On 4 May 2015, at 5:25, David Conrad wrote: > [This has drifted a ways off from DNS terminology, so I changed the subject]
Thanks! Note that if we look at the original discussion, my only point is that we should not claim ccTLDs are only allocated according to ISO 3166, but that there is freedom to create two letter codes -- and that this has happened. > Sorry, I'm a bit confused. > >> What ISO 3166/MA said was that the _only_ difference between the two, and >> the reason why they reacted on .EU, was that ICANN referred to .EU being >> reserved as a reason for registration. > > ICANN argued that EU should be registered because .EU was reserved? Yes. >> 3166/MA is specifically reserving things that other organization uses. I.e. >> 3166/MA want to have an exceptionally reserved code of EU just because ICANN >> has decided to use EU as a ccTLD. > > I presume a "does not" is missing there, i..e, "3166/MA DOES NOT want to ..." No, 3166/MA did not mind allocating a code just because of the use as a ccTLD. >> To be able to do that, they do not want ICANN to register a ccTLD of .EU >> with an argument that 3166/MA has reserved EU. That would end up being a >> circular reference. > > Understandable. if the EU ISO-3166 hadn't been exceptionally reserved (like > AC, UK, etc), I would have presumed that ICANN would have simply rejected a > request to register .EU. In those days it was only reserved. In reality, EUR was reserved due to the use as a currency, and the two letter reserved code was EU. My point is that what ISO 3166/MA did not like was exactly what you say. It would have been much better if ICANN would "just" have allocated EU just because EU so wanted. >> A *registration* in 3166/MA on the other hand is something that happens >> because someone else (UN Statistics Division, UPU or someone) decides to use >> a code, and 3166/MA decides to actually register the code. That can be >> referenced by anyone (except the one that have done the registration). > > The 2-letter code "EU" was not in use by "someone else" prior to ICANN being > asked (presumably by the EU) to create it as a ccTLD? Correct. > Or is the parenthetical the issue: that the only "someone else" was the EU > itself, so 3166/MA refused to register that code in their internal > registration table and someone argued that ICANN's use constituted a valid > "someone else"? They could not register it as it was not in use "by someone else". Only EUR was. If ICANN would have "just" allocated EU for use as a ccTLD, without referencing ISO 3166/MA, EU could (my guess) have become a *registered* code, and ICANN would have been just like UPU and others one of the organizations 3166/MA reference. I.e. 3166/MA is very careful with it not being the ones that register codes. Other organizations do make the decisions. They only keep the registry (which of course in turn require careful determination and decision making -- also quite sensitive). Similar wording I have heard from....hmmm....IANA ;-) But I do understand ICANN want to keep referencing 3166 "as much as possible". That way to get a ccTLD one need a few things to happen: 1. Someone must allocate the code for some specific use (or at least make it clear a code should be there, but it does not matter which one it is) 2. ISO 3166/MA must recognize this [1] decision and register the code 3. ICANN detect that ISO 3166/MA has recognized the allocation in [1] and approve the ccTLD given a request is sent in that fulfils whatever other requirements there is Now, my point is that we can simply say that what can happen is also [3] directly, and leave it open for discussions between ICANN and whoever else to just allocate a code. Why limit "ourselves" when we know exceptions will happen? Although we can say it WILL be an exception. > Sorry for the confusion. This IS confusing! :-) I bet Jaap can tell you quite a number of (horror) stories.... Patrik P.S. When .EU was created as a ccTLD I did get quite a number of phone calls and email from ISO 3166/MA, and even if I have kept my email very very organized (I thought) there are a number of them that I unfortunately can not find anymore. Reason for me to be contacted was the various hats I did wear at the time plus the fact Swedish Standard Organization is one of the members of 3166/MA so they did reach out to me. But I see over and over again that it would be better for me to instead of relying on my memory reference actual email... :-( But I do remember one asking basically "Do ICANN know what they are doing?" to which I responded sort of "Yes, they just are not clear enough about terminology, so please relax a bit."
signature.asc
Description: OpenPGP digital signature
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
