Re: acknowledging reviewers better (was Re: Diversity considerations)

[Benjamin Kaduk <[e-mail]>]

On Tue, Oct 02, 2018 at 09:33:34PM +0200, Toerless Eckert wrote:
> On Tue, Oct 02, 2018 at 05:23:01AM -0400, John C Klensin wrote:
> > Michael,
> > 
> > Agreed, but note that, if you want such acknowledgments to be
> > useful for some of the purposes you and others have argued they
> > would help with, you may need to figure out how to differentiate
> > between a review that addresses the document substantively and
> > makes a contribution to its quality and one that has substantive
> > content equivalent to "I'm the designated Fubar team reviewer. I
> > know nothing about the subject matter of this document, but I
> > found a comma out of place in the middle of the second paragraph
> > of Section 6.6.6".
> 
> I got reviews about the quality of the language being bad, so i always
> weelcome good language review feedback, and often there are more people
> willing to do this in the broader IETF review than in a WG. I am curious
> though how much of that language review should happen through the
> formal IETF review or through RFC editor. The RFC editors are IMHO
> (no offense to IETF community meant) degrees better on language
> review than any IETF review i've seen. But of course i worry about what
> is an appropriate low bar to pass on to them. I also felt bad about 
> the language i passed on to them, so i compared the latest draft
> version and final RFC from a couple of "native" english speakers RFCs,
> and felt a lot better about my own documents afterwards (grin).

The RFC Editor is pretty cautious about making changes, certainly ones that
might change the technical meaning, but also for more broadly
rewording/reordering sentences or changing the organizational structure of
a document.  In some cases this broader sort of change might help a
document out, but even if the RPC staffer editing the document knows that
it would be better prose, the change will still not be proposed to the
authors as it's out of scope for their task.

It's also true that everyone suffers from "review fatigue" -- if there's a
lot of things to change, some will get missed in the time available.  More
nits will make it through to the final version if there are more in the
input to the process.  How important the final value is is a subjective
question, of course, and not necessarily even one that we need to achieve
IETF consensus on.

> I think it will be quite common that specific areas will have no
> reviewer available who is familiar with subject matter. I would
> encourage those reviewers to not only help improve the language but also
> not be afraid to ask "dumb question" when they do not understand the
> document. In parenthesis, because i think there are no dumb questions.
> Just badly written documents. Aka: non-experts can very well help
> to improve the readability of documents by forcing authors/editors
> to insert explanations that experts most likely will have thought to
> be common sense. But IETF work gets more and more compartmentalized,
> so that expectation is actually less and less true. 

This is also true -- even in my IESG ballots I will frequently ask for
clarification on a point here and there.  I go in assuming that it's just
my confusion and lack of familiarity with the subject matter, but
occasionally the text is actually incorrect.  (And when I do ask for
clarification, most of the time I only want it in email, and not as a
change to the document.)

-Ben