Re: Tibetan half digits

From: dat@pobox.com
Date: Fri Oct 25 1996 - 17:46:47 EDT


   From: unicode@Unicode.ORG
   Date: Fri, 25 Oct 96 11:29:39 -0700

   [[ bunch of stuff omitted to start with ]]

> That's what I gathered from the Tibetans too, but in the case you give,
> /3 = 2.5, which would imply that there is some notional value, /1 = .5, /2
> = 1.5, /3 = 2.5, /4 = 3.5 ... don't those always work?

   But not all implications are necessarily facts. I think the implication you
   mention may NOT be the case, except on some purely theoretical level that
   never appears in usage... But someone needs to ask the Tibetans. I don't
   know the answers to all the corner cases.

Right; if /9 = 8.5, an interesting case is /0 = -.5 (one would assume, instead
of /0 = 9.5!). Luckily, the theory of putting them before all the other digits
(/0 is the first zero, /1 is the first one, etc.) seems to work just fine, as
long as you don't have to worry about mixing decimal fractions and /-notation).

   It may be that this is over-analysis -- trying to fit these things into a
   decimal digitary mold for which they're not suited or intended. Maybe they
   don't occur alone. I don't know.

The intersting case to me is are they ever mixed in with other decimal
fractions, such as:

   0.08 0.4 /0 0.6 .7 00.9 etc. ???

True numerical sorting would seem to be out of scope for an alphabitization
base purely on local character comparison (one at a time); which is the goal
here? In practice it might be reasonable to lexocographically sort as follows:

  /0 0.0 0.4 0.6 0.7 0/0 00 01 etc.

           rick

-- 
david taffs (dat@pobox.com) // "...this saturation effect indicates increasing
// subjective indifference to increasingly large distortions..." - Sarnoff Labs
// comments packed by volume; soMe unSeTTLiNg maY oCCur DUrInG sHiPmeNt...
assert (this->ideas != employer.ideas && this->opinions != employer.opinions);



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:32 EDT