19:00:55 <gundalow> #startmeeting Ansible Contributor Experience Working Group
19:00:55 <zodbot> Meeting started Thu Dec  6 19:00:55 2018 UTC.
19:00:55 <zodbot> This meeting is logged and archived in a public location.
19:00:55 <zodbot> The chair is gundalow. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:55 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
19:00:55 <zodbot> The meeting name has been set to 'ansible_contributor_experience_working_group'
19:01:02 <gundalow> #chair felixfontein
19:01:02 <zodbot> Current chairs: felixfontein gundalow
19:01:06 * gundalow waves, who's around
19:01:28 <felixfontein> me
19:01:30 <felixfontein> :)
19:01:37 <felixfontein> and I guess dag didn't join to ignore us ;)
19:01:52 * gundalow waves to dag, you joining?
19:02:08 <dag> owh, sorry, I just opened the laptop (and this is automated :p)
19:02:22 <dag> I have to do my finances and indoor soccer at 21h
19:02:30 <felixfontein> lol ok :)
19:03:11 <gundalow> :(
19:03:43 * gundalow threw a few topics on the agenda https://github.com/ansible/community/issues/390#issuecomment-444976170
19:04:00 <gundalow> felixfontein: what would *you* like to talk about?
19:04:15 <dag> I'll be lurking ;-)
19:04:24 <felixfontein> good question, I'm currently finishing dinner, so I didn't thought much about it yet ;)
19:04:46 <felixfontein> it's good to hear that the author ping bug has been fixed in ansibot
19:05:04 <gundalow> yup
19:05:08 <gundalow> #topic Ansibulbot
19:05:11 <dag> Not necessarily contributor experience, but we still have an important issue with documentation (people end up on v2.4 and v2.5 docs and there's no easy way to go to latest or stable)
19:05:51 <dag> and it is very easy to fix with a basic solution
19:05:52 <felixfontein> ah, you mean having a more or less obvious note with a link to latest
19:06:03 <felixfontein> note=box, banner, ...
19:06:23 <dag> felixfontein: at the top at least a link to latest and stable, and a simple ErrorDocument that sends you to an sorry-page and a search form
19:06:25 <gundalow> dag: I believe adding banners is on acozine's list for early 2019. Also we've added some code to so that `/latest/` is the preferred version. So Google should be learning
19:06:49 <bcoca> doc meeting might be better for that, but they are aware of issues of cross version linking and content, just need a good solution to appear
19:06:51 <dag> gundalow: great to hear ! I didn't know
19:06:58 <felixfontein> hi bcoca!
19:07:05 <gundalow> #chair bcoca dag
19:07:05 <zodbot> Current chairs: bcoca dag felixfontein gundalow
19:07:10 <gundalow> ok, so back to Bot
19:07:19 <gundalow> #info Bot no longer spams "Thank you"
19:07:23 <dag> but it's taking too long, we've been discussing the same thing 2 years ago ;-)
19:07:36 <dag> and this is really basic stuff (sorry for the rant)
19:07:39 <gundalow> #info Bot should be pinging author again
19:07:42 <bcoca> dag: i dont disagree, but we also had 3 diff doc 'teams' during those 2 years
19:07:43 * dag back in lurking mode
19:07:50 <felixfontein> better too much 'thank you' than no author ping... ;)
19:07:51 <acozine> dag: gundalow: bcoca: we may have found someone with the right CSS skills and the time to devote to the problem
19:08:01 <bcoca> woot!
19:08:03 <gundalow> #chair acozine
19:08:03 <zodbot> Current chairs: acozine bcoca dag felixfontein gundalow
19:08:13 <felixfontein> hi acozine!
19:08:22 <felixfontein> that sounds awesome!
19:08:32 <acozine> FYI, I'm in a TW meeting right now too, so my attention will be intermittent
19:08:51 <gundalow> #info Bot planning https://etherpad.openstack.org/p/ansibot
19:09:06 <acozine> yeah, I finally found the right person to ask . . . we just don't have that skillset on the docs teams themselves
19:09:51 <gundalow> #info Need to put together a plan for the bot, particularly to guide both the creator and potential reviewers to the next steps
19:10:56 * felixfontein is reading
19:14:16 <dag> acozine: thanks !
19:16:31 <gundalow> felixfontein: dag Any thoughts?
19:16:46 <gundalow> Feel free to edit the doc, that's why it's in Etherpad :)
19:17:09 <felixfontein> the typescript bot looks nice!
19:17:31 <gundalow> oh, I've not looked into that much, lets do that now
19:17:39 <gundalow> wonder if their is a better example
19:17:57 <felixfontein> I've just looked in the thread you linked to
19:18:03 <felixfontein> but that already looks really neat
19:18:15 <felixfontein> it would be interesting to see it in action when something doesn't work that smooth, though :)
19:18:44 <gundalow> #info typescript-bot https://github.com/DefinitelyTyped/DefinitelyTyped/pull/30936
19:19:35 <gundalow> #info Example of failed CI https://github.com/DefinitelyTyped/DefinitelyTyped/pull/31126
19:19:43 <felixfontein> DefinitelyTyped is a good thing to compare to Ansible, at least to Ansible modules
19:24:08 <gundalow> I can't find the Bot source/docs anywhere
19:25:44 <felixfontein> it belongs to Microsoft, so maybe it's not open sourced (yet)?
19:26:36 <gundalow> maaaaaaybe
19:27:00 <gundalow> Thoughts on `Messaging` Line 30-58 on https://etherpad.openstack.org/p/ansibot
19:27:08 <gundalow> ie the Bot adding a single comment
19:27:23 <dag> That's a big one for me !
19:27:24 <felixfontein> I like that
19:27:34 <felixfontein> currently, I think you get sometimes three comments for a new issue?
19:27:42 <felixfontein> (or was it PR?)
19:27:43 <dag> Currently, there's also a big comment about matches the product, which I think is confusing/scary
19:27:54 <gundalow> I think it was 4/5 :(
19:27:55 <dag> contributors don't know what to do with that IMO
19:28:12 <gundalow> dag: Which case?
19:28:27 <dag> gundalow, many of them
19:28:29 <gundalow> hum, wonder if the component matches more than $n we should skip
19:28:50 <gundalow> dag: I mean, I'm not sure which problem you are reffering to
19:28:59 <dag> gundalow: yeah component matches, that's what I mean
19:29:17 <dag> that comment is lost on most people, it's better to add a comment if you don't find a match IMO
19:31:54 <gundalow> Adding comments at line 27-30
19:32:54 <dag> if the match is wrong, let people evaluate and report it through bot_broken
19:33:33 <felixfontein> getting rid of bot_broken seems to be complicated, so I'm really trying to avoid
19:33:47 <gundalow> felixfontein: how so?
19:34:16 <gundalow> info http://dash.tannerjc.net/ansibot_command_report?periods=12 shows the various commands people have given the bot over the last 12 months
19:34:43 <felixfontein> the last (and I think only) time I tried to get a bot_broken removed from an issue (or PR), it took quite some people probing and trying, and it took some time for that to get resolved
19:35:02 <misc> gundalow: you forgot the # , no ?
19:35:24 <gundalow> In order: shipit, needs_info, rebuild_merge, bot_status
19:35:32 <gundalow> #chair misc
19:35:32 <zodbot> Current chairs: acozine bcoca dag felixfontein gundalow misc
19:35:38 <gundalow> misc: maybe, where?
19:36:01 <gundalow> oh
19:36:04 <misc> gundalow: @gundalow> info http://dash...
19:36:15 <gundalow> #info http://dash.tannerjc.net/ansibot_command_report?periods=12 shows the various commands people have given the bot over the last 12 months
19:36:19 <gundalow> misc++
19:36:19 <zodbot> gundalow: Karma for misc changed to 3 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
19:36:28 <felixfontein> ah, I can't find the issue anymore
19:38:26 <dag> felixfontein: debugging ansibot is indeed a special skill
19:39:56 <gundalow> Hopefully that's getting a bit easier webknjaz added http://sentry.io so we are capturing better logs now, helps us see how often certain failures are occuring
19:39:58 <felixfontein> ah, I think this one was it: https://github.com/ansible/ansible/issues/21188 the label never got removed in the end
19:43:42 <gundalow> oh, that is strange
19:43:53 <gundalow> Any other thoughts?
19:45:14 <felixfontein> sorry, got sidetracked
19:45:31 <felixfontein> (had to answer the door)
19:45:43 <gundalow> It's cool :)
19:48:34 <jtanner> bot_broken can not be removed via commands
19:48:47 <jtanner> once it's set, it's permanent until the comment that set it is removed
19:50:14 <bcoca> the needs_revision seems to be getting applied to often, even when no request for changes/merge/ci issues exist
19:50:33 <bcoca> i have looked at code, but was unable to figure out
19:57:55 <jtanner> example?
20:01:34 <felixfontein> gundalow: I think I added all my thoughts to the etherpad
20:02:42 <gundalow> felixfontein: Thank you :)
20:09:19 <gundalow> Cool, unless anyone has anything else, guess we are done
20:09:22 <gundalow> Thanks y'all
20:09:27 <gundalow> #endmeeting