IETF
dnssdext@jabber.ietf.org
Wednesday, July 31, 2013< ^ >
Room Configuration
Room Occupants

GMT+0
[06:12:11] spencer.shepler joins the room
[06:12:16] spencer.shepler leaves the room
[06:56:10] Tim Wicinski joins the room
[07:01:20] mcr joins the room
[07:01:44] danyork joins the room
[07:02:09] danyork has set the subject to: DNS-SD Extensions BOF - IETF 87
[07:02:33] yone joins the room
[07:02:37] <danyork> I'll jabber scribe
[07:02:45] <danyork> mcr - I assume you're remote, eh?
[07:03:07] <mcr> yes, I am remote. I'll mic:
[07:03:38] <danyork> great
[07:05:49] <danyork> Currently in the chair's slides - http://www.ietf.org/proceedings/87/slides/slides-87-dnssdext-2.pptx
[07:06:20] <danyork> Slide 5 - Drafts
[07:06:32] <danyork> key is requirements draft
[07:06:47] <danyork> That draft is http://tools.ietf.org/html/draft-lynn-mdnsext-requirements-02
[07:07:23] <danyork> Tim Chown is speaking
[07:07:59] <danyork> Slide 6 - dnssdext - mail list
[07:10:03] <danyork> Ralph Droms explaining why the "mdnsext" name was kept
[07:10:10] <danyork> Slide 7 - RFC 5434 - reminder
[07:10:58] <danyork> Ralph walking through the goals of the BOF
[07:11:52] Ning Kong joins the room
[07:12:39] <danyork> Slide 8 - Presentation
[07:13:04] <danyork> Summary: Stuart Cheshire is going to walk through the requirements - and then there will be a discussion about forming a WG
[07:13:30] jaap joins the room
[07:13:48] g.e.montenegro joins the room
[07:14:02] Ted Lemon joins the room
[07:14:07] <danyork> Can you hear Stuart?
[07:14:13] <danyork> Slides: http://www.ietf.org/proceedings/87/slides/slides-87-dnssdext-3.pptx
[07:14:14] cabo joins the room
[07:14:15] ctg1701 joins the room
[07:14:27] <danyork> Slide 2: Goals
[07:15:24] <mcr> I can hear Stuart.
[07:15:27] <danyork> Thanks
[07:15:38] <danyork> Audio stream is at: http://ietf87streaming.dnsalias.net/ietf/ietf876.m3u
[07:15:41] <mcr> I wish people would upload pdf, it's so much easier to figure out where I am in a browser... :-)
[07:15:57] <danyork> Slide 3 - Usability
[07:16:04] <Ted Lemon> Do you want me to convert it for you, Michael?
[07:16:11] <danyork> mcr: Me, too.  As a scribe working with PPTs is a serious pain in the neck
[07:16:23] <danyork> Slide 4 - change since draft-01
[07:16:29] <danyork> Slide 5 - Initial Experiences
[07:17:40] John Jason Brzozowski joins the room
[07:17:45] <danyork> Stuart relaying experience that iPads are being used on campuses but native solution is limited to single subnet.
[07:18:15] <danyork> Stuart - Collobos came out with solution that implements part of the Hybrid proposal
[07:18:55] <danyork> Stuart - "Presto Print Server" translates from URF/IPP to printer-native protocol
[07:19:26] <danyork> Slide 6 - IT Dept Cooperation
[07:19:52] <danyork> Stuart - IT group that supports iPads is different from group that supports DNS
[07:20:02] <danyork> Slide 7
[07:20:14] <danyork> Slide 8 - Recursive DNS Server Choices
[07:20:53] Andrew Sullivan joins the room
[07:21:16] <danyork> Summary - it works find with Unbound or BIND, but doesn't work with Windows Server 2008 DNS
[07:21:34] kerlyn joins the room
[07:21:57] joel jaeggli joins the room
[07:21:57] <danyork> Stuart - sites had to choose between iPad printer support and using Windows Server 2008 DNS - some chose to implement Unbound, some didn't
[07:22:11] <Andrew Sullivan> Why hope those bugs will be fixed?
[07:22:37] <Andrew Sullivan> we've had more than 10 years of problems with the unknown RRTYPE support in many of these products, and the problematic one here is one of the best known examples
[07:23:17] <kerlyn> Sol'n: bind9
[07:23:18] Eliot Lear joins the room
[07:23:23] matthijs joins the room
[07:23:36] <danyork> Slide 10 - REQ1
[07:24:10] john.levine joins the room
[07:24:11] <danyork> (that was me at the mic asking if the solution is really not to fix Windows 2008 DNS but instead to move to Windows Server 2012 DNS - or to Unbound or BIND)
[07:24:19] <danyork> Slide 11 - REQ2
[07:24:50] <danyork> If anyone remote wants a mic relay, please preface with "mic:"
[07:25:02] <danyork> Eliott Lear at mic
[07:25:34] <danyork> Eliot - not sure what is explained here matches the text of the draft
[07:25:40] ctg1701 leaves the room
[07:26:45] Dan Wing joins the room
[07:27:10] <danyork> Stuart is clarifying
[07:27:14] <danyork> Ted Lemon at mic
[07:27:34] <danyork> Ted - the tree could also be a graph or mesh
[07:27:44] <danyork> Harald Albrecht at mic
[07:28:13] <danyork> Harald - there is additional knowledge about B and C that is not written down
[07:28:16] <kerlyn> case C is to cover homenet, as I recall, so intentionally general
[07:28:30] <danyork> kerlyn: do you want a relay?
[07:28:54] <danyork> Harald - "How local is local?"
[07:28:57] <kerlyn> sorry, don't know what that means :-/
[07:29:12] <danyork> kerlyn: sorry... do you want me to relay that to the mic?
[07:29:16] <kerlyn> sure
[07:30:08] <danyork> Ray Bellis at mic
[07:30:26] <danyork> Ray - +1 to arbitrary topology
[07:30:39] <danyork> (?) at mic
[07:30:53] <danyork> (?) - doesn't want constraints on C
[07:31:09] <cabo> Toerless Eckert was at mic
[07:31:13] <danyork> thank you
[07:31:31] fneves joins the room
[07:31:41] <kerlyn> is that Toerless at the mic now?
[07:31:47] <danyork> Toerless - doesn't like the word "home"
[07:31:49] <Ted Lemon> yes
[07:31:51] <danyork> kerlyn: yes
[07:31:53] <kerlyn> thx
[07:32:15] <kerlyn> +1 on controlled flooding in some situations
[07:32:40] Kent Shuey joins the room
[07:32:44] <mcr> mic: the Starbucks is a variation of network B, and it needs to be in scope, because it's a case of a network with hostile peers.
[07:33:48] <mcr> mic: (and the homenet may have hostile teenagers on some links)
[07:34:05] <danyork> Toerless - does name change of BOF affect what we do here - or we shouldn't let the change of name do so
[07:34:29] <danyork> mcr: I am in line
[07:35:26] <danyork> (?) at mic
[07:35:39] <danyork> (?) - naming needs to be precise
[07:35:49] <mcr> mic: the point isn't that the Starbucks hasn't got an IT guy.  The point is that their IT guy didn't get to configure my laptop, and I still want to find your laptop while there, and I want my tablet to find my laptop.
[07:35:50] <danyork> Stuart - naming on slide was not normative, just an example
[07:36:12] <danyork> John Jason Brzowski at mic
[07:36:33] <danyork> JJB - will the text allow for a range of topologies?
[07:36:54] Chris Griffiths joins the room
[07:37:06] <danyork> i am next
[07:37:37] mcr is now known as Michael Richardson
[07:37:37] Michael Richardson is now known as mcr
[07:37:39] bhoeneis joins the room
[07:38:35] <danyork> stuart responding that Starbucks may in fact firewall customers from each other
[07:38:42] <danyork> Slide 12 - REQ3
[07:39:19] <danyork> Peter Koch at mic
[07:40:08] <danyork> Peter - as a network grows, it becomes more multi-dimensional
[07:40:46] <danyork> Peter isn't sure there are clear-cut boundaries between D and E
[07:41:42] mcr is now known as Michael Richardson
[07:42:00] <danyork> Erik Kline at mic
[07:42:39] <danyork> Erik - should we support one discovery scope or multiple scopes?
[07:43:35] <danyork> Stuart - multiple discovery scopes
[07:43:39] <danyork> Russ Mundy at mic
[07:44:22] markus.isomaki joins the room
[07:44:22] markus.isomaki leaves the room
[07:44:35] <danyork> Russ suggests that the likelihood of having separate administrative domains in D is at least as high as E
[07:45:07] <danyork> Stuart - I don't distinguish between D and E myself - we wanted to make it clear in document that both cases were being covered
[07:45:12] AndersBrandt joins the room
[07:45:15] <danyork> Russ at mic again
[07:45:54] <danyork> Russ - what is the degree of certainty that is desired for correctness in the requirements?
[07:46:28] <danyork> Stuart - printing on the wrong printer at home may not be a big deal, but it may be at a university
[07:46:41] <danyork> Ralph Droms - there are different kinds of administrative domains
[07:46:44] Michael Richardson is now known as mcr
[07:46:44] mcr is now known as Michael Richardson
[07:47:03] <danyork> Toerless Eckert at mic
[07:47:15] <danyork> Toerless - requirements are okay but way too weak
[07:47:42] <danyork> Toerless - we need a way to specify scope in service announcements
[07:48:01] <Ted Lemon> Wow.   Furious IPv6 nginx debugging session later: http://www.fugue.com/dnssdext.pdf
[07:48:33] <danyork> Tim Chown speaking
[07:48:39] markus.isomaki joins the room
[07:49:30] <danyork> Ted - where is that set of slides being presented?
[07:49:43] <danyork> (it was interesting to read)
[07:49:45] <kerlyn> "zone" is perhaps a better term
[07:49:55] <Ted Lemon> Argh.   Sorry.
[07:50:04] <Ted Lemon> That was from the language session on Saturday.
[07:50:15] <danyork> Ah... it was interesting, though. :-)
[07:50:23] <Ted Lemon> Sigh.
[07:50:33] <kerlyn> ted: http://www.ietf.org/proceedings/87/slides/slides-87-dnssdext-3.pptx
[07:51:06] <danyork> Toerless still at mic
[07:51:32] <danyork> Yi Yang at mic
[07:51:38] <Ted Lemon> Okay, same URL, different slides.
[07:51:45] <kerlyn> relay: at the *end* of this segment, please ask stuart to clarify that Apple's IPR statement was meant to apply to his hybrid proposal and *not* the requirements draft
[07:51:45] Michael Richardson is now known as mcr
[07:52:17] <danyork> kerlyn: okay, will relay
[07:52:22] <kerlyn> thx
[07:52:33] <Eliot Lear> Kery, is the declaration not clear that it covers his hybrid proposal?  The requirements one is sort of obvious.  How can you have IPR on requirements?
[07:52:58] Suz joins the room
[07:53:22] <kerlyn> I believe Apple's legal dept just named the wrong draft; but stuart can clarify
[07:53:35] <danyork> yi - discovery scope may change but IP address may stay the same
[07:53:40] <danyork> Matthew Gast at mic
[07:53:59] <danyork> Matthew - differentiation between D and E is really about the complexity of the IT organization
[07:54:07] ayourtch joins the room
[07:54:38] <danyork> Matthew - use cases seem to confuse complexity of the IT organization with the market they are in
[07:54:40] <kerlyn> +1: I see these use cases as spanning a continuum (?) of IT support as well as security requirements
[07:55:35] <kerlyn> security != simple (as least not yet)
[07:55:45] <danyork> Dave Thaler at mic
[07:55:52] <kerlyn> *at
[07:56:11] Markus Stenberg joins the room
[07:56:20] <danyork> Dave - there are several things here that can be split out as different questions that *could* be requirements
[07:56:26] <Markus Stenberg> can someone make these comments stop? small-medium-large networks, administrated-not
[07:56:38] <Markus Stenberg> I think Stuart's presentation is interest but this has now ratholed half an hour
[07:56:47] mcr is now known as Michael Richardson
[07:56:55] <kerlyn> tell the chairs...
[07:57:02] <danyork> Dave - is the requirement that there be an arbitrary levels of scope?
[07:57:02] <Markus Stenberg> I was hoping they were here ;)
[07:57:23] <Eliot Lear> markus, i was concerned about that as well, but then I see we have another 1.5 hours and only 2 or 3 more requirements
[07:57:57] <danyork> Dave - is the scope topological?
[07:57:58] <Andrew Sullivan> "BoF discussion expands to fill the time alloted"
[07:58:49] <Andrew Sullivan> Fortunately, like usual, Dave Thaler is making a bunch of useful distinctions that can be used to stop hand-waving.
[07:59:18] <danyork> Dave - can the entity who configures the network set the scope? is the person offering the service able to configure the scope? can the person configuring the resolution able to set the scope?
[07:59:24] <kerlyn> Real goal is the WG formation question.  Most of this could be done on the mailing list.  Seems like some are just hearing these req's for the first time and discussing paint colors...
[08:00:10] Wendy Seltzer joins the room
[08:00:39] <Andrew Sullivan> that was not pain colour, though.  That was "here are the requirements you're _really_ looking for", and it goes directly to the question of whether a WG is reasonable to form
[08:00:45] <danyork> Peter Koch at mic
[08:01:08] bhoeneis leaves the room
[08:01:13] <ayourtch> the "scope" can be defined as:( location metric type, location metric value. )The network can announce the collection of these, the nodes can filter on these.
[08:01:49] Michael Richardson is now known as mcr
[08:01:49] mcr is now known as Michael Richardson
[08:01:52] <danyork> Anyone other than kerlyn want a question or comment relayed?  I'm in line after the next speaker (after Peter)
[08:01:53] Bernie - ucom.ch joins the room
[08:02:19] <danyork> (assuming you're remote - people in the room can get up to the mic themselves ;-) )
[08:02:23] <ayourtch> with "location type" being "building", "physical 3d coordinates", "room", "place in the network", etc.
[08:02:30] <kerlyn> please don't relay my comment until after REQ5
[08:02:48] <kerlyn> at the very end
[08:02:50] <Andrew Sullivan> @dan: well, some of them can :)  The room layout is not ideal for getting to mic, as people have complained about.
[08:02:51] <danyork> ah, right
[08:03:20] <Dan Wing> need a circular room?
[08:03:29] <danyork> kerlyn: Ah, right... I'll wait on your comment
[08:03:33] <ayourtch> i am sitting far, so you can relay mine. :-)
[08:03:42] <danyork> Andrew Sullivan: Yes, I see you over on the edge
[08:04:17] g.e.montenegro leaves the room
[08:04:21] <danyork> ayourtch: okay, I can do that :-)
[08:04:23] <danyork> Amine Choukir at mic
[08:04:37] g.e.montenegro joins the room
[08:04:39] <danyork> ayourtch: and your full name is?
[08:04:43] <ayourtch> andrew yourtchenko
[08:04:44] <ayourtch> sorry  :)
[08:04:46] <danyork> Ha!
[08:05:13] <kerlyn> re: "scope" or "zone", it may not be unreasonable to constrain this based on topology.  For example, a scope may be equivalent to a subnet
[08:05:26] <danyork> Yi Yang at mic
[08:05:44] <ayourtch> kerlyn: thats my point, it i a question of defining what "distance" means, which also depends on what "location" means
[08:05:46] <danyork> Amine back at mic
[08:05:55] g.e.montenegro leaves the room
[08:06:09] <Andrew Sullivan> We do have the "loc" RRTYPE in the DNS, though I think it's only lat and long
[08:06:14] keith_nm joins the room
[08:06:22] <Andrew Sullivan> we'd need maybe 3dloc, so we could get height as well :)
[08:06:35] g.e.montenegro joins the room
[08:06:51] Michael Richardson is now known as mcr
[08:07:02] <mcr> and relloc, so that's its a +- against another loc, for a mobile device.
[08:07:09] <mcr> er, nomadic device.
[08:07:35] <danyork> Slide 13 - REQ4
[08:07:49] <danyork> Stuart - we shouldn't require a "flag day"
[08:07:59] <danyork> Slide 14 - REQ5
[08:08:11] <Andrew Sullivan> we still need height, though, so that you can be on floor 13 and not get the thing down the hall, only on floor 3
[08:08:13] <danyork> Dave Thaler at mic
[08:08:33] <danyork> Dave - we don't want to break ANY link scope protocols
[08:08:39] <danyork> Eliot Lear at mic
[08:08:39] <kerlyn> there are already some precedents where "scope" and "link" are equivalent; e.g. ".local" and ".ARPA"
[08:08:40] <Tim Wicinski> I agree on the elevation piece
[08:09:35] <danyork> Toerless Eckert at mic
[08:10:22] g.e.montenegro leaves the room
[08:10:37] g.e.montenegro joins the room
[08:11:53] mcr is now known as Michael Richardson
[08:12:12] Michael Richardson is now known as mcr
[08:12:12] mcr is now known as Michael Richardson
[08:12:12] Michael Richardson is now known as mcr
[08:12:12] mcr is now known as Michael Richardson
[08:12:38] roessler joins the room
[08:13:17] <danyork> That was me channeling kerlyn
[08:13:23] <kerlyn> tyvm
[08:13:36] <danyork> Richard Kelsey at mic
[08:13:44] ayourtch leaves the room
[08:13:54] Andrew Yourtchenko joins the room
[08:13:56] <danyork> Richard - there is a 6lowpan use case that doesn't appear in the requirements
[08:15:03] <danyork> Ralph speaking about what he captured as feedback
[08:15:43] <danyork> Slides - we're going back to the "Chair's slides" deck - slide 10 is currently being displayed "The draft charter" - http://www.ietf.org/proceedings/87/slides/slides-87-dnssdext-2.pptx
[08:15:47] g.e.montenegro leaves the room
[08:15:51] g.e.montenegro joins the room
[08:16:11] <danyork> Eliot Lear at mic responding to Ralph
[08:16:31] g.e.montenegro leaves the room
[08:16:32] g.e.montenegro joins the room
[08:16:54] Michael Richardson is now known as mcr
[08:18:58] <danyork> Slide 11 - Description/Problem statement
[08:19:08] <kerlyn> relay: at slide 12 I have an add'l goal to propose;  
4. To consider extensions to DNS-SD that improve its suitability for M2M and/or Aggregated Services Discovery.
[08:19:27] <danyork> kerlyn: okay, will relay for slide 12
[08:19:33] <danyork> Toerless Eckert back at mic
[08:20:00] <danyork> Toerless wants to make sure solution we support will work in larger environments
[08:20:43] <danyork> Peter Koch at mic
[08:21:30] mkgillmore joins the room
[08:21:55] <danyork> Peter asks about 3rd bullet and reference to ZigBee standard
[08:21:56] mcr is now known as Michael Richardson
[08:21:56] Michael Richardson is now known as mcr
[08:22:34] <kerlyn> "Multi-link" should also cover "multi-technology", e.g. LLNs
[08:23:36] <kerlyn> SEP2 is shipped
[08:24:14] <danyork> Andrew Sullivan at mic
[08:24:30] <danyork> Andrew - also questioning bullet 3
[08:24:44] <kerlyn> Not really - it expands it to multi-link subnets
[08:25:23] <danyork> Erik Nordmark at mic
[08:25:45] <danyork> Erik - doesn't see results of scope discussion in this text
[08:26:31] <danyork> Ralph answering Erik
[08:26:57] mcr is now known as Michael Richardson
[08:28:03] <kerlyn> I don't think its easy to decouple "scope" from topology and also be zero config in the A, B, C cases
[08:28:15] <danyork> Toerless at mic
[08:28:30] <danyork> kerlyn: want that relayed to the mic?
[08:28:51] <kerlyn> no, I think the person at the mic expressed it
[08:28:59] <danyork> ok
[08:29:17] <danyork> Slide 12 - dnssdext goals
[08:29:22] Dan Wing leaves the room
[08:29:32] <danyork> kerlyn: I'll raise your fourth goal
[08:29:37] <kerlyn> thx
[08:29:55] <danyork> Dave Thaler at mic
[08:31:12] <Ted Lemon> I'm confused—it says "service discovery" in the first sentence of bullet 1.
[08:31:40] <kerlyn> Rationale: in the ZigBee Smart Energy Profile 2 Application Specification it became clear there was a need for
1) fine-grained discovery, e.g., of different REST APIs within a server, and 2) a discovery-based method for a
host to discover its default settings for other services it depends on.
[08:31:51] <Ted Lemon> scope expansion seems like a bad idea.
[08:31:57] <kerlyn> kk
[08:31:59] Michael Richardson is now known as mcr
[08:32:13] Dan Wing joins the room
[08:32:28] <danyork> Peter Koch at mic
[08:32:44] <danyork> Peter has issues with goal 3
[08:33:10] <danyork> Peter - goal 3 implies architecture
[08:33:43] <danyork> Peter - architecture shouldn't be done in a BCP. Let's get rid of this requirement.
[08:33:47] <danyork> Eliot Lear at mic
[08:33:47] <Suz> +1 o Peter.
[08:34:43] <Suz> but also giving some consideration to the problem it's aimed at needs to be included.
[08:34:47] <danyork> Eliot - number 2 has a nit about scaling from home to wide area
[08:34:56] <danyork> Alessandro Vesely at mic
[08:35:19] <danyork> Alessandro also questions #3
[08:36:41] <danyork> Stuart Cheshire at mic
[08:37:00] mcr is now known as Michael Richardson
[08:37:53] <danyork> Any more questions from remote folks?
[08:38:10] <danyork> Andrew Sullivan at mic
[08:38:41] <danyork> (Stuart asked whether we are building on MDNS and DNS-SD or are we building something brand new?)
[08:39:44] <danyork> Eliot Lear at mic
[08:40:25] <danyork> Eliot - there are always tradeoffs - it is up to the group to have that discussion
[08:40:31] <danyork> Toerless Eckert at mic
[08:41:03] <danyork> Any other comments?
[08:41:19] <danyork> Next - slide 13 - deliverables
[08:41:44] <danyork> Harald Albrecht at mic
[08:42:02] Michael Richardson is now known as mcr
[08:42:41] <danyork> Tim Chown acknowledges that better words are needed for document 2
[08:42:49] <danyork> Eric Osterweil at mic
[08:43:37] <danyork> Eric - I worry about the BCP document
[08:45:57] <danyork> Harald Albrecht at mic
[08:46:05] AndersBrandt leaves the room
[08:46:56] <danyork> Harald - what architectual work is still required?
[08:47:03] mcr is now known as Michael Richardson
[08:47:03] Michael Richardson is now known as mcr
[08:47:07] <danyork> Dave Thaler at mic
[08:48:33] <kerlyn> ?? services are named
[08:49:02] <danyork> Dave - keep deliverables focused on *service discovery*
[08:49:15] mcr is now known as Michael Richardson
[08:49:15] Michael Richardson is now known as mcr
[08:49:15] mcr is now known as Michael Richardson
[08:49:15] Michael Richardson is now known as mcr
[08:49:15] mcr is now known as Michael Richardson
[08:49:15] Michael Richardson is now known as mcr
[08:49:46] <danyork> +1 to Dave's suggestion
[08:49:50] <danyork> (from me)
[08:49:56] <danyork> Toerless at mic
[08:50:18] <danyork> Toerless - wants mention of "service announcement and discovery"
[08:50:38] <danyork> Toerless - would like to see #2 split
[08:51:27] AndersBrandt joins the room
[08:51:50] <danyork> Brian Dickson at mic
[08:52:05] mcr is now known as Michael Richardson
[08:52:13] Dan Wing leaves the room
[08:52:32] <danyork> Eliot Lear at mic
[08:52:46] <danyork> Ray Bellis at mic
[08:52:50] <kerlyn> relay: please ask the speakers to take their comments to the mailing list
[08:53:12] <kerlyn> IOW, reiterate on the mailing list
[08:53:43] <kerlyn> np ;-)
[08:53:58] <danyork> Slide 14 - Scope of charter?
[08:54:07] Dan Wing joins the room
[08:54:13] <kerlyn> hmmmm
[08:54:17] <Ted Lemon> We should do a barbershop hum.
[08:54:26] <danyork> :-)
[08:54:36] <kerlyn> yes, four part!
[08:54:44] <Ted Lemon> silence on no
[08:54:53] <danyork> First question - no hums against
[08:55:17] <danyork> Next hum - is the problem statement clear enough and well-scoped?
[08:56:23] <danyork> Ralph - rough consensus with some dissent
[08:56:39] <Andrew Sullivan> I was dissenting.  I don't know how to answer this question in the face of, "Well we're going to restate the problem later."
[08:56:46] <danyork> Next hum - how mature is our understanding of the requirements?
[08:57:07] Michael Richardson is now known as mcr
[08:57:11] g.e.montenegro leaves the room
[08:57:17] <Ted Lemon> I thought he meant "assuming that we address the stuff we know how to address."
[08:57:28] g.e.montenegro joins the room
[08:57:39] <danyork> Ralph - Rough consensus we agree
[08:57:47] <danyork> Next hum - can we solve problem?
[08:57:56] <Andrew Sullivan> I just have no idea whether we can solve this problem
[08:57:59] <danyork> Ralph - rough consensus
[08:58:06] <Ted Lemon> That is, assuming that we address the stuff we know how to address, are you willing to take it on faith that we can get to a good charter.
[08:58:10] <Andrew Sullivan> I don't know whether it is solvable
[08:58:16] <Ted Lemon> And/or make us get to a good charter.
[08:58:20] <danyork> Next hum - are the three goals and deliverables appropriate?  (as modified)
[08:58:20] <kerlyn> ajsaf: so would you argue against a WG?
[08:58:24] mcr channels Newhart's town hall meetings.
[08:58:44] <Andrew Sullivan> no, I just think we should be prepared to fail
[08:58:56] <Ted Lemon> Yes.
[08:58:58] <danyork> Tim - Rough consensus
[08:59:10] <Ted Lemon> That's okay.   We can't succeed if we don't try, and we can't learn anything if we aren't willing to fail.
[08:59:11] <danyork> General question - what is the draft charter missing?
[08:59:16] <danyork> Dave Thaler at mic
[08:59:32] <mcr> Andrew, what requirement do you think will cause us to fail?  Or to put it another way, if we removed some requirement, could we succeed?
[08:59:37] <Andrew Sullivan> I think this is useful and important work.  I think we should give it a whack.  But I think there's a real possibility that, given DNS and mDNS as actually deployed, we're stuck.
[08:59:42] <danyork> Dave - scope management and discovery should be mentioned
[08:59:46] <danyork> Any remote comments?
[09:00:19] <kerlyn> ajsaf: have faith
[09:00:20] <danyork> Who would be willing to author documents?
[09:00:23] <mcr> Andrew Sullivan, so if we could greenfield, we would have a better opportunity to succeed?
[09:00:37] mcr volunteers to work on requirements document.
[09:00:42] <danyork> 10 people on requirements
[09:00:51] <Andrew Sullivan> The problem is that the existing DNS-SD approach depends on names.  Some of those are DNS names and some are mDNS names, and these namespaces are unfortunately at odds.
[09:00:53] <kerlyn> We cannot strand tens of millions of deployed printers and other services
[09:01:14] <Andrew Sullivan> I completely agree
[09:01:14] <danyork> about 10 people on service discovery document
[09:01:23] <mcr> kerlyn, nevermind the downside, I want to understand where Andrew thinks that hard part is.
[09:01:51] <danyork> about 7 or 8 on the informational document
[09:01:51] <danyork> lots of people offering to review
[09:02:08] mcr is now known as Michael Richardson
[09:02:08] Michael Richardson is now known as mcr
[09:02:09] <danyork> Question - Do we want to move forward towards a WG?
[09:02:15] <mcr> WG!
[09:02:26] <kerlyn> mcharlesr: I agree we need to document one or more examples of the difficulty.  I have already spoken to Andrew about this.
[09:02:31] <kerlyn> Stay tuned
[09:02:32] <Andrew Sullivan> the hard part is simple.  In mDNS, "Andrew's Printer" is a legal name, and "AndrĂ©'s Printer" is a legal name, and under DNS these two are legal but don't meet the LDH rule, and under IDNA2008 the latter one is not allowed.
[09:02:34] <danyork> No hums against (that I could hear)
[09:02:38] <Ted Lemon> Andrew, I agree with this and think it needs to be made explicit.
[09:03:13] yone leaves the room
[09:03:13] <mcr> what is the "LDH rule" (I think I don't know the TLA)
[09:03:18] <Ted Lemon> Wow, finished early!
[09:03:24] keith_nm leaves the room: IETF87, Berlin
[09:03:31] Ning Kong leaves the room
[09:03:32] <kerlyn> ajsaf: so mDNS allows to convert labels to IDNA, one at a time, rightmost first
[09:03:33] mkgillmore leaves the room
[09:03:33] <Ted Lemon> Leave Dallas Hastily, Michael.
[09:03:35] <danyork> Chairs - thank you for participation
[09:03:35] Markus Stenberg leaves the room
[09:03:40] <danyork> Session is concluded
[09:03:40] Bernie - ucom.ch leaves the room
[09:03:40] cabo leaves the room
[09:03:44] Ted Lemon leaves the room
[09:03:49] <kerlyn> see you on the mailing list
[09:04:18] Kent Shuey leaves the room: offline
[09:04:25] AndersBrandt leaves the room
[09:04:33] Andrew Sullivan leaves the room
[09:04:47] <danyork> For the log, the person in "[09:30:41] <danyork> (?) at mic" was Bernie Hoeneisen
[09:04:58] mcr leaves the room
[09:04:58] Michael Richardson joins the room
[09:05:02] joel jaeggli leaves the room
[09:05:13] fneves leaves the room
[09:05:28] <kerlyn> dyork: thanks for your scrivening
[09:05:40] matthijs leaves the room
[09:06:07] kerlyn leaves the room
[09:06:17] roessler leaves the room
[09:06:40] Chris Griffiths leaves the room
[09:07:30] Chris Griffiths joins the room
[09:08:42] joel jaeggli joins the room
[09:08:50] danyork leaves the room
[09:10:03] Tim Wicinski joins the room
[09:10:23] fneves joins the room
[09:10:24] Tim Wicinski leaves the room
[09:10:38] fneves leaves the room
[09:11:00] Tim Wicinski leaves the room
[09:11:27] Andrew Yourtchenko leaves the room
[09:12:40] Wendy Seltzer leaves the room
[09:13:45] john.levine leaves the room
[09:13:46] Dan Wing leaves the room
[09:14:40] Chris Griffiths leaves the room
[09:17:28] jaap leaves the room
[09:18:40] markus.isomaki leaves the room
[09:19:13] Tim Wicinski joins the room
[09:19:37] Dan Wing joins the room
[09:19:55] Tim Wicinski leaves the room
[09:22:38] Dan Wing leaves the room
[09:23:59] Wendy Seltzer joins the room
[09:25:04] Dan Wing joins the room
[09:26:07] g.e.montenegro leaves the room
[09:30:02] Dan Wing leaves the room
[09:34:37] g.e.montenegro joins the room
[09:35:58] Chris Griffiths joins the room
[09:37:10] Wendy Seltzer leaves the room
[09:37:34] Ning Kong joins the room
[09:38:01] g.e.montenegro leaves the room
[09:46:11] John Jason Brzozowski leaves the room
[09:47:34] jaap joins the room
[09:49:44] Chris Griffiths leaves the room
[09:50:06] Eliot Lear leaves the room
[09:55:40] Eliot Lear joins the room
[09:56:00] Ning Kong leaves the room
[09:56:30] Bernie - ucom.ch joins the room
[09:57:30] jaap leaves the room
[10:00:42] cabo joins the room
[10:01:52] joel jaeggli leaves the room
[10:02:01] joel jaeggli joins the room
[10:07:29] Tim Wicinski joins the room
[10:08:45] jaap joins the room
[10:09:04] joel jaeggli leaves the room
[10:10:35] joel jaeggli joins the room
[10:13:02] Eliot Lear leaves the room
[10:17:41] Tim Wicinski joins the room
[10:18:02] Tim Wicinski leaves the room
[10:19:20] jaap leaves the room
[10:19:30] Tim Wicinski leaves the room
[10:27:43] Tim Wicinski joins the room
[10:38:40] Tim Wicinski joins the room
[10:39:00] Tim Wicinski leaves the room
[10:39:40] Tim Wicinski leaves the room
[10:42:09] danyork joins the room
[10:45:47] Ted Lemon joins the room
[10:47:12] Bernie - ucom.ch leaves the room
[10:49:13] cabo leaves the room
[10:49:39] Ted Lemon leaves the room
[10:52:49] Markus Stenberg joins the room
[10:54:14] Ted Lemon joins the room
[10:54:33] Bernie - ucom.ch joins the room
[10:56:46] Markus Stenberg leaves the room
[10:59:33] Ted Lemon leaves the room
[11:00:21] jaap joins the room
[11:01:36] Ning Kong joins the room
[11:04:17] Dan Wing joins the room
[11:08:55] cabo joins the room
[11:09:21] Wendy Seltzer joins the room
[11:23:23] Wendy Seltzer leaves the room
[11:41:51] roessler joins the room
[11:50:23] jaap leaves the room
[11:53:13] roessler leaves the room
[12:04:59] John Jason Brzozowski joins the room
[12:06:13] danyork leaves the room
[12:09:34] danyork joins the room
[12:10:09] roessler joins the room
[12:19:57] cabo leaves the room
[12:27:47] Dan Wing leaves the room
[12:33:19] John Jason Brzozowski leaves the room
[12:33:59] joel jaeggli leaves the room
[12:35:46] John Jason Brzozowski joins the room
[12:37:51] Ning Kong leaves the room
[12:39:26] Eliot Lear joins the room
[12:39:31] Michael Richardson leaves the room
[12:42:56] Eliot Lear leaves the room
[12:51:28] danyork leaves the room
[13:03:32] jaap joins the room
[13:05:35] Eliot Lear joins the room
[13:09:40] mkgillmore joins the room
[13:13:37] danyork joins the room
[13:13:57] John Jason Brzozowski leaves the room
[13:14:03] John Jason Brzozowski joins the room
[13:17:57] Bernie - ucom.ch leaves the room
[13:21:17] jaap leaves the room: Replaced by new connection
[13:21:19] jaap joins the room
[13:27:28] joel jaeggli joins the room
[13:31:02] Bernie - ucom.ch joins the room
[13:35:31] Bernie - ucom.ch leaves the room
[13:37:29] John Jason Brzozowski leaves the room
[13:42:02] mkgillmore leaves the room
[13:44:13] John Jason Brzozowski joins the room
[13:57:58] roessler leaves the room
[14:02:23] Eliot Lear leaves the room
[14:05:15] Eliot Lear joins the room
[14:12:44] jaap leaves the room
[14:13:10] danyork leaves the room
[14:17:59] danyork joins the room
[14:22:38] joel jaeggli leaves the room
[14:26:11] joel jaeggli joins the room
[14:40:21] joel jaeggli leaves the room
[15:11:52] joel jaeggli joins the room
[15:14:34] Eliot Lear leaves the room
[15:16:13] joel jaeggli leaves the room
[15:23:35] danyork leaves the room
[15:35:17] jaap joins the room
[15:36:18] jaap leaves the room: Replaced by new connection
[15:36:19] jaap joins the room
[15:46:11] joel jaeggli joins the room
[15:53:33] danyork joins the room
[16:36:30] joel jaeggli leaves the room
[16:43:28] roessler joins the room
[16:45:11] roessler leaves the room
[16:48:53] roessler joins the room
[16:53:11] joel jaeggli joins the room
[16:53:39] Eliot Lear joins the room
[16:54:31] Eliot Lear leaves the room
[16:56:38] Ning Kong joins the room
[17:00:00] Ning Kong leaves the room
[17:01:02] Eliot Lear joins the room
[17:06:05] roessler leaves the room
[17:10:01] roessler joins the room
[17:22:58] jaap leaves the room
[17:28:24] roessler leaves the room
[17:41:01] jaap joins the room
[17:41:20] jaap leaves the room: Replaced by new connection
[17:41:21] jaap joins the room
[17:41:30] Eliot Lear leaves the room
[17:45:34] roessler joins the room
[18:01:16] roessler leaves the room
[18:03:10] John Jason Brzozowski leaves the room
[18:04:03] joel jaeggli leaves the room
[18:07:18] roessler joins the room
[18:07:26] roessler leaves the room
[18:11:44] roessler joins the room
[18:16:57] John Jason Brzozowski joins the room
[18:23:27] danyork leaves the room
[19:18:02] roessler leaves the room
[19:42:22] jaap leaves the room
[19:48:47] John Jason Brzozowski leaves the room
[20:38:32] joel jaeggli joins the room
[21:10:35] roessler joins the room
[21:27:56] roessler leaves the room
[21:32:06] roessler joins the room
[21:33:09] roessler leaves the room
[21:34:23] roessler joins the room
[21:35:28] roessler leaves the room
[21:45:22] Eliot Lear joins the room
[21:53:23] jaap joins the room
[21:57:33] roessler joins the room
[22:07:19] roessler leaves the room
[22:22:01] Eliot Lear leaves the room
[22:23:41] jaap leaves the room
[22:59:36] John Jason Brzozowski joins the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!