> What I am interested in is whether or not example.co.uk and co.uk are > administrated by different entities, and where the separation occur, and > that that change in administration signifies that any x.co.uk and y.co.uk > are not considered part of the same domain, but separate domains, and that > as a consequence a.x.co.uk, b.x.co.uk, and x.co.uk are in the same domain > and under the same administrator. > > To put it in a different way: What the suggested specification file > distributes is the vertical boundary separating domains acting effectively > as TLDs (e.g. co.uk) and domains that are not acting as TLDs > (example.co.uk). I have been calling these TLD-like domains "subTLDs", but > others are calling them "Effective TLDs", "public suffixes" or > "registry-like domains", and yet others call a domain immediately below > such domains a "Base domain".
Yngve It's not at all clear to me if your draft has any solution for the many unofficial "registry-like domains" that exist, particularly if your protocol expects each TLD to know about each and every such domain that might exist within its namespace. For example, I happen to know that 'demon.co.uk' has a separate sub-domain for each customer, and should probably therefore be included in your list - it's "registry-like". However there's a (large?) unknown number of similar domains that I don't know about. kind regards, Ray -- Ray Bellis, MA(Oxon) Senior Researcher in Advanced Projects, Nominet e: [EMAIL PROTECTED], t: +44 1865 332211 _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
