19:02:46 <SotK> #startmeeting storyboard
19:02:47 <openstack> Meeting started Wed Aug 22 19:02:46 2018 UTC and is due to finish in 60 minutes.  The chair is SotK. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:02:48 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:02:50 <openstack> The meeting name has been set to 'storyboard'
19:03:04 <SotK> #link https://wiki.openstack.org/wiki/Meetings/StoryBoard Agenda
19:03:20 <SotK> once again I forgot to update it until now, sorry
19:03:34 <SotK> #topic Announcements
19:04:47 <diablo_rojo> o/
19:05:02 <SotK> welcome diablo_rojo
19:05:10 <diablo_rojo> fungi, here too?
19:07:33 <SotK> do we have anything to announce?
19:07:43 <diablo_rojo> Requirements migrated last week?
19:07:55 <diablo_rojo> I think thats the only one
19:08:10 <fungi> oh, yep, i'm around
19:08:30 <diablo_rojo> Requirements was the only one you migrated last week correct?
19:08:36 <fungi> afaik, yes
19:08:44 <diablo_rojo> Cool.
19:08:49 <diablo_rojo> Oh hello fatema_ :)
19:08:59 <fungi> i'm scheduled to migrate some project called "slogging" a week from tomorrow
19:09:08 <fatema_> hello, sorry I'm late
19:09:30 <SotK> hello fatema_ :)
19:10:01 <fatema_> Hi, diablo_rojo SotK
19:10:06 <diablo_rojo> Oh yes that's the other one I was trying to remember
19:10:11 <diablo_rojo> fatema_, no worries
19:11:43 <SotK> #topic Migration Updates
19:12:16 <diablo_rojo> I saw that Searchlight and Freezer are kind of reorganizing and I suggested now might be a good time to migrate. Haven't heard anything back from them yet.
19:12:22 <diablo_rojo> Still nothing from keystone.
19:12:31 <diablo_rojo> And others are holding out for attachments.
19:12:57 <diablo_rojo> I thiiink thats all the updates I have
19:13:56 <SotK> cool, hopefully you hear back from folks
19:13:56 <fungi> diablo_rojo: it looked like the new searchlight ptl agreed with your recommendation?
19:14:16 <fungi> at least based on my reading of the reply on that thread
19:14:23 <diablo_rojo> fungi, oh, okay cool- missed that in my inbox
19:14:28 <diablo_rojo> I'll take another look.
19:14:33 <diablo_rojo> Can do a test migration today
19:16:04 <fungi> speaking of test migrations, not sure if you saw my comment in #storyboard during the requirements migration testing but repeating the migration tool runs only pulls in new tasks/comments (and users), doesn't update any existing tasks whose statuses have changed in lp since the previous run
19:16:07 <diablo_rojo> I see the thread but I don't see my message nor a response- just ttx and Nguen and Trinhs messages
19:16:59 <SotK> ah, its mentioned in the first email of the searchlight thread
19:17:04 <diablo_rojo> fungi, would we want to update those though? We wouldnt want sb statuses to be overwritten if they had changed.
19:17:13 <diablo_rojo> OH
19:17:14 <diablo_rojo> LOL
19:17:16 <diablo_rojo> I see now
19:17:26 <fungi> diablo_rojo: http://lists.openstack.org/pipermail/openstack-dev/2018-August/133632.html
19:17:48 <diablo_rojo> He must have started a new thread that looks practically the same.
19:17:52 <diablo_rojo> Hence my confusion
19:18:08 <fungi> the mailman archive seems to thread it correctly
19:18:31 <diablo_rojo> fungi, probably just inbox getting confused
19:18:35 <diablo_rojo> or my brain
19:18:37 <diablo_rojo> or both
19:18:40 <fungi> but yeah, maybe he dropped reference headers
19:21:24 <fungi> as for updating task status, i guess it depends
19:21:32 <SotK> #topic In Progress Work
19:21:46 <diablo_rojo> A bunch of your stuff landed SotK :)
19:21:57 <SotK> I noticed, thanks for the reviews!
19:22:01 <fungi> in that particular case we spent a while trying to figure out why there were a lot more active stories in storyboard-dev for requirements when there was only one open in lp
19:22:18 <fungi> before i noticed that most of them had been resolved before the second migration run
19:22:32 <fungi> made it hard to compare the state between the test import and what was in lp
19:23:07 <diablo_rojo> Oh weeird
19:24:03 <fungi> basically during an earlier test import they had been active, so when rerunning last week their active statuses weren't updated to be resolved
19:24:18 <fungi> as the tasks already existed on sb-dev
19:24:56 <diablo_rojo> That should be a simple thing to add. We would just want to check that the status in sb isnt already resolved and reset it to active
19:25:37 <fungi> yeah, i don't know that it's a problem to be "fixed" so much as just a caveat to keep in mind when you're doing these
19:26:17 <diablo_rojo> I could go either way
19:26:38 <diablo_rojo> I don't think its that big a deal, but I can see how if it happens to a lot of stories, it could be a pain
19:26:44 <fungi> well, one way doesn't involve needing to do anything ;)
19:27:15 <diablo_rojo> I am almost always in favor of those ways unless there is some pressing reason to do otherwise
19:27:19 <fungi> so i'm in favor of spending our effort wisely and just leaving it the way it is unless it becomes a serious problem
19:27:24 <SotK> me too
19:27:30 <diablo_rojo> Works for me :)
19:28:03 <fungi> we've also had a couple more requests for making the project and project-group story list have smarter rules for inferred story state (not taking other projects into account when determining state)
19:29:50 <SotK> mhmm, I think that should be fairly doable
19:29:59 <fungi> as a workaround, i suggested using a task board with automatic lanes based on task states rather than stories
19:30:12 <fungi> but that's obviously suboptimal
19:31:37 <diablo_rojo> So if its completed in a project show it as done there and if its being shown in a project group and they aren't all done show it as active?
19:32:31 <fungi> basically in the project story listing ignore the states of tasks for other projects, and in the project-group story listing ignore the states of tasks for any projects not in that project-group
19:32:40 <fungi> at least i find that an easier way to reason about it
19:32:47 <diablo_rojo> Okay
19:32:53 <diablo_rojo> I think that makes sense
19:32:55 <SotK> yep, that's how I've understood it
19:34:21 <SotK> dhellmann pointed out an issue earlier too, where stories with many comments become a bit unusable in the webclient
19:35:25 <fungi> yeah, we dropped pagination a while back to avoid confusion, but that brings new problems if you have a story with >1k comments
19:35:50 <diablo_rojo> Popular story :)
19:35:56 <diablo_rojo> A release goal I would guess?
19:36:01 <fungi> popular with his script anyway
19:36:04 <persia> Even > 100 comments ends up being noticeably slow.
19:36:10 <fungi> the py3k-first release goal, yeah
19:37:02 <diablo_rojo> looks like mordred was talking in #storyboard about updates to make it less painful
19:37:12 <diablo_rojo> but also 'is not really here right now'
19:37:33 <SotK> I think that's pretty much the next thing I'm going to work on fixing after I've finished the project groups by name thing I've started
19:37:58 <SotK> and then after that probably writing a spec for attachments
19:38:12 <SotK> unless anyone beats me to it of course :)
19:38:14 <fungi> he had also mentioned that the lack of separation between a story state for only "todo" tasks and a story state with an assorted mix of task states made finding untriaged stories harder (at least i think it was dhellmann who pointed that out)
19:38:44 <SotK> oh yeah, I think it was
19:39:15 <SotK> it probably makes sense to look at that at the same time as improving how story state is determined when browsing by project/project group
19:39:22 <persia> There is probably scope for having a richer list of story states (derived from the small set of task states).
19:39:45 <fungi> yeah, for that, there were some nuanced combinations of states we probably wanted to take into account, so having some sort of matrix might make it easier to reason through
19:39:46 <persia> Key to a larger list is not actually letting it be set manually, to avoid the "too many choices" problem.
19:42:11 <persia> Do there exist any stories of how people did that sort of thing with LP, or did people not usually use bugtasks in LP?
19:42:54 <SotK> the discussion about it is in the irc logs from the 5th and 6th of August
19:43:08 <SotK> ah now I can link it because I have the internet again
19:43:15 <SotK> http://eavesdrop.openstack.org/irclogs/%23storyboard/%23storyboard.2018-08-06.log.html
19:43:34 <SotK> and also http://eavesdrop.openstack.org/irclogs/%23storyboard/%23storyboard.2018-08-05.log.html
19:44:45 * fungi is jealous of SotK's internet access returning
19:45:06 <SotK> heh
19:46:08 <SotK> I'm happy if we just address the states discussed in those logs for now
19:46:25 <SotK> maybe there would be value in a "Review" state when all tasks are in review too
19:46:55 <persia> I like the idea of mirroring the task status when all tasks are in the same status (or invalid+that status).
19:47:02 <fungi> persia: we use bugtasks in lp a lot... i think it's a combination of the fact that lp doesn't have non-project-specific lists of bug and it treating bugtask states as a priority ordered list so just displayed the highest state value of those across the various series bugtasks present for that project
19:48:21 <persia> fungi: https://bugs.launchpad.net/bugs/+bugs exists, but I agree that nobody ever uses it.
19:48:23 <fungi> new>triaged>confirmed>in-progress>merged>released>opinion>wont-fix>invalid or something like that
19:48:47 * SotK is happy with mirroring too, re-reading those logs I'm a bit concerned that "new" has unwanted meaning for that context
19:48:50 <fungi> ahh, yes i've used the non-project-specific bug listing option so infrequently that i didn
19:48:56 <fungi> 't realize it existed at all
19:49:23 <fungi> SotK: sure, i like "todo" for that instead
19:49:27 * persia has been using that bug tracker since before it was called "Launchpad", and so remembers lots of special bits that probably should have been removed from the interface years ago
19:50:05 <fungi> *cough* blueprints *cough*
19:50:23 <persia> One of the features I like about mirroring is that it is amenable to gaming.  For example, if one wants to have "TODO" be untriaged and "Active" be triaged, one can create a task "triage this" and mark it complete.
19:50:45 <fungi> wfm
19:50:49 <persia> blueprints is actually an entirely different project that was put into LP.  Also, yes :)
19:52:21 <fungi> yeah, i got a ton of backstory on how blueprints was developed and shoehorned in, none of which is likely germane to this discussion ;)
19:52:54 <persia> Also yes :)
19:53:48 <SotK> heh
19:53:53 <SotK> anything else in progress?
19:54:22 <diablo_rojo> Some of fatema_ 's stuff needs reviews
19:54:38 <diablo_rojo> Other than that- not really
19:55:15 <SotK> oh, I'd be really happy if https://review.openstack.org/#/q/status:open+project:openstack-infra/storyboard-webclient+branch:master+topic:bugfixes can get some reviews so that I can pad out the update email I've been drafting with more things
19:55:48 <SotK> https://review.openstack.org/#/c/589663/ would also be good to get merged if someone that isn't diablo_rojo has time to review it
19:56:08 * SotK plans to look at fatema_'s patches in his next reviewing session
19:56:39 * diablo_rojo adds bugfixes reviews to the list
19:57:27 <fatema_> thanks SotK
19:57:38 <fatema_> Is there time for questions ?
19:57:42 <SotK> sure
19:57:47 <SotK> #topic Open Discussion
19:57:56 <SotK> 3 minutes :)
19:58:03 <fatema_> :D
19:58:09 <fatema_> https://review.openstack.org/#/c/423876/ for this one
19:58:35 <fatema_> Is there a way to pick it up?
19:58:53 <persia> Is it worth doing?
19:59:11 <fatema_> I want to add simplejson to make it possible to use arrow function in coding
19:59:26 <persia> Alternately, is there another way to cause those requirements to apply iff the migration script is run?
19:59:29 <fungi> if it's worth restoring, any core reviewer should be able to set it to restored state
19:59:47 <fatema_> well in the discussion http://eavesdrop.openstack.org/irclogs/%23storyboard/%23storyboard.2017-01-23.log.html
20:00:00 <persia> For clarity, my opposition is only to LaunchpadLib: I have no objection to simplejson.
20:00:22 <fatema_> persia, that's fine by me
20:00:29 <SotK> we're out of time, lets continue in #storyboard
20:00:34 <SotK> #endmeeting