On Jul 14, 2006, at 9:59 AM, C. M. Heard wrote:
Very well said. As I said in my message of 18 June, my advice would be to make a relatively minor set of clarifications to BCP 9 (RFC 2026) and move on. It would also be OK for newtrk to refocus on its original charter of simplifying the standards track. But I would be very dismayed to see it focus on document relationships.
Few RFCs are stand-alone elements for developing interchange. When flattening document categorization, conveying levels of interchange remains problematic. Evolution of document relationships remain an element poorly handled by composing these sets within RFCs themselves, which are likely rapidly dated. A tracking system not encumbered with inclusion of normative language ensures a working-set can be tracked in a reasonable fashion. Much of the RFC review process and utilization depends upon understanding what is the intended set. The Name.Serial proposals as found in both the ISD and SRD proposals provides a means for both tracking this evolution, while also stabilizing references used to uncover document sets. Often as documents change, reference to the prior set may be considered by the community as Stable, whereas the latest set, as Current. Stable versus Current is too dynamic to be tracked by a highly formalized process. The IETF could publish a list of interchange categories as just Name.Serial on a web page, for example.
Once the process is understood to be broken, it should also be obvious that it would have also been fixed had this information been useful. Those actually using the information have been well served by the efforts focused upon providing document relationships. It would also seem more appropriate to categorize the document sets rather than individual RFCs. There could by a set that includes the single RFC, but that will likely be the exception and not the rule.
-Doug _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf