Stephane, On Dec 3, 2013, at 2:10 PM, Stephane Bortzmeyer <bortzme...@nic.fr> wrote: >> The issue here is that they are, in fact, using the DNS in the sense >> that they are using applications that expect to query a local DNS >> stub resolver > > No, no, no. Few applications query " a local DNS stub resolver" (dig > is an exception). Most applications query a local name resolution > service, through a library routine such as getaddrinfo.
Yes. Mea culpa. As I've already explained a couple of times in private email (wow, people actually read my email??), I actually meant "local stub resolver" (no DNS) and: "My point was exactly that the local stub resolver (that is, the library on the system that issues to name resolution requests, not the part of the name resolution system that does the work, be it DNS, mDNS, YP, or whatever) will be expected to distinguish between real DNS names and these pseudo-DNS names. This implies every time the IETF makes use of 6761, those tables of special labels is going to need to be updated (and I suspect the chances of this being done universally and consistently approach zero). This means an application that will work on one system (that has update the table) won't work on another (because the table hasn't been updated). I consider this sub-optimal." Haven't we been here before (e.g., .bitnet/.csnet/.uucp)? Ignoring that, other than aesthetics, what is the downside of <p2p>.alt or <p2p>.not-dns or <p2p>.arpa again? Regards, -drc
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
