[13:00:39] --- gorryf has become available
[13:01:26] --- gorryf has left
[13:01:33] --- gorryf has become available
[13:02:25] <gorryf> Welcome to the IPDVB WG at IETF-61!
[13:17:29] --- gorryfair has become available
[13:18:56] --- gorryfair has left: Disconnected
[13:27:06] --- gorryf has left: Disconnected
[13:31:19] --- gorryf has become available
[13:31:59] --- gorryf has left: Disconnected
[13:34:47] --- cgn has become available
[13:34:53] --- cgn has left
[13:40:26] --- gorryf has become available
[13:52:51] --- gorryf has left: Disconnected
[14:06:50] --- gorryf has become available
[14:09:58] <gorryf> Slides for the meeting today (if you wish to follow with jabber) have been uploaded to the web server at http://www.erg.abdn.ac.uk/ipdvb/meetings/11-11-02-IETF-61-Washington/
[14:17:48] --- cabo--tzi--org has become available
[14:21:53] --- cabo--tzi--org has left: Disconnected
[14:28:23] --- cgn has become available
[14:33:01] --- cabo--tzi--org has become available
[14:33:31] --- Suresh Krishnan has become available
[14:33:54] <cabo--tzi--org> Gorry presents agenda
[14:34:19] <cabo--tzi--org> main focus of time will be on new items
[14:34:24] <cabo--tzi--org> address resolution
[14:34:43] <cabo--tzi--org> XML receiver config
[14:35:15] <cabo--tzi--org> and charter milestones at the end
[14:35:23] <cabo--tzi--org> no new agenda items
[14:35:36] <cabo--tzi--org> IPR admonishment
[14:35:49] <cabo--tzi--org> read RFC3667/8
[14:36:32] <cabo--tzi--org> reviews WG objectives
[14:37:38] <cabo--tzi--org> and active IDs
[14:37:54] <cabo--tzi--org> (WG and individual)
[14:37:57] <cabo--tzi--org> four areas:
[14:38:22] <cabo--tzi--org> milestones (two completed: arch and ULE)
[14:39:03] <cabo--tzi--org> late milestones: AR Framework, AR Protocol; also neither arch nor ULE have been submitted to IESG
[14:39:45] <cabo--tzi--org> * Marie-Jose Montpetit on arch status
[14:41:08] <cabo--tzi--org> (see slides p10: main purpose)
[14:41:45] <cabo--tzi--org> p11
[14:42:12] <cabo--tzi--org> initially focus was on sat, now more general
[14:42:40] --- cgn has left
[14:43:22] <cabo--tzi--org> p12
[14:44:11] <cabo--tzi--org> discuss e2e mgmt reqts later with AR
[14:44:47] <cabo--tzi--org> p13
[14:45:14] <cabo--tzi--org> on list, there was some discussion on whether link sec should be in focus
[14:45:39] <cabo--tzi--org> p14
[14:46:43] <cabo--tzi--org> greg daley points to rfcdiff
[14:47:24] <cabo--tzi--org> Gorry: arch doc has cleared WGLC, one issue remaining:
[14:47:34] <cabo--tzi--org> should link encryption be mandatory?
[14:48:48] <cabo--tzi--org> cabo: arch should not mandate it, because if there is L3 security, you don't necessarily need link sec
[14:49:14] <cabo--tzi--org> Greg Daley: managing the keys may not be in everyone's business model
[14:49:32] <cabo--tzi--org> Gorry: both comments are in favor of the current text
[14:49:47] <cabo--tzi--org> * ULE (Gorry)
[14:49:49] <cabo--tzi--org> p15
[14:50:26] <cabo--tzi--org> p16 (nice colors)
[14:51:30] <cabo--tzi--org> MPE and the variant used by ATSC to the right, ULE to be efficient and extensible (e.g., encryption, bridging, header compression)
[14:51:32] <cabo--tzi--org> p17
[14:52:44] <cabo--tzi--org> would like to have implementation experience with extension headers
[14:53:36] <cabo--tzi--org> p18
[14:53:37] <cabo--tzi--org> p19
[14:54:59] <cabo--tzi--org> slide is wrong way around
[14:55:22] <cabo--tzi--org> gorry will send an updated slide for the minutes
[14:58:28] <cabo--tzi--org> cabo: can't you do both?
[14:58:58] <cabo--tzi--org> gorry: FEC gives you a high chance the pointers will be corrupted, too
[14:59:16] <cabo--tzi--org> Greg Daley: Can't we just check CRC and continue if it matches
[14:59:32] <cabo--tzi--org> gorry: discuss this on the list, check whether we can be more liberal
[15:00:32] <cabo--tzi--org> gorry: Bernhard recommends just entering idle state
[15:00:48] <cabo--tzi--org> Greg: MAY, but be aware of the consequences?
[15:01:01] <cabo--tzi--org> p20
[15:01:47] <cabo--tzi--org> p21
[15:01:58] <cabo--tzi--org> gorry: implementers, please speak up
[15:02:20] <cabo--tzi--org> p22
[15:03:10] <cabo--tzi--org> need to decide whether a standards-track document is required for extension or just an RFC
[15:03:55] <cabo--tzi--org> * Gorry: XULE
[15:04:14] <cabo--tzi--org> p23
[15:04:15] <cabo--tzi--org> p34
[15:04:24] <cabo--tzi--org> p24
[15:05:16] <cabo--tzi--org> if there are people that would like to work on security headers, please submit an Internet-Draft
[15:05:49] <cabo--tzi--org> * Marie: AR status
[15:06:05] <cabo--tzi--org> p25
[15:06:07] <cabo--tzi--org> p26
[15:07:39] <cabo--tzi--org> p28 (p27 skipped)
[15:08:42] <cabo--tzi--org> p29
[15:09:19] <cabo--tzi--org> what is the usage of the INT: IP/PID or IP/MAC?
[15:10:33] <cabo--tzi--org> p30
[15:10:40] <cabo--tzi--org> christmas vacation project?
[15:10:59] <cabo--tzi--org> martin Stiemerling:
[15:11:14] <cabo--tzi--org> Address translation vs. address resolution
[15:11:34] <cabo--tzi--org> Marie-Jose: how are the tables built, the mappings
[15:12:17] <cabo--tzi--org> Joerg Ott: may want to look at BEHAVE
[15:12:38] <cabo--tzi--org> Martin: Too specific, BEhAVE is more for Linux implementers
[15:12:53] <cabo--tzi--org> * Martin: XML for receiver AR config
[15:13:06] <cabo--tzi--org> (slides are missing from the web page)
[15:14:00] <cabo--tzi--org> this talk is not about a specific solution, but opens discussion
[15:14:07] <cabo--tzi--org> config ob dvb receivers
[15:14:15] <cabo--tzi--org> ip AR config
[15:14:21] <cabo--tzi--org> maybe other IP related config
[15:14:35] <cabo--tzi--org> (cf. DHCP)
[15:14:37] <cabo--tzi--org> addtl config: service related
[15:14:52] <cabo--tzi--org> should be based on XML, unless this is too low layer
[15:15:01] <cabo--tzi--org> Questions:
[15:15:07] <cabo--tzi--org> -- which config space
[15:15:09] <cabo--tzi--org> ...
[15:15:18] <cabo--tzi--org> related: layer 2
[15:15:20] <cabo--tzi--org> IPCDN
[15:15:24] <cabo--tzi--org> DOCSIS MIBs
[15:15:43] <cabo--tzi--org> DVB: SI tables, MHP
[15:15:46] <cabo--tzi--org> Others?
[15:16:24] <cabo--tzi--org> Greg Daley: I'm confused; we expect these devices to be IP devices; do you expect they can connect to the Internet?
[15:16:26] <cabo--tzi--org> (yes)
[15:16:35] <cabo--tzi--org> Why not use ND etc. instead of static tables?
[15:16:46] <cabo--tzi--org> more live
[15:17:31] <cabo--tzi--org> make use of existing protocols
[15:18:42] <cabo--tzi--org> Zhigao Chen (Panasonic): DVB edge receiver may have difficulty in sending, cannot use ND
[15:19:02] <cabo--tzi--org> Related areas: L3
[15:19:06] <cabo--tzi--org> NETCONF
[15:19:06] <cabo--tzi--org> DHC
[15:19:17] <cabo--tzi--org> (NETCONF is router config only)
[15:19:27] <cabo--tzi--org> IP techniques: ND, ARP...
[15:19:29] <cabo--tzi--org> Others?
[15:19:39] <cabo--tzi--org> Configuration Scenarios:
[15:19:57] <cabo--tzi--org> IP config (basic DVB config available), backchannel eg.. ISDN
[15:20:04] <cabo--tzi--org> Multicast config
[15:20:16] <cabo--tzi--org> broadcast config ("open bitstream" like TV)
[15:20:24] <cabo--tzi--org> security config (keys)
[15:20:37] <cabo--tzi--org> might even want a complete bootstrap
[15:21:33] <cabo--tzi--org> Issues:
[15:21:41] <cabo--tzi--org> who is in control of receiver?
[15:22:06] <cabo--tzi--org> operator (like MHP), owner at home, network operator (router manager)
[15:22:35] <cabo--tzi--org> NETCONF: single router
[15:22:40] <cabo--tzi--org> IPCDN 1E3
[15:22:51] <cabo--tzi--org> IPDVB up to 10E5 receivers must be configured
[15:23:56] <cabo--tzi--org> what are the parameters etc.
[15:24:07] <cabo--tzi--org> greg: is this AR config or just host config?
[15:24:42] <cabo--tzi--org> for AR, use ND -- there are existing mechanisms even for unidir links
[15:25:11] <cabo--tzi--org> very little to do with DVB, much with simplex links
[15:25:31] <cabo--tzi--org> gorry: I'm guilty.
[15:25:48] <cabo--tzi--org> AR means several things to people.
[15:26:21] <cabo--tzi--org> ND/ARP with IP/MAC; but also also mapping of IP/MAC on underlying channel -- this is also AR, but a different function
[15:27:10] <cabo--tzi--org> Joerg: some boxes will be controlled by operator etc.; we should focus on common mechanisms
[15:27:39] <cabo--tzi--org> keep this basic, restrictive, please.
[15:27:58] <cabo--tzi--org> otherwise there will be so many weird things....
[15:29:05] <cabo--tzi--org> gorry: it's XML...
[15:29:20] <cabo--tzi--org> Joerg: this has no bearing on the complexity...
[15:29:56] <cabo--tzi--org> Zhigao: ...XML...
[15:30:04] <cabo--tzi--org> Martin: I don't stick to XML
[15:30:22] <cabo--tzi--org> KISS!
[15:31:12] <cabo--tzi--org> Greg: Which ones are AR in the classic IP sense, and which ones are multicast mapping
[15:32:47] <cabo--tzi--org> gorry: Are you going to participate?
[15:33:00] <cabo--tzi--org> Martin: at least comments, if there is enough material, even a draft
[15:33:05] <cabo--tzi--org> * Marie-Jose
[15:33:11] <cabo--tzi--org> draft not in repository
[15:33:47] <cabo--tzi--org> p32
[15:34:24] <cabo--tzi--org> IP content providers don't know about SI tables etc.
[15:34:41] <cabo--tzi--org> Joerg: Program Guides
[15:35:25] <cabo--tzi--org> Marie-Jose: This is about config of IP addresses to DVB services
[15:36:37] <cabo--tzi--org> p33
[15:37:11] <cabo--tzi--org> work was done in ATSC and cable community
[15:37:50] <cabo--tzi--org> syntax must be extensible
[15:38:46] <cabo--tzi--org> p34
[15:39:50] <cabo--tzi--org> p35
[15:43:38] <cabo--tzi--org> * Gorry
[15:43:45] <cabo--tzi--org> p37, slightly updated
[15:44:13] <cabo--tzi--org> stage 1: mapping MPEG to IP/MAC addresses
[15:44:50] <cabo--tzi--org> 1b: IETF protocols: ND, ARP, DHCP
[15:45:29] <cabo--tzi--org> may be useful as informational
[15:45:44] <cabo--tzi--org> state 2: (if stage 1 does not satisfy all the needs)
[15:47:16] <cabo--tzi--org> * Joerg jumps in with "real-time slides":
[15:47:44] <cabo--tzi--org> "Internet Media Guides"
[15:48:43] <cabo--tzi--org> in the MMUSIC working group; are at reqts and framework stage; more sophisticated that just AR, can do push broadcast as well as bidir
[15:48:56] <cabo--tzi--org> remember SAP/SDP (Mbone); do this for the real world
[15:49:07] <cabo--tzi--org> 1-1.5 yr old
[15:49:47] <cabo--tzi--org> info delivery will be IP-based
[15:49:54] <cabo--tzi--org> encodings
[15:49:58] <cabo--tzi--org> data transmission
[15:50:00] <cabo--tzi--org> addressing
[15:50:04] <cabo--tzi--org> interaction and control
[15:51:02] <cabo--tzi--org> platform/network independent content provision
[15:51:13] <cabo--tzi--org> same content available via different networks
[15:52:18] <cabo--tzi--org> application: DVB-T/H, 3G/4G, WLAN, "regular internet"
[15:52:28] <cabo--tzi--org> content is just a type of service
[15:53:01] <cabo--tzi--org> IMG = "collection of multimedia session descriptions" (MMUSIC charter)
[15:53:54] <cabo--tzi--org> Content: whatever metadata you want (e.g., SDP, SDPng, ...)
[15:54:05] <cabo--tzi--org> stuff this in an envelope
[15:54:12] <cabo--tzi--org> Distribution...
[15:56:53] <cabo--tzi--org> three types:
[15:56:56] <cabo--tzi--org> ANNOUNCE
[15:56:59] <cabo--tzi--org> QUERY/RESOLVE
[15:57:07] <cabo--tzi--org> SUBSCRIBE/NOTIFY
[15:57:22] <cabo--tzi--org> (possibly complemented by a QUERY/RESOLVE)
[15:58:05] <cabo--tzi--org> ANNOUNCE: FLUTE
[15:58:12] <cabo--tzi--org> SUBSCRIBE/NOTIFY: SIP
[15:58:24] <cabo--tzi--org> QUERY/RESOLVE: straightforward
[15:58:33] <cabo--tzi--org> current work is ongoing on IMG envelope
[15:58:43] <cabo--tzi--org> MIME may help
[16:00:05] --- gorryf has left: Disconnected
[16:00:28] <cabo--tzi--org> Security may be interesting in the presence of "transceivers"
[16:01:24] <cabo--tzi--org> Metadata: to be defined based on domain-specific requirements
[16:02:07] <cabo--tzi--org> regionalization and personalization
[16:02:42] <cabo--tzi--org> Status:
[16:02:58] <cabo--tzi--org> FLUTE is done, Tampere, INRIA, TZI, Nokia have interoperable implementations
[16:03:18] <cabo--tzi--org> running implementation of QUERY/RESPONSE, looking at versioning, proxying, caching
[16:03:24] <cabo--tzi--org> SIP notifications are being worked on
[16:03:37] <cabo--tzi--org> Fujitsu(?) and Henning
[16:03:42] <cabo--tzi--org> IMG URIs/URNs
[16:03:58] <cabo--tzi--org> not much more rocket science needed
[16:04:18] <cabo--tzi--org> in many ways, describing how to use existing standards to do IMGs
[16:04:47] <cabo--tzi--org> Gorry: useful to find out what's going on in other groups
[16:04:54] <cabo--tzi--org> this here is about link layer
[16:05:16] <cabo--tzi--org> Greg: This work is probably very applicable, most of the requirements, even solution space
[16:05:33] <cabo--tzi--org> Gorry: fundamental differences: IP address for a service
[16:06:34] <cabo--tzi--org> Joerg: I didn't suggest to use this for IP stack config; but when programming comes in, use more high-level mechanisms
[16:07:02] <cabo--tzi--org> Greg: AR can be done with existing L3 protocols
[16:07:13] <cabo--tzi--org> As far as content goes, this looks very interesting
[16:07:41] <cabo--tzi--org> Marie-Jose: great minds meet, same types of services; Media guides are huge files; we have very small files
[16:08:31] <cabo--tzi--org> You are higher than us; we deal with much more closed networks; very little IP devices
[16:14:55] <cabo--tzi--org> (oops, was at the mike too long)
[16:15:17] <cabo--tzi--org> Greg: make sure we don't have to do this ...
[16:15:40] <cabo--tzi--org> Marie-Jose: You have code, I have code, what's missing is the deployment...
[16:16:09] <cabo--tzi--org> Greg: we have to figure out what to do with existing networks
[16:19:03] <cabo--tzi--org> Joerg: Some of this is higher-layer; this group should not look at IMG level -- IPDVB is INT area...
[16:20:18] <cabo--tzi--org> Martin: Maybe this can be marked as "pure IP bootstrap"
[16:21:12] <cabo--tzi--org> Zhigao: To ensure the receiver IMG/EPG, does it require a predefined IP address/port number or at least PI?
[16:21:16] <cabo--tzi--org> Joerg:
[16:21:30] <cabo--tzi--org> Obviously, you need some information to bootstrap
[16:22:52] <cabo--tzi--org> Gorry: We have introduced a fair amount of confusion
[16:23:02] <cabo--tzi--org> * milestone update
[16:23:10] <cabo--tzi--org> p42
[16:23:30] <cabo--tzi--org> AD will discuss placement of the milestones
[16:23:40] <cabo--tzi--org> with Gorry
[16:24:10] <cabo--tzi--org> We will be given some time as the problem is more complex than thought.
[16:24:11] --- Suresh Krishnan has left
[16:24:17] <cabo--tzi--org> Thank you for coming to the fridge :-)
[16:26:15] --- cabo--tzi--org has left: Disconnected