In article <20161207151238.gc27...@mx2.yitter.info> you write: >On Wed, Dec 07, 2016 at 09:56:11AM -0500, John R Levine wrote: >> I was thinking that you're the RDAP server for .FOO and someone who's >> screwed up his bootstrap asks about BLAH.BAR, with whom you have no >> connection. > >I will just point out that this is the _exact_ reason some of us >thought the bootstrap mechanism should have been SRV records in the >DNS, because it would have neatly solved that exact problem.
Even though I was as enthusiastic about SRV as you were, in this regard it's no better than the existing JSON kludge. If the client is working properly, it'll find the right server. If it isn't, it won't, and we all know that trying to add complexity to make stuff more fool-resistant is counterproductive. And in fact, the current kludge is slightly more flexible since it can have separate entries for .uk and .ac.uk which would be a challenge for SRV. >... In particular, for policy reasons it's important to >understand "this domain isn't available because someone has it", "this >domain isn't available because someone has something that prevents it >being registered", and "this domain isn't available to anyone for >policy reasons." Consider people doing compliance checks, who maybe >shouldn't have access to the SRS directly and who should only have >access to the RDDS. They still need to be able to see these >distinctions. I wouldn't be opposed in principle to returning an optional JSON blob along with the 404 that gives the client some hints along those lines. But it really is optional, and an empty response is still 100% compliant. R's, John _______________________________________________ regext mailing list regext@ietf.org https://www.ietf.org/mailman/listinfo/regext
