Hi Connie, I think the locale spec indeed aims to differentiate our script from Japanese and Latin; just that. A locale spec is designed to be coded into software, which ain't have no sense of languages. All it cares about is that the Arabic script uses Arabic shaping rules and the 06xx block of Unicode characters; period.
behdad who's getting tired of this discussion ;) On Sun, 13 Jun 2004, C Bobroff wrote: > Hi Michael again after a long time, > You've unfortunately been CC'd in the middle of a conversation on *locale > requirements* not unicode level encoding. > You are correct and encouraged to put Persian in with Arabic for unicode > purposes. > > At the level of the current conversation, however, modern standard Persian > is written in the *Perso-Arabic script.* Urdu is also written in the > Perso-Arabic > script. (Urdu is NOT written in the Perso-Arabic-Urdu script.) Arabic is > written in the Arabic script. Various North African > languages and dialects are written in a modified Arabic script. > > Please don't consider the letter "Beh." Think about the Yeh, the Keheh, > numbers 4,5,6, Heh+Hamzeh Above, ZWNJ, some punctuation, sorting. I'm not > talking about calligraphic styles here. It is ok to just say "Arabic > script" if you are simply differentiating it from Japanese and Latin. But > at the level of Locale specs, you need to be more precise so as to reflect > the additions and modifications of the original Arabic script from which > it was derived. > > Since this locale information is being written in Persian, it can be > assumed that the Persian readers know the script they are reading the info > in has some additions and modifications. However, for an internatinal > audience, (not the unicode level), it is necessary to make it clear that > modern Persian is not written in the same exact script as modern Arabic. > I don't think it is *too much* wishful thinking that non-Persian experts > will want / need to consult this document. > > Again, you got dragged into something without context. That's why I"m not > replying to you point-by-point. > > -Connie > > On Sun, 13 Jun 2004, Michael Everson wrote: > > > At 15:43 +0430 2004-06-13, Roozbeh Pournader wrote: > > >I wish to restate my position. I'm CC-ing Michael Everson, a Unicode > > >expert in script naming. Michael, would you please tell us if Connie is > > >right here? > > > > > >On Sun, 2004-06-13 at 00:49, C Bobroff wrote: > > > > > Yes, all those script are called Arabic in scientific circles. > > >> > > >> No, the others are, in scientific circles said to be in "Perso-Arabic > > >> script." > > > > Not since the 19th century. > > > > > > You can also say "a modified form of the Arabic script" but that > > > > is what is meant by "Perso-Arabic script." Just "Arabic script" only > > > > applies to the Arabic language. > > > > This is not correct. > > > > What Ms Bobroff is doing is confusing character and glyph, I believe. > > It us true that the Arabic script has many variant styles, but this > > does not mean that those styles are or should be encoded as different > > characters. The ARABIC LETTER BEH which is used in Arabic, Persian, > > Urdu, Pashto, Sindhi, Kurdish, Kashmiri, Malay, Balochi, Uzbek, > > Kazakh, Uighur, etc. is the SAME intrinsic character in all of them. > > It has right-to-left directionality. It has a nominal, initial, > > medial, and final form which connects to other letters. > > > > Arabic script can be written or otherwise displayed in a number of > > styles, such as Kufi, Nastaliq, Naskh, and Maghrebi. But all > > varieties are ways of writing the same essential characters, and > > because of that, it is correct to speak of only one "script". > > -- > > Michael Everson * * Everson Typography * * http://www.evertype.com > > > _______________________________________________ > PersianComputing mailing list > [EMAIL PROTECTED] > http://lists.sharif.edu/mailman/listinfo/persiancomputing > > --behdad behdad.org _______________________________________________ PersianComputing mailing list [EMAIL PROTECTED] http://lists.sharif.edu/mailman/listinfo/persiancomputing
