At 12:14 PM 12/13/01 -0500, Howard C. Berkowitz wrote:
>In the IPng effort, there was what I considered a useful exercise in getting the requirements of various mostly user communities, roughly RFC 1667-1680. I think this might be worth redoing, including getting current perceptions of critical ecommerce, 3G wireless, etc. I'd be willing to coordinate a solicitation if that makes sense.
I do not think that this is appropriate for an IRTF work item.
We're trying to put together what _we_ think are the correct
requirements. Going into various communities like that seems to
be more of a task for an hypothetical standardization effort;
thus better suited to the IETF.
>> Requirements For a New Inter-Domain
>> Routing and Addressing Architecture
>
>
>Perhaps something on the lines of "scalable routing architecture for the global Internet" might be less controversial.
Scaling is only one part of the system.
>> 3.2 Separable Components
>> The architecture MUST place different functions into separate
>> components.
...
>> The Internet Routing Architecture should be divided into
>> components, each doing a specified, targeted, job. The
>> Architecture Specification shall define each of these
>> components, their jobs, and their interactions.
>> Some thoughts to consider along these lines are
>>
>> o Making topology and addressing separate subsystems. This
>> may allow highly optimized topology management and
>> discovery without constraining the addressing structure or
>> physical topology in unacceptable ways.
>
>Let me be sure I understand. By topology, do you mean physical topology only, and by addressing, you mean logical? Should identifiers/naming be identified separately here, so we can start the address overloading jihad early? :-)
The requirement text says "Some thoughts to consider". It does
not mandate any such split, but merely offers some thought for
potential architects to consider.
>I assume there is some consideration of VPN addressing later in the document. If not, for a non-specialist reader, a few words might be appropriate.
We presume that the readers are specialists :-)
Frank Kastenholz
This archive was generated by hypermail 2b29 : Mon Aug 04 2003 - 04:10:03 EDT