Towards robust domain names
From W3C Wiki
(See the meeting report for background to this)
Existing sub-domains of
.arpa and their uses
needs filled in
Some reflections by HST
What is a domain name bound to
What I said at the meeting was that my (mis)interpretation of the
example.arpa idea was that publication of an RFC for e.g.
dx.doi.org.arpa established the binding for
dx.doi.org as a gold-plated domain, but left resolution to the existing DNS system. But there's a hidden flaw/gap in that story, I think I realised as I thought further about this this morning.
For naming systems such as the RFC system, and the binomial taxonomic system, what names are bound to is pretty clear. But for a persistent domain name, I realised it's not clear, at least not to me. It certainly can't be bound to an IP address, because those are not permanent. . .
Borrowing from something Larry Masinter has said, is it right to think of the binding as a (description of a) service? That's pretty much right for
dx.doi.org, where the service is a resolution service for DOIs. Would it be right for e.g.
w3.org? In that case would the service be a resolution service for TR pages?
Robust domains and DNS
example.arpa as a new, robust, domain name, and therefore expecting the DNS system to handle the normal volume of lookup requests, is something we will need to explore very carefully. The IAB may well have concerns about this. . .
(Add your thoughts above in a section with your initials)
Back to Workshop wiki front page