On 17 Mar 2001, Gregg Reynolds wrote:
> Not considered a "character" in Arabic; it's not part of the abjad.
> Doesn't sort at all, in other words. That's in Arabic, mind you, not
> computerese; Arabid doesn't do "alphabetic" sorting. Although I guess
> you could say it always sorts after the same form without the teh
> marbuta; kitAba# comes after kitAb.
I don't believe that Arabs don't do alphabetic sorting. So where have the
arabic sorting tables in ISO 14651 and UTS #10 come from?
I have a Persian translation of Al-Munjed-ul-Abjadee Arabic dictionary at
home that uses alphabetic ordering, which uses something like Persian
alphabet ordering, but with Waw after Heh, so they have not changed the
ordering for Persian.
Would someone with Arabic language knowledge comment?
> In the two Arabic dictionaries I checked briefly, daa'erat was placed
> in the midst of a bunch of other d-w-r words, some of which terminate
> in teh marbutah. The ordering is based on morphology and semantics,
> not on alphabetic ordering. For example, teh marbuta is often
> suffixed to a generic noun to make it a unit noun: Sadm = verbal noun
> meaning roughly the action of striking a blow or colliding; Sadma# = a
> single blow or strike, etc.
I also have another Arabic to Persian dictionary (I can't remember the
name, it's written by an Iranian), that is sorted based on Arabic word
roots, something like what you say.
> Point of interest: all the great Arabic lexicons, of which there are
> many, sort based on radical structure. But they disagree on how to
> order the radicals. Most go first to last (this is used by most
> dictionaries today). But one of the best known, "Lisaan al-Arab",
> uses last-first-middle ordering.
Ok, it seems that it should be changes to "most" :) Would you look to see
if you can find "Al-Munjed-ul-Abjadee"?
--roozbeh
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:20 EDT