John Cowan wrote:

>
> Peter T. Daniels scripsit:
>
> > Hunh? They have something other than QWERTY? Substituting £ for $ is an
> > isomorphism.
>
> The letters and digits are in the same place; the remaining characters are not.
>
> If you don't like that example, take the hexagonal French AZERTY keyboard
> vs. the Canadian French QWERTY one.

And the relevance of this to the Indian problem is that if the nine
scripts (BTW does that figure include Urdu and/or roman?) already have
standardized and widely used typewriter keyboard layouts, then they
ought to be emulated by the software. But additionally, if the scripts
were isomorphic, anyone could use whichever of the Indian keyboards they
were familiar with to type in any of the other scripts. Or, there could
be automatic conversion between scripts -- which could be convenient for
presenting Sanskrit texts for use throughout India (the books say that
Skt is written in local scripts, though over here we only ever see Skt
in Devanagari).

But, from the materials on typewriter history that I have (the book by
Beeching with Olympia's huge catalog of keyboards, and my father's
salesman's handbook for Remington typewriters from the late 1940s), I
don't think there were typewriters for many of the scripts of India.
There wasn't even an Oriya computer font until I made the one used in
WWS. (Braj Kachru told me that my chapter for *Language in South Asia*
can't be typeset in India, where they're having the book done because
they thought it would be cheaper than hiring, say, me.)
--
Peter T. Daniels grammatim@...