--On Tuesday, May 11, 2021 10:53 +0100 Stewart Bryant
<stewart.bryant@xxxxxxxxx> wrote:
>...
> The AD can edit the text and provide a solution that works.
And which has IETF consensus? Again, disclaimers aside (see
Spencer's note), the more we formalize this -- or, to some
extent, view lists of errata and fixes as more than guidance to
the next WG to come along to work on the document -- the more we
create a confusing situation about what implementers are
actually expected to believe. So, if there is significant
controversy about whether the actual correct solution is, I'd
rather see the notes be explicit about that, rather than
describing "a solution that works" rather than an agreed-upon
best solution.
> If there is a problem but no solution that would be quite
> worrying.
Isn't that at least part of what "hold for document update" is
for. I.e., we've identified the problem but determining which
solution is good enough to get (or even presume) IETF consensus
is going to take more study and group involvement.
I don't know how many of them there have been in recent years
but, as someone generating a report about something I claim is
an error, I think it is important that I be about to say "this
particular section or provision is clearly wrong because...
But I either don't have a clue about how it should be fixed or
can propose three alternatives that seem reasonable". That
would be consistent with many comments we've seen on this (and
other) IETF lists over the years in which someone who flags a
problem doesn't really care how it is resolved as long as the
decision is clear and well-documented.
I'd hope that, unless it was possible to reject the claimed
error, any of those cases would go swiftly and efficiently into
"Hold for document revision" rather than, e.g., an AD trying to
figure out the right answer on their own.
john
best,
john
