ECRIT Interim Meeting Date: Jun 03, 2009 Minutes by: R. Marshall Chairs: Marc Linsner & Hannes Tschofenig Wiki page at: http://trac.tools.ietf.org/wg/ecrit/trac/wiki Local start time: 10:06 AM Pacific Agenda: Attendees: James Polk Keith Drage Hannes Tschofenig Greg Burdett Gabor Bajko John Elwell Karl Heinz Randall Gellens Richard Barnes Robert Sparks Roger Marshall Spencer Dawkins Marc Linsner Brian Rosen Cullen Jennings Milan Patel Dan Romascanu Webex Session: Notes of Meeting Discussion: SLIDE 1-4… SLIDE 5 – review the status of drafts… Sos-parameter Lost-sync WGLC is complete, waiting for a couple of implementers feedback since it is an exp Location Hiding Specifying Holes in LoST service boundaries (correction: Cullen to talk to Martin Thomson) (more details are in the link listed in the slide 5) James: Why isn’t there a milestone for RPH here? Hannes: Discussion with AD’s – in short, we need to have a wg rechartering. James: yes, for -00, but I’m on -05 Cullen: not unique circumstance, some drafts are not going to go forward until rechartering happens. James: Other drafts (after) have been brought in. Cullen: I’m a little lost myself on RPH – I didn’t realize that we needed to recharter for RPH. James: I don’t think we need to update the charter. Cullen: IESG will have to approve the recharter. Keith: Which situation are we now in – I believe the wg asked for a milestone addition at Dublin Hannes: RPH is listed as from October, other drafts from June – it was up to Jon and Cullen as to whether these fit within the existing charter, they were considered within the charter. …Skip to info in mail message sent out with milestone update (see SLIDE 9) James: Other drafts seem to be sneeking in. Cullen: What do you mean. We have the charter, then there is the goals & milestones, then there is the Internet drafts. Keith: One of the AD’s (presumably) will give us the determination of whether this draft is within scope, or not. Cullen: Some hesitancy here (on the phone) in light of new AD’s coming in… If Jon said it was within scope, then I’ll go with it, but I want to go back and look at what was decided. Hannes: Recounting some history, when asked, Jon may have agreed to allow additional drafts to be submitted, possibly because the phonebcp & framework draft completion was right around the corner. Hannes: Secondly, there was a milestones listing mistake, which is why I asked to have the milestones updated. Brian: I think that we’re only waiting on one critical issue – the applicability statement – the other issues, I don’t think are critical enough to hold up the document. Next Discussion Item – PhoneBCP and the text added based on SF ECRIT discussion: Hannes: SLIDE 6, Marc asked for consensus, but has noted that there is no clear consensus. James: Just one or two people who don’t want to take it out. Brian: More than just one or two. Brian: It’s a chair call, as to whether there is rough consensus. Hannes: The question in the consensus call that Marc issued was not whether people agreed with the applicability statement, but rather is the present wording acceptable. Randy: I’m saying, ask a different question – we have a draft out there, can that version go forward? Keith: The version with the statement is fine with me, could be clarified, but has a caveat. Hannes: Maybe we should resubmit the document without the statement, Brian: As you have to do is revert. Randy: We should ask whether the current version of the draft can go forward. Hannes: what is the previous version Randy: If we go back, we know from SF that there are problems with the previous version Hannes: You believe there are problems… [skip] Marc: Current version is -09, we could ask the simple question, should we ask do we go forward with -09 or -08? Randy: But, -08 is the same as in SF, and we will get back into the controversy again as in SF. Hannes: We can’t avoid the controversy, Randy: Let’s not call it an applicability statement, it’s just a note of text. Marc: We’ve got ____ people who didn’t like the changes from -08 to -09. Randy: My point was that the question as worded, … Marc: My proposal of 3 minutes ago… Brian: I think I mis-spoke, since there are other changes as well. You’d have to have a more detailed question to differentiate the two. Hannes: The compromise text was premature, done by a small group. Brian: It was done on the list – not done by a small group of people. Cullen: Consensus is something taken at a point in time, and over time, may change – a good thing – since that’s how agreements are reached… it sounds like the chairs are saying, it sounds like we didn’t have consensus at that time. Randy: We had a lot of debate on the mailing list, yes during the ietf meeting, but also in the week following. Randy: I think there was consensus that people weren’t content with -08 Hannes: Marc: (reading the consensus question from email) Marc: that vote, if by plurality was 10 for, ________ against, etc. Hannes: If there is something in the document that prevent ES from working in certain environments, then we should discuss it, though it comes a little late… Keith: It’s not late, these 3GPP documents have been available for a few years. Hannes: 3GPP doesn’t put these kinds of statements (e.g., about IETF shortcomings) in their documents Randy: I think we’ve got some emotional baggage tied in with this. Richard: It doesn’t say, “X is a situation, in which things won’t work” Hannes: (reading the applicability statement) Randy: All it does it to alert the reader James: This is applies to IP networks Keith: Who says that 3GPP is not an IP network Randy: It’s not an applicability statement, doesn’t limit a solution in any way. Most documents in the IETF are self-limiting. This document sets itself up as being “the single way” that all emergency calls happen. Hannes: The primary focus of this document is around access providers providing location, and doesn’t take into consideration other models, such as the communication service providers to provide, etc. Richard: This document allows for other stuff, but just doesn’t map it out… Hannes: There’s a big difference, it doesn’t talk about emergency registration, etc. Keith: You wouldn’t get a 3GPP LoST server, for example… Keith: A Service Provider implementing the 3GPP standard, wouldn’t implement a LoST server, because it doesn’t mention one. Marc: (voice breaking up) Richard: What do change in the document to eliminate the notion that other models can exist? Keith: To a certain extent, some have suggested that tried that, Richard: If I recall, Stephen (Edge’s) comments were fairly general, not specifically pointed changed suggested. Do you think this kind of thing can be done? Keith: I’d have to go back through it. Randy: I’d have to do another detailed read of the document. Spencer: I think the thread indicated that… 3GPP Keith: Sometimes CT1 vs. SA2 need pointed questions asked. Hannes: We’ve had many conversations with 3GPP, it depends on which questions are asked… which group. Brian: 3GPP… Hannes: I think it’s even worse than that, in that some of the communication models are very different. E.g., the split of services vs. being provided by the same provider. Hannes: What is different to the enterprise environment is the interaction between roaming and local… Brian: We should be asking, Is it all right that the service provider queries LoST, if it’s not all right… Keith: not the question – the question I am raising, is that if different addressing info is used in a 3GPP network, since 3GPP looks for specific call marking, and the call may not get to the PSAP at all. Brian: The main difference is due to the location being done by the UA or not. Roger: It’s a difference between “core” vs. “edge” routing based on location. Marc: If we exclude all other standards development organizations out of this, then it’s not overarching. Keith: I’d have to go back and look at the document. Marc: Can you look at it within a week? Keith: Other ietf documents are explicit as to what they apply to… Randy: This statement is helpful, is not harmful to have it here. Hannes: But if it’s not a big deal, why do you care so much about this text? Keith: All docs have a scope, what does the document do, and what is the document applicable to. Hannes: Most of the documents are written in such a way as to be generic building blocks… maybe I’m wrong about this assumption, but if what you point out the UA LoST lookup problem is not adequately covered, maybe we should read through again, and makes sure the document says what it needs to say. My opinion is that this is what the document says… Brian: The underlying protocols support either model where a UA gets location & queries LoST or a Proxy gets location & queries LoST, but the phonebcp doesn’t really talk about the latter, other than in a fail-case scenario. Roger: Phonebcp is really just a profile for edge-based routing based on location, there is no comparable profile document that describes a proxy-inserted location & routing model architecture. Brian: Yes, but the underlying protocols support it. Roger: yes. Marc: Doesn’t the title PhoneBCP suggest the edge context? Brian: As to the applicability statement, I don’t think this wording really matters. Keith: I’ve already referred to wording in the abstract that gives the impression of a universal solution. [skip] Hannes: Randy: Would like the question asked – are people happy with the existing document? Marc: If we could have a -10 w/o the statement, then ask do you (wg) want -09 or -10, would that work? Randy: No, I don’t think so. Randy: Ask, given that other documents are being held up, is the present document acceptable? Cullen: you can’t leave it binary then, it either needs… James: Ask the question, do you want something, but not what’s in -09? John: I think that James’ suggestion could break the (apparent) deadlock. Richard: suggest a rev -10 w/o text – two independent consensus calls. Randy: There is a difference between a preference and a strong objection. Brian: ok, suggest asking the question, who is strongly opposed to -09 and who is strongly opposed to -10? James: does that take into consideration whether people want something, but something different? Brian: no, James: right, then it becomes a two-step process. James: I strongly oppose the suggestions. Brian: you need to start with something non-prejudiced. Marc: I will discuss with the AD’s. Summary: NO CLEAR CONSENSUS REACHED /end of mtg. at 11:48 AM Pacific