Re: [Enum] IETF, ENUM and NGN

[John Curran <jcurran at istaff.org>]

At 10:05 AM +0100 6/30/04, Jim Reid wrote:
>AFAICT nobody is suggesting Infrastructure and User ENUM would use the
>same tree. The constraints and data models for these things are
>completely different. However some people seem to think that if these
>two flavours of ENUM use the same domain name, this means they both
>have to use the same tree with one model for ownership, administration
>and delegation. That doesn't necessarily follow. What some operator
>does with its private, internal-use-only version of e164.arpa and how
>they populate and manage that is nobody else's concern, apart from the
>usual regulatory issues like universal service obligations, fair
>competition and so on.
>
>What matters is the operators agree on a domain name for
>Infrastructure ENUM. This might as well be e164.arpa since (a) it's
>vendor neutral; (b) already an agreed IETF standard; (c) not
>controlled by any nation. It's hard to see how any other choice of
>domain name could be better (technically or politically).

Hmm...  If some operators really and truly intend to run a data tree
which is not publicly visible, than I see no need for any standard in
this area.  The operator (or set of operators which share such a data
tree) can simply configure an appropriate domain of their choice.

I don't believe that there needs to be a distinct Infstrastructure ENUM
and User ENUM data tree.  The only strong reason given for separate
models is privacy and I'll assert that it is possible to put information in
the tree for a given E.164 number that consists of nothing more than
a record which directs one to the appropriate service provider (as noted
by Mike Hammer earlier) and that this will indeed prove sufficient to
provide connectivity without raising privacy concerns.

The operator community should be encouraged to put some nominal
information for their number blocks in the Public ENUM tree, even if this
is nothing more than a single wildcard record pointing to a default server
for that operator.  As numbers port, it should be very straightforward
to insert an appropriate specific DNS record to direct queries to the
new provider as appropriate, including the degenerate case where a
knowledgeable user (or user with knowledgeable party under contract)
arranges to have richer connectivity information put in place under their
entry.

/John


_______________________________________________
enum mailing list
enum at ietf.org
https://www1.ietf.org/mailman/listinfo/enum