16:00:48 <krotscheck> #startmeeting StoryBoard
16:00:49 <openstack> Meeting started Mon Jan  5 16:00:48 2015 UTC and is due to finish in 60 minutes.  The chair is krotscheck. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:50 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:51 <ttx> o/
16:00:53 <openstack> The meeting name has been set to 'storyboard'
16:00:54 * krotscheck peers at the bot
16:00:57 <yolanda> hi
16:01:04 <CTtpollard> hello
16:01:22 <krotscheck> anteaya mentioned that there were bot problems, so here’s hoping we’re good.
16:01:22 <krotscheck> Hey there!
16:01:31 <krotscheck> #topic Actions from last week
16:01:34 <anteaya> krotscheck: hi
16:01:44 <anteaya> anteaya: bots seem to be working right now
16:01:50 <krotscheck> anteaya: Yay.
16:02:04 <anteaya> off to a great start
16:02:42 <krotscheck> Ok, so from last week: I did a bit of pestering fungi for storyboard-dev, and he’s updating the puppet module to accomodate it. https://review.openstack.org/#/c/144367/
16:02:52 <krotscheck> So as soon as that passes jenkins, people can look at it.
16:03:06 <krotscheck> Oh, right: Agenda - https://wiki.openstack.org/wiki/StoryBoard#Agenda
16:03:38 <krotscheck> rcarrilocruz isn’t here, we’ll skip him for now since he’s still investigating socket apis.
16:03:51 <krotscheck> CTtpollard: Did you manage to ping persia about docs?
16:04:06 <CTtpollard> I made him aware that you asked of him
16:04:15 <krotscheck> Cool.
16:04:20 <krotscheck> persia: You here?
16:04:24 <CTtpollard> I've only just got back from Holidays today
16:04:57 <krotscheck> I’ll take silence as a no.
16:05:37 <krotscheck> I’ve taken a look at documenting the event processing engine, and while trying to set up a data dictionary realized that our data is crazy inconsistent. Thus I’m working on https://review.openstack.org/#/c/143744/ to help with that.
16:05:50 <krotscheck> But that’s not ready yet.
16:06:05 <krotscheck> And, honestly, I’m thinking I should table that temporarily because email is now 2 months overdue.
16:06:46 <krotscheck> Any thoughts on that?
16:06:47 <yolanda> krotscheck , do you need help on that one?
16:07:01 <krotscheck> yolanda: Which one, the data dictionary or the emails?
16:07:07 <yolanda> data dictionary
16:07:20 <krotscheck> Yeah, any help would be appreciated.
16:07:23 <yolanda> or the other if you prefer
16:07:35 <yolanda> coming back from holiday on wednesday
16:07:41 <krotscheck> The goal is to make all of our events spit out a consistent set of data that we can work off of.
16:07:58 <yolanda> do you have some spec for it?
16:07:59 <krotscheck> For instance: Comments don’t include the relevant story id. This feels like an oversight.
16:08:27 <krotscheck> yolanda: The patch https://review.openstack.org/#/c/143744/ started as a very large battery of tests that proscribe desired behavior.
16:08:30 <yolanda> actually the commit message is quite accurate
16:09:01 <yolanda> cool, i can collaborate with that on wed
16:09:11 <krotscheck> Righto.
16:09:16 <krotscheck> Neat. thanks.
16:09:23 <krotscheck> Moving on:
16:09:42 <krotscheck> #topic Urgent Items: Deployment broken.
16:09:42 <krotscheck> That’s fixed, I believe.
16:10:00 <krotscheck> #topic User Feedback
16:10:02 <yolanda> yes, latest patches went live
16:10:13 <krotscheck> I just got another from jeblair https://storyboard.openstack.org/#!/story/2000095
16:10:44 <yolanda> mm, seing that as "No title"
16:10:55 <krotscheck> ANyone have a thought on how to address that?
16:11:04 <yolanda> ok, that was a bug, now i can see it properly
16:11:18 <krotscheck> Argh, it’s still doing that?
16:11:23 <yolanda> looks like, yes
16:11:25 <jeblair> yolanda: yeah, that sounds like the bug where it takes too long to async load the story data
16:11:37 <jeblair> (or fails)
16:11:37 * krotscheck thought he fixed that.
16:11:59 <yolanda> i had firebug disable so i could not see the error :(
16:12:21 <krotscheck> Yeah, same here.
16:12:22 <ttx> I think the suggestion on the bug is the right way to fix that: "the link portion of the field should consistently be a navigation link, and only the edit icon should begin editing the field"
16:12:35 <krotscheck> That UI is starting to get crowded :/
16:12:42 <krotscheck> But I agree.
16:13:04 <krotscheck> I’ll add a comment to that effect on the bug.
16:13:08 <rcarrillocruz> hey folks
16:13:18 <CTtpollard> hi
16:13:19 <yolanda> makes sense to me
16:13:23 <rcarrillocruz> sorry, but tomorrow it's bank holiday and i was leaving early today with wife and baby
16:13:24 <rcarrillocruz> !
16:13:39 <krotscheck> rcarrillocruz: No worries, we’ll catch up with you next week.
16:14:14 <krotscheck> Any other user feedback?
16:14:16 <yolanda> krotscheck, actually there is an edit icon next to the project name
16:14:54 <ttx> yolanda: yes, so this is not adding any extra space
16:15:06 <yolanda> so clicking that should allow editing of project, but i think it will be still a bit confusing
16:15:20 <krotscheck> That’s going to be super interesting to figure out which nuance feels natural. Tabbing from field-to-field, vs clicking, vs. linking....
16:15:22 <yolanda> i would expect that all entries in inline task edition behave the same way
16:15:42 <yolanda> so if you click on task and you can edit it, i would expect the same behaviour for project
16:16:31 <krotscheck> I agree.
16:16:46 <jeblair> i think things that look like hyperlinks should act like them, and things that look like edit buttons should act like those... so i think changing the project link as described will make sense, and perhaps the task "link" should just stop looking like a link and only have an edit button
16:17:23 <jeblair> i had no idea that you could tab through things and edit without selecting.
16:17:27 <jeblair> i think that may be an anti-feature
16:18:14 <ttx> also, it's not that often that you /change/ a project
16:18:17 <jeblair> oh
16:18:40 <jeblair> also, everytime i tab through, I'm adding another 'James E. Blair updated "Links to projects are not consistent".' to the history
16:18:47 <ttx> You add extra tasks to cover extra projects, but changing the project for an existing task ?
16:18:52 <jeblair> even without changing anything.  sorry.  i'll file a story for that.
16:19:45 <ttx> I'd argue it's fione that the edit icon is so small, since swiytching project names is a non-use-case.
16:21:05 <krotscheck> So how would you maintain a consistent set of controls, while still accomodating this? Assume that every editable field must behave the same way.
16:21:12 <ttx> hmm.
16:21:28 <ttx> I'll admit that switching assignees (or task titlke) will happen much more often.
16:21:54 <jeblair> krotscheck: accomodating what?  are you asking about the tab thing?
16:21:55 <yolanda> so we could just remove the feature of editing the project of a task
16:21:56 <ttx> i would still follow the suggestion
16:22:03 <krotscheck> jeblair: I’m asking about ttx’s question.
16:22:06 <yolanda> and project is just a simple hyperlink
16:22:19 <ttx> text is a consistent hyperlink, icon triggers quickchange.
16:22:53 <ttx> (the UI needs to be consistent between logged-in and not-logged-in, and those hyperlinks have values)
16:23:00 <ttx> value*
16:23:26 <krotscheck> jeblair: As for the tabbing being an anti-feature, it’s absolutely magical when you’re entering lots of tasks for a story.
16:23:56 <krotscheck> And it’s definitely a step up from the multi-click nonsense of expand-to-see-form that we had previously
16:24:12 <jeblair> krotscheck: yeah, though it breaks expected keyboard navigation behavior, so i wonder if we could try to accomodate both
16:24:40 <jeblair> story regarding erroneous timeline entries after tabbing out of a field: https://storyboard.openstack.org/#!/story/2000097
16:24:54 <krotscheck> What if we make the entire row editable?
16:24:54 <krotscheck> That is:
16:25:14 <krotscheck> Everything is a hyperlink. The row has an edit button. If you click it, all fields become editable.
16:25:37 <krotscheck> Click on save, and the whole thing saves.
16:25:44 <ttx> krotscheck: that would work
16:25:56 <krotscheck> that also fixes the story that jeblair just filed.
16:25:56 <krotscheck> s/fixes/addresses/
16:25:57 <ttx> saves icon space for sure
16:26:00 <yolanda> makes sense to me, actually looks clearer
16:26:07 <jeblair> krotscheck: would that still be magical enough when entering lots of tasks?
16:26:32 <krotscheck> jeblair: Entering lots of tasks ends up being a “Here’s a brand new editable row, click on save and we’ll give you a new one” thing.
16:26:39 <jeblair> krotscheck: gotcha
16:27:15 <krotscheck> Alright, let’s give that a shot and see how people feel about it.
16:27:54 <krotscheck> #action krotscheck switch task editing to toggleable-by-row.
16:28:20 <krotscheck> Any other user feedback?
16:29:38 <krotscheck> #topic Discussion Topics (Branch Support)
16:29:45 <krotscheck> Didn’t we finish this one?
16:29:49 <ttx> I think we did
16:30:00 <krotscheck> Cool.
16:30:03 <ttx> next one is task milestone @ https://review.openstack.org/#/c/139626/
16:30:15 <ttx> and then I have bug types up at https://review.openstack.org/#/c/129267/
16:30:39 <krotscheck> Alright, I’ll tag those on the end. I’m pretty sure we’ll get to task milestones.
16:31:13 <ttx> not a lot of reviews on those specs yet, I'll wait a bit before revving them
16:31:21 <krotscheck> #topic Paging (jedimike)
16:31:27 <krotscheck> jedimike is not here.
16:31:51 <yolanda> not sure about jedimike availability, i think he was having some doctor appointment these days
16:31:52 <krotscheck> #topic i18n
16:32:43 <krotscheck> Ok, so Aleksey’s been starting to contribute to the api codebase, and he’s done a lot of improvements on basic plumbing. In particular, he did this: https://review.openstack.org/#/c/142503/
16:33:12 <krotscheck> My question is: Is this the right time to have a discussion about i18n?
16:33:45 <krotscheck> …and to basically decide whether/what our long-term strategy might be.
16:34:40 <CTtpollard> I'm not sure I know enough about internalization to comment
16:35:26 <CTtpollard> or spelling for that matter it seems....
16:35:31 <krotscheck> Personally, while I definitely think that adding the translation hooks is a good practice, I’ve yet to hear of anyone wanting to use storyboard in not-english.
16:35:52 <krotscheck> CTtpollard: Just switch your dictionary ;)
16:36:04 <krotscheck> Does anyone else have an opinion?
16:36:05 <yolanda> i'm using spanish or catalan if i have translations available, but i suppose english is de facto language
16:36:40 <krotscheck> yolanda: You speak catalan? Neat!
16:36:51 <yolanda> i'm catalan :)
16:37:16 <krotscheck> yolanda: Have you ever worked with an app whose UI and errors are catalan/spanish, but whose user-generated content is english?
16:38:00 <yolanda> yes, for example i have jenkins in spanish but all tasks, logs, etc...are english
16:38:08 <krotscheck> How does that work for you?
16:38:19 <yolanda> it's natural to me and used to that
16:38:46 <krotscheck> Huhn.
16:38:51 <yolanda> i see jenkins messages in spanish but i assume that all the content is going to be in english
16:39:16 <krotscheck> That’s fascinating. I’ve always made the assumption that the UI translation should match the content.
16:39:49 <yolanda> doesn't need to
16:39:57 <krotscheck> How important would you rate it?
16:40:44 <yolanda> well, i assume that is not possible to have content in spanish that is generated by english people, i have grown with that so I don't see that like a problem
16:41:07 <yolanda> but i'm used to have the UI for ubuntu and all the available apps in spanish because is more natural to me
16:41:42 <yolanda> also for example the locale is important
16:41:48 <yolanda> as date formatting, numbers, etc... is different here
16:42:10 <krotscheck> yolanda: Where would you put internationalization on StoryBoard’s roadmap?
16:42:42 <ttx> (FTR as far as openstack is concerned, since all content will be in english... I rate it pretty low priority)
16:42:50 <yolanda> krotscheck, i don't see it like an urgent feature, but what i would do is at least to have it ready
16:43:13 <yolanda> the more messages you add, the more effort you need to put later to acomodate
16:43:54 <yolanda> so what i would do is at least to enable gettext as is done in the backend, so all messages are ready to be translated, but we only have english
16:44:40 <krotscheck> That seems fair. How about we add this to the roadmap in the “unplanned” section?
16:44:46 <jeblair> i think it's important for storyboard to have it eventually, but i don't think it's an early priority.  i would put it in mvp 1.3.1 or later.
16:44:49 <krotscheck> Or should this get mapped to a specific version?
16:45:53 * krotscheck is not-so-subtly trying to make yolanda do core-like roadmap things :D
16:46:12 <yolanda> heh :)
16:46:46 <yolanda> so i think jeblair comment makes sense
16:47:02 <krotscheck> Ok, I feel like everyone’s in the “not now, but let’s keep it in mind as we move forward."
16:47:27 <krotscheck> yolanda: can you do me a favor and add it to the roadmap?
16:47:27 <krotscheck> https://wiki.openstack.org/wiki/StoryBoard/Roadmap
16:47:34 <yolanda> sure
16:47:36 <krotscheck> Thanks
16:47:53 <krotscheck> #topic Discussion (python3)
16:48:05 <krotscheck> Similar thing: Aleksey did a bunch of work here: https://review.openstack.org/#/c/142733/
16:48:23 <krotscheck> Right now we’re on 2.7
16:48:48 <krotscheck> Does it make sense for us to aggressively upgrade python?
16:49:27 <jeblair> krotscheck: the thing with that is that unless we are actually testing under python3, it will bitrot
16:49:54 <krotscheck> jeblair: Point.
16:50:38 <jeblair> so i would say if someone wants to fix all the things to the point where we can enable python 3.4 tests, then sure, rip the bandaid off
16:50:59 <krotscheck> So we care about 3.3?
16:51:04 <krotscheck> s/So/Do
16:51:19 <ttx> I'd say no
16:52:04 <krotscheck> What I’m hearing is: “If someone wants to put in the effort, great, otherwise let’s keep shipping features"
16:52:18 <jeblair> yep
16:52:32 <krotscheck> Ok, moving on.
16:53:14 <krotscheck> #topic Discussion (Task Milestones): https://review.openstack.org/#/c/139626
16:53:14 <krotscheck> I may have broken the bot.
16:53:17 <krotscheck> There we go
16:53:26 <krotscheck> ttx? What’s up
16:54:00 <ttx> That once was updated to build on top of the approved task branch spec
16:54:13 <krotscheck> Just a rebase right?
16:54:28 <ttx> yes, still has 3 +1s
16:55:08 <ttx> so you might want to review it
16:55:29 <ttx> that said, we already have a sane pipeline of unimplemented features
16:55:34 <krotscheck> We do.
16:55:37 <ttx> so I don't think reviewing that is top prio
16:55:38 <krotscheck> But I will
16:55:50 <ttx> same for the story types spec
16:56:15 <ttx> That one was heavily rewritten to be more generally applicable and less hardcoded
16:56:21 <ttx> since that's the spirit of the times
16:56:29 <krotscheck> awesome.
16:56:44 <krotscheck> Are you comfortable switching topics to Open Discussion?
16:56:46 <ttx> I basically distilled the types we need for openstack into a series of properties
16:56:57 <yolanda> ttx, i added a comment about task mutation
16:57:07 <ttx> yolanda: yep, saw that
16:57:19 <ttx> yolanda: sounds valid to me
16:57:20 <yolanda> so it may be possible that not all states can mutate to others
16:57:41 <ttx> those properties can be combined to build custom types
16:57:58 * krotscheck has his interest piqued.
16:58:07 <ttx> so overall it feels a lot more flexible
16:58:35 <ttx> https://review.openstack.org/#/c/129267/
16:58:37 <ttx> for ref ^
16:58:45 <krotscheck> 2 minute warning
16:59:04 <ttx> krotscheck: it started as an alternative implementation, but then I liked it enough to replace the original one.
16:59:39 <ttx> I'm done, we can switch to open discussion
17:00:03 <krotscheck> ttx: That sounds like exactly the kind of thing that happens when people write specs, and why we like them so much :)
17:00:04 <krotscheck> Naah, we’re out of time.
17:00:04 <krotscheck> Thanks everyone!
17:00:04 <krotscheck> #endmeeting