PIM WG 

Agenda bits..

WG status update..
..things were said. (see slides)

mtid draft.
  Dimitri - new version has been submitted. Issue new version with rewording
  on its way. Incorporate more specific text regarding the parsing of the
  messages. And other work required..

LC pim registry draft. No response. 
	Toerless - I provided feedback
	LC back to the list again.

4601 revising?
	following slide

	Stig -
	Is there any part of the spec that is non-interoperable?
        We would eed an implementation report.

	Adrian - don't be afraid of going to draft standard. New rfc showing
        what an implementation should look like. Most of the errata are
	relatively small. Only a few days work.

pim-port 
	Stig:
	WG last call. long list of editorial changes. Some more substantial
         feedback from Dimitri. Editorial issues have been fixed.
	Some additional state but less message processing. 

	What should be changed in the draft?

	Dimitri:
	Clarify wording around reduced control traffic

	Shep:
	It's in the introduction.

	Dino:
	Can you offer text? (to Dimitri)

	Ice:
	Incremental updates are a side-affect of reliability and not the goal.

	Stig:
	Still not clear what change is needed to clarify this.
	
	Stig;
	Issue 4 slide
	What happens when we get a message with unknown type?

	Dimitri:
	Instance ID represents an instance of what?

	Stig:
	Instance ID types will need to be defined within documents

	Stig:
	..trying to describe the start-up behavior - start in PORT or wait
        until the TCP connection comes up
	..describing other outstanding issues w/ no comments

	Name Unknown
	Most people today are using BFD (keep alives)

	Stig:
	If we determine a need for keepalives then we'll need to define this.
        Anyone think there i is a need for TCP keepalives? (one 'yes' in the
	room)

PIM-Neighbor-reduction - Ice

	Ice - going over slides..

	Dino:
	Maybe PORT can be combined with this

End