Re: draft-06 plus last-call changes

On Sep 9, 2004, at 5:35 PM, Mike Brown wrote:
> Roy T. Fielding wrote:
>> it is moving RFC 2396 to Obsolete status.
>
> Thanks, but where is Obsolete status formally defined?
> It is not mentioned in RFC 2026 or its errata.

I have no idea -- I don't think 2026 accurately describes how the
RFC editor function works anyway, which is why there are other
working groups working on that.  I do know that the term Historic
is only placed in the Status section when a protocol is put out
to pasture (i.e., no further specification will take place).

>> That means implementors
>> are requested to implement according to the new RFC
>
> Requested, not required...OK.

All IETF specs are voluntary documents.  They usually reflect what
is in the best interests of the developer to do, but standards cannot
(and should not) prevent innovation from occurring -- they merely
state what is standard.

> It's just that in the absence of a formal definition of Obsolete (and 
> maybe I
> just didn't look hard enough for one), the mere declaration that RFC 
> 2396 is
> Obsolete does not imply that much to me for the purpose of updating an
> existing implementation of, say, RFC 2616.
>
> I *think* I am supposed to make it conform to the 2396bis as best I 
> can, but
> I'm just hesitant to assume *anything* without seeing an explicit 
> statement
> (in the spec) to that effect.

I think you should make it work as best you can, taking into
consideration the advice within the latest specification of any
given technology.  AFAIK, there is no advice in the latest spec
that you wouldn't want to follow, whereas I know of quite a lot
of advice in RFC 2616 that is just plain wrong outside of a
very limited scope that may (or may not) have been applicable
six years ago.  That specification will be revised as well,
eventually.

....Roy

Received on Friday, 10 September 2004 01:10:58 UTC