22:02:18 <gabrielhurley> #startmeeting horizon
22:02:20 <openstack> Meeting started Tue Sep 24 22:02:18 2013 UTC and is due to finish in 60 minutes.  The chair is gabrielhurley. Information about MeetBot at http://wiki.debian.org/MeetBot.
22:02:21 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
22:02:23 <openstack> The meeting name has been set to 'horizon'
22:02:29 <gabrielhurley> Hello folks
22:02:33 <julim> hi there
22:02:34 <david-lyle> Hello
22:02:34 <lsmola_> hello
22:02:38 <lblanchard> hello
22:02:46 <gabrielhurley> #topic overview
22:03:16 <gabrielhurley> So, the main business at hand is the RC
22:03:47 <gabrielhurley> The goal is to get our RC1 buglist down to zero within a week, cut the RC, and reopen trunk for Icehouse next week
22:04:03 <gabrielhurley> our velocity lately supports the idea that we can do this :-)
22:04:44 <gabrielhurley> the remaining 20-something bugs are split equally between in progress and confirmed, and none of them strike me as exceptionally hard problems, just quick diagnose-and-fix issues.
22:04:59 <gabrielhurley> so I think overall we're in good shape as long as we keep up the pace this week.
22:05:22 <gabrielhurley> two other items of business...
22:05:36 <gabrielhurley> 1. I've been asked to remind people about the OpenStack User Survey: http://www.openstack.org/blog/2013/09/participate-in-the-openstack-user-survey-by-september-30/
22:05:48 <gabrielhurley> the Horizon team has particularly valuable insight in this area
22:06:04 <gabrielhurley> even if we don't always consider ourselves "users", we consume more of openstack on a more regular basis than most people
22:06:08 <gabrielhurley> so make your opinions heard
22:06:28 <gabrielhurley> the results definitely matter to the foundation and to the TC/PTLs
22:08:11 <gabrielhurley> 2. PTL nominations are this week (most of you probably already know that). Myself and david-lyle are both candidates. I just wanna make sure everyone knows that I think having multiple people running for PTL is a great thing and shows the health of our community! Whether he is elected or I am we're all gonna keep working together, so I don't want anyone worrying about voting one way or the other. Vote for who you think
22:08:54 <david-lyle> +1
22:09:19 <gabrielhurley> Any questions on any of those items?
22:10:15 <gabrielhurley> Cool. Feel free to ask at any point if anything comes to mind.
22:10:19 <gabrielhurley> #topic bugs
22:10:43 <gabrielhurley> There are 6 unassigned bugs (I just claimed two). It'd be great to get assignees on those sooner rather than later just so we know everything's covered
22:10:51 <gabrielhurley> https://launchpad.net/horizon/+milestone/havana-rc1
22:11:12 <gabrielhurley> I don't have any specific items on the list that concern me. Does anyone else want to raise specific bugs?
22:12:37 <gabrielhurley> I'll take that as a no. ;-)
22:12:45 <gabrielhurley> again, if anything comes to mind, feel free to jump in at any point
22:12:50 <gabrielhurley> #topic open discussion
22:12:50 <david-lyle> I've finally been able to play with the trove panels on a live backend and finding issues, I think with the bug fixes I posted the base use case for instances will work, I have not tried backups yet
22:13:09 <gabrielhurley> david-lyle: I reviewed your patch earlier. It looked good to me.
22:13:42 <david-lyle> Thanks, I think we can do better in Icehouse, but this will allow it to work in Havana
22:14:14 <gabrielhurley> yep. this is a super-early-adopters feature
22:14:14 <lsmola_> gabrielhurley, if I may, I have one regarding https://blueprints.launchpad.net/horizon/+spec/ceilometer-hosts-measurement
22:14:22 <gabrielhurley> sure
22:14:30 <lsmola_> gabrielhurley, just a little introduction
22:14:57 <lsmola_> gabrielhurley, the Tuskar project just merged to tripleo, so we are kind of Tripleo-UI now
22:15:10 <gabrielhurley> yeah, I heard that in the project meeting before this
22:15:13 <gabrielhurley> seems sane
22:15:19 <gabrielhurley> there's certainly common interests
22:15:37 <lsmola_> gabrielhurley, now I was talking about tracking hardware in Horizon with mrunge
22:15:53 <gabrielhurley> the idea of a Hardware dashboard in the BP is good in my mind, as long as we can bring everything together meaningfully
22:16:05 <lsmola_> gabrielhurley, seems that you are planning some sort of infrastructure tab
22:16:06 <gabrielhurley> Triple-O/Tuskar is the missing piece to that
22:16:14 <gabrielhurley> me personally?
22:16:20 <lsmola_> well
22:16:26 <lsmola_> the Horizon
22:16:42 <david-lyle> I think that was part of the IA proposal jcoufal pu forward
22:16:45 <lblanchard> lsmola: that was an idea that jcoufal and I had :)
22:16:48 <david-lyle> s/pu/put
22:16:54 <lsmola_> we will be making such a tab in Tuskar
22:17:12 <david-lyle> sorry, and lblanchard
22:17:23 <lblanchard> david-lyle: no worries!
22:17:25 <lsmola_> and it seems, as part of tripleo, we will likely merge to Horiyon UI at some point
22:17:26 <gabrielhurley> nope. when Nebula firs started out we added a hardware dashboard to Horizon in our fork because it's very useful. Nebula no longer uses Horizon as our primary interface so that's gone now, but since OpenStack is gaining hardware capabilities and has a lot more data in that arena now, I think it's great to shift that into the open core of the dashboard.
22:18:06 <julim> +1
22:18:39 <david-lyle> It makes sense, and since it will be service availability and role based, should be a good addition
22:18:50 <gabrielhurley> I personally think "Hardware" is a preferrable term to "Infrastructure" just so we don't get muddled in Infrastructure as a Service terminology.
22:19:00 <lblanchard> agreed!
22:19:03 <gabrielhurley> david-lyle: totally
22:19:30 <gabrielhurley> though if we get into the tuskar side of managing the services themselves...
22:19:32 <lblanchard> this will hopefully be a big piece to the IA discussion
22:19:35 <gabrielhurley> hardware isn't quite right for that...
22:19:43 <gabrielhurley> I'll leave the naming for the IA discussion
22:19:43 <lsmola_> so my point is, as we will be implementing that in Tuskar, we could prepare it for the merge
22:20:12 <lsmola_> I just wanted to track, what hardware statistics are available in that BP
22:20:36 <gabrielhurley> isn't that more of a question for the ceilometer team?
22:20:46 <lsmola_> gabrielhurley, well
22:20:50 <gabrielhurley> the data-gathering isn't Horizon's business
22:21:00 <lsmola_> yes
22:21:08 <lsmola_> it regards the showing of the data
22:21:09 <gabrielhurley> I suspect it's going to vary by deployment and hardware as well
22:21:33 <lsmola_> that concerns several pages with hardware based statistics and alarms
22:22:01 <gabrielhurley> there are two ways to go: design for an ideal case and know how to pare things out of the design that don't exist yet; or design for what we know we can have as a base guaranteed set and know how we scale to N.
22:22:10 <lsmola_> My point is, whether it is a good idea to start it in Tuskar, and then merge it to Horizon
22:22:54 <david-lyle> I think you have to
22:22:54 <lsmola_> or the Horizon will go parallel way, at cost of some effort duplication
22:22:56 <gabrielhurley> I think in the IA discussion we should see how much of the existing admin dashboard belongs there, and that'll help determine whether we should start things from our end or yours
22:23:31 <david-lyle> or is it all ceilometer based data? and not tuskar?
22:23:41 <gabrielhurley> If we decide we should have that grouping already, then there's no sense in waiting on y'all. If we've got nothing without you, then we shouldn't start it on our end.
22:23:56 <gabrielhurley> david-lyle: my impression is both
22:24:09 <lsmola_> well
22:24:30 <david-lyle> ok, so the ceilometer based data makes sense, but tuskar would be a harder dependency to justify
22:24:33 <lsmola_> it should be possible to show just Ceilometer stats, if no Tuskar API is available
22:24:51 <lsmola_> if the Tuskar API will be available, it will show more stats + management
22:25:30 <gabrielhurley> regrouping what we know from Nova and Cinder and Ceilometer about hosts gives us a starting point. I'm just not sure if it's enough to justify our reorganizing things in I right yet.
22:25:50 <gabrielhurley> Tuskar/Triple-O won't even graduate until the end of I at best, so we'd be looking at using their data in J
22:26:38 <lsmola_> yeah my assumption was that we would make it in I
22:27:29 <gabrielhurley> I don't want to make a decision here without having the IA discussion and seeing some proposals around what this would look like with and without TripleO/Tuskar. It may make sense for us to start building it on our own in I and giving you hooks to add into, or it may not. I'm not gonna know that for sure 'til the summit.
22:27:52 <lsmola_> ok that make sense
22:28:02 <gabrielhurley> the general concept totally has my blessing though
22:28:07 <lsmola_> ok
22:28:19 <lsmola_> it was just for starting a discussion
22:28:21 <gabrielhurley> we'll keep following up on this regularly to make sure we do the right thing
22:28:25 <gabrielhurley> absolutely. thank nyou
22:28:46 <gabrielhurley> I never want to discourage people who are passionate to work on something :-)
22:29:01 <lsmola_> As i poke about it, i have noticed a lot of similar stuff, that could be worth doing on one place, rather then duplicating
22:29:19 <lsmola_> gabrielhurley, cool, thank you very much
22:29:25 <gabrielhurley> I totally agree
22:29:28 <lblanchard> I'd definitely hope we could reuse design and code no matter where it's implemented to start!
22:29:58 <gabrielhurley> I had hoped to do more IA improvement work in H but the quantity of new features/APIs/projects we had to support turned out to be way beyond what I'd expected
22:30:10 <gabrielhurley> this has been a heck of a release cycle
22:30:17 <gabrielhurley> OH
22:30:24 <gabrielhurley> that reminds me of something I was gonna mention
22:30:43 <gabrielhurley> I'm gonna put together the release notes for H this week, and they're pretty impressive.
22:31:02 <lsmola_> lblanchard, yeah but as the Tuskar will get merged to Horizon rather late, it might make sense to start implementing the common things to Horizon
22:31:12 <gabrielhurley> I had a call with the OpenStack marketing folks this morning to tell them about what they could advertise and prmote as the new hotness in H, and it was not a short list. :_0
22:31:18 <gabrielhurley> wrong smiley... :-)
22:31:32 <lsmola_> cool
22:31:47 <lblanchard> awesome!!
22:31:49 <gabrielhurley> So I'll get a review up for those release notes soon, and would love to get eyes on it to make sure I didn't miss stuff that people care strongly about
22:32:45 <gabrielhurley> that's all I've got. Anything else?
22:33:08 <amotoki> i am a topic about importing translation.
22:33:09 <lsmola_> lblanchard, will zou have time this week for some basic wireframes for statistic pages?
22:33:47 <lsmola_> amotoki, sorry, go ahead
22:33:58 <gabrielhurley> amotoki: please share
22:34:23 <amotoki> daisy and i are working on translation. i will propose a patch to import translation with 100% completed.
22:34:41 <amotoki> when is RC1 cut date?
22:35:07 <amotoki> i will make sure to import them before it.
22:35:09 <gabrielhurley> it's not a hard date
22:35:34 <gabrielhurley> the earliest possible would be Monday, but I really don't expect it until Wednesday or Thursday next week.
22:35:48 <amotoki> i will propose a patch this week.
22:35:54 <gabrielhurley> even if the RC1 list was all green I'd probably tell Thierry not to cut it until after we have our team meeting next week
22:36:25 <gabrielhurley> amotoki: how about this: open a bug for a "final" translation update and target it to RC1. we'll make sure it's the last one to get closed
22:36:41 <gabrielhurley> that way there's no way we'll cut the release without doing it and the translators have the most possible time
22:36:52 <amotoki> sounds nice to make sure to track them.
22:37:20 <amotoki> daisy prefer to including ~100% translated languages in RC1 and removing old translations.
22:37:36 <gabrielhurley> I will defer to the translation team on that decision
22:37:45 <amotoki> does it make sense to you all?
22:37:47 <gabrielhurley> and if users complain, I will direct them to the translation team. ;-)
22:37:54 <gabrielhurley> I've heard arguments on both sides
22:38:19 <gabrielhurley> I personally don't know which I think is worse: "my language just isn't available" or "my language is partially/badly translated and now I'm confused"
22:38:51 <amotoki> yeah, there are several discussion in i18n ml.
22:39:09 <gabrielhurley> I'm sure. As far as I can tell it's an ongoing debate in open source translation in general
22:39:25 <gabrielhurley> so somebody just makes a decision and we go with that until we're proven wrong
22:39:32 <gabrielhurley> I'm good either way
22:40:09 <amotoki> me too. daisy and i have a bit different opinion already :-(
22:40:18 <amotoki> anyway, i will keep on eyes to make sure to import them into each RC release.
22:40:27 <gabrielhurley> amotoki: all you need to do is find more people that agree with you then. ;-)
22:40:33 <gabrielhurley> sounds perfect
22:40:35 <gabrielhurley> thank you
22:40:43 <lblanchard> lsmola: In response to your question…I hope to post a set of wireframes for review to the UX community by the end of the week. They will be focused on visualizing details around nodes, racks, and resource classes…but my hope is that a lot of the concepts could be reused on the Horizon Overview pages eventually. So everyone, please feel free to comment and help me improve them!
22:41:13 <gabrielhurley> awesome!
22:41:16 <lsmola_> lblanchard, very cool, thank you
22:41:29 <lblanchard> lsmola: no problem! I enjoy it :)
22:41:49 <lsmola_> great :-)
22:41:54 <lblanchard> I should also give a shoutout to julim who has been helping me improve them!
22:42:11 <julim> thx lblanchard. happy to help!
22:42:20 <lsmola_> julim, thank you very much
22:43:01 <julim> hope the wireframes make sense to all when you all see them… lblanchard is doing a great job
22:43:11 <lsmola_> gabrielhurley, one last thing from me
22:43:11 <gabrielhurley> I was wondering who julim was... :-)
22:43:18 <gabrielhurley> go for it
22:43:32 <lsmola_> gabrielhurley, I have started the conversation about the Alarming in Horizon
22:43:41 <gabrielhurley> yep
22:43:49 <lsmola_> gabrielhurley, would be glad if you could look on it
22:44:12 <lsmola_> jd__ already gave me very helpful feedback
22:44:21 <gabrielhurley> will do
22:44:56 <lsmola_> And that's all from me, I am very glad that the Ceilometer integration is moving :-)
22:45:04 <julim> julim = Ju Lim at Red Hat
22:45:07 <lsmola_> gabrielhurley, thank you very much
22:45:12 <lblanchard> gabrielhurley: haha I should have introduced julim to the team :) She joined the UX team @ Red Hat a few weeks back and is focusing on use case / customer research work
22:45:18 <julim> part of UXD team
22:45:53 <julim> sorry… have 2 young kids fighting for keyboard access… limited typing now
22:46:06 <lsmola_> hehe
22:46:09 <lblanchard> lol
22:46:20 <gabrielhurley> lsmola_: was there something particular about the alarms? The blueprint looks pretty straightforward.
22:46:32 <lsmola_> gabrielhurley, it's on mailing list
22:46:37 <gabrielhurley> julim: welcome
22:46:49 <gabrielhurley> lsmola_: ah, I missed that one somehow. Lemme go find the message.
22:46:59 <lsmola_> gabrielhurley, cool, thank you
22:47:15 <gabrielhurley> weird. my mail filter for horizon messages on the list missed that one.
22:47:18 <gabrielhurley> I see it now
22:47:31 <lsmola_> maybe ceilometer stole it :-)
22:47:53 <gabrielhurley> haha
22:47:54 <gabrielhurley> dunno
22:47:58 <gabrielhurley> I'll respond to it though
22:48:07 <gabrielhurley> lblanchard: thanks for the intro
22:48:10 <lsmola_> that would be cool, thank you
22:48:10 <gabrielhurley> julim: glad to have you
22:48:18 <julim> glad to be part of the team.
22:48:30 <gabrielhurley> okay. other topics?
22:50:36 <gabrielhurley> cool. well, good meeting folks. happy bug hunting and we'll check in again next week!
22:50:48 <gabrielhurley> hpoefully we'll be cutting our Havana RC!
22:50:55 <gabrielhurley> #endmeeting