Romanization and Nagari output

Thomas B. Ridgeway ridgeway at EDU.WASHINGTON.HACC.BLACKBOX
Wed Oct 23 17:13:04 UTC 1991

Dear partners,
  Jonathan Silk writes (in part):
> . . .  the great advantage of Romanization is
> that it enables easy analysis: we can split words, even break
> compounds with hyphens, capitalize names, use italics and bold, etc.
> It is true that one can use bold, italic and underlined Devanagari,
> but at least for me it makes sometimes a rather odd impression (not
> bold but the other types of odd fonts).In very many cases this is
> more than enough reason to use Roman script.
> Another problem in which I have no suggestions but only questions
> concerns trading data and OCR.  The technology to input Nagari
> in a number of different ways certainly exists.  BUT, for example,
> Madhav Deshpande's beautiful Chiwriter Devanagari produces ASCII
> code which is, as far as I know, still unconvertable to Romanized
> Skt, and vice versa.  I do not know what kind of code the Nagari
> programs for the Mac produce, but isn't it likely we will
> encounter the same problems?
  These are excellent points, with whose sensibilities I agree entirely,
but which do not, I believe, bear on the matter as to whether one should
print one's hardcopy output in Devanagari.  [I am skipping over JS's
comments on Buddhism/Sanskrit as I have no opinion; regarding OCR: no data
which originates from a computer transcription should have to be scanned---
the original data file should be made available].
  *Many* of the systems currently available for output of devanagari
do so in association with an input method or encoding system which
substantially reduces the *data* value of the encoded text: i.e. it
is in a proprietary system-specific form, it records vowel marks and
consonant marks on separate lines, . . .  For representation as data
a reasonable Romanization, or any other essentially phonetic transcription
system, is much to be preferred.
  Ahem.  There is nothing to say, however, that the data cannot be entered
in a reasonable form, perhaps even the proposed CSX [Classical Sanskrit
Extended] encoding system, then converted to the form appropriate for
the software which will produce the printed copy.  It is, for instance,
perfectly feasible to adapt the recently available as freeware Nagari
font of Frans Velthuis for use with CSX encoding.  If nobody else
does it first, I will endeavor to do so by Christmas, as an exercise for
- - - - - - - - - - - - - - - - - - - - - - - - - - - -
Thomas Ridgeway, Director,
Humanities and Arts Computing Center/NorthWest Computing Support Center
35 Thomson Hall, University of Washington, DR-10
Seattle, WA 98195   phone: (206)-543-4218            *  Ask me about  *
Internet: ridgeway at      *    Unix TeX    *
- - - - - - - - - - - - - - - - - - - - - - - - - - - - -

More information about the INDOLOGY mailing list