Apology to the group

[l.m.fosse at easteur-orient.uio.no]

>
>Finally, I do have some advice for those trying to sort Sanskrit words in
>roman transliteration sets on computers.  The answer is an artificial 'third
>party' transliteration, consisting of alpha-numeric characters (eg. a0, a1,
>a2, ..., b0, b1, ..., c0, ...) that CAN be sorted by the computer.  The
>trick is to 'global change' each of the characters of the text or list into
>the correct order in the alpha-numeric 'dummy aphabet'.  However, the order
>in which the global changes are made is crucial to every character being
>transliterated correctly.  I will work out this process with those who have
>already requested help and post suggestions to the list at a later stage.

David, this seems absolutely brilliant! There are, by the way, programs
that can make such a search and replace operation very, very easy! The Lexa
suite, which can be obtained from the University of Bergen, contains such a
program (LREPLACE) which I constantly use to convert from one
transliteration system to another. LREPLACE uses a list file with the
search item in one column and the exchange item in another column. If you
work out the sorting order for a dummy sort, using LREPLACE (or a similar
program) makes the whole process very quick and easy.

Lexa (version 6) can be obtained from the Norwegian Computing Centre for
the Humanities, Harald H}rfagres gt. 31, 5007 Bergen, Norway, tel: +47 55
21 22 54/55/56; fax +47 55 32 26 56. Email: icame at hd.uib.no. It costs about
NOK 700,- which includes several (necessary) manuals. The program as such
is public domain and is, I believe, available from a server in Bergen, but
you need the manuals.

Best regards,

Lars Martin



Lars Martin Fosse
Research Fellow
Department of East European
and Oriental Studies
P. O. Box 1030, Blindern
N-0315 OSLO Norway

Tel: +47 22 85 68 48
Fax: +47 22 85 41 40

E-mail: l.m.fosse at easteur-orient.uio.no