Joe, On 2008-06-28 08:31, Joe Abley wrote: > > On 27 Jun 2008, at 15:57, David Conrad wrote: > >> On Jun 27, 2008, at 12:21 PM, SM wrote: >>>> I believe an RFC that provides an IETF-defined list of names (beyond >>>> the 4 in 2606) and/or rules defining names the "Internet technical >>>> community" feels would be inappropriate as top-level domains would be >>>> quite helpful. >>> Do you mean as in RFC 3675? >> >> No. I feel an RFC that creates a list (or defines a rule) that >> identifies what names would be inappropriate for top-level domains >> would be quite helpful. > > Personally, I think that any such list (even one that was not static, > but existed in the form of an IANA registry) would always be incomplete. > > A better approach, I think, would be for proposed TLDs to pass technical > review through some suitable body who could consider each case on its > merits. I think all the external evidence is that ICANN is deeply reluctant to set up mechanisms that require the application of common sense (a.k.a. judgment) as to whether or not a particular domain name may be registered. I see no reason to expect this to be different now they have opened the floodgates to greed at the TLD level too. So I think that any such technical review process is doomed. The best we can do is proceed under the second paragraph of section 4.3 of RFC 2850, i.e. designate specific TLDs as reserved for technical reasons, and so instruct IANA. Furthermore, I believe this is not only the *best* we can; it's essential that we do so, although translating 'example' into every script and language may be going a bit too far. So I believe that 2606bis is very necessary. > >> A couple of examples: >> >> - a label consisting of all numbers >> - the label "local" >> >> There may be others... > > There will always be others, in my opinion, which is why I think the > idea of a list of bad ideas is dangerous. Just because things are not on > the list of bad ideas doesn't mean they are good ideas, but that's now > how people will interpret it. Unfortunately that's true, and that may mean cranking 2606bis repeatedly. But the alternative (inserting the IETF in a TLD approval process) is pure lawyer-bait and would no doubt send the IETF's insurer apoplectic. Brian _______________________________________________ Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf