On Mon, 2004-12-06 at 11:52 -0500, John R Levine wrote:
If they do, that's great. If they don't, we're no worse off than we are
now. But you're still making the implausible assumption that you can
design a signature scheme that survives list mutation without also letting
through a lot of forgery, breaking existing list clients, or more likely
both.
Actually, at this point I'm not saying that at all. I'm saying _either_
use RFC2822 and make it survive, or accept that you can only use the
'most recent sender' and use RFC2821 for that instead of gratuitously
banging your head against a brick wall and trying to use RFC2822
identities.
I guess I'll have to help you guys come out of your time warp and show you
all the messy stuff that mailing lists do these days to help you
understand why surviving lists is a myth. Mailman and ezmlm run an
insignificant fraction of lists, you know.
Please do. Real statistics on the amount of mailing list traffic which
wouldn't survive the (sig+count) signature currently described by IIM
would be a very useful addition to the debate.
--
dwmw2