Re: Hashing, and Ted's concerns about wastefulness

> 
> At 22:06 03/02/99 -0500, Al Gilman wrote:
> >At 01:32 PM 2/3/99 -0800, Ted Hardie wrote:
> [...]
> >>Remember that typical hash functions care about control characters, so
> >>we inherit the whole CR vs. CRLF kinds of problems.  Normalizing it now
> >>is worth the time, not so much for the "seperately authored" case, but
> >>so that the internal representations used to generate the hashes are
> >>consistent across platforms with different conventions.  
> >>
> >
> >If you can wait a while (time measured in months, not years), the work
> >coming forward as character model and InfoSet for XML will give you a good
> >foundation so we can have international tags and you can still normalize
> >text if you wish prior to hashing.
> >
> >I am not sure we are going to agonize ourselves through the character set
> >problems any faster than they.
> 
> I basically agree with Ted here.
> 
> I think the syntax is sufficiently constrained that character set problems
> are not the issue.  I think we're just talking about whitespace
> normalization (including CRLF if present).
> 
> The basic syntax is easy:  all whitespace not in a quoted string can be
> stripped.

Having spoken out before against normalization in general (as a
significant complication with small benefit), I'll chime in now to say
that this specific variant sounds reasonably simple, and that my
opinion on it is pretty neutral.

  Bill Newman
  [email protected]