[lisp] Ceasefire in the namespace & EID used as RLOC skirmish

[Robin Whittle <rw at firstpr.com.au>]

Short version:         No more LISP messages from me for a week.


At:  http://www.firstpr.com.au/ip/ivip/namespace/#20090319

are links to the 15 messages I have written to this list in the
last week trying to get substantial responses to these concerns:

  1 - Claims that LISP involves separate namespaces for EIDs and
      RLOCs are not true for any version of LISP which could be
      a practical solution to the routing scaling problem.

  2 - Claims that an address could be used as both an EID and as
      an RLOC, in contradiction of this being ruled out by
      http://tools.ietf.org/html/draft-farinacci-lisp-12#page-8

        "EIDs MUST NOT be used as LISP RLOCs."

  3 - My attempt to improve the draft Charter in a number of respects:

        http://www.ietf.org/mail-archive/web/lisp/current/msg00286.html

        http://www.ietf.org/mail-archive/web/lisp/current/msg00298.html
        http://www.ietf.org/mail-archive/web/lisp/current/msg00309.html
        http://www.ietf.org/mail-archive/web/lisp/current/msg00335.html

This also links to other people's messages in these threads.


While Sam initially liked some of my suggested changes, he did not use
any of them.  His own replacement text gained criticism and no support,
I think.  (Links to the versions and suggestions are in:
http://www.ietf.org/mail-archive/web/lisp/current/msg00370.html)

While I may have prompted Dow into suggesting changes which deal more
clearly with the question of namespaces, I have so far failed to prompt
anyone who has made statements, written presentations etc. contrary to
1 and 2 above to argue their  case, present a detailed practical
example etc. in any substantial way.

These three questions are clearly directly relevant to LISP development
and to the final text in the Charter, since we need to distinguish LISP
from other protocols, primarily HIP, which have a stronger claim to being
a true "Locator Identifier Split" architecture, but which cannot be a
practical solution to the routing scaling problem precisely *because*
they involve the use of a new namespace for EIDs and/or RLOCs.

No-one has argued why these questions are not important to LISP in
general or to the wording of the Charter.

Yet, so far, no-one has responded in a substantial manner.


I will concentrate on paying work for a while and will maintain
radio silence for a week.

  - Robin