Re: Netscape vs. Digest

On Tue, 27 Aug 1996, Ari Luotonen wrote:

> 
> 
> In Netscape's defence, I implemented Digest for a late beta of
> Navigator 2.0 (I think that version was called SimpleMD5), but just
> days before that beta was due to go out the spec submitted to the IETF
> changed and became unstable judging by the amount of discussion that
> suddenly bursted out in the respective mailing list.  Consequently, I
> had to remove the support that I had added.  We had all due eagerness
> to support it asap but it backfired when spec kept changing.
> 

This isn't quite fair.  Anyone can propose any crazy idea they want on
a newsgroup and many people did.  However, I don't believe that there
have been any drafts of the spec (the current one is 04) which would
not interoperate with "SimpleMD5".  There were some additions, but all
were optional. This was done despite the fact that (IMHO) there were
suggestions for (incompatible) changes which would have improved the
specification.

The motivation for this (at least on my part) was to preserve
compatibility with Spyglass client implementations (some of which also
showed up in some versions of MSIE).  People at Spyglass had done
much of the early work on digest and I thought they deserved that
consideration -- even though when I took over as editor of the spec
they said do whatever is best and they would accomodate if need be.

John Franks 	Dept of Math. Northwestern University
		john@math.nwu.edu

Received on Tuesday, 27 August 1996 14:14:46 UTC