I can see the point. A subsection of the introduction entitled "But why
DNS?", and listing out your points plus a few more (like "DNS already
exists and is ubiquitous", "it distributes in a global manner", "like
wow, it works well"). A bit of a paeon to Vixie's serendipitous insight
at a time when few would have thought of it.
FYI, it wasn't Vixie. He asked me to submit an erratum to 5782:
Eric Ziegast, who was a system administrator for Vixie Enterprises
where the RBL was hosted, created a DNS-based distribution scheme
that quickly became more popular than the original BGP distribution.
I agree that this isn't a great place for a justification of DNS
distribution of BL/WL beyond the obvious fact that we've done it that
way since 1997, and there isn't anything else available off the shelf
that will do a lookup in a single UDP round trip and has widely
deployed caches.
I've been doing some simulations of DNSBL cache behavior, to compare
conventional record-per-IP DNSBLs with other approaches. That may
evolve into an I-D and an RFC, so the merits or lack thereof of
various publication techniques could go in there.
R's,
John
PS: "paean"
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
http://www.irtf.org/mailman/listinfo/asrg