| *** alexismonville has joined #storyboard | 01:06 | |
| *** alexismonville has quit IRC | 02:23 | |
| *** jtomasek has quit IRC | 04:42 | |
| *** matthewbodkin has joined #storyboard | 07:43 | |
| *** openstackgerrit has quit IRC | 08:03 | |
| *** openstackgerrit has joined #storyboard | 08:04 | |
| *** jtomasek has joined #storyboard | 08:07 | |
| pedroalvarez | Zara: answered to https://review.openstack.org/#/c/355554/7 | 10:44 |
|---|---|---|
| pedroalvarez | matthewbodkin: ^ it still looks weird to me, I'm not sure what is the problem though. | 10:44 |
| Zara | ah, my browser doesn't do that :( | 10:45 |
| Zara | thanks, I've changed vote | 10:45 |
| matthewbodkin | hmmm | 10:47 |
| *** fay has joined #storyboard | 10:48 | |
| *** fay is now known as Guest86071 | 10:48 | |
| *** alexismonville has joined #storyboard | 10:50 | |
| Zara | (also tried in chrome emulated views, and the scrollbars look fine, but the dash seems unusable on mobile. I think people do use it so my guess is the emulation isn't representative.) | 10:50 |
| *** faybrocklebank has quit IRC | 10:51 | |
| Zara | oh, I just didn't notice the 'you may need to refresh the page' hint | 10:51 |
| Zara | still can't reproduce. :/ | 10:52 |
| pedroalvarez | hm... | 10:53 |
| pedroalvarez | maybe that emulator hides the scroll bars | 10:54 |
| Zara | yeah. I just tried emulating a bunch of phones and tablets in chrome, and there were no visible scrollbars for any of them | 10:54 |
| Zara | well, it has 'enable text autosizing' | 10:55 |
| pedroalvarez | from phone scrollbar isnot visible, but i can scroll | 10:55 |
| Zara | so maybe zoom is affecting it | 10:55 |
| pedroalvarez | probably same behaviour from emulator | 10:55 |
| *** alexismonville has quit IRC | 10:55 | |
| pedroalvarez | ok, not annoying at all from phone | 10:56 |
| Zara | ooh, is it the emulator getting confused then? :D | 10:57 |
| Zara | (if nothing else we'll learn which emulators are representative for use in the future) | 10:58 |
| pedroalvarez | No, emulator let's you scroll too | 10:58 |
| pedroalvarez | but emulator emulates as well the invisible scrollbars | 10:59 |
| pedroalvarez | so, well emulated I guess | 10:59 |
| openstackgerrit | Matthew Bodkin proposed openstack-infra/storyboard-webclient: Move 'Save' button up in 'Preferences' page https://review.openstack.org/359119 | 11:00 |
| pedroalvarez | Zara: anyway, I happy to not -1 the patch if this is not a problem for anyone | 11:07 |
| Zara | okay, I'm going to +1 but not +2 since I don't know that my testing environment is representative, and that should mean we get another pair of eyes on it | 11:08 |
| matthewbodkin | when I access storyboard on my actually phone there's no scrollbar and even if I try I can't scroll | 11:08 |
| Zara | are icons cut off that you need to scroll to? | 11:09 |
| matthewbodkin | nope | 11:13 |
| Zara | I'll leave the +1 as a +1 then :) | 11:33 |
| *** Guest86071 has quit IRC | 11:35 | |
| *** Guest86071 has joined #storyboard | 11:35 | |
| *** jtomasek has quit IRC | 11:37 | |
| Zara | (btw, for context on the comments on the 'move save button up' patch-- we try to keep the 'what's displayed' separate from 'where it's displayed', so that it's easier to change one or the other. so while our html templates might say 'put this on the left', they don't say where the 'left' is relative to the edge of the screen; that's set in the css/less.) | 11:39 |
| Zara | it makes things easier to manage when all the little changes add up and changing the placement of one thing can break everything else. | 11:39 |
| Zara | because you don't have to work all over the codebase to fix it. | 11:40 |
| * persia decides to be more fussy if style="" sections are seen in HTML templates based on that guidance | 11:41 | |
| SotK | might be worth looking at how the :last-child or :last-of-type CSS pseudo-classes work to add a margin to the last `row` on every page | 11:42 |
| Zara | persia: sure, I believe style="" is rarely used in our html templates. | 11:43 |
| Zara | 4 instances, I think | 11:44 |
| * SotK thinks those 4 instances should go away | 11:45 | |
| Zara | some were in worklists so they might have been odd things | 11:46 |
| Zara | https://github.com/openstack-infra/storyboard-webclient/search?p=2&q=style&utf8=%E2%9C%93 | 11:47 |
| persia | I also think that, but based on wanting to separate these things, if I encounter it in patches, I'll complain, rather than just feeling vaguely dirty. | 11:47 |
| persia | Only 4? That's significantly cleaner than many projects. | 11:47 |
| * persia blinks at a 180s post delay | 11:47 | |
| SotK | o.O | 11:48 |
| Zara | (one in story search, one in story list, one in new worklist, one in worklist detail) | 11:48 |
| *** jtomasek has joined #storyboard | 11:49 | |
| Zara | yeah, I think it was designed that way from the start, rather than being tidied up after the fact | 11:49 |
| Zara | I suspect storyboard in general is a good environment for best practices in web-development, though I don't have a nice sample to check against. | 11:51 |
| openstackgerrit | Matthew Bodkin proposed openstack-infra/storyboard-webclient: Add a margin to the bottom of all pages https://review.openstack.org/359119 | 11:56 |
| Zara | \o/ | 12:06 |
| Zara | (ugh, forgot you can't do 'tasks matching tag and tasks matching status' for an auto worklist, as tags are story things and statuses are task things) | 12:23 |
| persia | Oh, ugh. That's going to be a commonly-requested feature. | 12:25 |
| SotK | yeah | 12:25 |
| * persia isn't sure of the best way to do things though: do tasks automatially inherit all tags from a story? | 12:26 | |
| SotK | I think that is the only sane approach we can take | 12:26 |
| Zara | + matching on stories gets stories rather than tasks-in-stories, which often isn't what I'd choose | 12:26 |
| SotK | I think tags are the only issue with that though right? | 12:27 |
| Zara | well, that's the same thing really | 12:27 |
| persia | Zara: Could you expand on that? | 12:27 |
| Zara | snap, hah | 12:27 |
| Zara | I was thinking that, if I do 'stories matching $condition', the worklist will be a list of stories, whereas sometimes I want all tasks that are in stories that meet that condition. so if you could filter tasks by properties of their stories then it wouldn't matter. | 12:28 |
| SotK | and I think that the only one that is useful that we can't do is tags | 12:29 |
| SotK | ? | 12:29 |
| Zara | yeah, I think so | 12:29 |
| Zara | since projects and project groups are common to both | 12:29 |
| Zara | possssibly story status (when combined with other filters), if you wanted to check that completed stories were actually complete, but that's too theoretical for me to think it worth spending much time on | 12:30 |
| SotK | I don't think that is very useful, since the story status is directly derived from the task status rather than stored in the db separately | 12:31 |
| Zara | gah, I meant the other way round, eg: for stories in project foo with story status !merged, display tasks | 12:32 |
| Zara | so you can spot if something had actually been merged on that list | 12:32 |
| SotK | ah, and people haven't been updating things you mean? | 12:33 |
| Zara | yeah | 12:33 |
| Zara | I like that I said the complete opposite of what I meant before... | 12:34 |
| SotK | you could emulate that with `Tasks where Project is foo and TaskStatus is not Merged` | 12:34 |
| Zara | yeah | 12:35 |
| Zara | I was trying to think if there was any difference but I don't see how | 12:35 |
| SotK | only difference is that you wouldn't see tasks related to other projects that are affected by the stories involved, which is probably not useful anyway | 12:35 |
| SotK | (since you could just add `Tasks where Project is bar and TaskStatus is not Merged` if you care about another project too) | 12:36 |
| Zara | yeah, I will decide to only care about tags, then | 12:36 |
| * SotK strongly thinks that things will rapidly descend into madness if we start making it possible for `Stories where Project is foo and StoryStatus is not Merged` to return a list of tasks anyway | 12:37 | |
| Zara | I'm happy to not introduce any more complexity than necessary | 12:38 |
| Zara | and yeah, you'd need to let the user choose | 12:38 |
| Zara | and then horror awaits | 12:38 |
| Zara | because that's just the same as choosing 'tasks' anyway | 12:39 |
| SotK | yep | 12:39 |
| * persia got distracted, and is delighted to have nothing to add at the conclusion of the discussion | 12:41 | |
| Zara | so far testing notifications suggest everything's working-- including the recent events templates, this time! :D | 12:57 |
| SotK | \o/ | 12:57 |
| *** alexismonville has joined #storyboard | 12:59 | |
| Zara | now I can see them, I'm not sure about the appearance of the cards in teh timeline (since saying 'position' on the cards makes it take up more room, but it's hard to get useful info from it since it's not in context. but either way it's good that the api side all seems to work fine | 12:59 |
| Zara | *the | 13:00 |
| Zara | I'm also not sure how much use it's going to get; I'm planning to turn it off anyway so I'm not sure I have much input on how they should look. | 13:00 |
| SotK | it looked even more meaningless without being card-shaped to me, and without position being there they will just be "a card changed" which is entirely useless | 13:01 |
| Zara | yeah, I wondered about moving 'position' outside the card body, but really it's in my bucket for 'people who want to use it can demo the ui they'd find most useful' | 13:03 |
| Zara | I don't know the right approach to it. | 13:04 |
| Zara | (and I think I'd end up leaning toward doing everything visually, so showing worklist before and after, and having 'private' with a checkbox ticked or unticked, etc, which would be complicated.) | 13:05 |
| persia | I'm missing some context here, but are these about notifications for worklist change events (like add card, move card)? | 13:05 |
| Zara | yeah, in the recent events list on the dashboard | 13:05 |
| SotK | Zara: not possible unless the worklist and its entire contents are tracked in the event info (which they currently are not) | 13:05 |
| persia | I think those are most interesting when one has many stakeholders to which one is beholden. For example, if the PTL increases the priority of some task within the set on a project in which you work, that might be interesting if your supervisor has recently deprioritised it, etc. | 13:07 |
| persia | Saves actually looking at all the lists on a regular basis. | 13:07 |
| Zara | SotK: yeah, I can see that getting awkward. it's motivated from a place of 'represent the thing in the same terms as its surroundings', so emails, it should be a textual representation, on a ui, visual, etc. | 13:07 |
| persia | Folk with clearer direction and goals likely don't care about seeing the updates. | 13:07 |
| SotK | (Zara: Also I worry that showing the full worklist may lead to it appearing to be editable, same with checkboxes) | 13:08 |
| Zara | SotK: yep, I think it's complex in a bunch of ways so it'd only be worth looking into if there was a good incentive. | 13:09 |
| persia | +1 on not showing the full worklists in recent events | 13:10 |
| persia | If people want to see it, there should be a navigation element that takes them away from the dashboard. | 13:10 |
| Zara | (SotK: if it tracked position and total of elements, you could have a dummy list that just displayed the diff in position of the element) | 13:11 |
| Zara | greyed out items, something like that, idk | 13:11 |
| Zara | *greyed out and textless | 13:11 |
| SotK | that doesn't really provide anything superior to a number though | 13:11 |
| SotK | and uses a lot more space | 13:11 |
| Zara | yeah, it gives context, though. right now the number doesn't communicate much. | 13:12 |
| Zara | and stops a user having to pause to parse it. but I don't think it's sufficient benefit for the effort. | 13:13 |
| SotK | the useful information from the context is seeing *what* it was moved above/below, which a bunch of blank cards don't help with | 13:13 |
| Zara | I was thinking for situations where a worklist suddenly grew, but tbf other notifications would cover that. though we're assuming a user knows which end of the list the position is relative to. and I still think it's way easier to skim it laid out than to find at the numbers and work out the difference. | 13:16 |
| Zara | s/at// | 13:16 |
| Zara | let's have all the ascii worklists | 13:17 |
| Zara | http://paste.openstack.org/show/562462/ | 13:22 |
| persia | I think a large portion of the confusion is because the view is new. If users only ever see notifications with numbers, they will mostly ignore the numbers, until they ask "what does this number mean", after which they will probably say "cool" | 13:22 |
| Zara | they're about three times the size of most of our notifications | 13:28 |
| Zara | and the relevant bit of information is at the bottom of the card, which the user has to compare manually to get the meaning from. | 13:29 |
| persia | Hrm. True. | 13:30 |
| * persia will look more closely before having more opinions, but had expected something like "worklist foo changed, task bar in position i, was j. | 13:30 | |
| Zara | an example in my test instance http://imgur.com/a/07YZw | 13:34 |
| Zara | that shows the difference in size | 13:34 |
| Zara | once the card title gets to about 9 words, it wraps, so they can be bigger. | 13:36 |
| persia | And this is why I should look at more things in browsers than text editors. That's massively confusing. | 13:37 |
| *** davidlenwell has quit IRC | 14:36 | |
| *** davidlenwell has joined #storyboard | 14:37 | |
| *** fay has joined #storyboard | 14:41 | |
| *** fay is now known as faybrocklebank | 14:41 | |
| *** Guest86071 has quit IRC | 14:43 | |
| *** fay has joined #storyboard | 14:45 | |
| *** fay is now known as Guest72725 | 14:46 | |
| *** faybrocklebank has quit IRC | 14:49 | |
| *** Guest72725 is now known as faybrocklebank | 14:49 | |
| *** jtomasek has quit IRC | 15:08 | |
| *** davidlenwell has quit IRC | 15:27 | |
| matthewbodkin | I've just been using https://www.browserstack.com/start to test out the sidebar for length and if any icons are cut off or if there is a horizontal scrollbar and all was good on every device | 15:27 |
| matthewbodkin | I don't know how reliable this is but just thought it was worth a mention | 15:28 |
| *** davidlenwell has joined #storyboard | 15:40 | |
| *** matthewbodkin has quit IRC | 15:59 | |
| *** bethwhite_ has quit IRC | 16:44 | |
| *** faybrocklebank has quit IRC | 17:06 | |
| Zara | realised I should actually put review notes in review... | 18:14 |
| Zara | oh yeah, re: browserstack-- nice to see it works in another place, hopefully there'll be more reviews soon! we generally prefer to use open source tools; from the discussion above it looks like the emulation in chromium was representative after all, so it should be possible to switch to that. | 18:22 |
| Zara | + if there's something that that tool does that isn't available there, would be good to check if there's another open source thing that offers it. | 18:28 |
| Zara | (so also, I'm being hypocritical when I link to github; I do it because it's much quicker to query and find a relevant bit of code than other options I've found so far, but it is a faux-pas. http://codesearch.openstack.org/ is the best alternative, it's just a longer route to navigate) | 18:37 |
| Zara | which sometimes isn't feasible in a realtime irc conversation, but can be a good option for linking in review comments | 18:38 |
| *** jtomasek has joined #storyboard | 19:18 | |
| *** jtomasek has quit IRC | 19:20 | |
| *** jtomasek_ has joined #storyboard | 19:20 | |
| *** aswadr_ has quit IRC | 20:31 | |
| *** alexismonville has quit IRC | 20:44 | |
| *** alexismonville has joined #storyboard | 23:39 | |
| *** alexismonville has quit IRC | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!