IDNA Comparisons

John C Klensin klensin at jck.com
Fri Jul 17 18:37:29 CEST 2009



--On Friday, July 17, 2009 12:11 -0400 Eric Brunner-Williams
<ebw at abenaki.wabanaki.net> wrote:

> I've asked also John, and until I did, the idea that tactile
> encoding could have a feel that was distinct from the toAscii
> value was something I didn't appreciate.

Oh yeah.  I made that mistake too.  The answer I got (in
somewhat more colorful language) was that if one could
distinguish basic Latin characters well enough and quickly
enough from their embossed shapes in a size small enough for
finger-scanning, one really wouldn't need Braille.  

That conversation occurred back in the days when the
conventional wisdom about character recognition involved
artificial intelligence and the essential characteristics of
characters, so led to a discussion about whether ASCII in some
OCR-like font would work better than more conventional fonts,
i.e., if what one would want to do was character recognition,
conversion to ASCII, and then rendering on the tactile device in
that special form rather than simply representing the printed
characters in tactile form.  But that conversion makes it fully
as easy to produce Braille on the tactile device as to produce
shapes in that alternate font and Braille requires fewer pins,
bumps, or servos, so the answer ends up pretty much the same, at
least for those who can easily read Braille.

>...

    john





More information about the Idna-update mailing list