GGF NM WG Notes 02/01/05
Call began at 11:00 am EDT.

Attendees:
Mark Leese 
Jim Ferguson
Jeff Boote
Eric Boyd
Susan Evett (scribe)
Martin Swany
Jason Buraski (UDel)
Dan Gunter 
Tanya Brethour

AGENDA
Agenda:

1. Continue discussing the new/"normalised" schemas:
 + progress from Dan
 + any new examples (e.g. Iperf example from Mark)
2. Review Mark and Paul's progress in stabilizing the current schemas

Mark introduced the document he’d sent around just before the call -- he’d like the group 
to discuss how to represent traceroutes in the new schemas. Paul and Mark looked at the 
schema and could only find ‘hop count’ – in the old schema, all you could record was the 
number of hops for traceroute. Paul and Mark had an idea of how to represent ‘hopless’ 
count to represent the route of a traceroute. If you have a set of data, say RTT, for a series 
of hops along a path, you’d like to be able to put together a set of hop data together. (See 
page 4 in the document.) Mark asked the group to take a quick look and, if there weren’t 
any major problems, he’d take it away and figure out how to fit it into the revised 
schema.

Dan asked Mark if he knew how this affects the example in the Developer’s Guide (page 
7); also, how would this look if two of the hops had the same TTL? Mark said they’d 
thought about it, briefly, but hadn’t gotten into the nitty gritty as yet. (The group 
discussed, briefly, the need for some terminology clarifications – perhaps need a glossary 
of terms?) Mark thought they should stretch it to fit the example in the DG. He felt that 
the only real difference between the revised schema and the examples in the DG is where 
it relates to traceroute – he felt you could report back the values of hops A-B, B-C, C-D, 
etc. You could represent it with multiple meta-data inline.

Mark asked how the revised schema looked from the Advisor PoV; Tanya said she’d be 
happy to implement whatever the group decided in the Advisor and report back on how it 
worked. Martin felt the timestamp issue was still open; Paul was going to put in an 
attribute but didn’t get that in before he left, so Mark will put one in.

Mark asked Dan, in the response schema, if he wanted empty results returned so that he’d 
have record of no response (instead of no record that data was requested); Dan said, yes, 
he thought that would be useful for developers. He also wants the units recorded, even if 
they are the mean; Dan noted that, sometimes, the units can change so having the data 
reported allows this to be picked up.

Jeff commented on an item on page 6 – reports that ‘available bandwidth’ vs. ‘achievable 
bandwidth’ is there. Jeff wanted him to verify that both are available; if not, change to 
‘achievable’.  Mark said both are available, he just misstated it; he’ll change it in the 
document. (1)

Eric mentioned that the next meeting is at SLC; Mark will be presenting a final version of 
the schemas (first draft for formal approval), Martin and Dan will be presenting current 
state of the schemas, meeting is 2-6 pm on Wednesday 2/16 (possible segway into 
dinner). Milestones: formally have a NM WG schema pair, plan for going forward, early 
adopters of NGen schemas and some prototypes. Mark wanted to talk about how to get 
the material documented and publicized; do we want a GGF info document or just 
available from the web pages.

ACTION ITEM:
1.	Mark to change ‘available bandwidth’ to ‘achievable bandwidth’ on page 6.
2.	adsfkasdjf
3.	asdfasd


Next call is scheduled for ??, same time.  Eric will setup the call and send around the 
dialup information.
Call ended at 11:18 p.m.