 12 January 2009 (UTC)
Yes, that was something along the lines of what I was thinking; sorry if I gave the impression that I proposed to create it as a WPBM subtemplate, which would as you note make it difficult for other things to use it. Eventually of course we want all banners to use WPBM, but there are still things like all the stats tables that use the class colours, so we still need an external template. Happy‑melon19:50, 12 January 2009 (UTC)
(e/c)I think I agree. Whatever happens to WPBM, there will still be a need for a standalone class templates for various reasons. I guess it might make sense to combine them ... but it might also confuse things unnecesarily. (It's much easier to type {{A-Class}} than {{Class|class=A|type=htmltable|category=category}}!) But anyway, my most pressing need at the moment is to change the Project-Class colour to #C0C090 on Template:Cat class, because it's inconsistent at the moment. Martin19:56, 12 January 2009 (UTC)
Would someone mind filling in the rest of the defined scale colours into the CSS below? Just copy the syntax with whatever colour codes we've dreamt up over the years.
/* Skinnable CSS for assessment grades */.mediawiki.assess-fa{background:#6699FF;}.mediawiki.assess-fl{background:#6699FF;}.mediawiki.assess-a{background:#66FFFF;}.mediawiki.assess-ga{background:#66FF66;}.mediawiki.assess-b{background:#B2FF66;}.mediawiki.assess-c{background:#FFFF66;}.mediawiki.assess-start{background:#FFAA66;}.mediawiki.assess-stub{background:#FF6666;}.mediawiki.assess-list{background:#AA88FF;}.mediawiki.assess-na{background:#F5F5F5;}.mediawiki.assess-{background:transparent;}.mediawiki.assess-category{background:#FFA500;}.mediawiki.assess-disambig{background:#00FA9A;}.mediawiki.assess-image{background:#DDCCFF;}.mediawiki.assess-portal{background:#808080;}.mediawiki.assess-project{background:#C0C090;}.mediawiki.assess-redirect{background:#C0C0C0;}.mediawiki.assess-template{background:#FFCCFF;}.mediawiki.import-top{background:#FF00FF;}.mediawiki.import-high{background:#FF88FF;}.mediawiki.import-mid{background:#FFCCFF;}.mediawiki.import-low{background:#FFEEFF;}.mediawiki.import-na{background:#F5F5F5;}
I realise now, of course, that we've made a HUGE mistake with these class names not making them one case; that's probably going to bite us. I wonder if it's too late to change them to lowercase? Happy‑melon22:48, 12 January 2009 (UTC)
Yep, you were right! this time some parts of the graphic were changed. (before the Serbian Army graphics were just copied to wikipedia and the license changed to PD). Anyway, I checked now if the changes are correct (they were) and updated the original files on commons. Now the files on wikipedia are duplicates. I will also inform the graphics uploader that next time he shall not change the graphics and upload it to wikipedia, but to leave a not on the graphics talkpage, so that changes can be made there. thanks for your diligence! --noclador (talk) 15:35, 13 January 2009 (UTC)
Richard Cranor deletion
Is it really nescesary to delete my wikipedia entry? "Katana" is a bona fide TV Pilot release on STrike.TV...there are plenty of other pilots being released on Strike.TV neither you or your collegues have deleted, along with the writers/producers/directors associated with those titles. I think you are being a bit unescessarily extreme. —Preceding unsigned comment added by 67.183.138.27 (talk) 18:16, 13 January 2009 (UTC)
Renaming Image-Class to File-Class?
Can I get your input here? I idly made this suggestion earlier today on the assumption that it would just involve renaming the class template, but I guess there's a bit more too it than that? PC78 (talk) 00:55, 14 January 2009 (UTC)
Flagged Revision Problem
Hi!
Is the different implementation to the German WP in Wikipedia:Flagged revisions/Trial intended? Or just due to an oversight/misunderstanding?
In the German WP the latest unsighted edits remain visible to all readers (including IPs/not logged in readers), only the default display gets changed, but any reader can choose to read the latest version by clicking on zur aktuellen Version. This difference is crucial, because the German implementation does not violate or really restrict the anyone-can-edit principle, while the suggested English implementation does in a way. Could you please clarify? Regards--Kmhkmh (talk) 04:23, 14 January 2009 (UTC)
Multi-listen item
The Melonbot conversion of {{multi-listen item}} to {{listen}} - basically, a left-aligned non-floating template into a right-aligned floating template in a box - has broken the formatting of hundreds of articles. Please have the bot go through and revert itself. If you geive me a list of all the places it can't undo, I'll help put the remainder straight. Shoemaker's Holiday (talk) 10:17, 14 January 2009 (UTC)
Hi, and thanks for bringing this to my attention. However, following discussion at Template talk:Listen and the comprehensive update of the {{listen}} template, the functionality of the {{multi-listen start}}, {{multi-listen item}}, {{multi-listen end}} templates is deprecated in favour of {{listen}}. Multi-listen uses a nasty hack to add the loudspeaker, which is now strongly discouraged, and has various other disadvantages compared to the new {{listen}} template.
I was quite careful in constructing the regexes that were used in that conversion process, and did confirm every edit manually. Where the samples were left-aligned before, I attempted to ensure that they were left-aligned afterwards, eg. If I wasn't completely successful in ensuring that that was done, I apologise; if you show me some examples I can improve the regular expressions so that it doesn't happen in future.
With regards the conversion from a non-floating to a floating box, I don't believe this is an issue when the samples were correctly positioned in the article in the first place. WP:SAMPLE is quite clear: "It's better to insert the samples next to paragraph mentioning them to justify their fair use, instead of grouping them in the end of the article"; audio samples are not supposed to form their own section; they should be used to supplement the text inline. I agree that the conversion eg here leaves the article somewhat confused, but that's because the "audio samples" section should not exist in the first place - the audio samples should be distributed through the text, as I did with Bassoon. Having audio samples in a non-floated arrangement is undesirable encouragement for this sort of use. In the few situations where it is acceptable, as in Antonio Vivaldi, {{listen}} can still be used.
Frankly, the appearance and use of multi-listen item is very different. Look at, say, WP:FS, which uses it extensively - if that was made into listen templates, the description text would be forceably wrapped to fit the box width, and everything would be put in ugly boxes. Also, even if left aligned, the listen template still *floats* left, meaning that things are placed to the right of it.
And on and on for dozens, if not hundreds more examples. The thing that was broken by this change was a standard format used in classical music articles and elsewhere VERY WIDELY. Not to mention half the time it says it's replacing multi-listen with listen it actually failed, and thankfully kept the multi-listen, just adding format= instead. Don't fix that, it's the only thing preventing the problem from having spread to hundreds more articles. Shoemaker's Holiday (talk) 11:29, 14 January 2009 (UTC)
While we're at it
{{PromotedFSC}}, {{UploadedFS}}, {{CreatedFS}}. Thank-you templates should not have a big white box ruining a standard format shared with the Featured picture templates.
If Listen had been left with the same appearance as it had had a month ago, that would be one thing, but major formatting changes have happened to listen, not least changing it from left-aligned to right aligned.
And I'm rather upset about all this. I have 50% of all featured sound credits. I am one of the major people working on sounds on Wikipedia, and probably have put more sounds into articles than anyone else. You've basically ruined the layout for about half the articles displaying my work, and, if I fix it, which will usually require multi-listen item, I risk having your bot changing it back.
I don't want to be running around for the next month fixing accidental bot vandalism of almost all the articles I work on.
That said, I do understand you're trying to help, but this doesn't seem to have been thought through very much. We now have two templates with vastly different appearances and default layouts being interchanged. (Who depreciated multi-listen item in the first place, anyway? Why couldn't they depreciate something sensible, like {{Audio}}?) Shoemaker's Holiday (talk) 11:41, 14 January 2009 (UTC)
First of all, let me apologise if you feel that I'm somehow attacking your work here. That certainly wasn't my intention; I have huge appreciation for seeing media in articles, and an equal respect for the people who put them there. I first found myself in this corner of wikipedia when I was cutting and uploading sound samples for Wicked (musical) - the templates that were available were without exception awful for inline media, which is what is recommended by the manual of style and what I wanted to add to the article. One thing led to another and I eventually in desperation coded and implemented an improved version, which was generally popular. See Template talk:Listen for the discussion. So I'm certainly not out to get you or your work, I'm hugely grateful for what you do.
That said, the way audio files are used in many articles (not usually by you, but certainly by others) is simply wrong, according to the MOS. Media should not be put in their own section, or used to adorn the "references" or "external links" sections. They should be inlined through the text where they complement the article prose. As such, {{multi-listen item}}is deprecated because it has no valid use: when it is used on its own in a "media" section it is used incorrectly, and when it is used with {{sound sample box}} or a variant thereof, the functionality of {{listen}} is far superior. So while I agree that the appearance of some articles after the conversion is not ideal and is perhaps in some ways inferior, I don't agree that reverting them would be anything other than a step backwards. Where the conversion has caused a visual problem, that is indicative of a deeper underlying problem that won't be addressed by simply reverting to an older template. Media samples need to be rearranged to be more closely integrated into the text; even when the text is very minimal it is possible to do this effectively, eg.
I'd like to work with you here, not against you. I agree that there are issues outstanding, but I think that we need to keep moving forward to best resolve them, not jump backwards when the first step uncovers deeper problems. I agree, for instance, that my initial conversion of the featured sound templates wasn't optimal; extra functionality has since been added to {{listen}} which enables a much nicer display, as I've shown on {{uploadedFS}}. Where listen should duplicate the appearance of multi-listen item, it can do so, but this is not in fact desirable in the majority of situations. Happy‑melon12:12, 14 January 2009 (UTC)
I by no means thought you were attacking my work, I just wanted to point out the problems with the bot implementation, and explain why it's a problem.
That said, there is functionality that multi-listen item has that listen does not. Listen wraps the sound file in a floating box, and that has consequences regarding how wide the description can be on the page, and how it interacts with other parts of the page layout. Also, listen can only go up to ten, some of the media sections have significantly more than that.
While I have integrated sound files into the text before, e.g. Charles Gounod, Tosca, there is a limit to how many files can be presented that way. For the composer and performer articles, a gallery is often necessary, to give a broad sample of their work and career.
There's another aspect. I do a lot of work with media files, and understand the templates, but for new users, multi-listen item is going to be much easier than trying to set up a media section with Listen and having to both figure out the pos=left, but also needing to end it with {{-}}, a not particularly well-known wiki function call.
I honestly do think that too many articles have been damaged not to revert the bot; however, that doesn't mean that it couldn't run again once the bugs were worked out. Shoemaker's Holiday (talk) 12:46, 14 January 2009 (UTC)
I think we disagree fairly fundamentally on how media should be included in articles. I see very little difference between a list of media files and a gallery of images, they have essentially the same appearance and purpose. Galleries are explicitly discouraged by policy, with WP:SAMPLE saying that "It's better to insert the samples next to paragraph mentioning them...". Also, "[while] there's no limit of how many samples you could use in one article, but you have to put in mind that music samples serve as tools for a better understanding of the article, so insert only relevant samples". These policies to me indicate that there is rarely if ever a need to create lengthy lists of audio samples, and so there is no need for the 'unlimited length list' that is the one advantage of multi-listen item. While I have demonstrated on Antonio Vivaldi that such a gallery can be created using {{listen}}, these policies indicate that these media should really be spun out into an article such as List of works by Antonio Vivaldi, as with other composition lists; the media files can then be distributed through the list as is being done (slowly but effectively) with eg List of compositions by Johann Sebastian Bach.
As such, it should not ever be necessary for new users to use {{listen}} in its 'list' mode; I agree it is somewhat esoteric, but since its use in that fashion is to be discouraged, that is not a bad thing. Equally it should rarely if ever be genuinely a Good Idea to use more than ten media files in one place in an article. Full adoption of this template is likely to encourage proper use of audio files in articles, as it becomes easier to add audio inline than to create a gallery, as is currently the case with images. So I don't think your concerns are valid, except in the context of perpetuating the underlying problem of using media files in contravention of policy. I know you're not yourself responsible for these widespread improper implementations, but a lot of people are, and we should be doing everything possible to encourage audio files to be used properly, to enhance and complement articles, not to dominate them. Happy‑melon19:26, 14 January 2009 (UTC)
I'm sorry, but "I broke hundreds of pages, but it's alright because they deserved to be broken" doesn't really help matters much. The pages are still broken. Shoemaker's Holiday (talk) 01:20, 15 January 2009 (UTC)
Whether the result was to "break hundreds of pages" or to 'highlight a major style issue in hundreds of pages thereby encouraging it to be fixed', is entirely a matter of perspective. I don't think we're going to reach an agreement on this amongst ourselves, so outside commentary is probably desirable. Do you think we should continue this discussion here, or move to a more public forum (a template talk page, for instance - but which?)? Happy‑melon14:46, 15 January 2009 (UTC)
Normally I hate it when people butt in on a conversation on my talk page, but as you've asked for outside commentary here we go. I can see this issue from both sides, and as you're both trying to improve things and both are reasonable people I can't see why we shouldn't be able to sort this out easily :) Firstly I think we could make some headway if you could agree on any or all of the following statements.
People were consulted widely enough about this change to the templates, at an appropriate venue and given enough time to respond.
Image and sound galleries are generally undesirable.
Some of the pages look a bit uglier after MelonBot visited them than before.
Secondly I would be interested in how SMH thinks the pages could be best fixed using the new template system. For example, did this fix the formatting in his/her opinion? Martin18:14, 15 January 2009 (UTC)
Point 2: Not necessarily. In composer and artist pages they've traditionally been accepted, and lists have the advantage of always being in the same place, whereas images can push sideboxes around unpredicably on widescreen monitors.
Point 3: Well, yes. In many cases "completely broken" is a better description.
As for how to fix it: While something like your suggestion may be applicable in some cases, in most I'd say it should be reverted back to the multi-listen template until such time as revision is going to happen. Keeping pages broken is not a real option. Shoemaker's Holiday (talk) 18:32, 15 January 2009 (UTC)
H-M, can you address this point? SMH, note that just because you were not aware of a discussion, doesn't mean that it didn't take place ;) A discussion can be widely advertised and still miss people who perhaps should have been involved. It can be mildly annoying to propose something, get little input, and then later get complaints from people who didn't take part in the discussion ...
Well perhaps this needs to go be brought up at WP:IUP. If indeed they are deemed acceptable in certain cases, then the templates should allow for this.
SMH I think perhaps you are exaggerating here uneccesarily. Can you give me an example of a page which you think was "completely broken"? It looks to me that most of these can be fixed quite easily. Martin18:56, 15 January 2009 (UTC)
Thankyou for this contribution, Msgj. In light of Shoemaker's ANI thread, it seems sensible to move this discussion there. I have copied the thread above to that discussion, and will respond there. Happy‑melon20:49, 15 January 2009 (UTC)
