Hi Dave,
I agree with you about the counter-marketing text that will only turn
off potential implementers.
However I think this would be more appropriate as Experimental. Yes, we
give PS to specs that may not ever be implemented. But at least working
groups are (usually) good at avoiding publishing multiple specs to solve
the same problem. Here we do not have that gating function. If we
foresee multiple solutions being published for this problem space, which
is what I'm hearing, then Experimental is the better choice.
Thanks,
Yaron
Date: Fri, 09 Aug 2013 19:33:49 -0700
From: Dave Crocker<dhc(_at_)dcrocker(_dot_)net>
To: Stephen Farrell<stephen(_dot_)farrell(_at_)cs(_dot_)tcd(_dot_)ie>
Cc: Carsten Bormann<cabo(_at_)tzi(_dot_)org>, IETF Discussion Mailing List
<ietf(_at_)ietf(_dot_)org>
Subject: Re: Last Call: <draft-bormann-cbor-04.txt> (Concise Binary
Object Representation (CBOR)) to Proposed Standard
Message-ID:<5205A68D(_dot_)2020403(_at_)dcrocker(_dot_)net>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
On 8/9/2013 6:09 PM, Stephen Farrell wrote:
>So some kind of statement that CBOR is one point in a design
>space (as opposed to an optimal solution for some set of
>design objectives) would be worthwhile.
huh? a statement beyond the opening sentence of the introduction and
its third paragraph?
worthwhile to whom and for what? it's a spec (or perhaps a meta-spec.)
it provides a capability. it needs to specify the what and how well
enough to be usable.
while ietf culture permits specifications to have quite a variety of
commentary, historical or contextual discussion is not an essential part
of the document, and certainly not discussion cast in a manner to
denigrate the current spec.
Counter-marketing that has the current spec self-deprecatingly casting
itself as only one of many seems mostly worthwhile to get people to
avoid using it.
exp makes sense if there is doubt that it is technically workable, not
because its success in the market is questionable. we give ps all the
time to specs that have little clear market and go on gain small market use.
from what i've seen, this is a carefully researched and crafted
mechanism and i haven't noted anyone challenging it on basic technical
grounds.
d/
-- Dave Crocker Brandenburg InternetWorking bbiw.net