Bob, BN> I made almost exactly the same argument BN> for an "Architectural Considerations" section on a par excellent. two, independent occurrences of a crazy idea by unrelated lunatics probably means there is some merit to it. (by which i mean that any desire to dismiss the idea because it is easy to dismiss the proposer gets more difficult when there are multiple, independent proposers. And, yes, it must be saturday night, for this kind of writing.) BN> "Any specification that creates interesting features should be required BN> to specify its requirements on providers and its impact on consumers." BN> -- might be a bit too optimistic...I don't think it goes BN> quite far enough in noting that it will probably take BN> review by members of the small cadre of IETF architecture BN> experts to be sure that all relevant requirements and that's fine. the point of the exercise is to require specification authors to pay significant attention to the question. this actively creates a dialogue on the issues, rather than hoping that some clever architect notices this particular document. d/ -- Dave Crocker <mailto:dcrocker@brandenburg.com> Brandenburg InternetWorking <http://www.brandenburg.com> Sunnyvale, CA USA <tel:+1.408.246.8253>, <fax:+1.866.358.5301>