Unreversible Transliteration

[rcohen at sas.upenn.edu]

Your point is well taken regarding the reversibility problem of the
transliteration scheme which uses "ai" and "au." I work with Middle and
New Indo-Aryan languages which contrast the dipthongal "ai" and "au" with
non-dipthongal "a-i" and "a-u." Franz Velthuis' devanagari preprocessor is
not written to make the distinction. I traded messages with him on this
topic, and, with guidance from Franz, came up with a modification of the
C-code to handle the vowel collocations: a-i and a-u in devanagari. But,
you are right, we need to agree on a transliteration scheme which includes
such collocations.

-- 
Richard J. Cohen, Assistant Director, South Asia Regional Studies
University of Pennsylvania, 820 Williams Hall, Univ. of Pennsylvania
Philadelphia, PA 19104-6305, Tel: 215-898-7475; Fax: 215-573-2138
E-mail: rcohen at mail.sas.upenn.edu