Re: A basic question on encoding Latin characters

From: John Hudson (tiro@tiro.com)
Date: Thu Sep 23 1999 - 15:37:49 EDT


At 11:15 AM 23-09-99 -0700, Rick McGowan wrote:

>and Michael said...

>> The chief reason for wanting to do so now is font-related.

>Correction: It is rendering (text sub-system) related. Fonts themselves
>have nothing to do with it, Michael. Fonts are just numbered bags of
glyphs
>-- it's what you DO with the glyphs that makes the difference. If all your
>display system can do is string glyphs one-after the other corresponding
>one-to-one with codepoints, then you'll need every interesting graphical
>element as a unit of encoding. If all your string-manipulation routines
can
>do is blind binary comparision, then you're out of luck in other ways.

I agree. There are certainly good arguments, from a font technology
perspective, for building precomposed glyphs within a font -- e.g.
controlled positioning of diacritic marks at small ppm sizes --, but it
does not follow that these glyphs need to be encoded. The problem once
again facing font developers is not whether there will be _an_
implementation of 'on-the-fly' composition in text processing and font
display systems, but just how many implementations we might have to deal with!

John Hudson

Tiro Typeworks
Vancouver, BC
www.tiro.com
tiro@tiro.com



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:53 EDT