[Geopriv] review comments on draft-thomson-geopriv-lis-discovery-00.txt

http://www.ietf.org/internet-drafts/draft-thomson-geopriv-lis-
discovery-00.txt

Overall, I like this draft very much, and favor its adoption by this
working group. I appreciate James and Martin taking the time to
separate out the LIS discovery mechanisms from their candidate
proposal, HELD. One of the issues I think this working group has is
that many people feel that to accept HELD or RELO or etc, they have
to take the bad with the good. This draft separates the issue of LIS
discovery so that this good idea of HELD may be applied to other
protocols. Such ideas enable compromise and progression of good ideas.

My comments:

1) The term LIS needs to be defined in the terminology section,
especially since there is a reference to RFC 3693 where the terms LS
and Target are defined. This might also help clarify the scope of
the work and reduce some of the heartburn the security folks my have.

2) The introduction may want to expand upon #1 a little more by
specifying the scope of the work to the application of L7-LCP enabled
LIS environments. In other words, if you are already getting your LI
from DHCP or LLDP-MED, this is not applicable.

3) Remove the registration for HELD to make this totally protocol
neutral. Instead, note that whatever protocols do use this must
create an IANA registration for U-NAPTR.

4) Specify the order of the discovery methods. Predictability is a
good thing. I suggest this order: 1)DHCP, 2) U-NAPTR via configured
domain (either static or dynamic), and 3) U-NAPTR via reverse IP.
That's just a strawman proposal, but I think the methods do need to
be ordered.

5) Create an appendix with examples. Examples always help.

6) There needs to be some text around how a client would query its
location over DHCP vs. querying the LIS location over DHCP. In other
words, explain how a client uses both. See #2.

7) Pat yourselves on the back for a job well done.

-andy

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Fri, 16 Mar 2007 07:23:27 -0400