New TeX fonts for Indian languages

[Dominik Wujastyk]

Mani Varadarajan said:
> 
> 
> One of the problems with Velthuis's otherwise excellent
> Devanagari font is the lack of hyphenation tables.


Actually, Velthuis's devnag.c pre-processor was updated several years
ago to insert discretionary hyphenation points after all syllables.  The
skt.c pre-processor from Charles Wikner does the same.  My dev2e.sty
(q.v. for documentation) package for LaTeX2e contains [hyphenation] as
an option when loading.

Unfortunately, Frans hasn't been very pro-active in promoting the latest
changes to his system.  But the INDOLOGY supplementary gopher has the
new version.

Incidentally, this scheme for hyphenation may seem rudimentary, but it
works surprisingly well, in my view.  It also reflects accurately the
line-breaking habits of manuscript scribes.

In transliterated Sanskrit, scholars have implicitly developed
etymologically-based hyphenation rules, which are much harder to encode
for an automatic system.  (But work on this problem for German and other
languages has been reasonably successful.)  But while, say,
"buddhava-cana" looks very odd in romanization, it looks okay in
Devanagari (in my view), and it is in the tradition of manuscript and
early book production.

Best wishes,
Dominik