Hi, "Scott O Bradner <[email protected]>" wrote: > encouraging a report is fine
Agreed. > retracting the code points seems to add more confusion than it is worth > unless the code space is very tight Disagree. From my experience at IANA, trying to figure out who to contact to remove a code point gets harder the longer the code points are not being used. Unless the code space is unlimited, I'd argue that you want to deallocate as soon as an experiment is over. I'd even go so far as to say that the original proposal for experimental code points should have explicit revocation dates (which can, of course, be refreshed similarly to IDs). > and I see no reason to obsolete the experimental rfc or move it to historic > status unless the report is that some bad thing happens when you try it out - > updating the old rfc is fine Having been involved with RFC 6563, I think in general it is quite useful to signal "hey, you really don't want to implement this". If this can be done by updating the old RFC, that's fine. > and I agree with Elliot about the nature of research - it is very common to > not reach a conclusion that something is bad (as in bad for the net) - and > that is the only case where I think that an experiment should be flagged as a > don't go there situation Agreed, with the proviso that limited resources (whether they are scarce or not) should be reclaimed. Regards, -drc
