User talk:Stephen G. Brown/Arabic alphabet

Arabic alphabet
Me again. Now then, we have Arabic, and we have Persian شاهین. I know that (in some fonts anyway) the is formed slightly differently, but I would regard these two words as being spelt the same way. Is this wrong? What I'm getting at is that it seems intuitive to me that the two words should be on the same page. There are probably other words which are formed in exactly the same way in each language, yet they end up on different pages. Is this a function of the templates? Is there any way around it? Widsith 09:39, 16 April 2006 (UTC)


 * Well, until very recently, they were indeed spelt precisely the same in Arabic and Persian. However, Persian now has a separate letter for, namely ی. When the Persian letter ی is used in initial or medial positions, it automatically assumes two underdots, just like the Arabic letter. It’s only in the final and isolated forms that the Persian letter has no dots. To make things more confusing, Arabic also has a dotless form, , called "alif maksura" (ye-shaped alif), and this is a different letter from the identical-appearing Persian one (the Arabic "alif maksura" only has final and isolated forms, and cannot appear initially or medially). Unfortunately, this isn’t a matter of templates or even fonts, they are different letters entirely.
 * The final confusion is that this Unicode-based separation of Arabic and Persian is still so new that many online Persian dictionaries, glossaries, and other resources, even some older keyboards, still use the Arabic letter rather than the Persian one. The upshot is that there must be two separate pages, but the Arabic page should also include some mention of the Persian word, either as a full entry or as a "See also" or something of the sort. The word for book is a another good example, which is in Arabic and کتاب in Persian (Arabic initial  versus Persian initial ک) ... I originally made two separate pages for this, and also made a complete Persian entry on the Arabic page. However, someone else objected to the duplication and removed the Persian from the Arabic page.
 * So this matter isn’t quite settled yet. Words with / ی or  / ک must have separate pages, but we need to agree on some format for directing Persian-searchers from the Arabic page to the correct Persian page. —Stephen 11:04, 17 April 2006 (UTC)


 * This is very helpful, thank you. Widsith 11:13, 17 April 2006 (UTC)


 * I side with Stephen here. I think if writers of Arabic and writers of Persian feel that they use different "yeh" and different "ka" then we should certainly put the words under the spelling they would use. Where entries for the words in both languages only differ by such letters, there should be a "disambiguation see also" at the top just as we do for Latin-language articles which differ only by capitalisation and/or diacritics.
 * But the keyboard vs. Unicode issue is a strong one. Unicode has provided extra variants to distinguish "s" and "t" with comma from "s" and "t" with cedilla - I believe out of demand, but despite that demand, nobody is using the newer Unicode letters. There are similar issues with ligatures such as the Dutch "ij" which exists as a single character but in reality everybody uses two characters to express it.
 * I feel strongly that Unicode should enhance a computerized dictionary, not fight with it. But the problem with Wiktionary is we control only the content, and not the machinery. The Wiki software was developed for an encyclopedia and there are no developers interested in Wiktionary and no Wiktionary contributors interested in becoming developers or raising funds to hire one to add features we need, such as allowing a search for Arabic or Persian "ka" or "yeh" to return results to pages which use either variant. So for now you have to search for the exact (Unicode) spelling the article uses, or use redirects and/or "disambiguation see also". Even explaining these issues is difficult! &mdash; Hippietrail 15:52, 18 April 2006 (UTC)