20:01:28 <david-lyle> #startmeeting Horizon
20:01:29 <openstack> Meeting started Wed Feb 11 20:01:28 2015 UTC and is due to finish in 60 minutes.  The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:01:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:01:33 <openstack> The meeting name has been set to 'horizon'
20:01:40 <david-lyle> hello everyone, who's around
20:01:44 <gary-smith> hi
20:01:49 <jgravel_> hello
20:01:54 <robcresswell> Hi again
20:01:54 <gugl2> hi
20:01:56 <sambetts> Hey!
20:02:00 <lhcheng_> o/
20:02:04 <TravT> hello!
20:02:09 <crobertsrh> hello/
20:02:26 <mrunge_afk> hey!
20:02:44 <TravT> mrunge: lol
20:02:50 <TravT> i was just thinking that exact thing
20:03:12 <david-lyle> so general items...
20:03:26 <david-lyle> we posted k-2 last Wed
20:03:56 <david-lyle> 9 bps implemented and none of them were above medium
20:04:21 <david-lyle> plenty of good things made it, but it really leaves us over-tasked for k-3
20:04:28 <david-lyle> https://launchpad.net/horizon/+milestone/kilo-3
20:04:34 <david-lyle> which needs major pruning
20:05:08 <david-lyle> we really need to target the highs for k-3
20:05:14 <david-lyle> getting them in
20:05:24 <david-lyle> the glance v2 maybe not as much
20:05:55 <TravT> jpichon: is leaving this week and it didn't get as far along as hoped
20:06:31 <david-lyle> i'm going to prune the list and shift priorities a bit
20:06:46 <david-lyle> but for reviewers, highs are the target
20:07:04 <david-lyle> mediums will be very good to have, and lows are opportunistic
20:07:12 <sqchen> david-lyle: we need approval on the plug-in extension to more launchinstance towarrd k-3
20:07:17 <david-lyle> so again priorities will likely shift
20:07:39 <david-lyle> hold on sqchen we'll get there
20:08:11 <david-lyle> any questions about any of that?
20:08:49 <david-lyle> I assume everyone has heard by now, but next release is Liberty
20:08:55 <david-lyle> so there's that
20:08:59 <mrunge> do we now have some common review guidelines for angular based blueprints?
20:09:26 <david-lyle> mrunge: we have coding guidelines for JS
20:09:36 <mrunge> we have?
20:09:42 <mrunge> cool.
20:09:45 <david-lyle> that address angular as well. Beyond that not really
20:09:49 <david-lyle> tqtran: link?
20:09:55 <david-lyle> rbertram:
20:09:56 <david-lyle> ?
20:09:57 <tqtran> let me look it up
20:10:00 <mrunge> would help to know that
20:10:03 <TravT> https://wiki.openstack.org/wiki/Horizon/Javascript#AngularJS
20:10:06 <david-lyle> thanks :)
20:10:14 <rbertram> right
20:10:32 <r1chardj0n3s> we need something in there about module namespace I think
20:10:33 <tqtran> dang Travis beat me to it
20:10:38 <TravT> it isn't very fully, i think.
20:10:48 <david-lyle> fully?
20:10:50 <tqtran> yeah, there are things missing in there
20:10:52 <TravT> full
20:10:53 <rbertram> yeah, we just got it started
20:10:56 <mrunge> you mean, guidelines and complete?
20:11:06 <TravT> yes, there are some things that could be added.
20:11:18 <rbertram> the angularjs guidelines - we just got the ball rolling
20:11:31 <tqtran> like naming modules, docs, tests, etc...
20:11:36 <tqtran> file structure
20:11:49 <david-lyle> so, just a few things...
20:11:56 <mrunge> hint: if you want your code being reviewed, complete guides will help there
20:11:57 <tqtran> yeah just a "few"
20:11:59 <rdopiera> sorry for being late
20:12:14 <david-lyle> mrunge: we need to shape that
20:12:19 <sqchen> I think how plugin extension works is part of it.
20:12:47 <TravT> some patterns in horizon are getting worked out over the course of this early development.
20:12:48 <sqchen> the extension is about angular
20:13:19 <TravT> but i think to Dave's next couple of topics, we should undertake improving those guidelines.
20:13:20 <david-lyle> it's unrealistic to think we'll have all the answers and a perfect implementation on the first pass, we need something to iterate on
20:13:36 <tqtran> yep, they are brand spanking new, so we haven't add to docs yet
20:13:45 <david-lyle> what we want to block is decisions that are going to make that difficult
20:14:08 <david-lyle> ok, let's move to the agenda, since we are
20:14:20 <david-lyle> agenda can be found at https://wiki.openstack.org/wiki/Meetings/Horizon
20:14:39 <david-lyle> #topic Angular work status
20:14:52 <david-lyle> So there has been a lot of good work in this area
20:15:11 <david-lyle> and the beginning of the API layer for angular code is starting to merge
20:15:23 <david-lyle> a reusable wizard widget has also merged
20:15:38 <TravT> help panel merged
20:15:38 <david-lyle> but we need more reviews and consensus
20:15:46 <david-lyle> TravT: thanks, yes
20:16:03 <tqtran> table is in the process of merging as well
20:16:09 <david-lyle> Most of the angular patches have demo patches that go along with them
20:16:17 <david-lyle> which is extremely helpful
20:16:20 <david-lyle> thanks for posting those
20:16:56 <david-lyle> but the are a couple of larger things blocking merging more of the clientside code
20:17:03 <david-lyle> well 3 things
20:17:12 <david-lyle> more of the REST API layer merging
20:17:39 <david-lyle> a way to get configuration settings to the client without exposing too many internal details
20:17:53 <david-lyle> and an entry point to the policy engine
20:18:11 <david-lyle> the first, I think most of is progress if not just waiting for reviews
20:18:24 <david-lyle> the second, tqtran has started and needs to fine tune
20:18:43 <david-lyle> and the third, I will start to address
20:19:02 <david-lyle> the good news is, I think the launch instance work can progress without policy for now
20:19:14 <david-lyle> the identity panels are another story
20:19:48 <david-lyle> but I feel like the progress is moving too slowly to make k-3 at this point.
20:19:53 <david-lyle> not due to work on patches
20:19:59 <tqtran> the identity panels internal code itself is almost ready, we just need the policy check in place
20:20:19 <david-lyle> but review lag, concensus building, design decisions
20:20:39 <david-lyle> to address that, I'd like to propose a virtual sprint
20:20:53 <david-lyle> #link https://wiki.openstack.org/wiki/VirtualSprints
20:21:01 <david-lyle> essentially a sprint without the travel
20:21:28 <david-lyle> but a way to get people together and heads down for an extended period of time
20:21:49 <david-lyle> part would be IRC, but I would advocate a hangout as well going on in concert
20:21:57 <tqtran> +1 I like the idea
20:22:11 <sambetts> +1
20:22:13 <sqchen> +1
20:22:15 <david-lyle> I think if we try to do something in person, we will miss k-3 for sure
20:22:17 <TravT> +1, also it is basically what is going on.  this just helps to formalize it
20:22:20 <rbertram> +1 but what is number of people limited in hangout?
20:22:24 <ongk> +1
20:22:41 <tqtran> rbertram: we'll find out =)
20:22:51 <david-lyle> I will post an etherpad as determine interest
20:22:57 <TravT> 15
20:23:02 <david-lyle> please sign up if you are actually interested
20:23:14 <mrunge> yeah 11-15 is hangout limit
20:23:25 <mrunge> not sure, if a hangout really helps
20:23:48 <david-lyle> we can decide on the duration, but I would advocate 4-5 hours for a couple of days at least
20:23:52 <rbertram> I'm a big fan of hangouts - not just see eachother, but share demos - also code walkthroughs
20:23:59 <david-lyle> mrunge: voice helps a lot
20:24:08 <david-lyle> whether it's hangouts or not
20:24:14 <tqtran> makes explaining things easier
20:24:15 <david-lyle> visual sharing helps too
20:24:24 <TravT> yes, hangouts are very helpful.
20:24:33 <david-lyle> I don't really care what the technology is
20:24:53 <david-lyle> but typing in IRC can take forever
20:25:21 <mrunge> depends on the noise in the channel, I'd say
20:25:26 <sqchen> hangouts is helpful
20:25:26 <mrunge> but let's try it out
20:25:38 <TravT> HP has a tool that scales higher that we can use as well, but would rather not force another plugin
20:25:38 <david-lyle> people type slow and their input gets lost
20:25:52 <ongk> could always do a hangout for each topic
20:25:59 <david-lyle> I just want people to participate and communication to be clear
20:26:01 <TravT> to that end, I want to invite people to a hangout right after this IRC.
20:26:07 <TravT> we are talking about the rest API
20:26:24 <TravT> with code discussion
20:26:54 <TravT> if you are interested PM me.
20:27:07 <TravT> i'd rather not have the room in the logged channel right now
20:27:12 <david-lyle> #link https://etherpad.openstack.org/p/horizon-kilo-virtual-sprint
20:27:14 <TravT> i need to create a dif one
20:27:51 <david-lyle> I want this to happen soon. Thinking next week
20:27:56 <david-lyle> or late this week
20:28:02 <tqtran> or we can start an unofficial one this week
20:28:26 <david-lyle> but people with managers may need a little time to coordinate dedicated time
20:28:31 <mrunge> so, you're basically taking this to a non-public room, is that right?
20:28:35 <TravT> no
20:28:38 <TravT> it is public
20:28:44 <mrunge> you said, you want no log
20:28:50 <mrunge> as example
20:28:53 <tqtran> if people are interested, they can join, we can post link in etherpad?
20:28:54 <TravT> just i need to create the new room url
20:28:55 <david-lyle> there is a sprint IRC room
20:29:02 <TravT> for the hangout
20:29:10 <robcresswell> Me and Sam have time, for those also EU based
20:29:25 <david-lyle> mrunge: the reality is in face sprints aren't logged either
20:29:40 <david-lyle> this is really just about getting unstuck
20:29:50 <david-lyle> summits aren't either
20:30:14 <david-lyle> but we will have the etherpad for content, IRC for logs, and patches with the code as a result
20:30:29 <mrunge> well, you're just creating small circles, excluding folks, who can not attend at that time,for whatever reason
20:30:51 <david-lyle> mrunge: no I'm not
20:31:01 <david-lyle> I'm asking interested parties to join
20:31:01 <mrunge> it's the proposal to do so
20:31:08 <david-lyle> no it's not
20:31:10 <david-lyle> listen
20:31:15 <david-lyle> it's open
20:31:21 <david-lyle> it's a way to coordinate time
20:31:28 <david-lyle> it's not exclusive
20:31:47 <david-lyle> time has not been decided nor dates
20:31:47 <mrunge> it's separate from official
20:31:55 <david-lyle> so are mid-cycles
20:31:55 <mrunge> and limited to 10-15 participants
20:32:06 <david-lyle> mrunge: not necessarily
20:32:13 <tqtran> mrunge: we can have multiple hangouts according to topic
20:32:23 <david-lyle> hangouts is one tool, there are hundreds of possible tools
20:32:41 <tqtran> i doubt we'll need more than 15 people in one hangout, anymore than that and it'll start to get distractive anyhow
20:32:49 <mrunge> let's move forward
20:33:05 <david-lyle> but there are two advantages to in-person sprints, 1 coordinated time together (focused), 2 you aren't typing everything
20:33:19 <david-lyle> and 3 social actities
20:33:24 <david-lyle> *activities
20:33:28 <david-lyle> we won't have 3
20:34:01 <david-lyle> we also won't require travel, time away from families, or the expense
20:34:58 <sqchen> also we can demo.
20:35:21 <david-lyle> I don't think we can afford another cycle of almost angular
20:35:38 <david-lyle> people will lose interest, me included
20:35:44 <mrunge> yupp
20:35:45 <david-lyle> :D
20:36:03 <btully> “almost angular”?
20:36:19 <btully> (sorry i’m new here) ;)
20:36:32 <tqtran> ok then its decided, lets move forward
20:36:33 <david-lyle> btully: you are new, but we've been building to allow angular for 3 releases now
20:37:06 <btully> gotcha, thanks
20:37:18 <david-lyle> it's amazing how hard it is to get tools approved and in, and an agree path forward with distros, devs, deployers, etc
20:37:31 <lhcheng_> david-lyle: how about sending out the etherpad link in the ML in spirit of open-ness?
20:37:51 <david-lyle> lhcheng_: I will. I will also post it on the virtual sprints wiki page
20:38:04 <david-lyle> I created it about 10 minutes ago, so haven't had a chance
20:38:11 <david-lyle> but good suggestion
20:38:11 <TravT> ok, I have the hangout url created.  anybody is welcome after this meeting.  topic is rest api with walkthrough. if you can't get in. PM me and we'll have to switch to the HP room that allows more people.
20:38:22 <lhcheng_> david-lyle: great!
20:38:29 <TravT> https://plus.google.com/hangouts/_/gv53nv5va2nh4xhekedzhbarhea
20:38:39 <mrunge> TravT, thanks. can we have it recorded?
20:38:50 <david-lyle> TravT: ask Piet about that one
20:38:54 <TravT> mrunge: yeah, i'll try
20:39:11 <mrunge> thanks! (It's almost 10 in the evening here...)
20:39:12 <david-lyle> we need times and dates
20:39:37 <david-lyle> just a hanging open room is not much of an advantage over IRC
20:39:46 <mrunge> ...
20:40:02 <david-lyle> ok
20:40:08 <TravT> i like the weekly irc scheduled
20:40:17 <TravT> but often the hangouts are more ad-hoc
20:40:24 <david-lyle> #action mail/post sprinty stuff
20:40:35 <david-lyle> #undo
20:40:36 <openstack> Removing item from minutes: <ircmeeting.items.Action object at 0x9937350>
20:40:56 <david-lyle> #action david-lyle mail/post sprinty stuff
20:41:28 <david-lyle> #topic Summit planning
20:41:48 <david-lyle> so the summit format is changing a bit
20:42:06 <david-lyle> see #link http://lists.openstack.org/pipermail/openstack-dev/2015-January/054122.html
20:42:38 <david-lyle> there is not the choice of traditional summit sessions and smaller sprint sessions
20:42:51 <david-lyle> And the friday sprint will still be held
20:42:58 <david-lyle> half or full day
20:43:16 <mrunge> any option to have a sprint earlier?
20:43:18 <david-lyle> additionally, the pods are going away
20:43:36 <david-lyle> mrunge: yes
20:43:54 <mrunge> I mean, on last summits, the most valuable and productive day was Friday
20:44:00 <david-lyle> the choice of the first two is where normal sessions were tradionally held
20:44:08 <david-lyle> *traditionally
20:44:25 <david-lyle> so structure will be:
20:44:33 <david-lyle> 1st day: cross-project topics
20:45:00 <david-lyle> 2nd and 3rd day: mix of traditional design sessions and sprint sessions
20:45:01 <tqtran> mrunge: I agree, Friday was the most productive.
20:45:17 <david-lyle> 4th day: half or full day sprint
20:45:30 <david-lyle> so we have two things to decide
20:45:51 <david-lyle> how many sessions do we want, and should they be sprint or traditional
20:46:07 <david-lyle> and on Friday do we want a full day, or just a half
20:46:16 <david-lyle> how much together is too much?
20:47:17 <david-lyle> I think we should probably have 1 or 2 traditional sessions, but I'm open beyond that
20:47:23 <TravT> +1 to the idea as a whole
20:47:41 <david-lyle> We have to decide a bit earlier than usual because it will effect room layouts
20:47:49 <TravT> how soon?
20:47:57 <mrunge> like: today?
20:48:32 <david-lyle> #link https://etherpad.openstack.org/p/horizon-liberty-summit
20:48:53 <david-lyle> not today, not sure the exact timeline, but I've been asked to provide estimates
20:49:08 <mrunge> I would like to see an upcoming cycle session as a sprint (or so) as early as possible
20:49:36 <mrunge> just to use the rest of the summit to get that started or pushed earlier
20:49:58 <david-lyle> mrunge: I agree
20:50:23 <TravT> mrunge: david-lyle: this dovetails into a topic i added to the agenda
20:50:41 <TravT> david-lyle: may I?
20:50:46 <david-lyle> just a sec
20:50:51 <mrunge> ideally, we could separate into smaller groups for the rest of day 2 and the next few days to actually drive it
20:51:20 <david-lyle> so I think this will be an email thing, or etherpad driven exercise
20:51:33 <david-lyle> #action david-lyle post more email about summit ideas
20:51:53 <david-lyle> #topic Pre-Summit Meetup (TravT)
20:51:55 <david-lyle> ok TravT
20:52:02 <TravT> We (HP) would like to host a couple day pre-summit contributor meetup between the end of K-3 and the summit.
20:52:11 <TravT> There are actually about two months between the last day of K-3 and the design summit that we could use to get momentum rolling for the "L"iberty release.
20:52:38 <TravT> This would help us to really prep for the summit and feed into the summit sessions. It may also help to decide which topics need to be fishbowl topics at the summit.
20:53:06 <mrunge> TravT, so, something right after Kilo release?
20:53:06 <TravT> The goal would be for the meetup to have some contributor focused sessions on some big topics
20:53:17 <TravT> yeah, so k3 is like March 19th
20:53:24 <TravT> summit is May 2?
20:53:44 <TravT> I’ve started an ether pad / doodle to start getting input on participation interest: https://etherpad.openstack.org/p/horizon-liberty-presummit-meetup
20:53:45 <mrunge> May 18-22
20:54:17 <TravT> Demos, white boarding, code sharing, and sub group breakouts would be encouraged.
20:54:23 <david-lyle> TravT: where are you thinking?
20:54:47 <TravT> Primary locations my management is interested in are Fort Collins, CO or Seattle, WA
20:55:13 <r1chardj0n3s> this kind of meetup would have been useful this month :/
20:55:36 <r1chardj0n3s> I'm unsure it'll be that useful immediately after k3
20:55:54 <tqtran> haha that is very true
20:55:59 <david-lyle> r1chardj0n3s: it would, but you never really know what stage you'll be at when determining to schedule
20:56:09 <TravT> if people want to do it sooner...
20:56:13 <david-lyle> I fully expected us to be a lot further
20:56:21 <r1chardj0n3s> it'd end up discussing stuff we would need to discuss at the L summit anyway
20:56:41 <mrunge> r1chardj0n3s, yes, right
20:56:45 <TravT> well, if we look at many releases
20:57:07 <r1chardj0n3s> BTW, I cannot attend on the dates in question - PyCon is right in the middle
20:57:10 <TravT> if the branch is cut at k3, why can't we start working seriously towards the next release pre-summit
20:57:43 <david-lyle> TravT: we can and the summit always feels too late
20:58:06 <david-lyle> you're theoretically doing your planning halfway through the first milestone
20:58:16 <david-lyle> oops
20:58:47 <TravT> maybe we'll start getting things merged sooner in the cycle.
20:59:04 <mrunge> TravT, that is usually the case
20:59:12 <mrunge> this cycle was.... different
20:59:29 <david-lyle> the reality of the situation is that the same people are finishing K that are looking to L
20:59:30 <mrunge> in the past, we had larger portions merged for milestone-1
20:59:46 <david-lyle> which makes parallel paths difficult
20:59:47 <TravT> i just put up some proposed dates
20:59:54 <TravT> others can be added to etherpad
21:00:02 <TravT> and we can continue this discussion
21:00:06 <david-lyle> Thanks TravT
21:00:27 <david-lyle> looks like we've reached our time
21:00:28 <mrunge> TravT, you're proposing to meet for a full week?
21:00:35 <TravT> that is up for debate as well
21:00:35 <david-lyle> 4 weeks
21:00:44 <TravT> i was thinking 6 weeks in tahiti.
21:00:52 <mrunge> +1
21:01:00 <TravT> but a few days at least
21:01:05 <mrunge> and HP is willing to sponsor that?
21:01:10 <mrunge> ;-)
21:01:11 <david-lyle> Thanks HP!
21:01:16 <TravT> lol
21:01:17 <mrunge> yes, thanks
21:01:32 <david-lyle> have to get up to a billion you know
21:01:40 <mrunge> count me in, Barbados would be cool, too
21:01:51 <mrunge> ... (or whatever)
21:02:06 <david-lyle> Thanks everyone, look at the sprint links and provide input.
21:02:09 <david-lyle> #endmeeting