On Mon, Apr 3, 2017 at 1:07 PM, Thomas Corte <thomas.co...@knipp.de> wrote: > Hello, > > On 03/04/2017 17:20, Hollenbeck, Scott wrote: > >> “However, EPP wasn't designed for high-volume, lightweight availability >> checking.” >> >> That statement is patently false. It may be that some server implementers >> constrain clients, but that’s not a protocol limitation. The whole reason >> the <check> command exists is so that clients can perform high volume, >> light weight availability checks. > > Agreed, which is we've consistently advised registrars and customers to > use EPP <domain:check> (rather than the Whois) to implement availability > checks.
There aren't certainly use cases where EPP is the right solution here, especially in registry/registrar models where the registrar has a piece of the information influencing availability. > The general problem with services such as the Whois or RDAP in this > context is that these services are only suitable for telling which > domains are definitely *unavailable*, but not which ones are available. > There are plenty of cases in which a <domain:check> correctly yields > avail="false" but Whois or RDAP services return "not found" results, > which are easily misinterpreted as "domain available". There's nothing to do about Whois in this case, but RDAP can handle this easily (hence the extension). > Whois/RDAP are services for obtaining information about existing domains, > not about availability. I believe that any attempt to add availability > check capability to these services violates the "separation of concerns" > principle. Availability is information about the domain, so it falls neatly in the concern. -andy _______________________________________________ regext mailing list regext@ietf.org https://www.ietf.org/mailman/listinfo/regext
