<TLDR>:
Do both. "This Internet-Draft will automatically expire on <date>,
185 days after publication. It may be invalidated for other reasons
before this date. For more information about the IETF's document
management system and the current status of this draft, go <here>."
</TDLR>
"...this document introduces... "active" and "inactive" in the IETF
tooling.
It appears that the author(s) are attempting to legislate against
stupidity. This has been shown to rarely work as planned. What is the
intended audience for this change? Who will benefit from it?
Right now, every external repository that retains our drafts
includes, as part of the document itself, a statement that "This
Internet-Draft will expire on <date>." No other information is needed
for a careful reader to conclude that, if the current date is, in fact,
later than the printed date, the IETF no longer considers this draft to
be valid. Maybe it got replaced by a newer draft, maybe it got accepted
and published as an RFC, maybe it got abandoned as a bad idea or even
simply as too much work.
Those familiar with the IETF may be more likely to go to the source
than refer to a possibly obsolete backup in an unmaintained library. A
click-happy reader will go see what's behind the link. This change is
not intended to help those people. They don't need our help. No, this
proposed change is intended to help those who meet two specific
qualifications: 1) they don't care about its status because THIS is the
document they are going to use, and 2) they don't care that there is a
place where they might be able to see if there is a newer version.
The proposal wants to replace this text with an online registry
which holds the current status of all IETF drafts. I submit that those
people who see an expired date today and don't care, probably won't care
much about a status registry and all the extra effort needed to check it
in the future, either.
If this proposal is adopted, it might be better to use a different
label instead of 'inactive'. Perhaps 'replaced' would be better, as
that makes it clear exactly WHY this particular document is no longer
'active'. On a slightly different tack, is there any chance at all of
these (or similar) labels actually increasing the confusion? Are these
the only two possible options so the status is a simple binary choice?
If I post draft-wills-rswg-need-more-meetings-00, it will be labeled
'active'. Other people jump on it, correcting my grammar and
complicating things with unnecessary logic, and before long we are up to
-12. In each case, the Datatracker automatically changes -00 through
-11 to 'inactive' as it gets replaced by a newer version. That part of
the system will be automated and can be expected to work well.
With the current proposal, all older versions will be labeled
'inactive'. To me, this particular label implies that it could be
re-activated by someone who wanted to update it. Someone may decide to
noodle with it. Can the 'active' status be re-instated? As the author,
I should be able to update this status without having to post a newer
version. Unless the WG adopts it, in which case it's someone else's
headache. How do we keep me from deciding that we've wandered too far
off the subject and dusting off -03 and updating its status back to
'active'? Then we have both -03 and -12 active, unless the Datatracker
automatically sets -12 to 'inactive' when I update -03.
No, we have to have at least one more status 'Obsoleted by newer
version', and that particular status should be permanent and read-only.
If we are to make this change, the status should be a text field. This
gives us other possibilities like 'Published as RFC XXXX', 'RFC Pending'
or 'Holding as historical reference'. Who gets the authority to / gets
stuck with the responsibility of maintaining this field?
I submit that the current system is not perfect, but that the
proposed system in its current form is in no way better. It is merely
more complicated and thus more prone to failure in unexpected and
spectacular fashion. It certainly does nothing to solve the problem.
It merely shifts 'didn't comprehend the document' to 'didn't go online
to check something'. Why go to all the trouble of making this change?
Well, because it _is_ in some ways a good idea. It makes useful
data more easily accessed by the rest of the world. Perhaps the correct
solution is to leave the current system in place, maybe with the ability
to adjust the expiration date, but add the new one beside it. Implement
the new status registry, and then put a pointer to it into all newer drafts:
"This Internet-Draft will automatically expire on <date>, 185 days
after publication. It may be invalidated for other reasons before this
date. For more information about the IETF's document management system
and the current status of this draft, go <here>."
-Sandy Wills
On 1/24/24 04:21, Lars Eggert wrote:
Hi,
I was asked to AD sponsor draft-thomson-gendispatch-no-expiry-03, and am willing to do so.
Before I initiate a formal last-call, I'd like to do a substitute of a WGLC and ask interested IETF participants to give this a final read and indicate whether they are OK with seeing this go forward.
Please send feedback along those lines by Feb 4.
Thanks,
Lars
--
Sandy Wills
(727) 267-8037
Sandy@xxxxxxxxxx
