On Mar 5, 2005, at 7:39 AM, Florian Weimer wrote:
I think this summary is misleading, given that the document is silent on a very important issue: standards which attach special semantics to parts of the DNS tree. By "special semantics" I mean special stub resolver or resolver which is triggered by the appearance of some magic labels or domain names, indepedent of actual zone data. (An extreme example: "Resolvers MUST return the address 192.0.2.1 for IN A queries of the name www.example.com.")
This is extremely annoying for those who use Internet technology on private networks. Publishing such a document and not addressing this class of mistakes could mean that the IAB even endorses such poor standards.
I see this document as addressing applications (or more generally, automata) that assign special meaning to DNS names where those meanings are not specified by the standards. It's intended for implementors.
To the extent that standards should not assign special meanings to DNS names (or subtrees) I suspect that belongs in a different document.
Keith
_______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf