ietf-nntp AUTHINFO SASL protocol choices

[Jeffrey M. Vinocur]

On Wed, 27 Mar 2002, Charles Lindsey wrote:

> Almost. What I had in mind was that if the SASL extension needed a line
> length of 1024, then it MUST also implement (and report in LIST EXTENSIONS
> accordingly) a LENGTH of at least that amount.
>
> I.e., in the document, you
>
>     A. Define a LENGTH extension with parameter 'n' (default and minimum
> allowed 512) which any implementation can choose to provide as it thinks
> fit.
>
>     B. "Any extension which requires the use of a command line length in
> excess of 512 MUST also support the LENGTH extension, with parameter
> greater than or equal to the length it requires."

Ah, ok.  Three things:

- We have to convince Stan and whoever expressed the opinion he relayed.

- So to be sure I understand -- you're suggesting we leave the strict 512
  byte limit in the NNTP spec, unmodified, and then override it here?

- Do we really want to put not-really-related extensions in the same
  document?  It makes giving it a good title kinda hard, if nothing else.
  Why not put it in a separate document?


By the way, was there any discussion at IETF about putting a limit on
server responses, as somebody suggested here when I raised this question
originally?


-- 
Jeffrey M. Vinocur
jeff at litech.org