-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
From: "Derek Atkins" <warlord(_at_)MIT(_dot_)EDU>
Just to play devil's advocate, why do we need yet another compression
algorithm?
OK... I'll argue the other side for you ;-).
Why did we need any more ciphers after adding 3DES? Performance, for one.
BZ2 produces smaller output for many common forms of input.
If it didn't have any obvious advantage, I'd agree with you; but, it does.
It's certainly more useful than two stylistic variants of the same thing,
which is what we have now.
Turning the question around: what's the harm?
Adding another optional compression algorithm won't create any problems
we don't already have. (ZLIB is supported by GnuPG but not PGP.)
The preference system is the architected solution to this problem.
As David notes, moving from one implementation to another requires
being able to rewrite preferences, but that's always been the case.
Now, given that compression can be done by the end-user, it's hard to
argue that *any* algorithm is a have-to-have. (This seems to be the
essence of Werner's argument, although he makes it in terms of
PGP/MIME, which I don't find compelling for the pure OpenPGP spec.)
But since the architecture includes compression, I'd say that
including a popular and more powerful algorithm is worthwhile.
-----BEGIN PGP SIGNATURE-----
Version: PGP Personal Privacy 6.5.3
iQA/AwUBPwWrnec3iHYL8FknEQK+kgCfatRR21tyFSM5Oy0T5XO9hr7fTB0An14+
5fVxGAopM2XWaJY5E/OlodxJ
=l/o7
-----END PGP SIGNATURE-----