---------------------------------------------------------------------- Date: 30 Aug 1996 18:19:30 -0700 From: johnl@iecc.com (John R Levine) Subject: Re: New RFCs re incremental DNS update >Remember that the protocol between the registries and the root server >for the domain doesn't have to be the same as between all other name >servers -- they are, after all, in a special relationship. You bet. That's why I've been suggested a conventional database in the middle so we don't have to reinvent a fairly large transaction system wheel to get something usable. - -- John R. Levine, IECC, POB 640 Trumansburg NY 14886 +1 607 387 6869 johnl@iecc.com "Space aliens are stealing American jobs." - Stanford econ prof ---------------------------------------------------------------------- Date: 30 Aug 1996 20:01:26 -0700 From: "Dave Collier-Brown" Subject: Re: New RFCs re incremental DNS update John R. Levine saith: [in reference to incremental zone transfer] |You bet. That's why I've been suggested a conventional database in the |middle so we don't have to reinvent a fairly large transaction system wheel |to get something usable. I suspect we should proceed with prototypes of several designs, and, given one or more which meet our requirements, implement not less than two. I would also comment that this, as well as both the heavyweight and lighweight registry proposals (but possibly not the chaotic) really amount to different positioning of the committ mechanism in a distributed data management system. They're all equivalent, and I suspect one of my academic colleagues might be able to show the chaotic proposal is equivalent too. - --dave
