Re: Proposed WG: SSH, Secure Shell (SECSH)

[Robert Elz <[e-mail]>]

I would actually suggest changing the name, "secure shell" has
almost nothing whatever to do with what is being proposed here.
The name, of course, originates from the UCB "rsh" command,
"remote shell", whose purpose is/was to give a shell on a remote
system, with "remote" being the important big (rsh is not a shell,
just a method to get to one elsewhere).

ssh as a unix command name is nice, easy to type, and all that,
but by getting rid of the "remote" part, the essence of what is
being provided is totally lost, ssh doesn't give you a secure
shell (which most would interpret as a local command interpreter
from in which some element of security was provided) but a
secure connection to a remote shell, which may be totally insecure.

If the group could have the ssh acronym, it would probably be worth
keeping, just because of the inertia in the name, but as that one
is taken already, and can't be duplicated, it would probably be a
better idea to use the WG game to emphasise the true nature of the
work and the protocol, which isn't the "shell" at all, which really
has nothing whatever to do with anything (quite apart of the ssh
stuff doing much more than just providing access to a remote
shell).

A minimalist change might be to make it srsh (which at least
puts the "remote" part back), but getting rid of the shell entirely
and calling the group "secure remote communications" or something
would probably be better.

The mailing list would probably be better named after whatever
the final acronym for the group is, than "ssh".

The only other query I have is with respect to the April 97
(Memphis IETF) milestone... "decide on transport layer protocol".
Am I missing something here, or is that ambiguous, or is there
some reason why anyone would even consider (in the IETF) anything
different than TCP for this?   Is that decision really going to
take an entire WG meeting (even in one of the new 1 hour slots)?

Oh, and if the proposals are to be submitted in Sept 97 (which
given that ssh is fairly mature, and just needs to be documented
in IETF style) I don't see as being too ambitious, why is the
WG scheduelled to meet in Dec?  That would be too soon to consider
DS status, what else would the WG be doing (that couldn't be
better done on a mailing list).

kre