Re: key, character and glyph codes

Date: Sat Dec 19 1998 - 10:21:43 EST

       The AFII site can be found at

       AFII is not part of Unicode, but the site was being neglected
       (I gather) at its former location and the Unicode Consortium
       offered to maintain the site on their servers.

       The purpose of AFII is to maintain a registry, an international
       standard, of glyphs and their identifiers. This registry was
       developed in accordance to ISO 10036:1996.

       I have been wanting to get info about AFII, specifically to
       what extent the industry is paying attention to the glyph
       registry, but I haven't seen any indication that any major
       companies are all that interested. (If that isn't so, I'd be
       glad to see some of their representatives give such indication
       in response to this message!) In fact, I saw recently that,
       whereas Adobe had been making use of AFII glyph identifiers,
       they have specifically decided not to do so any further.

       My take on the lack of interest I have seen is this: people
       must not believe that there is any benefit for them to pay
       attention to this glyph registry. Indeed, the only ones who
       would benefit are those that are doing anything in which it may
       be useful to be able to refer to abstract glyphs in a
       font-independent way.

       One potential candidate for this are the makers of Postscript:
       Adobe have told us, in a recent message on this list, that
       Postscript deals with glyph identifiers and not character
       codes. They have also indicated, however, that they don't have
       any need for an international, font-independent registry of
       glyph identifiers.

       Another group that might be interested in such a registry are
       type designers: they need a way to document glyphs they have
       designed, and this registry should be useful to them. Adobe are
       involved in type design, but again don't seem to need this
       registry. I don't know about others such as
       Linotype-Hell/Monotype. It's quite possible that large
       foundries such as these already had their internal systems for
       identifying glyphs and didn't really need a new registry. (But,
       then, who promoted ISO 10036 in the first place?)

       There is a final group that this registry might benefit: those
       that are involved in developing algorithms for rendering of
       complex scripts (those involving reordering, substitution,
       etc.) Whatever technologies are used for handling such
       rendering (state tables in GX, TTO, etc.) the algorithms that
       do the rendering must operate on glyphs. Now, if there were a
       font-independent way to refer to glyphs, then an algorithm that
       has been implemented for a given writing system could
       potentially be used for any number of fonts that are designed
       to support that writing system, provided that all of those
       fonts use the same glyph identifiers assumed in the
       implementation of the rendering algorithm. This is where I see
       the biggest potential payoff. That is one reason why our
       organization will plan to use this registry or, if we find that
       this really is a lame duck, an equivalent of our own.

       Is there another possible benefit of this registry that I have

       Is anybody out there on this list making use of this registry?

       Have I been inaccurate in anything I've said? (I'm still
       learning myself.)

       Erik, I hope what I've said is useful for you. Like you, I'm
       wanting to find out more.

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:43 EDT