From: David Weinberg (davidweinb@googlemail.com)
Date: Mon Jan 28 2008 - 14:49:48 CST
I'm still waiting for an answer -- or a reason why none is forthcoming.
---------- Forwarded message ----------
From: David Weinberg <davidweinb@googlemail.com>
Date: 27.01.2008 22:37
Mark,
why UNGEGN and not ISO?
or: What is the purpose of these trans"literation" charts?
Only when we know the purpose, we can see whether they serve it well or do
not.
I agree with Jony that there should be hundreds of transcription charts --
one for each pair of languages.
But there should be only ONE transLITERation chart for each script pair.
David
2008/1/27, Mark Davis <mark.davis@icu-project.org>:
David, Jony,
These are not made of whole cloth. The goal of the transliteration
schemes is to follow established sources, deviating sometimes where
necessary for reversibility. In both of these cases, the sources are
the UN.
The sources are generally described in the comments in the source
file. So for Arabic, you'd look in:
http://www.unicode.org/cldr/data/common/transforms/Arabic-Latin.xml
and find a reference to the UNGEGN tables:
http://www.eki.ee/wgrs/rom1_ar.pdf
Similarly for Hebrew, which also follows UNGEGN:
http://www.eki.ee/wgrs/rom1_he.pdf
Now of course, there may be problems in the data. If you find any, you
can file a bug requesting a change, as described in the document. Or
if you would like to see some alternate methods added, you are free to
propose them (as described earlier in this thread).
Mark
This archive was generated by hypermail 2.1.5 : Mon Jan 28 2008 - 14:51:37 CST