Language committee/Archives/2009-06

From Meta, a Wikimedia project coordination wiki
For a summary of discussions, see the archives index.

Spanned discussions[edit]

The following discussions span multiple months and are archived in the first applicable archive:


Wikipedias Western Punjabi & Acehnese & Mari & Sorani[edit]

The requests for Western Punjabi, Acehnese, Mari, and Sorani Wikipedias were approved.

  1. Robin P. (SPQRobin)
    03 June 2009 14:39

    I'd like to inform you that the number of articles of the Punjabi Wikipedia doubled in a few weeks, from 1000 to 2000. Most articles are just a stub, but this does show that they really want to see their Wikipedia approved.

    So, may I ask if there is any progress in the approval of current proposals: Pubjabi Wikipedia, and also Acehnese Wikipedia, and probably also Mari and Sorani. It seems to me that (afaik) only their content needs to be verified, before they can be really approved.

    Thanks, Robin

  2. Gerard Meijssen (GerardM)
    03 June 2009 14:59

    <this user has not agreed to public archival.>

  3. Robin P. (SPQRobin)
    03 June 2009 15:22

    Yes, that could be added to the list as well, I think.

  4. Michael Everson (Evertype)
    11 June 2009 03:46

    <this user has not agreed to public archival.>

    How about it?

Wikisource Georgian[edit]

The request for a Georgian Wikisource was marked eligible.

  1. Michael Everson (Evertype)
    18 June 2009 10:09

    Sounds OK to me. (A proposer asked me.)

  2. Milos Rancic (Millosh)
    18 June 2009 11:37
    Michael Everson wrote:

    Sounds OK to me. (A proposer asked me.)

    +1

  3. Antony D. Green
    19 June 2009 02:45

    Me too.

  4. Gerard Meijssen (GerardM)
    19 June 2009 02:56

    <this user has not agreed to public archival.>

  5. Michael Everson (Evertype)
    19 June 2009 08:07

    Please tell them.

  6. Milos Rancic (Millosh)
    19 June 2009 08:11

    I'll do that.

Use LINGUIST List for language verification[edit]

No action was taken with regard to using LINGUIST List for language verification.

  1. Milos Rancic (Millosh)
    24 June 2009 06:34

    What do you think about asking linguistlist.org for checking validity of language? We won't need to wait long to verify random language.

  2. Gerard Meijssen (GerardM)
    24 June 2009 07:55

    <this user has not agreed to public archival.>

  3. Milos Rancic (Millosh)
    24 June 2009 08:40
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Of course, publicly. I didn't understand what and whom we need to ask?

Wikibooks Rajasthani & Classical Chinese; Wikipedias Algerian & Classical Mongolian & Hainanese & Hmong & Llanito & Manchu & Mapudungun & Mongol & South Azerbaijani & Southern Min written in Hanji & Sudanese Arabic & Teochew & Tokelauan & Wymysorys; Wikisources Ancient Greek & Baluchi & Classical Chinese & Coptic & Old Norse; Wikiversities Galician & Limburgish; Wiktionaries Aramaic & Gan & Kirmanjki & Konkani & Pitcairnese & Western Panjabi & Silesian & Yerkish[edit]

The requests for an Aramaic Wiktionary, Galician Wikiversity, Gan Wiktionary, Kirmanjki Wiktionary, Limburgish Wikiversity, and Silesian Wiktionary were marked eligible.

No decisions were taken on the requests for an Algerian Wikipedia, Ancient Greek Wikisource, Baluchi Wikisource, Classical Chinese Wikibooks, Classical Chinese Wikisource, Classical Mongolian Wikipedia, Coptic Wikisource, Hainanese Wikipedia, Hmong Wikipedia, Konkani Wiktionary, Llanito Wikipedia, Manchu 2 Wikipedia, Mapudungun 2 Wikipedia, Mongol Wikipedia, Old Norse Wikisource, Pitcairnese Wiktionary, Rajasthani Wikibooks, South Azerbaijani Wikipedia Southern Min written with Hanji Wikipedia, Sudanese Arabic Wikipedia, Teochew Wikipedia, Tokelauan Wikipedia, Western Panjabi Wiktionary, Wymysorys Wikipedia, and Yerkish Wiktionary.

  1. Milos Rancic (Millosh)
    24 June 2009 07:11

    Please, comment for now proposals for approval and rejection because those are my suggestions for the conclusion of particular processes. We should talk about "To by analyzed" group after solving simple groups. Below is listed around half of proposals under discussion.

    == Approval ==

    • Algerian Wikipedia [3]. If there is Egyptian Wikipedia, Algerian has much more sense because Algerian Arabic is much more distant from the Standard Arabic. The situation should be analyzed, but if it is analogue to the Egyptian (with the written form), we should approve it.
    • Mapudungun Wikipedia [10]. They have a couple of orthographies, but it shouldn't stop us to allow them to have the project.

    == Rejection ==

    • Rajasthani Wikibooks [1]. This is an open issue for a long time. I agree with Gerard (quote: Marwari is considered to be a great many languages. the codes that are given are being replaced by the ones in the ISO 639-3. For instance ethnologue in version 14 raj is split in eight and Marwari has an Indian and a Pakistani code. It is therefore not clear what the request is for. The ISO 639 has for Marwari different information which raises the same issues but differently. GerardM 16:46, 20 October 2007 (UTC)). My answer is to ask them to make proposal more precise.
    • Classical Chinese Wikibooks [2]. I agree with Shizhao (quote: yes, support closing this request. and Wikisource Chinese have to include Classical Chinese--Shizhao 07:53, 22 December 2008 (UTC))
    • Hmong Wikipedia [6]. The situation is similar to Rajastani.

    == To be analyzed ==

    • Classical Mongolian Wikipedia [4]. Dead language, but it should be checked how useful Classical Mongolian is.
    • Hainanese Wikipedia [5]. Generally, I think that it doesn't have a written form different from Min Nan, but it should be checked.
    • Llanito Wikipedia [8]. I wouldn't reject such proposals. I would prefer more discussions. See the article on Wikipedia [7].
    • Manchu Wikipedia [9]. It should be discussed more; preferably, with some linguists introduced in Mongolian languages.
    • Mongolian Wikipedia [11] should be, if possible, solved through the conversion engine. However, it should be analyzed.
    • South Azerbaijani Wikipedia [12]. It should be analyzed is a kind of conversion engine possible. Note that the prominent part of az.wp is from Iran and, thus, South Azerbaijani. So, I am very close to the rejection of the proposal.
    • Southern Min Written in Hanji [13]. Do we have any relevant non-nationalist cinologist? Students in Serbia, for example, don't know that there are different written forms (cf. Cantonese).
    • Sudanese Arabic [14]. I suggest that we should wait for 5 to 10 years.


    [1] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikibooks_Rajasthani
    [2] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikibooks_Classical_Chinese
    [3] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Algerian
    [4] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Classical_Mongolian
    [5] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Hainanese
    [6] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Hmong
    [7] - http://en.wikipedia.org/wiki/Llanito
    [8] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Llanito
    [9] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Manchu_2
    [10] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Mapudungun_2
    [11] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Mongol
    [12] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_South_Azerbaijani
    [13] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Southern_Min_written_with_Hanji
    [14] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Sudanese_Arabic

  2. Milos Rancic (Millosh)
    24 June 2009 10:59

    (Note, I used in the previous email "approval" instead of "eligibility". Here are the rest.)

    == Eligible ==
    There are no doubts about eligibility of the next projects:

    • Wikiversity in Galician [12].
    • Wikiversity in Limburgish [13].
    • Wiktionary in Aramaic [14].
    • Wiktionary in Gan [15]

    So, I marked them as eligible.

    == For eligibility ==

    • Wiktionary in Kirmanjki [16]. One of the Zazaki variants [17].
    • Wiktionary in Konkani [18]. 2.5 millions of speakers [19]
    • Wiktionary in Silesian [22]. ~56.000 of speakers in a developed country [23].
    • Yerkish Wiktionary [26]. "Yerkish is an artificial language developed for use by non-human primates. Yerkish requires the primates to use a keyboard to punch keys with so called lexigrams, symbols corresponding to objects or ideas." [27]. If it is a developed language, Wiktionary may be useful. But, the question will be related to the default interface localization (probably in English).

    == For rejection ==

    • Vilamovian Wikipedia [5]. Just 70 speakers [6]. My suggestion is opening Wiktionary + putting sources at ws, de.ws and/or pl.ws.

    == For discussion ==

    • Wikipedia Tokelauan [2]. Just 3200 speakers. Generally, I am for approval.
    • Wikipedia Teochew [3]. Again, Chinese (see [4]). And, again, we need a sinologist.
    • Wikisource in Ancient Greek [7]. Multilingual Wikisource should be used.
    • Wikisource in Classical Chinese [8]. Multilingual Wikisource should be used.
    • Wikisource in Baluchi [9] is eligible, but it has to be renamed into Wikisource Southern Baluchi.
    • Wikisource in Coptic [10]. Multilingual Wikisource should be used.
    • Wikisource in Old Norse [11]. Jesse marked it as eligible. I would say that the status should be same as for other classical languages: texts should go to the Multilingual Wikisource.
    • Wiktionary in Pitkern [21]. Why Wikipedia in Pitkern has been approved? It has less than 100 speakers [20]. It is much more reasonable to approve Wiktionary or so.
    • Wiktionary in Western Panjabi [24]. What do do with that? It looks to me like a political issue [25].

    [1] - http://en.wikipedia.org/wiki/Tokelauan_language
    [2] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Tokelauan_2
    [3] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Teochew
    [4] - http://en.wikipedia.org/wiki/Teochew_language
    [5] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Wymysorys
    [6] - http://en.wikipedia.org/wiki/Vilamovian_language
    [7] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikisource_Ancient_Greek_2
    [8] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikisource_Classical_Chinese
    [9] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikisource_Baluchi
    [10] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikisource_Coptic
    [11] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikisource_Old_Norse
    [12] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikiversity_Galician
    [13] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikiversity_Limburgish
    [14] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Aramaic
    [15] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Gan
    [16] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Kirmanjki_(Northern_Zazaki)
    [17] - http://en.wikipedia.org/wiki/Zazaki_language
    [18] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Konkani
    [19] - http://en.wikipedia.org/wiki/Konkani_language
    [20] - http://en.wikipedia.org/wiki/Pitkern
    [21] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Pitcairnese
    [22] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Silesian
    [23] - http://en.wikipedia.org/wiki/Silesian_language
    [24] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Western_Panjabi
    [25] - http://en.wikipedia.org/wiki/Western_Panjabi
    [26] - http://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Yerkish
    [27] - http://en.wikipedia.org/wiki/Yerkish

  3. Gerard Meijssen (GerardM)
    28 June 2009 04:41

    <this user has not agreed to public archival.>

  4. Milos Rancic (Millosh)
    28 June 2009 05:04
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    I think that multilingual Wikisource has another important characteristics: To provide the infrastructure for sources in extinct or moribund languages. I don't think that the problem is to see at one project a lot of texts in Ancient Greek, Classical Chinese and a lot of texts in a moribund languages. In other words, I think that "the rest of the languages" in the sense of free already created texts should go to the multilingual Wikisource.

  5. Robin P. (SPQRobin)
    28 June 2009 18:08
    Milos Rancic wrote:
    • Wikisource in Ancient Greek [7]. Multilingual Wikisource should be used.
    • Wikisource in Coptic [10]. Multilingual Wikisource should be used.

    Pages of these two are transferred to the Multilingual Wikisource.


    Milos Rancic wrote:
    • Wikisource in Classical Chinese [8]. Multilingual Wikisource should be used.

    There weren't any other Wikisources at Incubator other than the two above. This one just linked to a non-existing page, so I changed the link to point to the Multilingual Wikisource (also a non-existing page).

  6. Antony D. Green
    29 June 2009 01:22

    Wait, I'm confused. I know that el-Wikisource hosts Ancient Greek texts, and I would have expected zh-Wikisource to host Classical Chinese texts. Shouldn't those be the suggested targets rather than Multilingual WS?

  7. Milos Rancic (Millosh)
    29 June 2009 01:55

    <this user has not agreed to public archival.>

    The problem is that, by default, you are not able to write in any other language except in Modern Greek at Greek Wikisource and texts in Ancient Greek are of greater interest than just to those who know Modern Greek.

    The opposite argumentation would be that Modern Greek and Ancient Greek are genetically connected languages. I think that they are, as Sanskrit is related to many modern Indic languages. So, where should we keep texts in Sanskrit? Or we should introduce the rule: If there is just one successor of an extinct language, sources should be kept at the Wikisource in its modern derivative. Or, should Old Church Slavonic texts go to Bulgarian Wikisource?

    The only class of cases in which I think that it is not a big deal are languages which are not of greater interest, like, for example, Medieval Serbian is. However, Old Church Slavonic texts, if there is no WS (and I think that there is no cu.ws) should go to the Multilingual WS. However, I think that all texts in extinct language systems (like Ancient Greek or Medieval Serbian are) should go to Multilingual Wikisource.

    Of course, this is just my position.

  8. Antony D. Green
    29 June 2009 02:31

    The other option would be to relax the rule that a language has to have native speakers in order to have a Wikimedia Project in the case of Wikisources in ancient languages. That way, Sanskrit could have its own sa-ws and Ancient Greek its own grc-ws, without the requirement that there be native speakers. That rule makes sense for Wikipedias and Wiktionaries, but not so much for Wikisources (or for that matter, Wikiquotes), since the content added in the former is orignal, but the content added in the latter is not original. Such Wikisources could then use modern languages as their lingua franca, the way Latin Wikisource already does.

    Medieval Serbian should probably really be hosted at sr-ws, because as you said, anyone interested in Medieval Serbian is probably competent in modern Serbian. The same goes for Classical Chinese - anyone who can read it has almost certainly learned Modern Chinese first, so texts can safely be hosted at zh-ws. Likewise, Old English and Middle English should remain at en-ws. Languages like Coptic and OCS that have multiple descendents, or no descendants, or are of widespread interest outside their equivalent modern-language community can either have their own Wikisources or stay at oldwikisource, depending on how big and how active they are. At any rate, decisions should be made on a case-by-case basis.

  9. Milos Rancic (Millosh)
    29 June 2009 03:29
    Antony D. Green wrote:

    > The other option would be to relax the rule that a language has to have
    > native speakers in order to have a Wikimedia Project in the case of
    > Wikisources in ancient languages. That way, Sanskrit could have its own
    > sa-ws and Ancient Greek its own grc-ws, without the requirement that there
    > be native speakers. That rule makes sense for Wikipedias and Wiktionaries,
    > but not so much for Wikisources (or for that matter, Wikiquotes), since the
    > content added in the former is orignal, but the content added in the latter
    > is not original. Such Wikisources could then use modern languages as their
    > lingua franca, the way Latin Wikisource already does.

    This is one of the options and I agree that we should find one workable solution. It doesn't need to be strict, so, generally, I am supporting your idea.

    Projects affected by this rule would be: Wikisource and Wikiquote. Wiktionary shouldn't be affected because anyone may make a Sumerian dictionary on English Wiktionary or similar. Monolingual dictionaries, if they existed, like in Classical Chinese case, should go to Wikisource.

    Rules for such cases would be, approximately:

    • Extinct [synchronic] language systems known just by those who know modern variant of that language should go to the WS and WQ of the modern variant. (Cases: Classical Chinese, Medieval Serbian.)
    • Other extinct language systems with a lot of written materials known today should have separate WS and WQ if requested. (Cases: Ancient Greek, Sanskrit, Ancient Egyptian, Sumerian.)
    • Other extinct language systems should go to Multilingual Wikisource. (Cases: Old Church Slavonic, a lot of other languages older).
    • This rule doesn't affect classical languages which are still used as a communication tool. (Latin)
    • This rule implies that we need multilingual Wikiquote.

Compendium projects[edit]

No consensus was reached on the proposal for compendium projects.

  1. Milos Rancic (Millosh)
    28 June 2009 18:41

    May we suggest that Incubator becomes compendium.wikimedia.org, too? In other words, something like http://incubator.wikimedia.org/wiki/wp:xyz:<Main Page> would have redirect as http://xyz.wikipedia.org/wiki/<Main Page>.

    It is really unrealistic to keep project on efforts of one person. At the other side, that person would loose enthusiasm after long time of not approving the project.

    In the terms of our decisions, it would mean:

    • Eligibility + the beginning of the work on Incubator would mean turning on redirect engine (i.e. they would have http://xyz.wikipedia.org/).
    • Approval would mean making a separate project.

    It would require some technical work (from the side of WMF tech team), but I am almost sure that it is not a lot of work.

    What do you think about that?

  2. Gerard Meijssen (GerardM)
    29 June 2009 02:17

    <this user has not agreed to public archival.>

  3. Milos Rancic (Millosh)
    29 June 2009 02:26
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    How can be made a live project in one moribund language with less than 100 speakers?

    The other option, which you suggest is to leave them to fulfill formal requirements which they won't do. Just few hundreds of few thousands of languages may create fully independent projects.

  4. Gerard Meijssen (GerardM)
    29 June 2009 03:41

    <this user has not agreed to public archival.>

  5. Milos Rancic (Millosh)
    29 June 2009 04:03
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    In other words, your position is: "You are free to try Wikipedia even if you are one of 70 speakers, and we are not responsible if you fail to fulfill our requirements." I think that such position is irresponsible.

    I think that our responsibility is to tell to them one of two things:

    1. Please, don't try with Wikipedia, you have much more important tasks to do to keep your language. -- or
    2. OK, you want Wikipedia, we know that it is not realistic that you keep it, but we may offer to you aided project. If you become strong enough, we'll leave you on your own.

    How many projects in new languages became eligible and how many of them wasn't approved? And what are the reasons for that?

  6. Gerard Meijssen (GerardM)
    29 June 2009 04:49

    <this user has not agreed to public archival.>

  7. Michael Everson (Evertype)
    29 June 2009 05:14
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    I think that if there are even a few editors in a severely endangered language we should be more generous. If someone is interested in writing many articles or translating articles for his or her language we should support that. The cost of hosting the Wikipedia would surely be less than the work a few assiduous editors might put in.

    Always it should be on a case-by-case basis. Flexibility is human.

  8. Gerard Meijssen (GerardM)
    29 June 2009 05:20

    <this user has not agreed to public archival.>

  9. Milos Rancic (Millosh)
    29 June 2009 06:50
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    There are at least two points against your too formal arguments:

    • Localization is a messy job even for very experienced users. Usability of translatewiki.net is often at zero level: There is no any context for translation (I have bizarre translations in Serbian on default MW installations and I don't have a clue how to quickly find and fix it). I may imagine how one or two enthusiasts (let's say, one philologist and one blue collar worker) may be disappointed when they find such wall.
    • One project needs at least one or two experienced wiki contributors to make something like wikilink is. Out of Incubator a lot of energy of those two enthusiasts would be wasted on administrative tasks. Not to talk about understanding copyright. A year ago I had to remove urgently admin rights on Luxembourgian Wikipedia to their only bureaucrat because he started to delete images as he didn't understand copyright (he didn't want to see his images on Commons).

    By the way, I am not talking here about, let's say, ask for the recent ask Wikisource in Georgian. There are enough of speakers of Georgian and I don't think that it is necessary to give them any project without fulfilling formal requirements.

    Also, I am not talking about giving to anyone a project without sustainable amount of contributors. I am saying that we should allow them to have a virtual project inside of the area controlled by experienced Wikimedians, at Incubator.

  10. Gerard Meijssen (GerardM)
    29 June 2009 07:08

    <this user has not agreed to public archival.>

  11. Milos Rancic (Millosh)
    29 June 2009 07:27
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    My intention wasn't to say that translatewiki.net is a bad project; it is the best which we have. The problem is reality of demands to a random contributor, one of couple of dozens, hundreds, thousands of speakers of some small language -- who is probably not a computer expert. Features related to more language flexibility are not connected so strong with very hard work on localization on translatewiki.net.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Again, you are reading what you want to read. I said that two random persons, from one ~50-5000 big language community, interested in Wikipedia may be one philologist and one blue collar worker. Those two persons are not computer experts even they have a lot of will to work on Wikipedia.

    Also, one person from a linguistic community of 5000 persons, if knows to read, knows for sure at least one more language, because it is not so realistic that such there is a full set of primary school books for such language. Actually, I am quite sure that that person, who knows to read, learned in school non-native language. And that language  may be the default language for localization. Localization is not the content; actually, it is really unimportant thing if we are talking about preserving a language with 70 or 7000 speakers.

    The point is that you are asking enormous waste of time from people who should preserve their language, not to fulfill some bureaucratic demands (of Gerard Meijssen :P ).

  12. Gerard Meijssen (GerardM)
    29 June 2009 07:31

    <this user has not agreed to public archival.>

  13. Michael Everson (Evertype)
    29 June 2009 07:44
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Im going to have to side with Miloš here.

    Localization is **less important** than content. It's important. But less than content. Most speakers of minority languages in Russia, for instance, are certainly used to software interfaces in Russian. Even if the Wikipedia interface were completely localized into Hill Mari, ALL of the rest of the applications on the user's computer will certainly be in Russian, if not some in Russian and some in English.

    Content in Hill Mari is more important to the readers than the localization interface, since all readers of Hill Mari are bilingual in Russian. (If there are any who are not, they are probably not using any computers.)

    Which does a speaker of Hill Mari want more or want quicker in his language? An article about keeping honey bees, or about the rings of Saturn, or about petroleum? Or the localization of the Wikipedia's Date Formatting options page?

  14. Michael Everson (Evertype)
    29 June 2009 07:57
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    If you say so. I went there just now, for the first time, and could not even find the language I was looking for (Gaeilge, Irish) in the popup menu to the right. Now, I used to localize software into Irish. I have done a bit of localization on ga.wikipedia.org. I even know about two- and three-letter language codes and I ***STILL*** could not find my language in the popup menu on the first page of translatewiki.net

    I consider that to be strikingly unfriendly to the end-user, who is not likely to be a language and standardization geek like me. And I ***FAILED*** to find what I was looking for. Guess what I did next?

    I quit looking.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    As I say, the primacy of its essentialness is less than what you think, since nearly everyone in the world who wants to use a computer HAS TO DO SO ALREADY in one of twenty languages or fewer.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    I certainly got stopped on its home page.

    (I would like to keep sentiments like "resentment" out of this. We may all give and take criticism like a gentlemen.)


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Nice. Doesn't make it user-friendly.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    It's a tool, evidently with room for improvement.

  15. Gerard Meijssen (GerardM)
    29 June 2009 08:25

    <this user has not agreed to public archival.>

  16. Michael Everson (Evertype)
    29 June 2009 11:12
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Honestly? I don't see how this addresses the realities of what I wrote. You've simply gainsaid me by saying "we need both".

    I said, Mari speakers (just an example, no relation to the recent proposal) may not need a localized interface as urgently as you have just insisted, because ALL OF THE OTHER CONTEXTS on their computers are in Russian or English ANYWAY.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    That doesn't mean anything to me. I don't know what EMC2 is, nor how they evaluate the difficult of software localization into a language which DOESN'T have developed technical vocabulary for things like Open and Save and Cut and Paste and Undo. (I remember how difficult it was to determine the right text for "OK" on an Irish dialogue box, as well as the near-impossibility of translating Undo.)


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    What?


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Excuse me? What does this mean? It means NOTHING to me. It doesn't HELP me to know that there was a compromise. I STILL can't find my language in that list. The interface there is ***poor*** and the "compromise" is dysfunctional.

  17. Gerard Meijssen (GerardM)
    29 June 2009 11:33

    <this user has not agreed to public archival.>

  18. Michael Everson (Evertype)
    29 June 2009 13:24
    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Mixing alphabets is crazy


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Then you probably do not nee to choose it. (It says Al-Arabiya.)


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    That could be remedied.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    Also the voluntary nature of the interface with them.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    That can be and should be to be fixed.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    And there are rules in RFC 4066 about managing that.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    And such things exist.


    Gerard Meijssen wrote:

    <this text is quoted from a user who has not agreed to public archival.>

    That kind of research and design project would be an appointment requiring remuneration, now wouldn't it? But I tell you, as a specialist in languages and writing systems, if ***I*** can't navigate that popup menu, who can?