Template talk:Language/Archive 3

From Wikipedia, the free encyclopedia

Contents

[edit] Move

the current Template:Infobox Language is a private version of User:Netoholic and should be moved to his user space or to Template:Infobox Language new. He has no right to abuse the standard name to (in the long run) delete the history of the original template.

If needed a bot can change all article references from Language to Language Infobox. As Commander Kean told me, asking at Wikipedia:Bot requests gives fast help in general. Netoholic pointed out no bot should be used to do this, because a redirect is sufficient. But this is a different matter. This voting is only about having the true language infobox at Template:Infobox Language. Tobias Conradi (Talk) 23:29, 5 January 2006 (UTC)

[edit] result

not moved - please take destination template to WP:TfD if you wish to so, thanks WhiteNight T | @ | C 09:10, 11 January 2006 (UTC)

I don't think it should be deleted. Tobias Conradi (Talk) 21:14, 11 January 2006 (UTC)

moved to {{{Infobox Language}}} Tobias Conradi (Talk) 05:59, 7 February 2006 (UTC)

[edit] Template:Infobox Language

Netoholic has created this template as a version that avoids conflict with WP:AUM. However, he has done this with no discussion, and I am not happy with the result. I am posting this here so that those who use this template might be aware of what he's doing. --Gareth Hughes 18:40, 31 December 2005 (UTC)

I warned users on this page (#Wikipedia:Avoid_using_meta-templates) weeks ago, and noone was forthcoming. I am beginning to convert the "simpler" language articles into this new infobox. I have started with a pilot group of a few articles, but everytime I try and make progress, Garzo rollbacks my edits. Garzo's displeasure is irrelevant, because WP:AUM trumps anything else that this template has been hacked into doing. I'm not sure how much functionality can be packed into the non-meta-template version, but if even half can be converted quickly, that is an improvement. -- Netoholic @ 08:59, 1 January 2006 (UTC)

Garzo! Thanks for informing us! What Netoholic is doing is realy bad. He warned us, lol. You have my full support of reverts. Furthermore I am not even sure if WP:AUM really applies here, I raised the question on a different issue [1] (a template that's not edited because it is protected) but nobody could tell a reference in WP:AUM that's states how this could harm the servers. So if our subtemplates are stable soon, they may get protected. Tobias Conradi (Talk) 16:06, 1 January 2006 (UTC)

No one can deny that meta-templates (templates that have other templates in them) are harmful to the servers. It's documented in Wikipedia:Avoid using meta-templates. The sooner the non-meta Infobox Language can be fully implemented the better.--Commander Keane 18:07, 1 January 2006 (UTC)
Why this arrogance and agressive wording? (no one can deny - arrogance because you don't seem to have read my previous posting, aggressive because you speak of 'no one can deny' - as if we others are idiots.) From WP:AUM _I_ could NOT learn that meta-templates are harmful to the servers. EDITING meta-templates is harmful. The non-meta Infobox is quite bad design as of now. Why could Netoholic not implement his "improvements" in the language template and by doing so keep the history ... of Garzo's and other peoples great work? and go with the other editors. Why did he need to work where is work could not be seen and switch the templates without informing here? Tobias Conradi (Talk) 07:19, 2 January 2006 (UTC)
Wikipedia:Request_for_Arbitration#User:Netoholic Tobias Conradi (Talk) 07:35, 2 January 2006 (UTC)
  • (Tobias got the following on his talk) bad form: Dude, poisoning the well is a shitty tactic. Please remove this comment, since it's only put there to take a cheap shot at me. -- Netoholic @ 17:49, 2 January 2006 (UTC)
  • Thanks for the link to "poisoning the well". I didn't want to descredit you here with the RfA. So I am sorry if this impression was left. I thought it might be related. Especially after you defended your reverts with the words that you warned users (and now have some right of replacing?) and that "Garzo rolls back your progresses".
  • Please come here and try to present some ideas. I think we are all quite open to reduce the serverload. AFAIK there are more than 7000 ISO 639-3 language codes, so the effect of this template may be huge if we have it on 7000 pages one day. Tobias Conradi (Talk) 20:51, 3 January 2006 (UTC)
Meta-templates require extra calls to render. This is independent of editing a meta-template. From WP:AUM...
Here's some technical background which may be of use. Jamesday 07:52, 13 Feb 2005 (UTC) and clarified by LarryLACa 00:22, 11 November 2005 (UTC) (Clarifications show in []'s).
[The impact of meta-templates is two fold. a) They are more expensive to render. b) editing them has an immediate database affect, regardless of whether the pages using them are currently being accessed.]
[Re: Rendering impact: The first time a page is viewed, i.e. after it has been removed from caches, the page request causes several steps:]
* Each item in the base page portion is requested from the database (images and CSS aren't in the main part). The page you edit, each template, each template included in the template and so on. Two templates, two database records to be retrieved. One template on its own, one read, one template including another, two. Plus the one for the base page.
* Once that and the rest of what is called the parsing is done, the page is saved in the parser cache. That's kept in RAM in memcached.
* Finally the skin is applied and the page is passed on to the Squids, which cache it in RAM and on disk (to get larger capacity but at slower access time) for all who aren't logged in (will only be useful if it's the normal skin) and send it on to the person who originally requested it.
[So meta-templates are more expensive to render, and the impact shows up when the page is requested.].
Maybe I'm the idiot, but this looks clear to me. Meta-templates (even when if they are protected against editing) are to be avoided. Am I wrong?--Commander Keane 10:37, 2 January 2006 (UTC)
  • Harmful effects -> Server load -> When a meta-template is edited, many pages need to be updated in a single instance
  • if it's not edited it is less harmful, I assume.
  • the two templates = two DB calls stuff is _not_ specific to meta templates. one could also say: don't use two templates in one page
  • it is WP:AUM, not WP:DUM (don't use meta templates) Tobias Conradi (Talk) 20:23, 3 January 2006 (UTC)
I believe that we should improve the technical aspect of this template by phasing out its reliance on other templates. I approve of any suggestions that could help here. However, we have built this template on consensus over a long time: we are linguists and understand something of the complexity of the world's languages and the need for a flexible template to serve them. We were made aware of WP:AUM, and have been watching the debate there. I would like proposals of how to make this template more serve friendly made here, and trialled in full view, not concocted and imposed in secret. I am in full agreement with WP:AUM: just in case anyone misreads my actions. However, I see that it mandates no one user to apply the policy, and believe that changes should be made in the usual Wikipedia way: user agree and developing the change together. This is the natural progression of things. I have implored Netoholic to work with others, but he seems only capable of imposing his own ideas. I have reverted the implementation of the new template in articles as it has not been agreed. Read this well: I am not talking about WP:AUM, but the unilateral implementation of this template. Thank you for quoting the policy again above; I understand it and wish to see it implemented. However, as one who has been deeply involved in the develoment of this template, I wish to remain so, and repeat the request that we work together. --Gareth Hughes 17:14, 2 January 2006 (UTC)

Hey guys, please stop arguing. Template:Infobox Language was nothing more than a place for me to demonstrate how much can be done without resorting to meta-templates. There is simply no way to make these changes to the present template since it is so intricate and over-thought, and in such wide use. Unfortunately, it was discovered by Garzo/Gareth before I could properly introduce it and he, failing to assume good faith, interpreted that page's existence as some "secret" effort. Geez, the template is still marked "Inuse" because it still needs work. -- Netoholic @ 17:51, 2 January 2006 (UTC)

You implemented in articles, and you havea history of being argumentative and disruptive. This page was not informed about your 'demonstration': it looks very much the opposite to what you say. --Gareth Hughes 17:54, 2 January 2006 (UTC)
I implemented in precisely four articles, and no information was lost in any of them. Though it worked on those, the template was still not ready to show off. Now, please stop with the personal remarks. -- Netoholic @ 18:09, 2 January 2006 (UTC)
I think other contributors to this page know who I am, and know my history: that speaks for itself. I find it difficult to understand the 'not ready to show off'/'only implemented in four articles' confusion: if it's not ready it should not be in any articles. The nature and future of the template is the business of this discussion page and Wikipedia talk:WikiProject Languages: that latter page has had no discussion of the issue. I am annoyed because you are unilaterally changing (adding the thing to articles without discussion) a WikiProject's template without letting the project know or being a prior member of that project. It is useful to enlighten us about our failure to live up to certain policies and guidelines and to help us keep them, but is not useful to impose your will thus on other users. --Gareth Hughes 20:58, 2 January 2006 (UTC)

Wikipedia's in sore need of some server relief, and this template was the grand-daddy of all offenders of WP:AUM. I've put in place my best attempt at this template, without using meta-templates. Improve away. Unless you need my help, I'm going to pretty much be done with this. Please put your energies into improving this non-meta-templated version rather than squabble about who-did-what. -- Netoholic @ 22:37, 3 January 2006 (UTC)

If you want to get rid of the meta-templates you don't need to make a new template. Why anyone would want to make a completely new template to fix this problem and push it through so aggressively is beyond me. If the problem was just server load, the reverts of the current template would obviously been just as effective. But this strikes me as just being as tactless as it is unilateral. I understand Garzo's frustration and I would like to Netoholic to calm down and have some patience about such a major undertaking before we get enbroiled in yet another drawn out dispute resolution.
Peter Isotalo 12:44, 7 January 2006 (UTC)

[edit] New template

Thanks to the new template, the Ido page no longer displays the symbol it used to contain. Please fix it, thanks. Mithridates 22:19, 3 January 2006 (UTC)

There were a handful of complaints about the template doing really stupid things in the last few minutes: you can post complaints to user talk:Netoholic. --Gareth Hughes 22:36, 3 January 2006 (UTC)
Netoholic tried out a new technique to avoid using meta-templates. This technique uses a CSS kludge that will not work on Wikipedia mirrors (unless they use Wikipedia CSS). It broke a great load of templates. We have already discussed that the family colour is not implemented. However, this is the least of the problems. Ido's infobox went really funny. There is a nice question on talk:Macedonian language about what a lawngreen language is: apparently the template ended up classifying Macedonian as such a language. All the multiple ISO 639-3 codes disappeared and were replaced with bare template code. When I've worked on this template, I've tested everything fully before altering the template itself. Not only must the server load from these bad edits by fairly high, but this must be considered vandalism: wrecking a few hundred articles. --Gareth Hughes 23:20, 3 January 2006 (UTC)
  • it was probably not WP:Vandalism.
  • some of the subtemplates could be put back in the main, reducing harm-effect serverload, but actually WP:AUM also mentions harm-effect editibility. I think meta-templates can increase editibility. Garzo, can you protect the colorfile? And change the file only once a month? Maybe all color request should be placed in the talk of that file? Tobias Conradi (Talk) 14:44, 4 January 2006 (UTC)
Over at Wikipedia talk:Avoid using meta-templates there is a growing consensus that we start using WeebleCode, a kind of parameter kludge, rather than hiddenStructure, a CSS kludge. This could deal with every template:qif call in this template. I have been working on a different way of including certain of the sub-templates in the body of the main template, but it makes more sense to use WeebleCode throughout. The only drawback to this approach is that the parameter |if= (an empty parameter) would have to be added to every article to make the kludge workable. This can be done by a bot if we can get one. All colour changes have been discussed over at Wikipedia talk:WikiProject Languages so far, and there has been no vandalism of the array. Actually, vandalism of this array isn't much fun: you can swap the colours around, but, if you add anything else, everything goes white. I don't think we want to protect it at the moment. --Gareth Hughes 15:23, 4 January 2006 (UTC)
It is disingenuous to say the "only drawback" is the |if= (empty parameter). I'm not going to re-hash the arguments here (discussion is better held on WT:AUM), but some significant other problems exists with that method. -- Netoholic @ 16:11, 4 January 2006 (UTC)
I just returned from the WP:AUM talk page. The WeebleCode did look kind of complicated. Where in the articles does the if need to be added? Protecting: Fun or not, we should be able to state to any AUM-reviewer that we did our best. No problem if you protect it, colors are not changed every day. Why not protect? Or should we add the colors to the main temp? Tobias Conradi (Talk) 16:08, 4 January 2006 (UTC)
Again: why not protect. It is also listed at WP:AUM#Alternatives. Tobias Conradi (Talk) 18:51, 5 January 2006 (UTC)
I didn't want to be accused of 'stonewalling' again, but I shall protect the sub-templates. The WeebleCode is just another kludge: the |if= parameter can be added anywhere in the template call, just like any other parameter. There are different drawbacks to all three proposals: using 'meta-templates', hiddenStructure and WeebleCode. There has been no real discussion or evidence about the weighted pros and cons of each, so it's still difficult to decide. I proposed WeebleCode because a few people with more expertise than I reckoned it a better option. I have a couple of little tweaks in draft that could be added to this template, and should simplify it and reduce the number of sub-templates needed. So, in the meantime, I'll work on those to see if they are viable, and let you know when I think they're ready for a trial. --Gareth Hughes 19:24, 5 January 2006 (UTC)
sometimes I like using mirros. Look to the stonewall argument throught the mirror and you may see the one who posted it, to be the stonewaller. ;-) Tobias Conradi (Talk) 20:27, 5 January 2006 (UTC)

[edit] Template move to Infobox Language - blocked by Netoholics template

the folowing talk is copied from Tobias and Netohlics talk pages Tobias Conradi (Talk) 20:27, 5 January 2006 (UTC)

Hi Netoholic, I would like to move Template:Infobox Language to somewhere, maybe your user-space? Than we can move Template:Language -> Template:Infobox Language , what is a standard conform name for Infoboxes. I will ask for a bot than, to change the articles. Maybe we also find some other task the bot could in the same run. ... and yes, hopefully we can at least reduce some load made by the language. I think the language-project people don't want to be the bad guys that put unecessary server load. best regards Tobias Conradi (Talk) 18:45, 5 January 2006 (UTC)

I'd rather not move this. It will prove to be more useful for the coming conversion since we can migrate to Template:Infobox Language over time. -- Netoholic @ 18:50, 5 January 2006 (UTC)

That would destroy the whole history of the template. Lets bring the original box there and than look for changes. Please don't lock the improvement regarding naming, by your box, a box that currently almost nobody wants to implement. Let's seperate the issues on template-naming and on template-content. Tobias Conradi (Talk) 18:55, 5 January 2006 (UTC)

Please don't think naming is nearly as important as content. Doing a bot run to replace a single redirect is not critical or even desirable. What is important is the flagrant WP:AUM abuse taking place on Template:Language. -- Netoholic @ 18:59, 5 January 2006 (UTC)

I don't think the two issues have the same importance and never thought so. Take care of your usage of the word abuse. Why is it unwanted to have a bot help switch to using correct names? Tobias Conradi (Talk) 19:04, 5 January 2006 (UTC)

Single redirects are no big deal, and no one and no bot should find it necessary to replace them. It is just busywork with no gain. It just adds to the bandwidth and creates an unnecessary extra version in the page history, adding to database space needed. -- Netoholic @ 19:06, 5 January 2006 (UTC)

I thought to have read something diff for template redirects. Than we could switch to correct name without bot use? That would be very good. How about the css-version moved to Template:Infobox Language new? Tobias Conradi (Talk) 19:12, 5 January 2006 (UTC)

[edit] prepare RfC Netoholic

Netoholic wrong edits summary:

Tobias Conradi (Talk) 19:54, 6 January 2006 (UTC)

  • I am tearing my hair out: Netoholic has just re-added his conlang template to all these articles, and invented template:Infobox Sign language and added that to all the sign languages. There is enough evidence to show this user is disruptive and refuses to work with others. --Gareth Hughes 23:20, 16 January 2006 (UTC)

[edit] Conversion

I converted {{language/familycolor}} from a meta-template (it used to call {{switch}}) to a single template and made corresponding changes to how it is called in this template. Please let me know if this breaks any pages. If it works correctly that removes one level of meta'ing on this template. We can probably start removing these nested conditions one at a time. --CBD ☎ ✉ 22:04, 6 January 2006 (UTC)

I installed this re-written {{Language/familycolor}} into the call for {{Infobox language}}. Since this is called and set as a parameter for the call to the language template it avoids the problems of meta-templating, but gives the same family colors functionality. An example can be seen at Template talk:Infobox Language. I also made some minor formatting adjustments to make the new box look more like the current one. Let me know what else should be adjusted. --CBD ☎ ✉ 02:07, 7 January 2006 (UTC)
Since Netoholic reverted my changes and has slapped 'inuse' on the template again I have made a copy at User:CBDunkerson/Sandbox2. That page also shows samples of what the non-meta version looks like (on the left) against the current meta version (on the right). Currently my goal is to copy the existing Template:Language as closely as possible without using meta-templates at all. Still a work in progress, but please take a look and let me know what should be adjusted. If everyone agrees that the template has reached a usable state we can then move it into the proper location and make any adjustments to the articles calling it as needed. --CBD ☎ ✉ 16:48, 7 January 2006 (UTC)
what changes are needed in the articles? Tobias Conradi (Talk) 19:37, 7 January 2006 (UTC)
Nothing specifically yet. What I'm saying is that if we implement a new template with parameter changes (such as changing the 'state' parameter to 'official' as you suggest) then all the calls to the template in the individual articles will need to be updated to reflect any and all such changes. --CBD ☎ ✉ 20:38, 7 January 2006 (UTC)

[edit] What is going on?!?

Excuse me, but what is going on with this template? I think it's good as it is. Bomac 19:59, 7 January 2006 (UTC)

Short answer: WP:AUM says that the sort of meta-templating used to make this template work is bad for the servers and must be reduced. Netoholic has written a new Template:Infobox language template which doesn't use meta-templating, but also doesn't have the same capabilities. I'm trying to adjust a copy of his template at User:CBDunkerson/Sandbox2 to match the capabilities of the current template. Various edit wars and conflicting adjustments have gone on around all of this.
Longer answer: Read this page and associated links from here down. --CBD ☎ ✉ 20:36, 7 January 2006 (UTC)

[edit] ISO 639 - none

Garzo / CBD, can we add "none" or so to the ISO 639 code fields? E.g. Montenegrin language has no code. But the reader would not know whether it is simply missing or does not exist. I think it is interesting for discussion whether something is considered language by ISO or not. Even if SIL/ISO is not _the_ authority to decide this (or any) it is interesting what they think. Tobias Conradi (Talk) 21:39, 11 January 2006 (UTC)

The current chaos really prohibits doing anything constructive with the template. We could change the default of the parameter to 'none' instead of '—'. Most languages don't have an ISO 639-1 code, many constructed languages don't have an ISO 639-3 code. Because ISO 639-2 has a comprehensive set of collective codes every language should be able to find an appropriate one. I thought about having ISO 639-1 optional for all languages, and ISO 639-3 optional for constructed languages. However, this AUM cult has scuppered those plans. ISO 639-3 is still in draft format, and there are a good number of cases for new codes pending. --Gareth Hughes 23:01, 11 January 2006 (UTC)
default should be '-'. 'None' would be explicit statement that there is none. Don't care about the new codes, they can be added. Hey and this clowny Neto can't stop us from improving the infobox? Cheers Tobias Conradi (Talk) 23:26, 11 January 2006 (UTC)

Two wishes: I would like the explicite 'none' statement. AFAICS for ISO 639 1 & 2 we could do so already. For 639-3 a link would be created, this should be avoided. the second gets an extra section Tobias Conradi (Talk) 03:34, 20 January 2006 (UTC)

Go ahead and make updates. It looks like WP:AUM is going to be severely downgraded/changed or scrapped altogether. Brion, the lead developer, has said that the server load problems aren't that big a deal and will be handled on the dev side if/when needed. Meta-templates are still considered 'ugly' and to be avoided when possible, but better that than all the problems which the CSS hack has been introducing. The '|if=' parameter trick can still be used, but only if it is an easier/nicer way to do things than meta-templates like 'qif' would be. --CBD ☎ ✉ 12:16, 23 January 2006 (UTC)
OK. I think Montenegrin language is a bad example as it is usually considered to be a Serbian dialect. However, it would be useful to look again at how we present the codes. I've suggested that we might look at incorporating the Linguasphere language code into the template (discussion here). The ISO 639-1 codes are only given to major world languages, so many languages simply will not have one. The default text is an em-dash, but we might want to change that to read none, as not having a part-one code is the most likely reason why it's not in the infobox. I have thought about making this row removable, so that languages without a part-one code could skip it. However, it is perhaps useful to state that these languages do not have any code assigned them under part one. In part two, every language should be able to find a suitable code, as part two provides 'collective' codes. If an infobox is lacking a part-two code, it is most likely that the authors cannot find the appropriate code. Part three does not allow such collective codes and is attempting to be comprehensive. The real problem with part-three codes is with constructed languages: most will not have any code assignment in this part of the standard. We could make this field optional for constructed languages. In the meantime, I'll move the to be added notice from the default for {{{iso3}}} to {{{iso2}}} (it was added before the ISO 639-3 codes were added to articles), and make none the default for {{{iso3}}} and {{{iso1}}}. --Gareth Hughes 14:55, 23 January 2006 (UTC)

Don't ! If someone adds a new article and cannot look up the code/ is lazy he will not add it. for iso3 the default should be "---". "none" should only be displayed if some looked for the code and there was none. That's an explicit none than. for iso1 default none is ok, because all the articles do allready exist, the code set does not change, is small, can easily be looked up. Tobias Conradi (Talk) 04:22, 24 January 2006 (UTC)

Yes, I see the logic in that. I've changed the default for {{{iso3}}} to an em-dash, but left the option that when the parameter is equal to none it returns the text none. I hope that looks right now. --Gareth Hughes 13:24, 24 January 2006 (UTC)

[edit] script / alphabet

can we include script? maybe it could be done by using an array like for the families. So we stay flexible as for where the link goes and how to analyze the script variable. Maybe the IPA notice can later on depend on the script and can be removed. I once read someone thinking about codes for scripts - does anyone know about it? what about alphabet? Tobias Conradi (Talk) 03:34, 20 January 2006 (UTC)

I think the neutral term would be writing system. The easiest implementation of this would be to have an optional row that can be added to the infobox that allows a blank field to be edited (e.g. |script=[[Latin alphabet]]). This would allow the maximum flexibility, but would make automatic changing of the footer notice more difficult. A code-based array might work, but would be more complicated to put together. --Gareth Hughes 12:14, 23 January 2006 (UTC)
I really prefer array, especially after WP:AUM seems to have been downgraded. if "script" is incorrect the variable should be IMO "write", "writing", "writingsystem". I tend to "write" because it is short. Tobias Conradi (Talk) 04:26, 24 January 2006 (UTC)
I was thinking of 'Writing system' as the text to be displayed in the left-hand column: {{{script}}} seems like a good name for the parameter. I just wonder whether a code based array would be flexible enough. There are many languages with two or three different writing systems, or a different system used in different eras or places. On the positive side, many such languages have a single page covering the various differen systems (e.g. Mongolian alphabet). Some languages use a well-known system, but have a separate page on that language's implementation of the system (e.g. Welsh alphabet). I wonder how to fit this amount of information into a parameter. --Gareth Hughes 13:35, 24 January 2006 (UTC)
ok, array seems not to be (easily) feasible. Before starting let's maybe analyze what will be put in. for param name: are the writing system all called scripts? if not i would prefer no to use script, or only use it for scripts. Tobias Conradi (Talk) 03:17, 25 January 2006 (UTC)

List, please improve:

  • Latin
  • Latin modified
  • Latin with diacritics
  • Latin with extra letters (azeri)
  • Russian cirillic
  • Serbian cirillic and Latin modified (serbian)
  • Katagana, Hiragana (japanese)
  • ...

[edit] Template:Infobox Conlang

Please view discussion about this template on Wikipedia talk:WikiProject Languages#Template:Infobox Conlang. -- Netoholic @ 07:14, 12 January 2006 (UTC)

Moved thread replies. PLEASE keep the conversation on the WikiProject talk page. -- Netoholic @ 17:27, 12 January 2006 (UTC)

[edit] Announcement

I am kindly requesting all interested parties to review discussion regarding this template on my talk page: User talk:Ezhiki#Infobox:Language issues and User talk:Ezhiki#Infobox:Language proposed solution. I realize that I should have announced it a lot earlier than this, but I did not expect the discussion to take as long as it did. Your attention and comments will be much appreciated.—Ëzhiki (ërinacëus amurënsis) 16:32, 16 January 2006 (UTC)

[edit] maps

Image:Niger-Congo.png IMO we should upload an image with the project colors. Tobias Conradi (Talk) 03:46, 20 January 2006 (UTC)

I think the colors are fine, but here's the complete list of those maps anyways:
--Khoikhoi 01:29, 21 January 2006 (UTC)
I know when I was looking at this I was confused by the colour changes - probably indicates a bear of little brain, but there will be others like me out there. I like Tobias' suggestion if someone has time to do it. Gailtb 08:52, 21 January 2006 (UTC)

Interesting idea to highlight the main subject. Did not knew this feature before. Nevertheless I tend to prefer the project-color way. Is SVG format better? I am not an image professinal. And never did any SVG file. If we have a worldwide map with project colors in SVG we can then cut whatever regions / language groups we want. Tobias Conradi (Talk) 19:06, 21 January 2006 (UTC)

I created this series some time ago. I have been collecting comments over at the commons version. I have been meaning to update the series for some time now, so I might give the colors a shot. However, I must say I quite like the current coloring. As Tobias says, in the four individual maps, the main subject is highlighted. I think this is only confusing when one puts them next to each other, but normally they are seen in the context of their respective articles so then there's no problem. As for switching to the project colors, one of the problems would be that the distinction between Niger-Congo (A) and Niger-Congo (Bantu), which serves to show the extent of the massive Bantu family, would be lost. — mark ✎ 09:14, 27 January 2006 (UTC)
I might suggest using both methods. That is, use the 'highlight' method to make the language(s) being addressed more prominent, but leave the languages around it on the map in their project colors. This might require some adjustments to the color scheme to take visual presentation on a map into account (and remove the 'highlight' color from the normal scheme), but would help to standardize when looking at maps in different articles. Alternatively, you could use these 'highlight' maps as they are but then have an overall 'language distribution' map using the project colors when not highlighting any one language/group. --CBD ☎ ✉ 12:01, 27 January 2006 (UTC)

[edit] Templates for Deletion

I am happy to announce the following polls:

As I'm not looking forward to the next time Netoholic decides to spread these all over the article space, I thought we might as well vote them out of use. --Gareth Hughes 14:25, 24 January 2006 (UTC)

[edit] Pronunciation

I've created Template:Language/pronunciation and modified Template:Language to optionally include the pronuncation of a particular language.

The syntax added in template:language is simply

 {{if defined
   |test={{{pronunciation|}}}
   |call=language/pronunciation
   |1=1={{{pronunciation|}}}}}

See the effect in English language, German language and French language. --Hello World! 04:49, 27 January 2006 (UTC)

This works well; thank you. --Gareth Hughes 13:57, 27 January 2006 (UTC)
I added the substed equivalent of {{IPA}} so it should display properly for everyone, too. --Angr 14:43, 27 January 2006 (UTC)

Since I previously argued for it, I'm very happy to see this parameter. I wonder about a couple of small improvements:

  1. Would it be better to have the label as "Native name /pronunciation:" (ie move the native name down from its current position)? That way it would be absolutely clear what the pronunciation is of.
  2. When using it on pages, shouldn't there be / / or [ ] around the IPA (cf Hungarian language)?

Gailtb 19:52, 27 January 2006 (UTC)

Yes, Gail, I think that we might want a more descriptive left-hand column than 'pronunciation'. However, I feel that moving the native name out of the header reduces its importance (somewhen there was a debate that concluded that we should try not to place the native name below the English name, but beside it). On the second point, you are quite correct about proper transcription of IPA (I often get so worried about the squiggles that I forget what goes round them). I suppose we should aim for a fully phonetic representation, but in practice this might be untidy and divisive (which dialect do we use?). --Gareth Hughes 20:38, 28 January 2006 (UTC)

[edit] Propose tweaks to template

It seems that the template is triggered into sign language mode in one of two ways: by assigning a value to the "signers=" parameter or by familycolor=sign. Oftentimes even an estimate of the number of signers is just not possible — but when familycolor=sign and no "signers=" value is assigned, it seems that the family is still given as "sign language" rather than "unknown" (see Turkish Sign Language). I'm not sure how to change this myself, sorry, but I would like to see the family as "unknown" in both cases.

Also, I don't know how easy it is to make a further change, but I would prefer the when the family is unknown to just have the text "unknown" in the infobox, not "unknown" followed by the language name. (see German Sign Language for example).

Thirdly, the bottom bar of the sign language template has two links - sign language and list of sign languages - very nice! But the sign language link is written as "sign languages" so unless there is a reason for this I'd like to remove the "s". Thanks! ntennis 00:10, 28 January 2006 (UTC)

Things would work much better if sign languages used their own infobox, tailored specifically for their needs. -- Netoholic @ 00:43, 28 January 2006 (UTC)

I'm happy to go with consensus either way on this. I don't think others are opposed in principle to a seperate sign language template either, but what are the pros and cons? IIRC we went with a single template for simplicity and unity, and it does work well as is. I guess if there was a seperate template, it would require less esoteric syntax, and i could make the changes myself without bothering the brainy coders like Garzo! ntennis 01:02, 28 January 2006 (UTC)

You're part of making the consensus. Unfortunately, using one language template gives "unity" but falls very short on simplicity/elegance. A separate template would not require any esoteric syntax. Like you said, you could make changes as needed. The main language template likewise becomes much simpler, as well. Keeping the "look" the same is no different than other infoboxs are handled (the WikiProject keeps Template:Infobox Arcade Game and Template:Infobox VG in a common style). I had created a reasonable Template:Infobox Sign language, but that was prematurely nominated for deletion for somewhat emotional reasons. It's yours if you want to go ahead with it. -- Netoholic @ 03:07, 28 January 2006 (UTC)
The changes ntennis is requesting would be no more or less complex with multiple templates rather than a single one. He's asking for differences in the default presentation of sign languages. Easy enough to implement if people agree they make sense. --CBD ☎ ✉ 16:02, 28 January 2006 (UTC)
The fact that he has to ask someone else to make the changes because this template's code is so complex is the problem. -- Netoholic @ 00:45, 29 January 2006 (UTC)
Maybe it's because it's protected too. You probably haven't noticed but conditionals have moved a lot of the formatting out of the template calls and into the template. That means it is easier to modify the template contents. --Gareth Hughes 00:51, 29 January 2006 (UTC)
It's only semi-protected, and even that doesn't seem to have any reason. And what are you talk about... my version of separate sing language template didn't require any special "formatting" on the article side. -- Netoholic @ 05:50, 29 January 2006 (UTC)
Protecting important stuff (here: "used a lot" stuff) isn't that bad. Some templates just shouldn't be treated like articles when editing. There is nothing wrong with doing the change work in a sandbox on the side, presenting it on the talk, gathering consenus for it and then asking an admin to apply it (in the case of full protection). Revert warring forth and back, as you prefer and in-situ surgery is just bad for the servers, something which you usually care a lot. --Adrian Buehlmann 10:14, 29 January 2006 (UTC)
Yeah, this isn't really a problem. I've corrected template:language/signnotice so that it bypasses the redirect to sign language (that's point 3 dealt with). You are correct about the way the sign-language options are triggered. The initial idea was that these options would be turned on by having the parameter {{{signers}}} defined in the call. However, I then realised that there often isn't adequate data about the number of signers of a given language. That's where 'familycolor=Sign' comes in: it triggers the options without anything needed for {{{signers}}}. The 'genetic classification' box is designed to build itself out of available information. If it is given nothing else, it has text determined from {{{familycolor}}} with the text of {{{name}}} in bold below. The complete override for this is {{{family}}}. If that parameter is defined, its contents will be put into the box instead of the automatic stuff. The first line can be altered separately by using {{{fam1}}}. However, as we are looking at a global solution for sign languages, it seems a bit much to use these workarounds. As I understand it (correct me if I'm wrong), we want the automatic text in the 'genetic classification' box for sign languages to read simply "unknown" (and nothing else) as a default. I'll test it out, and if it works OK, I'll upload it. --Gareth Hughes 20:59, 28 January 2006 (UTC)
I didn't do the best job of it, but the infobox should now display as requested for sign languages. --Gareth Hughes 22:42, 28 January 2006 (UTC)

Brilliant! Thanks again. ntennis 23:22, 28 January 2006 (UTC)

Can I suggest getting rid of the term Genetic classification on this template? Guinnog 18:30, 5 February 2006 (UTC)

Can someone incorporate this change [2] (Alphabet) into the wiki machine? Ksenon 19:43, 5 February 2006 (UTC)

  • Guinnog, what do you suggest as an alternative?
    • 'Language family' Guinnog 16:24, 6 February 2006 (UTC)
    • On the grounds that we are not talking about genes but about language here. Guinnog 01:15, 7 February 2006 (UTC)
      • If I just changed it myself, would that cause any problems? I am a bit nervous of doing so. Guinnog 00:11, 9 February 2006 (UTC)
        • You could just change it: it should be straightforward enough. I suppose 'genetic' emphasizes that the relationship is one of common origin or genesis, and not just the lumping together of languages for convenience (areal classification) or tidyness (super-family hypotheses). --Gareth Hughes 00:26, 9 February 2006 (UTC)
          • Thanks, I'll try and work up to it. I understand your points, but 'genetic' has an undertone I don't like. Guinnog 00:44, 9 February 2006 (UTC)
  • Ksenon, which change do you wish to see incorporated where (elaborate)? --Gareth Hughes 13:16, 6 February 2006 (UTC)
I believe Ksenon was referring to this change to add an alphabet line. Might make sense, though I'd suggest making it optional. Does get complicated with languages which have more than one alphabet used though. 'Moldovan' and Sindarin come to mind. It's somewhat similar to the recent change to show the name of the language in its natural script and might be combined with that. --CBD ☎ ✉ 13:32, 6 February 2006 (UTC)

[edit] Bot

Could we get a bot to make the change from {{language}} to {{Infobox Language}}? — mark ✎ 13:20, 7 February 2006 (UTC)

There is no significant benefit in replacing that text, since the redirect leaves everything working fine. Update it later when there is some substantive change to the articles, or if a change is made to how this template is used. -- Netoholic @ 15:12, 7 February 2006 (UTC)
Thanks, it's good to know that. I'll just change it in articles when I come across it during normal editing. — mark ✎ 15:38, 7 February 2006 (UTC)

[edit] Pronunciation

Several infoboxes have pronunciation information; but there's no line listed for pronunciation in the gray box above. Is the IPA template automatically included in the pronunciation line, or does it need to be added separately? Angr/talk 05:57, 9 February 2006 (UTC)

Hi Angr. The sub-template that deals with this option is still living at template:language/pronunciation. The contents of template:IPA have been subst-ed into it, so there's no need to add it again. I'll add a line to the instructions about it. --Gareth Hughes 12:15, 9 February 2006 (UTC)
I've been trying to think of a better label for this parameter. My best idea so far is "Native name". What do you think? Of course the instructions will need to be clear about how this differs from the nativename parameter, but I think that can be managed. With the IPA label and format it seems very obvious that the information is pronuncation, so more useful to have a label saying what it's the pronuncation of. Gailtb 19:19, 12 February 2006 (UTC)

[edit] Writing system

I've finally got around to adding a sub-template that allows optional display of writing system. The change to the main template is this diff, and the sub-template that it calls is template:Infobox Language/script. The set up is basic: it produces an extra row below 'language family' with the left-hand column showing "Writing system:" and the right-hand column displaying the contents of {{{script}}}. To include the row, simply add |script=foo anywhere in an article's template call.

Next comes the question of how to manage the content of this parameter. I propose that we keep it very simple: the infobox should link to a article on the writing system, not be a mini-article on it. Some writing systems, like Latin, are used with only minor variation by hundreds of languages. Therefore, I suggest that the writing system for English be described as "Latin (English variant)". Any thoughts? --Gareth Hughes 19:00, 9 February 2006 (UTC)

Great idea, variants are also important (German, Scandinavian, Turkic, etc.). But I was also thinking of a parameter descriping it as either pictorial, alphabetic, iconographic, etc.? This would be helpful in classifying any language- ancient, modern, etc. Ksenon 19:51, 9 February 2006 (UTC)
I think it should be simpler still than Gareth suggested. I would say the writing system for English should just be described as "Latin", period. Angr/talk 19:54, 9 February 2006 (UTC)
Yes, I think keeping it simple is the way forward. There is no need to describe 'Latin' as 'alphabetic' if the link takes you to an article that says as much. However, I suggested the possibility of including variants because we do have some variant-specific articles for writing systems, like English alphabet. --Gareth Hughes 20:01, 9 February 2006 (UTC)
Yeah, but those can be linked to from the article. No need for it all in the infobox. Angr/talk 20:12, 9 February 2006 (UTC)
Egyptian language needs a template along with writing system info. Ksenon 05:28, 12 February 2006 (UTC)
I've put up a suggested template at talk:Egyptian language. --Gareth Hughes 13:46, 12 February 2006 (UTC)

Why do all of you talk about writing systems? Is the word script not equivalent? If not, can the article "writing system" be enhanced? Tobias Conradi (Talk) 13:59, 12 February 2006 (UTC)

I don't understand what you want. The article Writing system says the term "script" is equivalent, and it's a full-fledged article, not a stub. What "enhancement" are you looking for? Angr/talk 15:18, 12 February 2006 (UTC)
I wonder whether there is a diff between writing systems and script because you all use the term writing system which is btw longer than the term script. If there is a diff than I would like this to be stated in the article "writing system" - no big need to say so here. Tobias Conradi (Talk) 09:32, 20 February 2006 (UTC)
Finally someone did it [3] it seems we have to change this template. script != writing system Tobias Conradi (Talk) 03:33, 22 March 2006 (UTC)

The difference between scripts and variants of scripts is a bit like the difference between languages and dialects. What do we do with Irish, Fraktur, Coptic, Gothic..? --Ptcamn 15:51, 12 February 2006 (UTC)

I think that, when we have an article on a given language's writing system (English alphabet for example), we should link to that in the infobox. However, I think it is also important to say which general system is being used (in this case, Latin alphabet). If, for example we have an article about X language and the infobox says it' written in X alphabet, knowing nothing about X we can only assume that this is some special script used only for it. However, if the infobox tells us Arabic abugida (X variant), we know that the Arabic script (linked) is used, and we link the specfic X alphabet article that tells us that lines are drawn under various letters to represent the ejective consonants found in X but not in Arabic. Thinking about the examples above, we should say Latin alphabet (Irish variant), Fraktur isn't really a writing system but a variation of the German variant, Coptic alphabet (based on Greek) and Gothic alphabet. I would like to see us adopt a format that is both concise and clear. --Gareth Hughes 16:29, 12 February 2006 (UTC)
Okay, that makes sense. The reason Irish alphabet is a red link is that the article is at Irish orthography. Angr/talk 16:35, 12 February 2006 (UTC)
I've put a redirect in there, thanks. --Gareth Hughes 16:56, 12 February 2006 (UTC)
What makes Coptic an alphabet based on Greek, rather than a variant of Greek? It was treated as a variant by Unicode until recently. How do you decide what's a variant and what's a script while staying NPOV? --Ptcamn 20:13, 12 February 2006 (UTC)

Just to say I support the Latin alphabet (Irish variant) format - quickly gives information about whether it's an alphabet, abiguda or whatever, tells which system it is based on, and links to info on what the specifics are. I also agree with the use of "writing system" rather than script as the label - script can be used to mean handwriting as opposed to print, whereas writing system is unambiguous. Gailtb 19:11, 12 February 2006 (UTC)

Seven (eight including jinkim) letters of the Coptic alphabet are derived from Demotic Egyptian. Unicode's usage of the Greek block for Coptic was seen by many as a poor compromise. I think that Coptic writing is sufficiently different from Greek not to be considered a variant: many of the letter forms are of a different style, and the extra letters impede reading by anyone who does not know them. I think a couple of extra letters (like ɛ and ɔ used in a number of African languages) could be added without saying that this is a new alphabet, but more letters, different style and different phonemic mapping suggests that this more than a variant. --Gareth Hughes 20:36, 12 February 2006 (UTC)
I agree. Cyrillic is derived from Greek too, but it's considered a separate alphabet, not a variant of the Greek alphabet. Angr/talk 20:48, 12 February 2006 (UTC)

[edit] Language templates on TFD

I listed a whole bunch of templates that are basically copies for specific languages of the now deleted Template:ll. Since we got {{ll}} to be deleted by the votes of the people at this page, perhaps you all could vote again here. Thanks. --Khoikhoi 05:54, 20 February 2006 (UTC)

[edit] Proposal: Standard language typology description

Since you like to classify languages, please take a look Talk:Linguistic typology#Proposal: Standard typology description in articles about languages. `'mikka (t) 01:12, 22 April 2006 (UTC)

[edit] sub-templates

I think the Infobox Language template can be rewritten by including Template:!, so some (not all, I think) sub-templates such as Template:Infobox Language/agency can be disused and hence reducing the need to load so many sub-templates. What do you think? --Hello World! 14:53, 23 April 2006 (UTC)

[edit] Literacy

How about adding a "literacy rate" box next to script? --Ptcamn 01:41, 28 May 2006 (UTC)

I'm not sure that would be very meaningful. It's the kind of statistic that would normally be given for a country rather than a language. It might be possible for languages with a localised distribution, but for a language like French or Spanish which is spoken by very different populations across the world, I think it would be hard to get the info, and probably not very useful to anyone. Gailtb 09:04, 28 May 2006 (UTC)
I also think it would be hard to get the info. Also if you have it, what does it mean? I know of African societies that have a reported literacy rate of zero, yet they are writing in Ajami script or in an indigenous script. Literacy is a thoroughly problematic notion. — mark ✎ 13:51, 28 May 2006 (UTC)

[edit] OT ISO 639 redirects

little bit offtopic, but you may like: Category:Redirects from ISO 639. Tobias Conradi (Talk) 11:02, 15 June 2006 (UTC)

[edit] implementation quirks

I'd like to use the template on my wiki (ver MediaWiki: 1.7alpha r14796, ParserFunctions succesfully installed, "#if" works). However, the code doesn't get fully rendered. Any advice available? --tickle me 01:25, 24 June 2006 (UTC)

The first thing which comes to mind is to ask if you have also copied over or converted all of the 'sub templates'; Infobox Language/IPA notice, Infobox Language/agency, Infobox Language/aposteriori, Infobox Language/codelist, Infobox Language/creationdate, Infobox Language/family-color, Infobox Language/genetic2, Infobox Language/image, Infobox Language/map, Infobox Language/official, Infobox Language/pronunciation, Infobox Language/rank, Infobox Language/script, Infobox Language/statesregion, and Infobox Language/terminological. Without those this template would not work correctly. --CBD 11:16, 24 June 2006 (UTC)
It looks a little like one of the templates has not been copied correctly — some of them begin with a non-breaking space that is essential to their correct display. If the parser functions are working properly, I would suggest going through all the sub-templates and copying them afresh. — Gareth Hughes 11:23, 24 June 2006 (UTC)
"...not been copied correctly": Indeed, though I used only Ctrl A/C/V for failsafe copying, something ...happened. Works now, thx both! --tickle me 02:57, 25 June 2006 (UTC)
I must apologize that the templates are all very fiddley. The fact that something was lost in transition doesn't surprise me. Perhaps sometime soon we can do an overhaul here. I adapted the templates to be more flexible when the very first, basic functions were available — they can probably be made neater. I'm glad it's working now. — Gareth Hughes 11:49, 25 June 2006 (UTC)
Yes, finding out about the dependancies and the needed ParserFunctions1 wasn't easy. Some succinct documentation (...templates for dummies) and an import function that deals with all things necessary would be great. Hopefully, I'll find the time eventually... Don't get me wrong, please: All things considered, I'm glad this isn't payable, binary MS one-click stuff :-)
1 ...whose "if", needed for some templates, works only from MW 1.7+ ...without throwing up helpful errors msgs when installed on 1.6 downwards. I had 1.6 installed first, as newbies shy away from alpha builds.
--tickle me 19:45, 25 June 2006 (UTC)

[edit] Wiki Link

Would it make sense to have the Wiki of that language in the Infobox? Have the French infobox give a link to fr.wikipedia.org and the Latin one link to la.wikipedia.org? —The preceding unsigned comment was added by Cigarette (talk • contribs) 14:26, 20 July 2006 (UTC)

[edit] oc:modèl:lenga

Occitan wiki also has its template: oc:modèl:Lenga. Could you add it in interwiki? Thank you. —The preceding unsigned comment was added by 201.0.66.213 (talk • contribs) 10:29, 30 September 2006 (UTC)