This comes back to the question or why have maturity levels at all. Ideally, an
implementer should prefer to implement a mature standard over a less-mature
one. In practice, adopting the more advanced standard may give you an obsolete
protocol, rather than a more stable one. IOW the standardization level of a
document does not give a potential implementer any signal as to whether or not
this standard is in any sense of the word "good". And if it doesn't signal
anything to the "customers" of the documents, what's the point of having these
levels at all?
Suppose (picking a random recent RFC) I'm into the Host Identity Protocol, and
I would like to implement multi-hop routing. Searching the RFC database (or
Wikipedia) I see that there's an RFC for that: 6028. So if I want to implement
multi-hop routing, I would implement this, even though it is "Experimental".
Even if there was an older specification for the same thing at full standards,
I would find the appropriate mailing list and ask which one to implement. I
would *not* go for the older document just because it's a full standard.
Given this, why should the authors of RFC 6028 ever bother to advance it to PS,
DS or FS? If this specification needs updating anyway, they might take the
opportunity to also advance it, but otherwise what's the point.
At some point I found that RFC 4478 had 3 independent implementations that
interoperated with each other. I asked an AD about advancing it, and his
response was along the same lines: why bother? My time (and his) can be better
spent on other things.
So in answering you second question, I don't see any reason why things won't
keep sticking in PS or even Experimental forever.
I am also not sure that the bar for Experimental is significantly lower than
PS. The only difference is that without actual implementations deployed in the
field, the ADs feel better about letting a standard go to Experimental. So even
if Russ's draft gets adopted, we'll still be with three maturity levels, except
that Experimental will be the bottom rung (on second though, it already is)
-----Original Message-----
From: ietf-bounces(_at_)ietf(_dot_)org
[mailto:ietf-bounces(_at_)ietf(_dot_)org] On Behalf Of Andrew Sullivan
Sent: 27 October 2010 13:47
To: ietf(_at_)ietf(_dot_)org
Subject: Re: what is the problem bis
On Tue, Oct 26, 2010 at 03:24:55PM -0400, Phillip Hallam-Baker wrote:
The problem with the current, failed process is that there is absolutely no
correlation between the standards status of a protocol and adoption.
Why exactly is that a problem? That's not a rhetorical question. If
_that_ is what the problem is, I want to know why. I agree that two
maturity levels might help if it is the problem to solve.
Supposing it is a problem, why do you think that two maturity levels
won't just cause everything to stick at Proposed Standard?
A
--
Andrew Sullivan
ajs(_at_)shinkuro(_dot_)com
Shinkuro, Inc.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf
Scanned by Check Point Total Security Gateway.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf