Hi all,
On 12/20/24 5:50 PM, Christian Huitema wrote:
On 12/20/2024 3:21 PM, Jean Mahoney wrote:
Hi all,
A few comments about WG scope below --
On 12/20/24 3:47 PM, Christian Huitema wrote:
On 12/20/2024 11:50 AM, John C Klensin wrote:
--On Saturday, December 21, 2024 08:04 +1300 Brian E Carpenter
<brian.e.carpen...@gmail.com> wrote:
On 21-Dec-24 04:14, S Moonesamy wrote:
Hi Brian,
At 05:13 PM 19-12-2024, Brian E Carpenter wrote:
Why? What if it describes a serious problem? How do we know that
the 4 open reports from 2010 are of no value without looking at
them?
I doubt that a person would believe that the RFC Editor is taking
the 2010 errata reports seriously if there wasn't any action on
those reports for the last 14 years. I would not describe a
report which I receive as being of no value.
The bug is that it isn't the RFC Editor that is responsible for
taking
action, beyond notifying the stream. The new system needs to ensure
that
there is a clear chain of responsibility (and probably some nagging
mechanism when no action occurs).
I note that most bug-fix fora have a possible end state of "won't
fix".
There's no reason we can't have a "won't fix" equivalent for errata.
Such fora also usually have a "re-open" action if a "won't fix" bug
turns out to really matter. I think the analogy is quite strong.
If we replace "rejected" by "won't fix", then an error report from
2010 could be re-opened in 2030 if appropriate.
Brian,
Agreed, except that, IMO, one reasonable interpretation of "hold for
document update" is "won't fix in current version but will review
this (e.g., automatic re-open) if we ever come back to that
document". If that interpretation is, in fact, reasonable, they we
have been misusing the existing category and, absent other changes,
there may be no reason to assume that the people/process who have
left a report open rather than assigning it to "hold for document
update" would not simply leave it open rather than going to the
effort of assigning "won't fix".
All of this, IMO, brings us closer to the hypothesis of Stephen and
others that the current system is fundamentally broken and that it is
time to discard it and start over.
Agreed. The "errata" system might have been adequate for a technical
journal, but the stream of IETF RFCs is not a technical journal.
Stephen proposes to remove the "errata" process from the RFC Editor
web site, and to replace it by an issue/comment system geared towards
informing the standardization process and the next release of the
standard described in the RFC. I think that's broadly what we want.
[JM] I share this group's enthusiasm for such as system; however, I
think that defining such as system is outside of the scope of the RSWG
because such a system is about managing new work and not about the
RFCs themselves. Discussions to work out the details should happen on
a mailing list with a broader charter and a bigger subscriber list.
For errata system topics that are in scope for RSWG, I listed several
potential policy items in a separate thread. I also appreciate the
discussion on how to handle the backlog, and I think that's also in
scope.
Yes, the current errata system is clunky and difficult to interact
with if you are a verifier. Reports are currently hard to find if you
are a new reader. However, I want to make it better rather than
removing it completely from rfc-editor.org. Removing the reports won't
help the readers of RFCs, who can benefit from reading the reports
even if they haven't been verified yet.
The errata system could exist as the last step in this new system:
Reader query -> discussion -> do we need to fix it? -> Yes -> Can we
issue a patch for readers?
No - WG/RG/IAB/author creates a new I-D
Yes - Verifiers create an erratum through errata system
And the RSWG can create policies around this.
I think the consensus is a version of the process in which we *never*
issue a patch for new readers. That is, we rename errata as "issues",
and there is never a "verified" state.
[JM] What I meant by patch was like the inline errata -- a reader can
see the original text and the corrected text.
Thanks!
Jean
-- Christian Huitema
--
rswg mailing list -- rswg@rfc-editor.org
To unsubscribe send an email to rswg-le...@rfc-editor.org
