IETF
nwcrg
nwcrg@jabber.ietf.org
Friday, March 27, 2015< ^ >
Room Configuration
Room Occupants

GMT+0
[12:45:35] Meetecho joins the room
[13:21:53] Tadanori TERUYA joins the room
[13:57:10] Morten V. Pedersen joins the room
[13:57:16] Morten V. Pedersen leaves the room
[13:57:21] Morten V. Pedersen joins the room
[13:57:37] CedricAdjih joins the room
[13:58:13] Aaron Falk joins the room
[14:00:09] <CedricAdjih> Brian presents the IPR Notewell
[14:01:01] <CedricAdjih> … the agenda
[14:01:22] <CedricAdjih> First presentation: for Marie-Jose
[14:01:42] K Fouli joins the room
[14:02:03] ajsaf@jabber.org joins the room
[14:03:18] Bob Edge joins the room
[14:06:40] ajsaf@jabber.org leaves the room
[14:06:44] ajsaf@jabber.org joins the room
[14:07:47] <CedricAdjih> Victor is pointing to the Taxonomy document if you need some network coding background
[14:08:21] <CedricAdjih> https://tools.ietf.org/html/draft-firoiu-nwcrg-network-coding-taxonomy-01
[14:11:12] <CedricAdjih> Aaron Falk mentions that "code points" may have a different definition in the context of the IETF
[14:12:13] Aaron Falk was confused
[14:12:44] <CedricAdjih> discussion: there was clarification on  code points: they are not "points" (nodes) were "coding" occurs
[14:15:05] <CedricAdjih> Comment from Brian for accepting the document as a WG documents: people are encouraged to read the document and provide feedback on the mailing list
[14:15:23] Senthil Sivakumar joins the room
[14:15:24] <CedricAdjih> [Vincent is Presenting]
[14:18:01] Aaron Falk leaves the room
[14:18:47] <CedricAdjih> question from Victor Firoiu about are pending patents included, are there more ? Vincent answers probably there are more
[14:21:04] Ning Kong joins the room
[14:21:58] Ning Kong leaves the room
[14:22:58] Bob Edge leaves the room
[14:26:55] <CedricAdjih> Brian comment ("personal view") saying essentially that  nothing is done without cost,  IETF has some procedures to deal IPR, recognize it, but the good thing is that this is a research group so we don't need to revisit the issue too often, and we should not be too discouraged from the site
[14:27:41] <CedricAdjih> +comment from Marie Jose
[14:28:11] Aaron Falk joins the room
[14:28:54] <CedricAdjih> [Bhumip presnets]
[14:29:26] Tadanori TERUYA leaves the room
[14:32:15] <CedricAdjih> Brian, 2 questions:
[14:32:32] <CedricAdjih> are there major references suggested?
[14:32:39] <CedricAdjih> (bhumip: too early)
[14:32:45] <CedricAdjih> are there some prototypes?
[14:32:54] <CedricAdjih> (bhumip: also too early)
[14:33:04] <CedricAdjih> (Cedric Presents)
[14:50:12] <CedricAdjih> (Jinzhu Presents)
[14:51:47] Tadanori TERUYA joins the room
[14:52:59] <CedricAdjih> Morten questions:
[14:53:08] <CedricAdjih> systematic phase als?
[14:53:21] <CedricAdjih> Jinzhu -> yes
[14:53:36] <CedricAdjih> q: how to deal with very short flows?
[14:53:48] <CedricAdjih> a: not yet considered
[14:59:03] Tadanori TERUYA leaves the room
[14:59:12] Tadanori TERUYA joins the room
[15:00:16] <CedricAdjih> Morten: comment on the simulation of packets loss, essentially: this rather nice for analysis, but maybe not so realistic
[15:01:10] <CedricAdjih> Morten: what kind of packet loss? disabled retransmissions?
[15:02:25] <CedricAdjih> Brian: full slides are available
[15:04:27] <CedricAdjih> (comments from David Black)
[15:08:11] <CedricAdjih> (discussions about ECN bits, interaction between network coding loss recovery and congestion control)
[15:08:23] <CedricAdjih> (Morten presents)
[15:10:54] <CedricAdjih> MarieJose: do you intent to put the API in opensource?
[15:11:12] <CedricAdjih> Morten says yes
[15:11:47] <CedricAdjih> Vincent Rocca: why not API? documents for sure
[15:12:13] <CedricAdjih> Brian: we are a RG, why not do somthing like API
[15:16:52] <CedricAdjih> (discussion about building blocks, signalling, API,...)
[15:17:29] <CedricAdjih> (Frank presents)
[15:18:41] ajsaf@jabber.org leaves the room
[15:25:02] ajsaf@jabber.org joins the room
[15:26:38] <CedricAdjih> Brian: there is an opportunity to bring problems such as multi-path to the MANET working group (rechartering)
[15:36:05] <CedricAdjih> Brian comments that the NWCrg was formed as kind of "home base" for other WG/RG of the IETF
[15:37:05] <CedricAdjih> Frank answera by asking are we doing a good job at that?
[15:41:04] <CedricAdjih> Bhumip asked about implementation in the SDN testbed; Frank answered along: for now is more  of a prototype
[15:41:37] <CedricAdjih> MarieJose objected about the notion that NWCrg would not do a good job in dissemination (example of TAPS)
[15:46:21] <CedricAdjih> (comments of Aaron Falk, about the role of the NWCRG w.r.t. other groups, and expertise)
[15:48:13] <CedricAdjih> (JOnathan presents)
[15:49:35] Tadanori TERUYA leaves the room
[15:49:43] Tadanori TERUYA joins the room
[15:54:34] <CedricAdjih> (Vincent presents)
[15:57:47] <CedricAdjih> victor comments proper document is https://datatracker.ietf.org/doc/draft-irtf-nwcrg-network-coding-taxonomy/ ; and we are pretty much done with document — discussion about emphasizing more important parts of the taxonomy
[16:00:43] <CedricAdjih> Morten, about probably the identification of symbols (on the wire) could be separated from rest which should be more complex
[16:03:31] <CedricAdjih> Victor comments that we also have a wiki, — there are comments that github could offer more
[16:04:41] <CedricAdjih> Morten comments maybe step 4 could have higher priority
[16:06:14] ajsaf@jabber.org leaves the room
[16:08:00] <CedricAdjih> MarieJosé comments that starting from building blocks and then going for architecture is a valid approach; there might be many architectures, … (other use cases)
[16:08:58] <CedricAdjih> Morten, q about the intended use of the architecture document; Brian answers could be useful for outisder to understand how NC fits in
[16:09:46] <CedricAdjih> Victor comments step 4 also could be done before or same time as step 3
[16:10:25] <CedricAdjih> (Victor presents)
[16:16:10] <CedricAdjih> comments about the complexity of the building block view (Vincent, MarieJosé) - maybe it is too much
[16:22:30] <CedricAdjih> Brian commented on the difficulty of a general architecture as well
[16:23:16] <CedricAdjih> Morten proposes also to start rather from proposals and identifies blocks (rather from architecture)
[16:23:49] <CedricAdjih> Marie-José concurs
[16:28:24] Aaron Falk leaves the room
[16:29:18] Morten V. Pedersen leaves the room
[16:29:22] Morten V. Pedersen joins the room
[16:31:33] <CedricAdjih> Marie-José proposes a bottom-up approach for the next IETF starting from implementers
[16:37:59] Tadanori TERUYA leaves the room
[16:38:30] K Fouli leaves the room
[16:39:11] CedricAdjih leaves the room
[16:41:27] Senthil Sivakumar leaves the room
[16:46:30] Meetecho leaves the room
[16:52:05] Aaron Falk joins the room
[16:52:55] Aaron Falk leaves the room: Replaced by new connection
[16:55:07] Tadanori TERUYA joins the room
[17:04:31] Morten V. Pedersen leaves the room
[17:08:07] CedricAdjih joins the room
[17:11:58] CedricAdjih leaves the room: Replaced by new connection
[17:11:59] CedricAdjih joins the room
[17:28:02] Morten V. Pedersen joins the room
[17:31:29] Morten V. Pedersen leaves the room
[18:08:51] CedricAdjih leaves the room
[18:10:44] Tadanori TERUYA leaves the room
[21:23:13] Tadanori TERUYA joins the room
[21:38:15] Tadanori TERUYA leaves the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!