On 18/10/2021 05:52, Murray S. Kucherawy wrote:
On Fri, Oct 15, 2021 at 10:12 AM Murray S. Kucherawy <superuser@xxxxxxxxx>
wrote:
I've got a draft that seeks to update BCP 97, which is the guidance around
how we handle normative downward references. It's currently made up of
three separate RFCs and an erratum, so this will consolidate those into a
single document. The main mission here, though, is to update the guidance
around normative references to external documents, especially those behind
paywalls.
The draft is being sponsored by Erik Kline and can be found in the
datatracker here:
https://datatracker.ietf.org/doc/draft-kucherawy-bcp97bis/
[...]
I've captured the feedback from several of you so far into -01 of this
document, which is now posted. Please let me know if I messed anything up.
I noticed and read this I-D before this thread started and was wondering
where and how to comment.
As others have said, producing process documents is not the IETF's
finest hour and this I see as an exemplar of that. (I wonder if those
commenting on it are commenting on what they think it says as opposed to
what it does:-(
The terminology I find unhelpful; the document cries out for a section
on Terminology. In most places, I do not know what the document is
referring to. Thus when it talks of what must be in an RFC, I think
good, I can ignore all this as it is up to the RFC Editor to add
annotating text! It may be of course that the text here means to refer
to I-D, as at Last Call, as well but that is not what it says. Likewise
the word standard appears in many places. Most on this list will know
what a Standard is; is that intended or does it mean a document on the
standards track which most on this list will know is different? And the
focus seems to be RFC ignoring the role of I-D in the IETF process. The
document introduces the concept of source document and target document
for clarity and then fails to use them when clarity is called for. And
then most will be familiar with Normative and Informative; here the
author uses normative and informative. Are they intended to have the
same meaning? Who knows (apart from the author:-)?
To me it is symptomatic that an example is made of a MIB when the IESG
made YANG the standard for network management many years ago and the
example would have exactly the same force were it to refer to YANG
(RFC8407).
Tom Petch
-MSK
