Template talk:Infobox language/Archive 1

Archive 1

Archive 2

Archive 3

→

Archive 5

Pronunciation?

There has to be some room for pronunciation. Template needs fix or improvement. --Rrjanbiah 17:50, 27 Jan 2005 (UTC)

You have just broken this template. If you insert a parameter that shows through in the article space if it is not defined, then you have to define it in every single article. I cannot see the value of your change. I suggest you fix it. Gareth Hughes 12:14, 5 Feb 2005 (UTC)

Rather than revert your edit, I put it in comment tags. Now the template isn't broken. If you would like to remove the comment tags, first define the parameter in every article that calls the template. Gareth Hughes 12:21, 5 Feb 2005 (UTC)

Can we delete the improvement by Rrjanbiah: the addition of the pronunciation parameter? I'm not sure what we are supposed to do with it, and its inventor is showing no sign of implementing it. For example, for French language would the infobox render pronunciation with the IPA for French or Français? The latter would make more sense, but this raises more questions: what about non-standard pronunciations? Is there any mileage in this parameter? Gareth Hughes 11:20, 11 Feb 2005 (UTC)

I thought it'd already been reverted. It doesn't seem to be there anymore. — Gwalla | Talk 06:20, 12 Feb 2005 (UTC)

Keep up your good work. Indeed, a wonderful argument. --Rrjanbiah 06:30, 12 Feb 2005 (UTC)

I've made some pronunciation sound files of languages over att Japanese, Swedish, Russian and Mandarin. Why not try seeing how these fit into the template? We could make an attempt to request pronunciations from other languages.

Isn't a pronunciation in the standard high class varieties of most languages better than no pronunciation at all? I'm sure a lot of people are more interested in actually hearing the pronunciations rather than just reading rather cryptic IPA-characters. Peter Isotalo 20:45, Mar 20, 2005 (UTC)

The proposal was for a line in the infobox that gave the pronunciation of the name of the language. My main gripe about it was that the proposer added the line to the infobox template but did not update any of the articles to use it: in effect, breaking the template. I think it would be great to have a wider range of pronunciation files, but that was not what this proposal was about. Gareth Hughes 10:38, 21 Mar 2005 (UTC)

Alignment

Timwi's recent change of align="right" to class="floatright" broke the template's right-floating characteristic in Firefox under the Classic skin, and I'm not sure why since this seems like the correct way to handle this. Is anyone else having this problem? Bryan 00:35, 5 Feb 2005 (UTC)

Yes. Unsurprisingly, it breaks things in the Mozilla Suite as well. I'll fix it up... — Gwalla | Talk

Can we talk about what we want the temaplate to look like?

There have been some good updates of the template recently, but there has been little talk about here. As there are a few ideas about what we ant from this template, I thought it might be best to air them here and see what the consensus really is. The issues I've seen are:

1. Addition of pronunciation line in template to record the pronunciation of the name of the language. As the IPA template won't work inside another template, and the fact this would involve updating every instance of the template with very specific information, I suggested we shelve the idea.
2. Whether we should align, float or something else seems to be a technical display issue.
3. A few users seem to advocate border="0": as it has been consistently reverted, I assume we don't want to get rid of the border.
4. Flamurai's change to CSS throughout the template is good, but this has got rid of cell borders. Do we want cell borders?
5. Mustafaa suggested on the project page that, as most languages do not have anything worthwhile to add to the Official status section of the template, we allow a variation without this section. If we can combine the two modes into this template (call a parameter to collapse/expand the template) without having to fix every template immeadiately, this would be the best path to follow.

Any thoughts? Gareth Hughes 12:05, 28 Feb 2005 (UTC)

• It's difficult to understand at a glance without cell borders. I think we want them. — Gwalla | Talk 19:49, 28 Feb 2005 (UTC)
• I like Mustafaa's suggestion. I also like cell borders (if light & max 1px). — mark ✎ 20:26, 28 Feb 2005 (UTC)
• I've added pronounciations of language names in the introductions at Swedish, Russian, Japanese and Mandarin. I'd love to see what the template would look like if we added these, especially if we can keep the general design of the audio template with the little speaker-PNG. Peter Isotalo 14:04, Mar 17, 2005 (UTC)

Extinct languages?

Hi, I've started (barely) an article on Tunica, and I was wondering whether there's some way to indicate the fact that it's extinct in the template (or even if other folks agree that that's the place for it). I don't have strong feelings either way, except that the "Official status" fields seem out of place. --babbage 22:37, 12 Mar 2005 (UTC)

Yes, we have jsut been discussing what to do with the 'official' section of the template. I think it might be best to replace it with a number of modules that are loaded by parameter into the existing template — the current 'official' section, a section for extinct languages &c. However, that would require a parameter to be placed in every use of the template before a change could be made. Any thoughts? Gareth Hughes 23:09, 12 Mar 2005 (UTC)

What happened to the template?

I liked the color bars indicating the language family; where did they go? I don't much like the "See also" bit; it seems kind of intrusive, and breaks the theme of the template insofar as it does not describe the language. But maybe at a smaller font size it would look better... Any thoughts? - Mustafaa 20:47, 14 Mar 2005 (UTC)

The template used to show the colour bars in orange, but since it was changed to CSS they show white. It's really odd, but I think it's about parameter defaults. I agree that the see also line is unneccesary, and I would prefer to see each line of the classification box linked to an article instead. Gareth Hughes 21:34, 14 Mar 2005 (UTC)

Modular template proposal

{{{name}}} ({{{nativename}}})
Spoken in:	{{{states}}}
Region:	{{{region}}}
Total speakers:	{{{speakers}}}
Ranking:	{{{rank}}}
Genetic classification:	{{{family}}}
Language codes
ISO 639-1	{{{iso1}}}
ISO 639-2	{{{iso2}}}
SIL	{{{sil}}}

Official status
Official language of:	{{{nation}}}
Regulated by:	{{{agency}}}

Language extinction
Year of extinction:	{{{extinct}}}
Last known speaker:	{{{speaker}}}

Literary language
Literature:	{{{literature}}}
Golden age:	{{{gold}}}

I've put together a few ideas about how to proceed with this template. The idea would be to have a basic template (shown right).

I hope we can all agree on that as the basic format. It is the infobox we already have without the official status and see also sections. For language articles where we want to use the basic template, we would add the parameter module=none into the wikitext.

Then I suggest that we have a number of additional modules that can be inserted into the infobox. For those language articles for which the official status section is relevant, we would add module=official into the wikitext. When the change takes place, this would reinsert the official status section in its current position.

The section by itself is shown below right.

Once we've got the idea of being to insert a module into the infobox, we could create a few extra ones for specific purposes. I thought of two we could use, shown below right, with the code module=extinct and module=lit.

Other modules could be created as we see the need. The benefit of this modular approach, appart from increased flexibility, would be that all language articles are still using the same basic template, rather than different ones. Any thoughts, or should I just start setting it up? Gareth Hughes 12:15, 16 Mar 2005 (UTC)

The smaller templates for things like extinct or literary languages sounds like a very good idea to me. However, is it really practical to have just one basic template and a bunch of add-ons? Why not make two basic templates, one long for the bigger languages that actually have offical status and such, and one short for the smaller, but not extinct languages, that don't have any official status or such? Mustafaa suggested a short version without a information for ranking at Wikipedia Talk:WikiProject Languages. I think that one and the current would be enough for our purposes.

What do the rest you think? Do we need actual templates for the extinct or literary languages? Peter Isotalo 13:25, Mar 17, 2005 (UTC)

A few points. First, I like the modular proposal very much. It looks like the solution we need.

• As for official status and such: there are many countries that make a three-way distinction between 'official' languages, 'national languages', and the rest. Zambia for example has appointed seven national languages: Bemba, Nyanja, Kaonde, Lozi, Lunda, Luvale, and Tonga. However, English is the main official language (cf. Constitution of Zambia 1.1.5), used by the administration and until recently almost exclusively in education. South Africa is a similar example that I know of, and no doubt there are many more cases like this. I think it is very relevant that Bemba and Nyanja have a special status, contrary to, say, Yao (an endangered language of Zambia). What do others think about this? (Actually, there is something related to this that needs to be sorted out in the country box template as well. It always bothers me to see 'English, other languages' there; as if only English is relevant.)
• Peter Isotalo, what's the advantage of having two different basic templates over the modular approach? Having only one basic template seems more consistent to me.
• In my opinion, the ranking bit definitely doesn't belong in the basic template — by definition it will be relevant for at most 100 languages (out of a total of approx. 6000 languages now spoken). Additionally, there is no agreement whatsoever on the correct ranking, not even in the top ten! I'd say this is an unusable piece of information.

And in answer to your (Peter's) question, I really like the 'extinct' module. I don't understand the {{{literature}}} item of the 'lit' module fully, though, what would one expect to go there? — mark ✎ 15:31, 17 Mar 2005 (UTC)

• The modular idea means that each language article uses the same basic template, and clicking what links here on the template page brings up a list of every instance of the template. It also forces all variations to be consistant. If there were two templates, an aesthetic change could be made to one but not the other. I agree that either the language ranking should go, or just be kept for the official module. I think the official module could handle all sorts of variation in official status — the Bemba language article could have: official status Zambia (national language). The idea behind the lit module is to deal with those languages that are effectively extinct, but are still the languages of important literature, e.g. Sanskrit language. The {{{literature}}} box is intended to be a list of major literary works in that language. I wonder if there would be any use of an endangered languages module. Gareth Hughes 11:16, 18 Mar 2005 (UTC)

What will this modular approach do to the color scheme? Are we going to ditch it altogether and go for one color? Will there be coloring at all? Peter Isotalo 10:27, Mar 21, 2005 (UTC)

No, the colour bars are the same in this proposal: the CSS has not been altered, and the {{{familycolor}}} parameter is called from the article and applies to the main template and any modules that load into it. Gareth Hughes 10:41, 21 Mar 2005 (UTC)

Does anyone feel like trying this proposal (without the ranking) out yet? It's a lot more flexible than the current one and I don't see any reasons not to start applying it. Any objections? Peter Isotalo 21:21, Apr 18, 2005 (UTC)

Template design

The user 80.178.165.124 changed the table design back to the old one that seemed a lot less gracious to me with the following explanation in the edit summary:

"this change allows also browsers other than Microsoft Internet Explorer to view this template properly"

Being a user of Firefox for both Mac and PC this somehow doesn't make sense to me, since I have never had problems with the new table designs. Could someone comment this? Peter Isotalo 21:03, Mar 27, 2005 (UTC)

That user changed it again, and I changed it back. I've left a note on the user's talk page. The change was from CSS back to HTML. This would suggest that the user is using an old browser that isn't kitted out to deal with CSS. Gareth Hughes 13:25, 29 Mar 2005 (UTC)

Exactly. I'm using Firefox just like Peter Isotalo and I don't have any problems. — mark ✎ 14:45, 29 Mar 2005 (UTC)

Should we make natives bend over backwards?

Karmosin removed the italics from around the native name, and Mark Dingemanse put them back. In between times, I removed a work around I had added to the template in Aramaic language to prevent italicisation of the native name (I had placed a single quote at the beginning and end of the parameter string to force the text bold instead). I did this because the italicised Unicode versions of the Hebrew and Syriac alphabets just look rotten. The italics might look good for Latin script, but probably not for many others. Gareth Hughes 00:32, 3 Apr 2005 (UTC)

I changed it because it looked fairly awful for Chinese characters. Do we really need to keep it? Peter Isotalo 00:51, Apr 3, 2005 (UTC)

A, didn't think about non-Latin characters, that's pretty stupid. I should've remembered that I had come across some examples of screwed-up IPA (because of the italicization) myself. Changed back! — mark ✎ 06:53, 3 Apr 2005 (UTC)

Sign language template

I created a template:Sign language to begin to address some of the difficulties I was having fitting Auslan into the general language template. Here is an example of the template in action:

{{Sign language|name=Auslan|fullname=Australian Sign Language
|familycolor=silver
|states=[[Australia]]
|region=Northern dialect (NSW, QLD) and Southern dialect (all other states)
|signers=6500|rank=''Not in top 100''
|family=[[BANZSL]]
    '''Auslan'''
|nation= -
|agency= -
|iso1= -|iso2=sgn-AU|sil=ASF}}

It's a temporary fix, and more changes would make it more suitable for sign languages - but for now I wanted to keep it as close as possible to the spoken language template.

I think the modular idea above is a great one. If the current language template infobox was modified, using language such as "number of users" rather than "number of speakers", then I could erase the sign language template and use the general language template (with added modules if desired).

If we decide to use a seperate infobox for spoken languages and sign languages, then the current language template should be renamed "template: language (spoken)" or some such.

There are several other possible solutions. I would appreciate input from regular contributors to the languge project. -- ntennis 06:11, 3 Apr 2005 (UTC)

Good work with the template. I think you're right, I'd love to see a sign language template working out of the language template by the modular system. The major problem is the choice of wording: speaker makes sense for vocal languages and signers makes sense for sign languages. I find user a bit bland, and I think there is a difference between saying I speak French and I use French — ability and practice. However, a major rework of the modular idea could incorporate this. --Gareth Hughes 15:47, 3 Apr 2005 (UTC)

I would also like to see a module for 'dialects'. One point of confusion for me in the basic template is the 'region' category. Until I saw the examples I assumed it meant regions within the country or countries where the language is used. This stems partly from a the heirarchical organisation of information - general at the top, then more specific categories as we go down the list (as with 'genetic classification'). I'm not sure why there's a need to explain the wider geographical regions in the basic template if we've already stated the countries? Surely readers can click on the link to the countries to see where they are located? Whereas, especially for smaller languages, it would make sense to indicate what region the language is used within a country. eg kurdish in turkey. --ntennis 06:28, 3 Apr 2005 (UTC)

p.s I moved this comment from above as I was told new comments should always go at the bottom of the discussion page. I'm new to this!

Just changed the sign lang template to remove 'official status' and 'regulated by' sections. -- ntennis00:08, 6 Apr 2005 (UTC)

I've just adapted {{language}} so that it can be used for sign languages. Do you want to retire {{sign language}} and use this. To adapt the template for sign languages, simply use {{{signers}}} instead of {{{speakers}}}, and the rest of the template will change its wording appropriately. I've implemented this on Auslan as an example. --Gareth Hughes 16:42, 15 November 2005 (UTC)

1. default color for sign lang could be silver. 2. iso2 default could be sgn. I implemented both defaults in the sign-template. IMO would be good to drop the sign-template. Tobias Conradi (Talk) 11:45, 16 November 2005 (UTC)

I think the familycolor=silver default for sign languages would be useful. For ISO 639-2, New Zealand Sign Language has sgn-NZ, so it might be worthwhile keeping this string open (or at least only defaulting to sgn). --Gareth Hughes 12:22, 16 November 2005 (UTC)

yes. only defaulting. sgn-CC seems to be widespread. would be interesting to know whether there is any sgn-X where X not like CC (CC= country code) Tobias Conradi (Talk) 14:44, 16 November 2005 (UTC)

First steps towards modularisation

There seems to be interest in the proposal of a modular template to cover all language articles. The first step would be to agree what the basic template should look like (without any modules). This might be a little tricky if we want to include sign languages as well, but well worth a try. Then we need a rough idea which modules we would like to see, and how they are to be implemented. The first actual change would be to drop module=[module name] into the template of every article that the template. This is to stop the template 'breaking' when the template calls a non-specified parameter from the article. Once the articles are fitted with this parameter, the basic template and its modules can go online. The drafts of these templates could be kept on sub-pages here for us to work on. I shall get started with some of this work. If there are any concerns or doubts, please voice them now. --Gareth Hughes 10:10, 6 Apr 2005 (UTC)

Just how are you planning to incorporate Ntennis' suggestion? I agree with you that 'Used' and 'Number of users' sound a bit awkward for spoken language. Is there great harm in having two separate basic templates, one for spoken languages and one for signed languages? It would be best, of course, if there was the possibility of some [IF > THEN] structure changing 'speakers' to 'users' when a certain parameter, say, 'type=signed', is present. But I'm a noob with regard to templates, so I don't know what's possible. — mark ✎ 10:53, 6 Apr 2005 (UTC)

{{{name}}} ({{{nativename}}})
Countries:	{{{states}}}
Region:	{{{region}}}
Genetic classification:	{{{family}}}
Total speakers:	{{{speakers}}}
Language codes
ISO 639-1	{{{iso1}}}
ISO 639-2	{{{iso2}}}
SIL	{{{sil}}}

This is a draft of the basic template. I've removed ranking, as is said below, it can be factored into {{{speakers}}} for those languages in the top hundred. I've moved {{{speakers}}} down, so that it could be left out of the basic template, and added to the template of every spoken language, and replaced with {{{signed}}} for sign languages. This would be the easiest way to combine the two. --Gareth Hughes 22:27, 6 Apr 2005 (UTC)

Ranking

I suggest that the "Rank" field be absorbed into the "total speakers" field. It seems weird to have that field, when most languages only ever have Not in top 100. Therefore, for the 100 languages that are in the top 100, we can give the rank in parantheses after the number of speakers. dab (ᛏ) 16:57, 6 Apr 2005 (UTC)

Weren't we skipping the ranking altogether since the figures vary so much depending on if you're including second-language speakers or not? Peter Isotalo 21:11, Apr 10, 2005 (UTC)

I'd be happy scrapping it altogether. --Gareth Hughes 21:19, 10 Apr 2005 (UTC)

I agree that it's kind of pointless. The only reason I put it in the first version is because I was just convertintg the table in Japanese language to a template. — Gwalla | Talk 06:11, 11 Apr 2005 (UTC)

I concur; scrap it. — mark ✎ 07:50, 11 Apr 2005 (UTC)

yes, i really dont like this field. perhaps a little insulting to some (?). please remove. peace — ishwar  (SPEAK) 06:25, 2005 May 20 (UTC)

Going modular...

I thought I'd start a new section to keep things in order here. The draft design for the base template is available at template:language/Draft. You'll see a little bit of bare syntax showing through in the Genetic classification box: this is where the modules are attached. --Gareth Hughes 22:14, 18 Apr 2005 (UTC)

Syntax	Template
{{language/Draft|module=basic}}	{{language/Draft|module=basic}}
{{language/Draft|module=sign}}	{{language/Draft|module=sign}}
{{language/Draft|module=official}}	{{language/Draft|module=official}}
{{language/Draft|module=extinct}}	{{language/Draft|module=extinct}}

I've got four templates up and running on the draft version now. Have a look at the table to the right to see what they look like. I'm still unsure about how to do the extinction module. Any thoughts? These templates are ready to go live as soon as all the articles are updated with the module parameter. --Gareth Hughes 22:44, 18 Apr 2005 (UTC)

The modules are cool. The sign language module could have the color (silver) built in as all sign languages are meant to use this color.

I'd like to repeat my comment above about the region category as I find it confusing and unneccessary. If I'm asking a stupid question could someone please point out why? :)

Note, for example, Aramaic language: "Spoken in: Armenia, Azerbaijan, Iran, Iraq, Israel, Georgia, Lebanon, Palestine, Russia, Syria, Turkey.

Region: Throughout the Middle East, Central Asia, Europe, North America and Australia."

Why does Australia end up as a region and not a country? Why is North America a region but the language is apparently spoken in no countries of the region? I looked at the first 6 languages listed as using the template [1] and found the 'region' category either pointless, nonsensical, or contradictory. (The languages are: Arabic language, Afrikaans, Armenian language, Aramaic language, Albanian language, Basque language)

Also note Chinese language and the way Catalan language uses region as a kind of sub category rather than 'super' category (as I thought it was originally intended)-- ntennis 01:07, 19 Apr 2005 (UTC)

Those are some confusing cases indeed. I for one always use it as a specifier: see Nafaanra language and Nobiin language for examples. The problem arises when a language is spoken in a lot of countries. — mark ✎ 07:29, 19 Apr 2005 (UTC)

Is it possible to make links to pronunciations as a separate module? That would allow it to be applied to at least a few languages which might might catch on if people see it in action. We might get more pronunciation files recorded that way. Peter Isotalo 20:04, Apr 19, 2005 (UTC)

OK: the responses are generally favourable. I agree that I find the two geographical sections confusing. I think it might be best to have one geographical section (probably best titled region, as it's less specific). Then all the geographical information can be put in there, and it can be a specific, or not, as is appropriate -- I'll make this change. Coding silver directly into the sign-language template won't work with the way I've set it up, I'm afraid: the parameter still needs to be specified in the article (it would be possible to create another template, which called template:language and gave it the parameter). Adding a link to pronunciation files would be possible, but I'm not sure if an infobox is the best place for them. A well-rounded article for a spoken language should have a pronunciation section, and that would be the appropriate place for the files. If there were a draft version of how the pronunciation files would be linked in, it might be easier to understand how it might work. --Gareth Hughes 22:16, 19 Apr 2005 (UTC)

I mean pronunciation of the name of the language. Like Russian language and Swedish language. All other pronunciation files belong to the phonology section. Peter Isotalo 22:29, Apr 19, 2005 (UTC)

I didn't realise that you meant the pronunciation of the name of the language. Therefore, the infobox for Swedish might read: Swedish (svenska) . It seems obvious to put this in the header of the infobox, where the name is given, but I wonder whether this might look a little too cluttered. --Gareth Hughes 11:06, 20 Apr 2005 (UTC)

I decided to try implementing the new modular language template, and ran into a few snags. You can see the results here. IceKarma 22:48, 2005 Apr 25 (UTC)

I've created a new draft that resolves the issues I found. IceKarma 00:51, 2005 Apr 26 (UTC)

I'm not sure that that fix works properly: I'll look into it. --Gareth Hughes 13:32, 30 Apr 2005 (UTC)

I like this idea, but I'm a bit confused about how to add specific information to a language page. I would like to include an image of a seal on a auxlang page, so how would I include this as a module?

There were some teething problems with the draft of the updated template: the parameters didn't link through muliple layers properly. Also, it seemed that things were in a state of flux: new ISO codes and other proposals. I'll have another look at the issues, but we may have to consider using a series of templates instead of the modular design. Unless you understand how the template works, it is best not to alter it. There are many language articles that use the template, and a small change might break every article referencing it. --Gareth Hughes 10:19, 29 September 2005 (UTC)

Language codes

links for SIL codes

The template was recently changed to make the SIL code an external link to Ethnologue. This is, in principle, cool, but it broke at least a couple articles.

Ethnologue tends to be rather liberal in treating lects as distinct languages. For some dialect continua and clusters of very closely related languages, it seems to make sense to treat them together in a single article. When I created Fula language, I did this, rather than having an individual article on every variety Ethnologue considers distinct. I followed the example of Aramaic language and listed all the SIL codes in the infobox. When the template was changed, both articles (and possible others), were messed up. I know very little about templates and don't know the best way to handle this. --Chris Johnson 15:22, 29 Apr 2005 (UTC)

I've had major problems with very stubborn POVing in Scanian language because of SIL's inexplicable determination to classify some (but far from all) dialects of certain languages as seperate languages. The best example of where SIL clearly disagrees with speakers and linguists alike is Flemish. There are a lot of editors who seem to use SIL as a way to validate various regionalist claims of their own specific dialect being a seperate language and this is not made easier by the fact that the SIL code is included in the template.

Would anyone object to ditching the SIL altogether and sticking only to ISO-639 in the infobox? Peter Isotalo 15:49, May 1, 2005 (UTC)

In its 15th edition, Ethnologue uses the ISO-639-3 codes, so the SIL codes seem to be deprecated anyway. I don't like the first versions of ISO 639, as they tend to lump a lot of African languages together under one code (e.g. many obscure Niger-Congo languages are simply coded nic together presumably just because they're obscure — bigger languages tend to get their own code). But ISO 639-3 looks OK to me. — mark ✎ 16:28, 1 May 2005 (UTC)

It's just too difficult to find a work-around for infoboxes that don't have a single SIL code. So, I've commented out the external link in the template. If someone really wants a direct link to the relevant SIL page in the infobox, then it is probably best to place the link inside the parameter. --Gareth Hughes 10:20, 11 May 2005 (UTC)

ISO 639-3

Does anyone think that the ISO 639-3 should be added to the template? REX 19:29, 12 May 2005 (UTC)

Yes I think so. In fact, as I pointed out above, the SIL code and the ISO 639-3 code are identical since the 15th edition of the Ethnologue. The template could be adjusted to reflect that. — mark ✎ 19:48, 12 May 2005 (UTC)

If we change the template, we'd need to go through all the language articles and change them all too, since the 14th edition SIL codes don't match, but that's not too big a deal. I've also updated my Template:Language replacement draft to read "ISO 639-3" instead of "SIL". Someone please take a look at it, I've had no feedback whatsoever. ;) IceKarmaॐ 23:49, 2005 May 12 (UTC)

If the 14th edition SIL codes don't match the ISO/DIS 639-3 code then don't mix them up. We should do a entry in the table for both, and possibly remove the SIL ones once ISO 693-3 is official. ---moyogo 20:51, 2005 May 17 (UTC)

Why should we stick to using 14th edition SIL codes if the 15th edition of the Ethnologue is already published? It is only an additional benefit that the ISO/DIS 639-3 are going to be the same as the 15th edition SIL codes. I don't see a reason to keep the obsolete SIL codes. — mark ✎ 00:22, 18 May 2005 (UTC)

The ISO/DIS 639-3 has not been released yet, ethnologue uses it and labels it correctly, it is not refered to as SIL code. So as far as I understand SIL code means 14th edition codes, and ISO/DIS 639-3 means 15th edition codes. If we simply subsitute the title SIL for ISO 693-3 in the table, then a lot of codes will be wrong. The SIL 14th edition code could be useful for reference, but it's ok if it's faded out of all the articles. ---moyogo 09:05, 2005 May 19 (UTC)

any idea which SIL14 are not the same as SIL15? Tobias Conradi (Talk) 18:39, 6 August 2005 (UTC)

Should we start the transition towards ISO 639-3? I removed an extra row in the template earlier today, specifically for the new codes. I did this because no default string was defined, and I wasn't sure what the staus was. If we add a row into the template for the ISO 639-3 codes (we could add '/DIS' until they are official), and define a default string, then we could go through each article adding the new codes (most will be the same as the old SIL14 codes, but it is better to do this by hand). Once this has been completed, and ISO 639-3 becomes a full standard, we should remove the SIL section from the template. Then, at our own leisure, we can remove the SIL codes from the articles, which will have been rendered invisible by the template change. Would everyone be happy for me to get started on this? --Gareth Hughes 16:30, 10 November 2005 (UTC)

IMO the old SIL code should be present, since it is used in old texts, just as iso-1 sits here. The transition to iso-3 may be done painlessly, since you may put an unused variable "iso-3" in the template reference with no harm. When all language articles updated, it will be safe to add the var into the template itself. mikka (t) 22:06, 10 November 2005 (UTC)

i like the idea of mikka, but I think it's also nice to see what you do. Maybe a middle way can be walked. Start inserting some codes via an unused variable, but do not wait until the last lang-page to put it in live-mode. To see an empty variable may also encourage editors to add it? Tobias Conradi (Talk) 13:33, 11 November 2005 (UTC)

I believe the SIL14 codes are now obsolete, and SIL15 codes are officially ISO/DIS 639-3. Therefore, I believe that these codes should replace the old ones. They are mostly the same: there are only a few changes. However, note that the new codes are written lowercase rather than uppercase. ISO 639-1 is not obsolete. There has simply been a decision not to enlarge the codebase further. ISO 639-1 is used to refer to major world languages, rather than being a comprehensive language indicator. Some old texts will, undoubtedly, keep the SIL14 codes, but is it really worth us keeping these listed in the template in the unlikely event that someone comes across an unsupported SIL14 code that is different from its ISO 639-3? Surely the odds on this must be pretty slim. Perhaps, we can start the first step, and add ISO/DIS 639-3 to the template and articles, and then decide whether we want to remove SIL14 codes. --Gareth Hughes 13:46, 11 November 2005 (UTC)

IMO At present: don't touch the SIL14. I started: ISO_639-3#Differences_with_ISO_639-2_codes . This can cause problems. I don't know how many more are out there. ISO 639-2 often refers to what is a language collection in ISO 639-3. Furthermore ms, zh, ar, are examples of languages that have a 639-1 code but there is not unique match in 639-3. Tobias Conradi (Talk) 14:02, 11 November 2005 (UTC)

Arabic language the arb entry is wrong. "arb" is not "ar/ara". Tobias Conradi (Talk) 12:36, 13 November 2005 (UTC)

I added the variable "iso3" to some languages and saw that it was allready in in some more. IMO iso3 should be used for single codes (allowing link to ethnologue). For multiple codes maybe a new variable should be defined. (Ask SEWilco, I think he once mentioned arrays - maybe a new feature?) Tobias Conradi (Talk) 13:38, 14 November 2005 (UTC)

I've had a little look at how to implement parameter arrays. There seems to be a straightforward nuts-and-bolts implementation, but there is the possibility of using an array to simplify code entry in th article itself. A similar system could be implemented for the {{{family}}} parameter so that all those &nbsp;...<br> entires are no longer needed. --Gareth Hughes 16:37, 14 November 2005 (UTC)

OK. It's not big or clever, but I've added {{{iso3list}}}. It overrides {{{iso3}}} and replaces it with the text string of the former. This allows one to write an unlinked text string in the ISO 639-3 box. If individual links to seperate pages of Ethnologue 15 are required, they have to be fully written into the parameter definition. I'm sure someone could think of a more elegant way to do this, but it's a useful workaround. --Gareth Hughes 22:35, 14 November 2005 (UTC)

good work Tobias Conradi [[User_talk:Tobias Conradi|(Talk)]] 06:58, 15 November 2005 (UTC)

quality control

some editors add ISO codes that do not exist. Thanks to direct links to ethnologue this can be discovered fast. Example:

• http://en.wikipedia.org/w/index.php?title=Novial_language&diff=prev&oldid=28166338

if an iso3 is defined we should make sure there is allways a link to ethnologue. Tobias Conradi (Talk) 09:44, 17 November 2005 (UTC)

Better colors?

I think the colors suggested here: Wikipedia:WikiProject Languages#Colors in the infoboxes are too bright. I really think we could find a more pleasing color pallet, for example see List of Presidents of the United States. Basically I think a paler set of colors (like the colors used on the Main Page), would be more pleasing. Anyone else agree? Any chance we could choose better colors? (I have also initiated a discussion about this here: Wikipedia talk:WikiProject Languages#Infobox colors.) Paul August ☎ 03:11, Jun 1, 2005 (UTC)

I most defenitely agree! -Mariano 10:27, 2005 Jun 10 (UTC)

I think no colors at all would be better. Hot green is just fucking ugly (who chose that anyways?), and there's no reason to have colors in the first place. They don't signify anything like categories in the List of Presidents page, they just highlight headings which need no highlighting. Of course, given past precedents the current ugly color will remain for at least another year because every change will get reverted until some drawn out discussion and voting completes. (Unsigned comment by 69.208.113.209)

We need to have them because each different color represents a different language family. For the presidents page, they represent a political party. I think the colors are fine. --User:Hottentot

this is not logic. Tobias Conradi (Talk) 13:51, 15 November 2005 (UTC)

Need a field for script.

I would like this template to have a field for the script used in the written form of the language (Roman, Cyrillic, Devanagari etc). Is this a sound field to add? -- 128.46.143.159 22:34, 2 August 2005 (UTC)

some use more than one script. but inclusion would be nice. Tobias Conradi (Talk) 13:49, 15 November 2005 (UTC)