RE: Arabic Requirements of the BMP

From: Mansour, Kamal (
Date: Fri Jan 10 1997 - 18:37:42 EST

I wholly agree with Rick's (and previously, Ken Whistler) assessment of the
Arabic ligatures in Unicode (FBEA-FDFB). Let's not forget that Unicode is a character
set, not a font encoding. The listed Arabic ligatures are extremely
style-dependent, and cannot be faithfully rendered in all Arabic typefaces. It's
really too bad they made it into the standard.

If we think of these ligatures as characters instead of glyphs, then they can
each be canonically decomposed into constituent base characters. The process
I'm describing is analogous to decomposing the Latin ffi-ligature into f + f +
i. Meaning is preserved, while shape is not.

Kamal Mansour
Monotype Typography
From: Rick McGowan on Fri 10 Jan 1997 13:43
Subject: Re: Arabic Requirements of the BMP
To: unicode@Unicode.ORG

Michael said, regarding the enormous Arabic ligature set...

>> Well, font implementors like me will do the stuff, anyway.

Not necessarily so. I think it really is a big waste zone. Have you talked
to Thomas Milo lately? Decotype (those who kindly implemented all of the
ligatures to print the standard!) has really world-class Arabic support &
fonts. I don't think they're rushing out to supply all of these "useless"
ligatures to end-users, and I doubt they'd assert their usefulness.


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