Re: [ietf-smtp] draft-freed-smtp-limits

[John C Klensin <[e-mail]>]

--On Friday, August 4, 2023 13:58 -0700 Dave Crocker
<[e-mail]> wrote:

> On 8/4/2023 1:48 PM, John C Klensin wrote:
>> I'm deliberately not considered adding additional limit
>> specifications to Section 4 because I have no way to evaluate
>> whether Ned would have liked those parameters or how they were
>> described.
> 
> Other than typos, that's true for any changes that might be
> made now.  In strict terms, that means that no changes of
> substance can be made.

And, indeed, I'm trying very hard to avoid changes of substance.
I do not believe that, e.g., adding the acknowledgments section
I added to be a substantive change, nor do I consider  fixing
the typographical and grammatical errors you and others spotted
to be substantive.  If others disagree, I'd rather hear it now
than in IETF LC.

> It is appealing to seek so pure a goal for the document, but
> is it practical?

I think it is but that may involve one quasi-substantive
decision that I mentioned in my note earlier today via-a-vis the
caching stuff.  You mentioned an issue there November; Murray
has mentioned some others in the last few days, including a note
posted a few minutes ago; and, as I have reflected on your
comments and your, I've come up with others.  My (increasing)
inclination right now is just to drop that subsection: it is
clear to me that Ned would not have wanted it to be there and
wrong (or even badly ambiguous) and it is arguably more of a
comment than substantive in any event.  It seems to me that the
alternative is to figure out (i.e., guess) what he intended to
say and then write considerable more text to clearly explain
that.  I don't want to go there.
 
> And to the extent there is a challenge in attempting to divine
> what Ned would have agreed with, I suspect there is little
> doubt he always wanted a specification to be as good and as
> practical as is reasonable.

Agreed.  See above.

best,
   john