chandOnuSeelanam.
STADIGAD.US.ORACLE.COM (STADIGAD@us.oracle.com)
17 Nov 95 10:31:11 -0800
I have made several points in my previous article on so called chando-ware.
Elaboration on one of the points here.
The program is bound to miss some laghu ( I ) to guru ( U ) non modifications
because of sanskrit and telugu mixing rules.
When a telugu word is followed by a sanskrit word or vice versa,
there is no sandhikriya and there is no I->U modification in these
cases. In the examples,
manDala mEle nokka dwija maatruDu ......
~~~~~~
and
dhairyamu vilOlambayye praNambulaa......
~~~~~~~
there is no sandhi at the underscored positions. Note the words
okka (telugu) and dwija(sanskrit) in the first and ayye(telugu)
and praaNa(sanskrit).
Softwares may not be able to detetect and cope with these kind of
linguistic barriers.
As Pillalamarri garu suggested, special syntactic sugars may be
applied for these cases. But frankly I can not subscribe to this
idea of an extra blank. It goes against the idea that space should
not be used for any other purpose and any amount of blank space is
tech'lly identical to one single space. Also, I do not think it is
solving the original problem any elegantly. I wish to use blanks and
tabs, even form feeds for my text formatting and resist any attempt
by softwares to assign any special syntactic and semantic significances
to white space characters under any pretext. A syntactic marker for
the ease of recognization is one thing and formatting and presenting
the text of matter is another. One should not step over the boundaries
of the other.
- Syamala Rao.
~~~~~~~~~~~~~~~