Re: Arabic Requirements of the BMP

From: Terry Allen (
Date: Mon Jan 13 1997 - 14:53:43 EST

Rafik Belhadj wrote:
| As far as I know, the Arabic ligatures were first defined in the ECMA
| (the European Computer Manufacturers Association) Arabic Task
| Group -ATG-.
| In this Group, there were representatives from Bull (myself),
| DEC, HP, IBM, Siemens and others.
| We had meetings with Arabic countries standards organisations (ASMO, Egypt,
| Tunisia, Jordan, Syria, Saudi Arabia, etc.).
| In these meetings, I raised John's question. The answer was: nobody
| using computers, because there was no computer fonts for these ligatures,
| but there are non-computer fonts for some of them.
| The Arabic writing is cursive, and people really want to present the
| written text by hand, by printer or by any mean in the same way.
| >If there is nobody, this is worth knowing for the developers of
| >Unicode and ISO/IEC 10646.
| These ligatures are used in books, newspapers, etc.

But to the extent they can be formed by rule, they need not be
code points in a coded character set (although it might be useful).

What prompted my question was my curiosity about whether they represent
all the conceivable ligatures or whether they represent a union of fonts
that include ligatures. In short, are they needed for Arabic typesetting?
if not, are they a convenience for Arabic typesetting?

| >> > What was the specific requirement? I think the Arabic section
| >> > is a mess and I can only imagine that it is the union of several
| >> > fonts. Is that so? what were the fonts?
| I do not agree with this assertion. The Arabic part of ISO-10646
| is not a mess - at least for the Arabic Language part of it.
| People of ATG are Arabic skilled people,
| one of them became the Convenor of SC2 .
| The ECMA/ATG work was welcommed by ISO.

I'm sure they are, but someone, at some point, agreed that
whole words and phrases should be 10646 "characters," to wit,
FDF0-FDFB. What was the rationale for this? Was it related
to the rationale for all those complex ligatures?

| There are fonts for some of the ligatures (the most used).
| These ligatures were not chosen for " standard purposes ", but they are
| for User Requirements purposes: Users of the Arabic language
| need them.

In what context? are code points for those ligatures required
for typesetting?

| >> > And beyond that there are full words and a symbol for "place
| >> > of prayer" that I've never seen anywhere (rather like the
| >> > "hot springs" symbol; perhaps drawn from some guidebook?).
| >> >
| Some of these symbols came from Users Requirements of the
| Arabic language. Others came from Urdu and
| many non-Arabic languages.
| For these coming from the Arabic language users,
| you can see them in many religious books.

Yes, I've gotten that symbol (06E9) cleared up; I was misled by the
fact that it is not adjacent to the related 06DD and 06DE and by
the fact that the glyph seems to be incorrect.

There remains the question of why whole words are provided with
10646 code points.


    Terry Allen Fujitsu Software Corp.
"In going on with these experiments, how many pretty systems do we build,
 which we soon find outselves obliged to destroy?" - Benjamin Franklin
  A Davenport Group Sponsor:

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:33 EDT