Re: Name Compression. Comparison and Tweaks

From: Robert Brady (
Date: Fri May 12 2000 - 22:09:43 EDT

On Fri, 12 May 2000, Kenneth Whistler wrote:

> I know Juliucz looks askance at the creation of new, ad hoc,
> compression algorithms by the Unicode crowd. But if your problem is finding
> a memory- and computation-efficient way to store the entire Unicode name list
> (or some significant subset of it) in memory for rollover hints, or whatever,
> the approach I have outlined above results in a pretty good tradeoff and
> is rather easy to implement.

a) it is still a totally pointless re-invention of axel-based disc
locomotion. better compression algorithms already exist, and more to the
point, they have been tested and won't suffer odd bugs due to being
reimplemented thousands of times.

b) it is not as easy to implement as a few calls to zlib.


This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:02 EDT