Re: Nits in current drafts

[pgut001%cs.auckland.ac.nz@localhost (Peter Gutmann)]

Ben Harris <bjh21%bjh21.me.uk@localhost> writes:

>I think this is a bad idea for the following reasons:

None of those seem valid.  At the moment, the only signature mechanism is ssh-
xyz (the others are all ambiguous, so in effect they're xyz%vendorname.com@localhost
mechanisms).  Therefore if a weakness is found in the hash or signature
algorithm, all implementations will need to be fixed anyway.  Conversely, if
no weaknesses are found, there's nothing to do.  In both cases it's no change
from the current state of affairs.

(If there was some huge deployed user base that needs to be accomodated that'd
 be another issue, but debating the theoretical pros and cons of something
 that, in 23 revisions of the spec over a period of 8(?) years, nothing has
 ever used, seems rather pointless).

Peter.