Re: RE: AD review of draft-ietf-rddp-sctp-03

Caitlin,

thanks for the feedback.

On Jun 15, 2006, at 1:07, Caitlin Bestler wrote:
> Lars Eggert wrote:
>>          Should this be an RFC2119 "MUST" or "SHOULD" instead
>>          of "is undesirable to require support for?"
>>
>
> The intent is to describe a rationale. There is no intention to
> discourage
> any implementation from attempting to support both.

I should have written "MUST NOT" or "SHOULD NOT."

Anyway, your explanation clears things up - thanks!

> This is for two
> reasons: 1) There is no interoperability reason to discourage such
> an added feature, 2) the technical problems mentioned provide  
> sufficient
> discouragement on their own.
>
> So the intent is to state the rationale for the exclusion of a
> requirement
> (such as "An SCTP association MUST be able to support both DDP and
> non-DDP
> streams concurrently.")
>
> Any requirement that could be stated would be on ULP designers,
> something
> like "Upper Layer Protocol designers MUST NOT assume that an SCTP
> association will support DDP and non-DDP streams concurrently."
> I don't think that such wording is any clearer, and in any event
> creating MUST NOTs for upper layers is somewhat pointless.
>
>
> _______________________________________________
> rddp mailing list
> [email protected]
> https://www1.ietf.org/mailman/listinfo/rddp

Lars
-- 
Lars Eggert                                     NEC Network Laboratories

_______________________________________________
rddp mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/rddp