ietf-nntp Title of AUTHINFO USER/SASL spec

[Jeffrey M. Vinocur]

On Tue, 18 Mar 2003, Ken Murchison wrote:

> "Jeffrey M. Vinocur" wrote:
> 
> > Now that the AUTHINFO stuff is separated from TLS, I have to figure out
> 
> Actually, why were they separated?  Since we know that they are both
> tightly coupled (USER/PASS can't exist without TLS), why not just have a
> single "NNTP Authentication and Security" document?

When it was a single document, it felt to me like two separate documents
lumped together.  (I think Russ agreed.)  It's a lot easier for the reader 
to get a handle on two separate specs, in this case.

There's no real reason to have them together (so long as the AUTHINFO USER
stuff can reference TLS where appropriate), and some benefit to having
them separate.  For example, the one can be implemented without the other,
the TLS draft can be published when it's done without having to wait for
SASL (which I expect will involve more debate), one spec can be updated at
some future date if necessary, without the awkwardness of "this obsoletes
section 7 of ..."  and such.


> Authentication is authentication.  I don't see the need to have SASL in
> the title.  I'm fine with "NNTP Authentication" or "NNTP Extension for
> Authentication", which is what SMTP does in RFC 2554.  

I think "NNTP Extension for Authentication" is a good suggestion.  

Any thoughts on rephrasing "Using TLS with NNTP" to "NNTP Extension for 
TLS" or something?  Consistent titling might be nice.  (Of course, then 
the obvious title would be draft-ietf-nntpext-nntp-tls instead of the 
current ...nntpext-tls-nntp, if anybody cares about such things.)


-- 
Jeffrey M. Vinocur
jeff at litech.org