The the case of Berber this is not true: it is the same language written
with 2 scripts (actually 3 as Arabic is also used). The mapping is not perfect
for now, but there are works to correct this and adopt a single convention in
each script (but with a question about whever the Tifinagh script used in
Berber will be enough to display Berber texts written in Latin: will it
accept new glyphs or diacritics to exhibit the missing letters used in the Latin
script which unifies the whole set of languages using some variant of the
Tifinagh script?
Perhaps my simple and not-worth-belabouring point would have been clearer if I had characterised transliteration as writing the words of a language in different *orthographies*, which may or may not involved a change of scripts. In the case of transliterating Russian names in English newspapers, you have two languages and two orthographies. In the case of Berber, you have one language and two or three orthographies. The fact that there is only one language is as irrelevant as a lot of the other things in this thread: what matters is the different orthographies.
An orthography is a set of *characters* used to write a language according to a set of conventions. A typography is a set of *glyphs* used to articulate a text in a given orthography. From this is follows that the proper place to performs transliteration -- the appropriate place to shift from one orthography to another -- is in character processing, not glyph processing.
John Hudson
Tiro Typeworks www.tiro.com Vancouver, BC [EMAIL PROTECTED]
I sometimes think that good readers are as singular,
and as awesome, as great authors themselves.
- JL Borges
