Regards/Ngā mihi
Brian Carpenter
On 08-Jul-26 08:09, Mirja Kuehlewind (IETF) wrote:
See inline.
On 7. Jul 2026, at 20:19, Eric Rescorla <[email protected]> wrote:
On Tue, Jul 7, 2026 at 9:38 AM Mirja Kuehlewind (IETF) <[email protected]
<mailto:[email protected]>> wrote:
The discussion on which tag applies would be at least based on something
written and consensus based and not just on differing strong opinions about how
to do it right. That should make the discussion easier and also create less of
them. That’s the goal! Maybe it wouldn’t avoid it entirely but I don’t think
that a good reason to not do it as I don’t think there is a perfect solution
(and probably there don’t need to be one; sometimes you simply need to have a
discussion and make a decision).
I'm sorry, I don't agree. My view is that adding more tags with detailed guidelines is just as
likely to create more debate, not less. If we are to do something here, I would instead replace
both Obsoletes and Updates with some variant of "look over here" or "you might be
interested in", thus avoiding any debate.
That’s why we also added the “see also” tag. Why do you think that “amends” and
“extends” are not clear?
As SM pointed out, there is some existing language in 2223bis [1] (ironically, an expired
"Obsoletes" draft) that actually strikes me as very good:
Updates: The new document cannot be
used alone, it can only be used in conjunction with the
earlier document.
Obsoletes: The new document can be used alone as a
replacement for the obsoleted document.
Also, the meaning of "Extends" is quite broad and can mean many different things
[RFC6709]. So can "Amends" (is it a correction, a clarification, or an actual change that
changes what should be on the wire?).
I'm slowly coming round to the view that "Updates" with a mandatory explanation may be
the best solution (whereas "Obsoletes" is complete on its own).
On balance, I'm not a fan of "See Also" - that's what Informational references
are for.
[1] https://datatracker.ietf.org/doc/html/draft-rfc-editor-rfc2223bis-08#page-13
Brian
Having some potentially confusing tags on old RFCs (that could be revised
if needed) doesn’t seem to be worse than what we have right now: one tag that
is not defined at all and a mismatch of use and interpretation as everybody
makes up their own requirements.
I think we just have different opinions.
-Ekr
On 7. Jul 2026, at 17:02, Eric Rescorla <[email protected]
<mailto:[email protected]>> wrote:
On Tue, Jul 7, 2026 at 7:54 AM Michael Richardson <[email protected]
<mailto:mcr%[email protected]>> wrote:
Mirja Kuehlewind \(IETF\) <[email protected]
<mailto:[email protected]>> wrote:
Agreed!
I think that it would be brilliant if an AD review told the WG:
"Yeah, that document is not an Updates-123, it's an Updates-545,
right?"
WG: Oh yeah, we didn't change our meta-data as our I-D evolved into
being
better at interoperating with old thing FOO.
This seems like the optimistic view. The pessimistic view is that it will
usher
in a whole new round of arguments about whether it should be Updates-123 or
Updates-545.
> Also my question is what’s the harm I trying it. If it doesn’t
help, we
> stop it again. However, I think there is a chance to at least
improve
> the situation even though we might not fix all issues entirely.
+1
In addition to the concern above, namely creating new arguments about
which tag applies, if it turns out not to work, we end up with a bunch of
confusing and no-longer-used tags in perpetuity.
-Ekr
--
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]
--
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]
