--On Thursday, June 10, 2010 15:39 -0700 Ned Freed
<[email protected]> wrote:
>> In general, the current IESG is quite good with not
>> nitpicking a document to death if warned/asked in advance. I
>> also haven't heard any suggested changes to the
>> pre-evaluation document of 5321bis, so I don't think
>> publishing them as RFCs would actually be much more work.
>
> You might want to ask the authors of that document if they
> would be happy
> having it published in its current form before making such
> assumptions.
I think Ned's implied question is important so, as one of said
authors, let me try an answer in both short and longer form.
Short form: ROFL! And that isn't with happiness and joy.
Longer form:
As both of you know, the informal agreement was that Barry would
assemble the pre-eval document and I would concentrate my
efforts on 5321bis itself. I've looked through the various
iterations on the pre-eval document, but I've concentrated on
the equivalents of the task list --what we are and are not going
to do. I've paid about zero attention to the more descriptive
text and even less to how things were phrased, etc. Barry
writes more than well enough for me to be very confident that
whatever he put in was good enough for the purpose.
But, if one turns the document into something that is going into
the RFC series with my name on it, I start behaving just about
the same way Ned describes himself as behaving. I want it right
and I want it up to whatever I try to establish as a standard
for my permanent/ professional writing. That means a complete
editing pass on the I-D, discussions with Barry about possible
changes, and probably another pass or two. Then we get to
AUTH48. Perhaps we would get lucky this time but, on almost
every RFC on which I've been an author or co-author, AUTH48
results in identifying some changes that I'm not happy about on
first reading. Many of those lead to extended discussions about
stylistic issues, norms, and comprehensibility. Sometimes ADs,
co-authors, WG Chairs, authors or editors of related documents,
and an occasional shepherd get into the act and we don't all
agree with each other, much less the RFC Editor, at least
initially. Patches the RFC Editor wanted to make by changing
a few words turn into rewritten paragraphs that several people
want to go over. Three to five iterations on a document after
the first AUTH48 are pretty common (and I've experienced larger
numbers). We usually emerge with a much better document and I
almost always end up reflecting on the professionalism and good
tempers of the RFC Editor staff for not deciding I take far more
effort than my documents could possibly be worth, but that is
how it goes. It is hugely time-consuming.
Alexey, you've been following the AUTH48 process on the IDNA2008
documents since they went into that state a week ago. There are
some things about those documents that make them unique, but
maybe there are some things about every document that makes it
unique. I'll spare the rest of you the details but... it is
taking a lot of everyone's time to make sure the published forms
are as clear, precise, and comprehensible as the future readers
deserve.
In addition, the effect on YAM of the extra time an RFC effort
for the pre-eval documents would cost is fairly clear for me and
probably others: the WG is only going to get so many cycles in a
given month and time spent doing editorial work on the pre-eval
docs is time that translates directly into slippage of the
schedule for getting the standards-track document finished, at a
ratio that is probably somewhat worse than one week per week.
best,
john
_______________________________________________
yam mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/yam
