Re: [e2md] Problem statement, Requirements and Use Cases

As Penn wisely points out, this can be discussed in detail in the appropriate forum at the appropriate time, however, ....

The ITUs OID looks like a good option.  It seems to meet the following requirements/advantages:

1) It already exists; no need to invent and debate a new scheme, approach, documentation set, etc.
2) It is hierarchical; organizations can optionally create sub-OIDs (or sub trees) under their main OID if they need to.
3) The sub-identifiers under the main identifier can be opaque; they need not have a known, or the same known, meaning to all parties that may use/see them
4) There is an existing urn that _might_ be useful "urn:oid"
5) If the set of existing OIDs already allocated under one of the existing OID trees (e.g. 1.3.6.1.4) are not appropriate then it seems that an organization (e.g. IANA) can be formally delegated the responsibility of dolling out (similar to the way that it doles out Enterprise Numbers) the OIDs under a new OID tree.  IOW, the OID tree it centrally managed, but the management of sub-trees can be delegated.
6) They are cleanly parseable by a computer or a human; their form is standardized and well structured.

A disadvantage, however, is that ITU OIDs can theoretically be fairly long (in contrast to the 4 characters that comprise a north american OCN or SPID, or the handful of digits that comprise the IANA Enterprise Number).  But it is not clear to me whether in practice, given our use case, if they will ever be as long as they can theoretically be.  The spec for our usage would just need to address this point.

http://www.itu.int/ITU-T/asn1/uuid.html
http://www.oid-info.com/#oid

Ken

-----Original Message-----
From: e2md-bounces@ietf.org [mailto:e2md-bounces@ietf.org] On Behalf Of Tony Rutkowski
Sent: Wednesday, May 19, 2010 11:52 AM
To: PFAUTZ, PENN L (ATTCORP)
Cc: E.164 To MetaData BOF discussion list
Subject: Re: [e2md] Problem statement, Requirements and Use Cases

On 5/19/2010 11:03 AM, PFAUTZ, PENN L (ATTCORP) wrote:
> Use RFC 3761 technology provide for the Carrier of Record for an E.164
> number (as defined in RFC 5067) to publish the mapping of
> that E.164 to a globally defined identifier, specifically an IANA
> Enterprise Number as defined in RFC 2578.
>

Hi Penn,

Good choice.  However, the text needs tweaking to read:

   Use RFC 3761 technology to provide for the Carrier of
   Record for an E.164 number (as defined in RFC 5067) to
   publish the mapping of that E.164 number using an assigned
   Carrier of Record Object Identifier (OID)) such as an IANA
   Enterprise Number as defined in RFC 2578.  Ref. Rec.
   ITU-T Recs. X.660 and X.680.

The changes clarify what identifier is being specified and
includes reference to the authoritative OID standard.  It
also allows for the possibility of using other assigned OIDs
for a Carrier of Record.  Although more than 35,000 Enterprise
Numbers have been registered with IANA, there are many
carriers who have OIDs that were registered with national
authorities for Carrier of Record purposes.  As long as there
is a valid, globally unique OID, some flexibility should exist.

For example, ATT has the OID 1.3.6.1.4.74 from IANA's
OID block, but it also has 0.7.7223 that it uses for some
services in Argentina.

best,
tony
_______________________________________________
e2md mailing list
e2md@ietf.org
https://www.ietf.org/mailman/listinfo/e2md

This e-mail message is for the sole use of the intended recipient(s)and may
contain confidential and privileged information of Transaction Network Services.
Any unauthorised review, use, disclosure or distribution is prohibited. If you
are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.