13:59:59 #startmeeting hubs-devel 13:59:59 Meeting started Tue Sep 12 13:59:59 2017 UTC. The chair is mizmo. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:59:59 Useful Commands: #action #agreed #halp #info #idea #link #topic. 13:59:59 The meeting name has been set to 'hubs-devel' 14:01:51 * sayan is back, but is also having food 14:02:09 mizmo: I added a few points to run the meeting: https://pagure.io/fedora-hubs/pull-request/384#_1,33 14:02:19 though not merged into README 14:03:01 .hello wispfox 14:03:02 shillman: wispfox 'Suzanne Hillman' 14:03:18 .hello sayanchowdhury 14:03:19 sayan: sayanchowdhury 'Sayan Chowdhury' 14:03:27 .hello duffy 14:03:28 mizmo: duffy 'Máirín Duffy' 14:04:12 mizmo: can you chair us? 14:04:42 i would rather not sayan, im just not feeling well 14:05:17 mizmo: I mean chair us, shillman or me? 14:05:23 shillman and me 14:05:24 oh haha 14:05:30 #chair sayan shillman 14:05:30 Current chairs: mizmo sayan shillman 14:05:44 * mizmo sorry, operating on 4 hrs sleep :( 14:05:58 Ugh. 14:06:07 mizmo: sad :( 14:06:30 I was down last week because of jet lag, feeling well since Sunday 14:07:00 #topic Status Updates 14:07:02 i know the feeling, with kids instead of planes lol 14:07:16 haha yeah 14:07:38 mizmo: shillman: any thing to update from last week? 14:08:11 Nope! Still continuing my job hunt, and doing other design projects, and moving my website to wix. 14:08:14 sayan: we've both been talking with abompard about a possible outreachy internship for helping develop hubs widgets now that he put together a great workshop for flock that could be used to onboard new devs 14:08:22 Oh, right. Heh! 14:08:24 sayan: i think abompard is still working on the proposal for it 14:09:32 #info abompard is working on the proposal for outreachy on developing hubs widgets 14:10:00 ive seen a lot of positive feedback on the hubs flock sessions too 14:10:00 I was mostly down the last week so I did not much much, was most completing task which were left 14:10:06 we should probably do some kind of blog summary about all of it 14:10:14 mizmo: yeah 14:10:25 mizmo: I thought of dropping a mail last week, but I could not 14:10:26 sayan: do you have a link to the slides handy? 14:10:53 sayan: totally understand :) 14:11:00 mizmo: the slides are hosted here: https://sayanchowdhury.fedorapeople.org/_slides/flock2017/hubs/ 14:11:10 I can create a pdf version of it I guess 14:11:11 awesome thanks 14:12:04 So, the opentack folks have office hours, when they sit for 2-3 hours and work collaboratively 14:12:04 nah i think thats ok 14:12:41 we can start one, once a month 14:13:05 and a bi-weekly hubs updates blog we need to start 14:13:09 sayan: do you think thats a model we should follow? 14:13:18 that would be awesome 14:13:29 #idea monthly 2-3 office hours / hack session (like open stack) 14:13:35 #idea bi-weekly hubs update blog 14:13:41 i think to get new contribs we should do bounties 14:13:59 are we in a position to take on new contribs right now tho 14:14:11 yes, we are in a position 14:14:40 mizmo: we have bunch of easyfix issue https://pagure.io/fedora-hubs/issues?status=Open&tags=easyfix 14:14:57 awesome, so what i can do is write up some of the easy fix issues as bounties 14:15:02 Except for a easyfix issue, there should be a format: who to contact and where in code the bug should be solved 14:15:09 maybe i can put togehter a 'prize pack' for whoever takes each bounty 14:15:48 mizmo: +1 14:15:55 sayan: this is the template we used for design bounties: https://mairin.wordpress.com/2010/07/09/fedora-design-bounty-fedora-slide-deck-template/ 14:16:27 ill do something up like this :) 14:16:47 on other hand I will create a PR with whatever IRC work I did 14:17:03 The plan is have quick PRs rather than a big work 14:17:12 yeh 14:17:14 we can deploy it to the hubs-dev and see how things work 14:17:16 #idea bounties for easyfix in hubs issues 14:17:26 #idea quick PRs rather than large PRs 14:17:34 #action sayan to create the PR for the IRC work 14:17:35 this all sounds great :) 14:17:56 sorry I'm late! 14:18:06 its ok abompard 14:18:16 im here with only half a brain :) 14:18:17 * sayan waves to abompard 14:18:21 #chair abompard 14:18:21 Current chairs: abompard mizmo sayan shillman 14:19:00 abompard: so, I was discussing that we should focus on small PRs, and merge them 14:19:08 and probaly push to hubs-devel 14:19:14 sayan: yeah 14:19:20 and when things work we can do push to prod 14:19:36 abompard: also we need to start working on pushing hubs to production 14:19:42 I'm cooking a pretty big PR for you guys, but I think it can wait after prod 14:19:50 yeah 14:20:18 do you guys think it's in a state where it can be useful to people? 14:20:27 abompard: we don't know 14:20:46 abompard: the thing is until and unless people don't use, we don't know how people are going to react 14:21:15 sayan: yeah, maybe push to prod and advertise to small groups only first, like commops? 14:21:19 i think if we push something into prod at this point it relaly needs to be marked as experimental / playground - 14:21:27 mizmo: agreed 14:21:32 i dont know that we have enough for any one team to get the full experience without irc 14:21:54 what i would rather do - is maybe use say the design team as alpha testers 14:22:02 so we have a user population to get feedback from without doing a real launch 14:22:13 with IRC i think we could do design team pretty well 14:22:15 I'm just careful about not pushing to prod too soon, have people think "nah, useless to me" and not go back to it when we do improvements 14:22:23 abompard: ^^^ this, 100% 14:22:51 Agreed. 14:23:12 for me, it's we should have the basic features in and let people start using 14:23:20 sayan: i agree but not all users 14:23:30 mizmo: yes, probably a team 14:23:31 i think we should pick a small community 14:23:34 I think people got used to beta software having 99% features, like gmail when it came out 14:23:54 thanks google :( lol 14:24:47 one thing that came out from the commops discussion was that if people contribute 14:24:55 they would like there feature online 14:25:06 that would give them the enthusiam to continue working 14:25:24 that makes sense, so we could allow for the alpha testing team + any devel / contribs to have access, 14:25:35 yes 14:25:37 we can do that 14:25:45 a notification maybe for others that its coming soon 14:26:01 atleast we should start planning for production, becuase we need to start packaging things 14:26:20 +1 14:26:26 what is needed at this point for prod? 14:26:41 abompard: wrote the ansible scripts for the deployment 14:27:15 but we still install via pip, so we need to create rpm packages 14:27:20 sayan: but it's still downloading a huge amount of NPM libs from the internet. Not sure they'll like that in prod 14:27:27 for the ones which are not available 14:27:50 Not sure we have a lot of Python deps, but we do have a lot of JS deps 14:27:57 abompard: yeah 14:28:05 anyways, we need to remove those deps 14:28:10 and move to RPM 14:28:41 we can create an issue, to track the issues to be packaged 14:28:51 abompard: ^^ 14:29:04 s/issues/npm libraries/ 14:30:17 sayan: good idea 14:30:44 #action sayan to create a issue to track all the libraries (Python + JS) to be packaged 14:31:37 we can keep editing the issue, when we see new packages to be packaged 14:32:06 mizmo: abompard: shillman: next thing I wanted to discuss was how to get more code contributors? 14:33:37 #action mizmo to create easyfix bounties for hubs 14:33:42 ^^ thats my idea :) 14:33:59 the main challenges: 14:34:11 1) identifying things newbies can work on 14:34:25 2) getting newbies bootstrapped (flock workshop will be a huge help here) 14:34:40 3) getting newbies integrated into the team and keeping them 14:35:41 we also need to identify the source to find the newbies 14:36:21 like we can align with fedora-classroom effort to get more contributors 14:37:21 also GSoC and outreachy 14:37:26 * mizmo points to shillman :) 14:37:29 Yeah! 14:37:35 +1 14:38:21 when is the last date for project proposal in Outreachy? 14:38:34 :) 14:39:12 sayan: applications opened up on sept 7, 14:39:35 sayan: labbott would know when fedora is aiming to have all the proposals in by, i know we are still taking proposals but i dont know the cut off 14:39:47 i would guess at least by the end of sept so applicants have some time 14:39:55 Okay 14:40:20 In that case abompard and me can work on widget and have a detail plan ready 14:40:39 so that we have a good experience for the applicants 14:40:56 sounds good :) 14:41:09 oh! 14:41:11 yeah, I don't know how write this kind of plan, I'll be happy if we can do it together 14:41:21 abompard: what are you preparing for proposal? 14:41:23 Oh 14:41:31 i talked to the cornell student group working on a UX project for the semester for hubs, i gave them a list of commops folks to interview for their hub, so they're getting started on that i think 14:42:01 abompard: should we create a timeline for the student or let the student do it? 14:42:06 s/student/applicant 14:42:22 work together with them? 14:42:24 I dont know 14:42:34 shillman: that sounds good 14:42:59 we can create a list of task to work on, (every small task) and let applicant add the timeline to it 14:43:13 (list of task to complete a widget) 14:43:45 Your applicants may not be sure how long things will take. I recommend having a discussion with them and figuring out the timeline together. 14:43:52 That's what mizmo and I did. 14:44:04 +1 14:44:12 shillman: yes, we will be there for everything, but they need to put in some research 14:44:24 based on individual applicants interests / skillsets you can have a rough outline for the internship but fill in the week-by-week tasks based on individuals 14:44:31 and time would be bit dependent on the potential of the students 14:45:01 +1 14:45:28 *nod* 14:46:06 so first lets discuss the widgets that we should focus on, and then we start on things we need 14:46:21 and then when the applicants turn up we help them build there proposal 14:47:15 i think it might then make senes to wait until after outreachy application period to do bounties so the bounty doesn't compete with an intern's project 14:47:39 mizmo: yes 14:47:45 +1 14:48:12 #info wait until after outreachy application period to do bounties so the bounty doesn't compete with an intern's project 14:48:33 #action abompard and sayan to discuss the widgets for the outreachy proposal, and draft the outline of the proposal 14:49:18 next thing is should we pushing the PRs, issue updates to the mailing list? 14:49:29 mizmo: abompard: shillman ^ 14:50:21 yes definitely 14:50:35 sayan: the ML is kinda dormant at the moment, but if we push every PR update to it the signal-to-noise ratio may go way down. Maybe only the PRs? 14:50:55 s/the PRs/the new PRs/ 14:51:05 abompard: I don't know if we can do that 14:51:12 OK 14:51:39 abompard: for atomic-wg we get all the activity, so I don't need to seperately watch the projects 14:51:46 I just watch the emails in the list 14:52:04 ohh i was thinking we'd write summaries (human-made) not auto, i dont like hitting MLs with auto generated commits 14:52:53 okay, in that case a summary of the weekly issue, triaged? 14:53:39 weekly issues triaged, PRs etc 14:53:56 that'd be great 14:54:24 and there can be list of dates of who want to write (maintain in etherpad) 14:54:35 check the bottom of the page here: https://fedoraproject.org/wiki/Cloud#Meetings 14:56:43 abompard: mizmo shillman ^^ 14:57:05 sayan: it's very interesting 14:57:13 ah thats a good idea 14:58:07 they do a ticket for each actions? 14:58:35 that list is unmaintained now, as cloud moved to atomic, but we generally picked up the week we wanted to maintain 14:58:59 abompard: nope, it just that next week we will reiterate through the action items 14:59:09 to see what is done and what is not 15:00:13 #action sayan to create a etherpad page with the list of dates, and add it to the readme 15:00:32 we are at time. I am moving to open floor 15:00:37 #topic Open Floor 15:00:55 abompard: mizmo: shillman: anything you want to share in open floor? 15:01:12 Not really? 15:01:18 * mizmo noppers 15:01:42 nope 15:02:08 cool, I am then ending the meeting 15:02:11 #endmeeting