>A contractual requirement at this level of detail seems totally >crazy.
I'm afraid I agree. I see this in our other kinds of process specifications too -- we write rules for which you need to exercise sensible judgement, and then fret about what happens when someone uses bad judgement and try to write rules to prevent it. While it would be possible for a provider to win the RFC Editor contract, and fail to provide rsync, that'd be an exercise in bad judgement since they'd probably not get the contract when it was time to renew. While it's possible to tie down most (I dare not say all) loose ends in a protocol description, I can't imagine that you can do it in a process document or contract, and think you'll go crazy trying. Bill _______________________________________________ Ietf mailing list Ietf@ietf.org https://www1.ietf.org/mailman/listinfo/ietf
