This raises a meta-question in my mind.
How are we going to prevent this list simply repeating all the discussion
that took place in the NSRG and elsewhere?
The reasons why FQDNs are imperfect EIDs have been listed quite recently
(on one of the IPV6 lists I think).
Brian
Fred Baker wrote:
>
> At 12:07 PM 10/16/2003, James Kempf wrote:
> >An IP address acts in two roles:
> >
> >1) As a node or endpoint identifier to the transport layer for an end to end
> >conversation.
> >
> >2) As a locator for routing packets to a particular topological location.
>
> actually, I will argue that the endpoint identifier is the dns name. That
> has problems in that
>
> - it needs to be able to name a service and get the address of the most
> serviceable instance of a computer offering it to the requestor; it
> currently simply offers *an* address of a list of addresses from which the
> requestor needs to choose at random
>
> - it presumes a global address space (there is no concept of "source
> route to this gateway address and then to that private address", or the
> obvious generalization)
>
> - A list of other things I will think of just after I send this email
>
> But it in fact acts to identify the intended endpoint.
>
> _______________________________________________
> Saad mailing list
> Saad@ietf.org
> https://www1.ietf.org/mailman/listinfo/saad
--
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Brian E Carpenter
Distinguished Engineer, Internet Standards & Technology, IBM
NEW ADDRESS <brc@zurich.ibm.com> PLEASE UPDATE ADDRESS BOOK
_______________________________________________
Saad mailing list
Saad@ietf.org
https://www1.ietf.org/mailman/listinfo/saad
