14:00:50 <sgordon> #startmeeting telcowg
14:00:51 <openstack> Meeting started Wed May  6 14:00:50 2015 UTC and is due to finish in 60 minutes.  The chair is sgordon. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:52 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:55 <openstack> The meeting name has been set to 'telcowg'
14:00:59 <sgordon> #link https://etherpad.openstack.org/p/nfv-meeting-agenda
14:01:15 <DaSchab> hi
14:01:20 <aveiga> hello
14:01:31 <cloudon> hi
14:01:43 <matrohon> hi
14:01:56 <sgordon> hi all, apologies i will be a little distracted just for the start today
14:02:11 <dcw> hi
14:02:47 <sgordon> #topic use cases
14:02:51 <mkoderer> hi
14:02:57 <sgordon> #link https://review.openstack.org/#/c/169201/
14:03:12 <sgordon> #info Ralf updated service chaining use case to integrate feedback.
14:03:19 <sgordon> more reviews welcome there
14:03:27 <sgordon> and similarly
14:03:30 <sgordon> #link https://review.openstack.org/#/c/158997/
14:03:41 <sgordon> #info Calum updated vIMS use case to integrate feedback.
14:03:49 <sgordon> has anyone had a chance to look at these updates
14:03:52 <DaSchab> but status is Abandoned
14:04:02 <sgordon> DaSchab, wrong link?
14:04:12 <DaSchab> https://review.openstack.org/#/c/169201/
14:04:20 <aveiga> sgordon: no, it lists abandoned
14:04:32 <DaSchab> sorry copy and paste error
14:04:33 <DaSchab> https://review.openstack.org/#/c/158997/
14:04:46 <aveiga> ah
14:05:15 <cloudon> well, certainly didn't mean to abandon it...
14:05:36 <mkoderer> we can restore the change
14:06:01 <cloudon> try https://review.openstack.org/#/c/179142/
14:06:12 <aveiga> there's the right link
14:06:23 <mkoderer> cloudon: ah yep :)
14:07:19 <cloudon> know I have one set of feedback to consider but more welcome
14:09:35 <DaSchab> sorry, I have to leave
14:10:16 <mkoderer> sgordon pushed a document for end2end workflow
14:10:19 <mkoderer> https://review.openstack.org/#/c/178347/3
14:10:31 <mkoderer> I think we all have to have a look at this
14:11:38 <cloudon> it's just a skeleton at the moment, isn't it?
14:11:43 <aveiga> so I noticed something missing from that doc and something we never really talked about here
14:11:47 <cloudon> thought Steve wanted to flesh it out more bfore reivew?
14:11:57 <aveiga> what if someone comes along with a similar but slightly different use case?
14:12:19 <mkoderer> aveiga: we should merge things into one
14:12:24 <aveiga> I was thinking it might be better to add on to the existing use cases as a "separate example" so that we can keep the efforts together
14:12:28 <aveiga> mkoderer: +1
14:12:29 <mkoderer> aveiga: we can use the co/author flag for it
14:14:05 <aveiga> does anyone else agree/disagree with the merging method?
14:14:26 <cloudon> think this overlaps with some feedback I've had on my use cases - that I'm describing a specific instance of an SBC or vIMS rather than general implementations of them
14:14:51 <aveiga> cloudon: that's ok
14:14:57 <cloudon> have to say I prefer the model where use cases are as concrete as possible - think it provides greater developer impact
14:15:12 <aveiga> I don't think anyone would be able to cover fully an entire set of uses around a technology, so we have to start with something
14:15:19 <cloudon> "a theoretical app might need these things" vs. "I want to run app X and it needs Y and Z"
14:15:21 <adrian-hoban> +1 to the merging method proposed
14:15:45 <rprakash> #info rprakash
14:16:41 <sgordon> im down with merging
14:16:52 <sgordon> #info it might be better to add on to the existing use cases as a "separate example" so that we can keep the efforts together
14:18:16 <sgordon> #chair aveiga
14:18:17 <openstack> Current chairs: aveiga sgordon
14:18:34 <sgordon> #info use co-author to designate multiple authors
14:18:57 <sgordon> i think merging them makes sense in terms of the goals of this group
14:19:14 <sgordon> speaking with one voice to the wider community when proposing features to support this
14:19:23 <rprakash> can get the process of converting an use case to from text to gerrit review process? any links?
14:19:32 <sgordon> i think we can have a breakdown within the document though saying "hey these are specific examples"
14:20:14 <sgordon> rprakash, https://wiki.openstack.org/wiki/TelcoWorkingGroup/UseCases#Contributing_Use_Cases
14:20:21 <rprakash> thanks
14:20:41 <aveiga> rprakash: copy the template from that repo and make a new file with your use case
14:20:55 <sgordon> #action sgordon to highlight in workflow document how/when to collapse use cases into each other
14:21:00 <aveiga> we follow the same workflow as the specs for regular projects
14:21:16 <sgordon> yeah, that page is mainly a pointer to the "normal" how to contribute docs
14:21:24 <rprakash> ok
14:21:59 <sgordon> if you need any help please ask in #openstack-nfv - i dont proclaim to be a git master but hey we did at least help cloudon pick up the vIMS patch (eventually ;))
14:22:00 <sgordon> :)
14:22:19 <cloudon> (cough)
14:22:40 <sgordon> ok so the workflow document was mentioned above
14:22:44 <sgordon> and yes it is just a skeleton
14:22:49 <sgordon> #link https://review.openstack.org/#/c/178347/3/doc/source/workflow.rst
14:22:58 <sgordon> i think it's important to note i dont want to dictate here though
14:23:06 <sgordon> so even in skeleton form feedback is important
14:23:19 <sgordon> the rough process atm is...
14:23:20 <aveiga> sgordon: I don't mind checking in a patchset, as long as we don't break each others' git again
14:23:24 <sgordon> 1) submit use case
14:23:28 <sgordon> 2) review use case
14:23:33 <sgordon> 3) approve use case
14:23:45 <sgordon> 4) raise tracker bugs (telcowg project)
14:23:52 <sgordon> 5) assign tracker bugs
14:24:01 <sgordon> 6) create project bug/blueprints
14:24:05 <sgordon> 7) implement
14:24:09 <sgordon> 8) verify solution
14:24:13 <sgordon> EOM
14:24:21 <mkoderer> sgordon: oh we have to activate the bug tracker then :)
14:24:24 <sgordon> the key i think is who owns each piece
14:24:33 <aveiga> mkoderer: yes we do
14:24:39 <sgordon> and any additional info we want to highlight in each step
14:24:43 <aveiga> we're going to need to be able to keep track
14:24:47 <sgordon> yeah
14:24:51 <mkoderer> ok
14:25:20 <sgordon> i think it's important as we will invariably have a single gap coming from multiple use cases that in turn has multiple bits and pieces to be implemented across multiple projects
14:25:26 <sgordon> we need to keep track of that linkage somehow
14:25:34 <adrian-hoban> Will every telcowg tracker bug need to be related to a use case?
14:25:45 <aveiga> and be able to refer back to individual gaps if we find the same ones in multiple cases
14:25:52 <aveiga> file a bug, mark duplicate of another gap bug
14:26:01 <sgordon> right
14:26:06 <sgordon> adrian-hoban, currently yeah
14:26:17 <aveiga> adrian-hoban: probably, since we need to be able to trace where they came from
14:26:19 <sgordon> adrian-hoban, but it depends what is presented as a reason to have bugs that aren't
14:26:45 <sgordon> the only reason we are planning to use the bug tracker at all atm is to trace the linkage between project-specific work items and the original use case
14:27:08 <sgordon> raising bugs for other reasons isn't part of the workflow atm
14:27:12 <sgordon> open to proposals
14:27:53 <sgordon> adrian-hoban, was there something specific you were thinking of?
14:28:06 <aveiga> sgordon: do we want a master bug for "Gap analysis on use case X" that gets closed when bugs are filed or the use case is found to be supported by other fixes?
14:28:28 <sgordon> that actually makes a lot of sense
14:28:35 <aveiga> I just don't want us to lose a use case and forget to do the analysis
14:28:45 <sgordon> i have been thinking that step where we take a merged use case and raise the gap bugs is where there is a bit of work to do
14:28:54 <aveiga> yup
14:29:31 <sgordon> #action sgordon update workflow to include "master bug" for "Gap analysis on use case X" that gets closed when bugs are filed or the use case is found to be supported by other fixes.
14:31:38 <sgordon> adrian-hoban, sorry we were asking if you had a specific example
14:31:45 <sgordon> w.r.t. "<adrian-hoban> Will every telcowg tracker bug need to be related to a use case?"
14:31:54 <adrian-hoban> Sorry, I got disconnected
14:32:13 <sgordon> currently we've focused on creating bugs to track the linkage between use cases and gaps (assuming a M:1 relationship)
14:32:22 <adrian-hoban> Nope, not yet, but was thinking about the case where something may come up in OPNFV that may not have a use case yet defined here
14:33:05 <aveiga> adrian-hoban: that's more a matter of getting OPNFV reqs down to this group in the form of a use case doc, or updates to one
14:33:52 <sgordon> yeah, what i had put forward to the opnfv m/l was the idea of them using the same template for defining their use cases
14:34:03 <sgordon> and collaborating on any required updates to it
14:34:19 <adrian-hoban> aveiga: I think that's an expectation that has to be clarified.
14:34:53 <aveiga> adrian-hoban: yes, see sgordon's reply.  It has to be communicated, but I think he's taking care of that
14:34:55 <adrian-hoban> aveiga: I think that's an expectation that has to be clarified/set with the OPNFV folks so that the expectations on how to engage are clear.
14:35:04 <aveiga> agreed
14:35:12 <sgordon> im just searching to necro that thread again
14:35:22 <sgordon> but this would be a good item to discuss at the OPNFV day on the monday at summit
14:35:52 <aveiga> sgordon: +1
14:36:00 <adrian-hoban> sgordon: +1
14:36:04 <aveiga> is there a topic list or agenda for that item, btw?
14:36:31 <aveiga> or is it more of a walk up and ask questions ala pods kind of thing?
14:36:33 <sgordon> not yet
14:36:34 <sgordon> #link http://meetbot.opnfv.org/meetings/opnfv-meeting/2015/opnfv-meeting.2015-04-29-13.00.html
14:36:38 <adrian-hoban> aveiga: I was going to ask that too. I haven't seen one yet, but there are some proposals in the OPNFV community
14:36:48 <sgordon> in theory: 30 minutes high level overview, 30 minutes project overview, 30 minutes set the tone for the rest of the week (the session the telco WG asked for)
14:36:58 <aveiga> ok
14:37:01 <sgordon> then breakouts for OPNFV projects
14:37:11 <sgordon> the timing of the telco wg one is being played with a bit i think
14:37:21 <sgordon> as i have a presentation that clashes with the opnfv day start
14:37:31 <sgordon> (unrelated, one of my nova compute update sessions)
14:38:32 <cloudon> is the desire still to whizz through the use cases during the TelcoWG slot?
14:38:48 <sgordon> yeah so we have a 90 minute session effectively again
14:38:55 <sgordon> (i think 2 x 40 with a 10 min break or something)
14:39:15 <adrian-hoban> Is there an Etherpad for the TelcoWG slot?
14:39:35 <aveiga> #topic Vancouver Summit
14:39:36 <sgordon> atm just the section in https://etherpad.openstack.org/p/YVR-ops-meetup
14:39:56 <sgordon> line 131
14:40:01 <sgordon> #link https://etherpad.openstack.org/p/YVR-ops-meetup
14:40:21 <sgordon> that was just my notes from an earlier discussion in this meeting
14:40:35 <sgordon> i think 40 min on updates/process and 40 min on use cases likely makes sense
14:40:51 <sgordon> possibly in the form of breaking off into smaller groups
14:41:25 <adrian-hoban> What session is this a part of?
14:41:53 <sgordon> it's a session in the ops track
14:42:00 <aveiga> adrian-hoban: there's a session for Telco WG specifically
14:42:25 <sgordon> #link https://openstacksummitmay2015vancouver.sched.org/event/e813504daf6df803d90836e8949a0562
14:42:36 <sgordon> #info Telco WG Ops session - Wednesday, May 20 • 9:00am - 10:30am
14:43:02 <adrian-hoban> Thanks!
14:43:36 <sgordon> #action sgordon create session-specific etherpad for ops summit session
14:45:49 <sgordon> ok
14:46:09 <sgordon> thank you all for your time
14:46:40 <sgordon> i think we will have one last meeting before summit, the goal of which should be to finalize the above and send out a quick email to highlight it imo
14:47:06 <aveiga> thanks, sgordon
14:47:11 <sgordon> #endmeeting