Re: A Process Experiment in Normative Reference Handling

["Kurt D. Zeilenga" <[e-mail]>]

At 05:06 AM 3/31/2006, Brian E Carpenter wrote:
>Kurt,
>
>The phrase "maturity level" is from RFC 2026 and it means exactly
>what you want.

I disagree.  RFC 2026 refers to all RFCs, including Informational
and Experimental RFCs, as having a maturity level.  RFC 2026
makes a clear distinction between standard track maturity levels
and non-standard track maturity levels.  

>To grossly simplify John's proposal, it will
>allow a DS to refer normatively to PS documents and a Standard to refer
>normatively to DS and PS documents.

John's I-D, as presently worded, appears to not to make any
distinction between RFCs on the standards track and those not
on the standards track. In fact, the I-D does not include the
phrase "standards track" (except in the title of one of the
references).  Hence the confusion the list.

John's proposal also allows RFCs to normatively reference
I-Ds.

>We would need to decide whether
>that's a problem or a solution.

If John's proposal would be revised to be clear as to
disallow a standard track RFC from referencing non-standard
track RFCs, as well as be changed to disallow referencing
I-Ds, I could support the revised proposal.  As it stands
now, I cannot.


>    Brian
>
>
>Kurt D. Zeilenga wrote:
>>At 08:45 PM 3/30/2006, John C Klensin wrote:
>>
>>>While the more general discussion of, e.g., what to do about I-Ds, is interesting and worth pursuing, all the document itself proposes is _exactly_ the above: references to published documents at earlier levels of the standards track and, at IESG/ community discretion, reference to approved (and hence firm, modulo only editorial work) I-Ds.   The latter are "editing in progress", not "works in progress" in the usual sense.
>>
>>I don't read your I-D as being restrictive as my suggestion. You didn't qualify what you meant by "maturity", implying
>>that experiment allowed PS to normatively reference Informational
>>and Experimental RFCs as well as I-Ds.  My suggestion is not
>>to allow such references.
>>For clarity,  I can support an experiment that allows progression
>>(without the need for a variance) of a standard track RFC to DS or
>>FS that normatively references one or more standard track RFCs of
>>lessor standard track maturity.  I cannot support an experiment
>>that allows publication of a standard track RFC that normatively
>>references a non-standard track RFC or an I-D.
>>Kurt 
>>