Hi Mark and everybody,
Yes, Tahoma works (but is ugly and therefore not used on ar: for example) and is even more widespread than Unikurd Web. I just tried it on http://ku.wikipedia.org/wiki/Wikipedia:Fontexample. The words there should look similar with the only difference being the diacriticals.
I have not expressed myself correctly. The problem was not only the font, but the fact one cannot tell Wikipedia to use one font throughout an article if there are headings and lists. For the page name it doesn't work anyway. Something I forgot to mention was the edit window, words appear differently in the edit window and on the page or in the preview.
An automatic script converter would be a fine thing, but there are also dialect differences. So an article in the two main dialects Kurmanci and Sorani would exist four times after automatic conversion into both scripts. At the moment we have three dialects represented which are only partly mutually intellegible. Until now all Sorani is written in arabic letter, all Kurmanci and Zazaki in latin letter with links to each other if the same article exists in different dialects. We like it that way, there were no objections yet.
But because arabic and latin script are used for either of them, automatic script conversion would be fine on the long run. Technically it should be possible, as far as I know 1:1 conversion is possible.
Erdal
Hmm... How different are Zazaki and Kurmanci and Sorani? Is there any sort of unified standard already existing?
If possible, one of the best options would probably be to use a unified standard which either might already exist or which you might devise yourselves (or you could find somebody else to help devise it for you).
Of course this won't work if Zazaki and Kurmanci and Sorani are too different, but if not it would be good because otherwise there will need to be 6 versions of each article (Zazaki Latin, Kurmanci Latin, Sorani Latin, Zazaki Arabic, Kurmanci Arabic, Sorani Arabic) or something similar.
Best, Node
On Tue, 12 Oct 2004 10:42:20 +0200, Erdal Ronahi erdal.ronahi@gmx.net wrote:
Hi Mark and everybody,
Yes, Tahoma works (but is ugly and therefore not used on ar: for example) and is even more widespread than Unikurd Web. I just tried it on http://ku.wikipedia.org/wiki/Wikipedia:Fontexample. The words there should look similar with the only difference being the diacriticals.
I have not expressed myself correctly. The problem was not only the font, but the fact one cannot tell Wikipedia to use one font throughout an article if there are headings and lists. For the page name it doesn't work anyway. Something I forgot to mention was the edit window, words appear differently in the edit window and on the page or in the preview.
An automatic script converter would be a fine thing, but there are also dialect differences. So an article in the two main dialects Kurmanci and Sorani would exist four times after automatic conversion into both scripts. At the moment we have three dialects represented which are only partly mutually intellegible. Until now all Sorani is written in arabic letter, all Kurmanci and Zazaki in latin letter with links to each other if the same article exists in different dialects. We like it that way, there were no objections yet.
But because arabic and latin script are used for either of them, automatic script conversion would be fine on the long run. Technically it should be possible, as far as I know 1:1 conversion is possible.
Erdal
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Still wondering, is there any sort of unified standard?
Languages such as Sicilian, Sardinian, etc have only recently gotten standard versions.
Dialect differences have always been a problem for Sicilian and Sardinian, there was no single dialect that you could use without upsetting people or causing confusion.
So advisory committees were formed, and they formulated proposals for unified versions of these languages. Currently, the Limba Sarda Unificata (Unified Sardinian) standard is official in the Sardinian province of Nugoro, and it is being promoted everywhere. Its intention is to be a bridge between the different dialects, a sort of linguistic "least common denominator" which works for everybody.
It did cause some upset because everybody was most supportive of their own dialect, but it is a much better option than using any single Sardinian dialect.
Now, whether or not this can be done for a language depends on how different the dialects are. For some languages, you can just take the dialect of the capital or largest city and use that, but for other languages there is too much linguistic regionalism and such a move would cause problems with speakers of other dialects. In these cases it is normal for a country or political unit to try "language planning" and to formulate a sort of "least common denominator".
That approach generally works the best with languages that don't have really vastly differing dialects, but the dialects aren't close enough that people understand each other easily. There have been attempts to do the same sort of thing for groups of languages, for example Interlingua for Western Romance languages, and in some cases they have worked, in others they have failed.
best, node
ps even if this problem is not dealt with now, it will need to be dealt with by any future kurdish nation-state
On Tue, 12 Oct 2004 12:10:36 -0700, Mark Williamson node.ue@gmail.com wrote:
Hmm... How different are Zazaki and Kurmanci and Sorani? Is there any sort of unified standard already existing?
If possible, one of the best options would probably be to use a unified standard which either might already exist or which you might devise yourselves (or you could find somebody else to help devise it for you).
Of course this won't work if Zazaki and Kurmanci and Sorani are too different, but if not it would be good because otherwise there will need to be 6 versions of each article (Zazaki Latin, Kurmanci Latin, Sorani Latin, Zazaki Arabic, Kurmanci Arabic, Sorani Arabic) or something similar.
Best, Node
On Tue, 12 Oct 2004 10:42:20 +0200, Erdal Ronahi erdal.ronahi@gmx.net wrote:
Hi Mark and everybody,
Yes, Tahoma works (but is ugly and therefore not used on ar: for example) and is even more widespread than Unikurd Web. I just tried it on http://ku.wikipedia.org/wiki/Wikipedia:Fontexample. The words there should look similar with the only difference being the diacriticals.
I have not expressed myself correctly. The problem was not only the font, but the fact one cannot tell Wikipedia to use one font throughout an article if there are headings and lists. For the page name it doesn't work anyway. Something I forgot to mention was the edit window, words appear differently in the edit window and on the page or in the preview.
An automatic script converter would be a fine thing, but there are also dialect differences. So an article in the two main dialects Kurmanci and Sorani would exist four times after automatic conversion into both scripts. At the moment we have three dialects represented which are only partly mutually intellegible. Until now all Sorani is written in arabic letter, all Kurmanci and Zazaki in latin letter with links to each other if the same article exists in different dialects. We like it that way, there were no objections yet.
But because arabic and latin script are used for either of them, automatic script conversion would be fine on the long run. Technically it should be possible, as far as I know 1:1 conversion is possible.
Erdal
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Mark Williamson schrieb:
Still wondering, is there any sort of unified standard?
Not for all dialects. For Sorani, there is a de-facto standard. It is the dialect of Sulaymaniya, the main written language since tens of years. There is a lot of publiations in that dialect, also education. It is user in parts of Iraq and Iran. Educated Sorani speakers have less difficulty with Kurmanji, because they know roman letters.
The situation is worse for Kurmanji, which is the dialect most of the Kurdish speakers speak. Because most of them are in Turkey they are afflicted by the Turkish state politics which has persecuted the use of the language since the 1930's. Only in the last years there are some publications. The situation is even worse for Zazaki, because there are less speakers. In the south of Iran there are some more dialects (Gorani, Hewremani) which I cannot say a lot about.
For Kurmanji there are some standards, like a roman-based alphabet and a lot of grammar rules, which are commonly accepted. But details are still being discussed. These dicussions have often taken place in exile in Europe, which made them even more difficult. Standardization is a main goal and will proceed as the political situation in Turkey improves.
But the differences between Kurmanji and Sorani are severe, and a further standardization of Kurmanji, even if there is a tendecy towards closing in, is likely to make these differences permanent. Kurmanji speakers also generally do not know the arabic script so they cannot make use of Sorani effectively.
So in short:
1. Kurmanji is most widely used, but not effectively standardized. Latin script, moste speakers. 2. Sorani is second most widely used, has just become official language in Iraq, well standardized, arabic script. 3. Zazaki used by less speakers, poorly standardized. 4. Little convergence of the main dialects.
This is how I see it, others may see it differently.
For Wikipedia, we take everything as it comes and try to support everything technically. I have switched the Monobook.css to Tahoma, which is ok for all dialects.
So advisory committees were formed, and they formulated proposals for unified versions of these languages. Currently, the Limba Sarda Unificata (Unified Sardinian) standard is official in the Sardinian province of Nugoro, and it is being promoted everywhere. Its intention is to be a bridge between the different dialects, a sort of linguistic "least common denominator" which works for everybody.
That is an interesting example how things can be done.
even if this problem is not dealt with now, it will need to be dealt with by any future kurdish nation-state
If there is one. A real nation-state is not likely to develop. Even if the Iraqi part breaks away from Iraq, that is just a small fraction of Kurdistan (1/6). There will be no language problem, Sorani is standard there. But the Kurmanji problem can only be solved in the north, in Turkish Kurdistan. Kurdish politics are very very difficult.
Erdal
On Oct 12, 2004, at 1:42 AM, Erdal Ronahi wrote:
I have not expressed myself correctly. The problem was not only the font, but the fact one cannot tell Wikipedia to use one font throughout an article if there are headings and lists.
It should be possible to edit [[MediaWiki:Monobook.css]] and add a style rule.
-- brion vibber (brion @ pobox.com)
Brion wrote:
It should be possible to edit [[MediaWiki:Monobook.css]] and add a style rule.
How is that done? I don't know CSS. What we need is this:
Non-latin (arabic) script letters should be shown as "Unikurd Web" everywhere, if "Unikurd Web" is not installed, it should be Tahoma (as a second choice).
Is there an easy way to do it?
Erdal
wikipedia-l@lists.wikimedia.org