On Nov 14, Arthur A. Gleckler wrote:
> On Nov 14, 2006, at 12:40 PM, Eli Barzilay wrote:
>
> > In an attempt to make programs more robust, and supported by the
> > Fooish academy, the FSCII (Fooish Standard Code for Information
> > Exchange) encoding specifies that "pp" is the same as "p". Being
> > a technologically influential nation, FSCII is adopted by the rest
> > of the world. The R9RS is revised, and R10RS has an `apend'
> > function that is used to append lists.
>
> Hmm. This is certainly thought provoking. But FSCII hasn't been
> adopted by the world; Unicode has.
Right -- so throwing the decision on Unicode would be a good thing,
but what bothers me about it is:
* Unicode still changes. More frequently than RnRS even.
* The rules are complex: difficult to know them all, difficult to
implement, efficiency problems.
The fact that these arguments always go very fast to all kinds of
Unicode nuances and exotic language features is a good indication that
I don't want that to have an effect on my code.
> [...] Case-insensitive Scheme will be helping them further their
> knowledge of Hebrew!
Ha -- and I prefer Scheme programmers to focus on the code rather than
explore new cultures...
> Sorry, I don't know Hebrew, so I don't know what the right thing
> would be. But aren't there rules in the Unicode standard that say
> what to do here and in general for case?
To add on what John said: besides case folding (which doesn't apply to
Hebrew), you can have additional rules for sorting, which can put
final letter forms at the same place the non-final forms are. But
that's beyond what I know on Unicode. (And I really didn't want to
know anything about it.)
> Human languages are a beautiful and complicated thing, ...
(I agree with the complicated part. Beautiful or not is an argument I
can have with my wife. She's doing NLP.)
--
((lambda (x) (x x)) (lambda (x) (x x))) Eli Barzilay:
http://www.barzilay.org/ Maze is Life!
Received on Tue Nov 14 2006 - 17:54:40 UTC