Hi Eric,
Eric Rescorla wrote:
Since experimentation resulted in significant Internet deployment
of these specifications, the DKIM working group will make every
reasonable attempt to keep changes compatible with what is
deployed, making incompatible changes only when they are necessary
for the success of the specifications.
As I argued on the DKIM working group list, I think this text is a bad
idea. Part of IETF having change control of a specification is having
the ability to make changes, and the bar of "necessary to the success of
the specification" is way too high for that. Note that I'm not
suggesting that the WG shouldn't consider compatibility, merely that it
shouldn't be effectively prohibited by charter from making incompatible
improvements.
I don't read that as prohibiting incompatible changes, since
canonicalisation and hashing are two such changes. Incompatible
improvements that are not bug-fixes are discouraged though, and
I can see how that raises concerns. I can also see how apparently
encouraging incompatible improvements that are not bug-fixes
raises concerns.
Personally, I think each on-the-face-of-it-reasonable suggested
improvement has to be considered, but the more time passes and
the more the specifications are mature, the higher the bar is
raised. Since these specs. have been around a while and have
been implemented it seems reasonable to start this WG with a
higher bar than one where neither of those things are true. Its
obviously tough to write text saying the above that makes everone
happy since there are so many subjective aspects involved.
Clearly though (as pointed out) a new "rough consensus" has
to be established in the WG, and subsequently during IETF last
call and I do expect that process to involve questioning the
design decisions which were made prior to the WG getting
started. Whether such questioning results in changes
(compatible or otherwise), is something we can't know at this
stage.
I'd also note again that "compatible with what's deployed"
doesn't (once you've changed canonicalisation, which has
happened already) mean the same as on-the-wire compatible, so
the current text is actually less constraining than might
seem to be the case at first reading.
Lastly, the text does say that the working group will not
be unreasonable in attempting to maintain compatibility, so
as the Marx brothers said, there is a sanity clause after
all:-)
Basically, I think the current text is ok, even if it
looks a bit scary.
Stephen.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf