ietf-nntp AUTHINFO SASL protocol choices

[Jeffrey M. Vinocur]

On Thu, 4 Apr 2002, Russ Allbery wrote:

> Andrew Gierth <andrew at erlenstar.demon.co.uk> writes:
> 
> > That's not acceptable, at least for the first part of the exchange,
> > because the client doesn't know that the command will succeed or even be
> > recognised (_requiring_ the client to do LIST EXTENSIONS just forces the
> > extra round-trip again).
> 
> Oh, that's an extremely good point.  I hadn't thought of that.

Nor had I.  *sigh*  I really liked that idea, too.

(Now, there is still some benefit there.  Say we don't support the
optional initial response, then we cost one round trip, but after that can 
send the data without waiting.  Except that we still have the issue that 
some software may have trouble implementing anything except 
command-response...was that concern for server software or client 
software?)


> Okay, yeah, it looks like we're back to allowing really long command
> lines.

Ok, here we go again with the line length issues.  Can we make a decision
soon about what modifications will be made to the base spec regarding line
length?  

-- 
Jeffrey M. Vinocur
jeff at litech.org