> p.s. rather than adding more and more burdens to DNS, what we really > need to be doing is figuring out how to replace it with something more > robust and more flexible. (Yes, you'd have to arrange that DNS queries > and queries to the new database would return consistent results; you'd > also have to make sure that DNSSEC didn't break, but those are both > doable.) > > DNS is getting very long in the tooth, and is entirely too inflexible > and too fragile. The very fact that we're having a discussion about > whether it makes more sense to add a new RR type or use TXT records with > DKIM is a clear indicator that something seriously is wrong with DNS. > Adding a new RR type should not require a single line of DNS server or > client library code to be recompiled, nor any changes to the > configuration of any server not advertising such records. Hear, hear! The DNS features that are in operation and are required for a functioning domain naming service, are a subset of the DNS as defined way back in the days of that small research network with high hopes. Today's DNS is an absolutely vital and critical part of the world's telecommunication network. An architectural component in that role should be simple and robust. We need to create a new version of DNS that removes the cruft, matches what is actually used operationally, and is more robust, i.e. less reliant on distributed servers for its resilience. While I wish that the proponents of adding new RRs would notice that LDAP can do the same job without any need for registering a new schema with the IETF, I understand that people want to leverage a successful protocol. Therefore, I think that the IETF would be wise to define DDDBP (DNS Distributed Database Protocol) using the same basic DNS protocol but running on a different port number. Divert all the DNS enhancement work to DDDBP unless it clearly addresses a need for the domain naming service infrastructure. --Michael Dillon _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf