23:03:18 <reed> #startmeeting openstack-community
23:03:19 <openstack> Meeting started Wed Aug 28 23:03:18 2013 UTC and is due to finish in 60 minutes.  The chair is reed. Information about MeetBot at http://wiki.debian.org/MeetBot.
23:03:20 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
23:03:23 <openstack> The meeting name has been set to 'openstack_community'
23:03:41 <fifieldt> #link https://wiki.openstack.org/wiki/Meetings/Community#Agenda_for_next_meeting
23:03:47 <reed> #topic review action times from last meeting
23:04:29 <reed> I have contacted Hui Cheng, Bruce Lok and Yujie Du to recruite Ask moderators in ZH
23:04:40 <fifieldt> great!
23:04:57 <reed> no reply from Cheng, he must be busy with his startup
23:05:07 <reed> the others are interested and happy to help
23:05:12 <fifieldt> good news
23:05:46 <reed> I think it would be good to setup a training session with them and other volunteer moderators, not just rely on each of them reading the wiki page
23:05:48 <reed> thoughts?
23:06:12 <fifieldt> sounds good
23:06:16 <fifieldt> what format are you thinking?
23:06:38 <fifieldt> getting them on the live system working on real questions?
23:06:50 <reed> i am not sure what would be optimal
23:07:00 <reed> that sounds like a good idea actually
23:07:21 <reed> I was thinking also to invite moderators in other Q&A sites, like askubuntu or ask fedora
23:07:22 <fifieldt> I think there's enough on there that task one could be "find a question that needs help"
23:07:56 <fifieldt> mmm, good thinking - if they like the pain of moderating, maybe they want more?
23:07:57 <fifieldt> :)
23:08:18 <reed> it's more to pick their brains,  they can give us ideas and thoughts too...
23:08:30 <fifieldt> oh, right!
23:08:34 <fifieldt> that is a good idea
23:08:46 <fifieldt> perhaps we could also start a post on the CLS forum or g+ thing?
23:09:33 <reed> not sure who's on the G+ thing but yeah, spread the word as much as possible
23:09:44 <reed> a blog post would be good too
23:10:03 <fifieldt> indeed
23:10:10 <reed> the timeframe is  tight though
23:10:17 <fifieldt> make it a dual announce-chinese and ask for moderators?
23:10:39 <reed> IMHO two announcements is better
23:10:44 <fifieldt> ok
23:11:02 <reed> I was thinking moderators first because I'm afraid of getting too many questions at once and no moderators
23:11:36 <fifieldt> right, ok
23:11:45 <reed> for English I feel can deal more easily with high volumes, in ZH not sure we can leave the site unattended at the beginning
23:12:02 <fifieldt> I was thinking along the lines of motivation to join ... if there's no content, it's more difficult to get moderators
23:12:12 <fifieldt> not a strong feeling :)
23:13:09 <reed> a good one-two combo may work better then: time the training moderators thing first and start shooting  announcements soon after
23:13:47 <fifieldt> cool
23:14:01 <reed> actually, this may work well because there is a Cloud Open something conference cloud in China (Shanghai IIRC) mid september and Lauren wants to issue a press release, we may mention Ask in ZH there
23:14:11 <fifieldt> indeed
23:14:15 * fifieldt will be at that one
23:14:24 <reed> when is it exactly?
23:14:29 <fifieldt> 15-17 sept
23:14:46 <reed> ok, so the PR will be later next week or so
23:15:01 <reed> we need to have this session ASAP then
23:15:10 <fifieldt> yup
23:15:14 <reed> like one week from now?
23:15:25 <fifieldt> sure
23:15:40 <reed> in Asian time zone, I think
23:15:49 <reed> what time is it in China now?
23:15:56 <fifieldt> 0715
23:16:25 <reed> so one hour later? 0800... too early?
23:16:44 <reed> i fear that if we do it later we may totally miss the US east coast
23:16:50 <fifieldt> indeed
23:17:00 <reed> even so it's late for them
23:17:02 <reed> oh well
23:17:04 <fifieldt> I consider 0800 early, but it depends very much on the individual
23:17:18 <fifieldt> sometimes volunteers for these kind of activities can't meet during work hours
23:17:26 <reed> is HK in the same timezone as China?
23:17:30 <fifieldt> yup
23:17:33 <fifieldt> "CST"
23:18:40 <reed> k, so the plan is to:
23:19:17 <reed> - date for the event is Sep 4th, 5pm PDT
23:19:50 <reed> - invite moderators wannabe and people from around various communities, groups to give us advice
23:20:05 <fifieldt> would you consider asking people what an appropriate time is?
23:20:09 <fifieldt> using doodle or similar?
23:20:25 <reed> - format: brief introduction (from the wiki page) and real life intervention on questions
23:21:31 <fifieldt> format looks good
23:21:43 <reed> fifieldt, that may work too... it may require more time though
23:21:59 <reed> you need to give people time to see the survey and respond, and then announce the date
23:22:25 <fifieldt> choose a date, but ask for the time
23:22:42 <fifieldt> maybe it's like a 'registration' step
23:22:53 <fifieldt> you just give a couple of options
23:23:43 <reed> let's check with Lauren when she wants to have the PR ready to go out
23:23:48 <fifieldt> ok
23:23:54 <reed> maybe we have more time than that
23:24:13 <reed> or we just skip the announcement now and we do this training well
23:24:31 <fifieldt> yeah, that might be a better option
23:24:43 <reed> agreed
23:25:20 <reed> set the bar to do this well and if we make it to the PR deadline, well, otherwise you'll mention it during your talk
23:25:38 <reed> and we keep soft launching that way until big launch in HK
23:26:09 <fifieldt> reed, example poll: http://doodle.com/b7agkwpgk59mdke5
23:27:05 <reed> fifieldt, looking good
23:27:39 <reed> not
23:27:45 <reed> 5am PDT is out of question for me :)
23:27:52 <fifieldt> :P
23:28:18 <fifieldt> admin link is https://doodle.com/b7agkwpgk59mdke5699uem8w/admin?tmail=poll_admin_adminlink&tlink=adminbtn
23:29:21 <reed> ok, will work on it later
23:29:21 <fifieldt> we should probably move to the next topic
23:29:24 <reed> I think we have a plan
23:29:27 <fifieldt> agreeed
23:30:12 <reed> #info setting up a training session for moderators of Ask OpenStack, focusing on training new mods for Chinese language
23:30:51 <reed> #action reed to organize the event before Sep 14
23:31:02 <reed> next topic is...
23:31:13 <fifieldt> reed to draft a blog post to describe the training sprint, link to eventbrite, objectives and process ?
23:31:29 <reed> fifieldt, haven't done that, fell off the list for this week
23:31:33 <fifieldt> coolo
23:31:50 <reed> the main reason why it fell off the list is that I need to understand it a bit more first
23:32:00 <reed> do you have a clear picture of that whole effort?
23:32:00 <fifieldt> it is an ambitious project!
23:32:11 <fifieldt> sarob ran me through things
23:32:15 <fifieldt> one meeting where it was just us two
23:32:25 <fifieldt> it should be logged
23:32:27 <reed> i spoke with Jonathan about it yesterday, he gave me his view
23:33:10 <fifieldt> http://eavesdrop.openstack.org/meetings/openstack_community/2013/openstack_community.2013-08-14-22.59.log.html
23:33:20 <fifieldt> has the conversation which is the sum of my knowledge more or less
23:33:31 <reed> do you think you would be able to write the post easily?
23:33:54 <fifieldt> maybe not easily, but I could give it a start
23:34:12 <reed> that'd be good, it would speed things up
23:34:20 <fifieldt> ok, I'll get on that
23:34:51 <reed> I have commented on the out of date wiki pages :)
23:35:18 <fifieldt> thank you :D
23:35:32 <fifieldt> #info discussion on obsolete wiki pages continues on mailing list
23:35:32 <reed> #action fifieldt to give a first stab at drafting the blog post for the training sprint
23:35:58 <reed> so many threads :(
23:36:07 <reed> ok, we've exausted this topic I think
23:36:21 <fifieldt> +1
23:37:04 <reed> #topic status update for running projects
23:38:11 <reed> #info Activity Board Dash (the Bitergia charts) are getting integrated in stackforge, first step towards a tighter integration into the official Infrastructure
23:38:46 <clarkb> #link https://review.openstack.org/#/c/44057/
23:39:13 <fifieldt> :D
23:39:17 <reed> I'm working with them to create a roadmap for the Havana report
23:40:09 <reed> thanks clarkb... my attempts to learn puppet go in that direction too, my patch is in draft :) too embarrassed to show it
23:40:57 <reed> i'll share it with fifieldt
23:41:25 <reed> https://review.openstack.org/#/c/42998/
23:41:48 <fifieldt> :D
23:42:09 <clarkb> don't be embarrased :)
23:42:12 <reed> #info the Groups portal is still waiting for the server to be created. Should be ready any time now
23:42:17 <reed> #link https://review.openstack.org/#/c/42608/
23:43:01 <fifieldt> good, good
23:43:41 <reed> #info the Insights part of Activity Board is being updated to a new version. Some data (affiliation, especially) is not being updated... investigating
23:45:00 <fifieldt> goodo
23:45:32 <reed> #info Ask project: waiting for an estimate (sprint plan) from the Askbot folks for adding new features (including import questions from Launchpad) and fixing some small issues
23:46:19 <fifieldt> cool
23:46:22 <reed> fifieldt, what's cooking on your side?
23:46:35 <fifieldt> well, the next post for the ambassador program went up
23:46:44 <fifieldt> #link http://www.openstack.org/blog/2013/08/ambassador-program-specifics/
23:46:51 <fifieldt> and we already have 3 applicants
23:46:53 <fifieldt> brave people!
23:47:07 <fifieldt> I'm going to be writing a bunch of copy for an introductory pack
23:47:13 <fifieldt> that needs to be done before the end of sept
23:48:00 <fifieldt> there's some discussions going on in i18n and docs to get japanese operations guide online, and chinese translation effort for same is ramping up to get it from 80% to 100%
23:48:23 <fifieldt> the trystack survey is getting a lot more reponses now the link is on the login page
23:48:39 <fifieldt> I started looking at the kind of analysis that could be done with this data
23:49:08 <fifieldt> heat made it to transifex
23:49:19 <fifieldt> working to update the user survey for a big push in september
23:49:31 <reed> sounds good... when do you close the survey? or start looking into the results?
23:49:34 <fifieldt> that's probably the ongoing projects from me
23:49:41 <fifieldt> it's looking like the first week of october so far
23:49:43 <fifieldt> oh
23:49:43 <reed> the trystack survey I mean
23:49:44 <fifieldt> trystack
23:49:54 <fifieldt> any suggestions?
23:50:05 <fifieldt> let me see how many results we got
23:50:06 <fifieldt> moment
23:50:28 <fifieldt> sitting at 61 results so far
23:50:29 <reed> yeah, i don't think we need to rush it until we get the urgency to make any decision about trystack :)
23:50:41 <fifieldt> yeah, even keeping it open for another month is fine I think
23:50:48 <fifieldt> but closing some time before the summit
23:50:54 <reed> and propose a session for design summit in hk?
23:50:54 <fifieldt> straw man date: October 4th?
23:51:05 <fifieldt> there is already a couple of trystack sessions proposed
23:51:12 <fifieldt> so maybe will just provide the data to those speakers
23:51:17 <reed> good, we'll show up with data :)
23:51:43 <reed> fifieldt, ? when did the proposals for design summit open?
23:51:49 <fifieldt> oh, design summit
23:51:53 <fifieldt> sorry I meant conference
23:52:04 <fifieldt> we could do a design summit session, sure
23:52:05 <fifieldt> that's a good idea
23:53:02 <reed> #info let's look at the data from trystack survey once the design summit opens for proposals
23:53:16 <fifieldt> done
23:53:46 <reed> now, since you mentioned i18n, I think we have an issue to solve together with the infra team
23:54:11 <fifieldt> yes?
23:54:38 <reed> well... since transifex is not open source anymore we may want to look into alternatives
23:54:50 <reed> and alternatives mean also provide these alternatives
23:54:52 <fifieldt> right, indeed
23:55:02 <reed> and the main alternative at the moment is self hosted poodle
23:55:25 <reed> in order for the i18n team to evaluate it somebody needs to set that service up for testing
23:55:59 <reed> i'll bring this up with jeblair and fungi
23:56:09 <fifieldt> ta
23:56:13 <fifieldt> action it for the log?
23:56:25 <jeblair> o/
23:56:28 <reed> the other thing to bring up with infra is the request from the UX team for an askbot instance
23:56:46 <reed> oh, hello jeblair
23:56:58 <fifieldt> it's jeblair!
23:57:02 * fifieldt bows down
23:57:06 <jeblair> from what i've seen of poodle, it looks really promising and has improved from last time we looked at it
23:57:07 <reed> a couple of lines above, the i18n things
23:57:21 <fifieldt> s/poodle/pootle
23:57:26 <jeblair> i'd be happy to spin one up; i don't want to be premature though, so any requirements gathering from i18n would be great
23:57:40 <reed> right, jeblair, I have a doc from i18n ... wait
23:57:44 <jeblair> fifieldt: oh, crap, i'm only prepared to spin up poodles!
23:58:01 <reed> https://docs.google.com/spreadsheet/ccc?key=0Aqevw3Q-ErDUdFgzT3VNVXQxd095bFgzODRmajJDeVE#gid=0
23:58:14 * fifieldt has a vision of jeblair doing one of those spinning-plates-on-sticks routines, only with poodles instead of plates
23:58:15 <reed> it has the comparison done at the time when Transifex was chosen
23:58:38 * reed thinks this is the longest community meeting so far
23:58:55 <jeblair> i think the translation memory stuff has improved, which was the week point
23:59:43 <jeblair> weak point even
23:59:44 <jeblair> wow
23:59:45 <fifieldt> that's what I heard too
00:00:01 <reed> right, so I believe the next step would be to give a playground to the i18n team
00:00:07 <jeblair> it's the end of the day here but i promise i haven't started drinking yet
00:00:35 * reed wishes he was sipping salt water now but the bridge is closed
00:00:49 <jeblair> clarkb: did a lot to set up initial i18n jobs too
00:00:53 <reed> and BART takes too long :(
00:01:02 <jeblair> reed: you have 3 hours :)
00:01:50 <reed> jeblair, /me pessimist, would never make it :)
00:02:12 <clarkb> transifex isn't open anymore?
00:02:17 <reed> jeblair, do you think some at infra has spare time to setup this playground?
00:02:18 <fifieldt> mmm :s
00:02:26 <fifieldt> clarkb, nope :(
00:02:38 <clarkb> wow I thought the organization change was bad :(
00:02:48 <fifieldt> the claim is they were spending so much effort with the split repositories
00:02:55 <fifieldt> that it was "better for everyone"
00:02:58 <fifieldt> to just bring it all in
00:03:18 <jeblair> reed: my current very basic understanding is that it's a django app with a db, which is a pattern we already have in production
00:03:28 <reed> fifieldt, yeah, that's what happens when open source is just a marketing badge
00:03:32 <clarkb> jeblair: ya https://github.com/transifex/transifex
00:03:50 <jeblair> reed: so it should be a <1 day job to spin one up; then we'll find out what we don't know.  :)
00:03:52 <clarkb> the translation jobs themselves are fairly generic. They just need a way to push and pull from the translation upstream
00:04:05 <reed> clarkb, would not touch that, we're talking about pootle
00:04:23 <jeblair> reed: clarkb's referring to jobs that push and pull from transifex
00:04:26 <jeblair> currently
00:04:33 <reed> clarkb, https://github.com/transifex/transifex/issues/206#issuecomment-15243207
00:05:12 <jeblair> oh, i think i misunderstood, reed's "would not touch that" -> "would not touch old source code dump of transifex"
00:05:33 <reed> jeblair, indeed, that's what I was referring to
00:05:55 <clarkb> ya probably not a bad idea
00:06:02 <jeblair> http://docs.translatehouse.org/projects/pootle/en/latest/server/installation.html
00:06:05 <clarkb> unless someone forks it and fixes security issues and so on
00:06:15 <jeblair> looks just as simple as it should be, so i'm optimistic.
00:06:45 <jeblair> reed, fifieldt, maybe next week?
00:06:46 <reed> clarkb, if the TX team abandoned it I believe there is a reason and it's likely unmaintainable anymore :)
00:07:12 <reed> jeblair, take your time, just let us know when it's doable
00:07:51 <reed> i think that the i18n team wants to arrive to HK ready to discuss the alternatives and make a decision
00:08:12 <clarkb> is launchpad being considered again?
00:08:48 <fifieldt> having used launchpad for translations
00:08:48 <jeblair> clarkb: i doubt anything has changed to make it more attractive
00:08:58 <fifieldt> it's workable, but still sucks more than transifex
00:09:14 <clarkb> jeblair: ya
00:09:40 <reed> clarkb, not really... I don't think that LP has a bright future either, might as well stay on TX
00:10:21 <reed> so, I think we have a draft of a plan
00:10:24 <clarkb> reed: I think at this point TX has no future. If for no other reason than to make a statement that open source matters
00:10:53 <reed> clarkb, we can take that debate offline :)
00:11:36 <reed> #info infra team to investigate setting up a demo server based on pootle for evaluation by i18n team. Check again next week for a more precise timeline.
00:11:38 <jeblair> debate?  :)
00:12:27 <reed> the future of TX is debatable :)
00:12:35 <reed> next topic?
00:12:54 <fifieldt> <propose something>
00:13:03 <reed> oh one thing: did you (infra) get any request from UX designers for an askbot installation?
00:13:43 <clarkb> they mentioned that askbot is the sort of thing they would like. I don't know if the discussion went further than that. THey may have been asked to try a thread on ask.o.o?
00:14:08 <reed> ok, i'll reach out to jaromir
00:14:24 <reed> (I'm sure I butchered his name, sorry)
00:14:45 <reed> that's it from me
00:14:46 <jeblair> well, we know from previous experience it's a huge undertaking to set up askbot
00:15:08 <reed> jeblair, it "should" have changed now
00:15:17 <reed> but not sure in what sense
00:15:25 <jeblair> so i'm reluctant to say we should put time into it; especially since there's a good chance that storyboard will meet their needs as it develops
00:15:36 <jeblair> that might be a good thing to check in on in hk
00:15:58 <jeblair> but having said that, if someone pushes up a patch for an askbot server, we can totally run it.  :)
00:16:46 <reed> i'd like to have a fast overview of the recent changes in askbot code to see if they address the difficulties we had before
00:17:11 <reed> IIRC the obstacles were around the backend database
00:17:25 <jeblair> clarkb: if reed gets a summary of that, would you perhaps have time to review it?
00:17:27 <clarkb> backend database and indexing for search
00:17:33 <clarkb> jeblair: reed yup I can take a look at it
00:17:37 <reed> right, I remember correctly
00:17:40 <clarkb> s/it/a summary/
00:18:08 <reed> the changes are exactly around there: now haystack support is implemented and search goes through there (and solr)
00:18:18 <reed> so probably it's still a bloody mess but a different one :)
00:18:34 <clarkb> there was also an issue with localization iirc
00:18:42 <clarkb> where it had circular imports
00:18:51 <reed> love i
00:19:09 <reed> ok would a changelog be enough?
00:19:21 <clarkb> possibly
00:19:42 <clarkb> changelogs seem to vary in verbosity but it would be a good place to start at least
00:19:56 <reed> #action reed to look at the recent changes in askbot and pass data to clarkb. Low priority.
00:20:14 <reed> not sure about you but i'm baked
00:21:15 <reed> closing the meeting?
00:21:19 <reed> sarob, anything to add?
00:22:37 <fifieldt> all good from my end too
00:23:27 <reed> #endmeeting