15:02:06 #startmeeting storyboard 15:02:06 Meeting started Wed Jun 22 15:02:06 2016 UTC and is due to finish in 60 minutes. The chair is Zara. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:02:07 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:02:09 The meeting name has been set to 'storyboard' 15:02:12 sorry we're a little late 15:02:27 #topic Announcements 15:02:32 none on agenda, but... 15:02:43 review-dev is now talking to storyboard-dev 15:02:46 \o/ 15:02:46 via zaro's gerrit plugin! 15:03:03 I hope we can update it to do wonderful task status things 15:03:07 AT SOME POINT 15:03:14 thanks for that zaro, anteaya, pleia2, and others who were involved 15:03:17 yes! 15:03:30 yay 15:03:36 other things... 15:03:42 the storyboard bugsquash is ongoing! 15:03:46 yay 15:03:51 SotK has been squashing api bugs 15:04:04 yay 15:04:22 I've been listing all the ui bugs that I keep forgetting to list 15:04:30 will hopefully get round to actually *squashing* some 15:04:54 any announcements I missed? 15:05:03 to clarify we have the channel for 48 hours 15:05:09 wednesday and thursday 15:05:16 so the bug squash is two days 15:05:29 ohh, I had it down as 11:00 UTC to 11:00 UTC 22nd-23rd 15:05:32 but folks can spend as much time on it as they please 15:05:55 well if folks are around after 11:00 utc on the 23rd I'll still be around to help 15:06:02 SURPRISE EXTENDED BUGSQUASH 15:06:05 I had it down as two days 15:06:06 well it's business as usual for us anyway 15:06:08 so that's fine 15:06:08 woooo 15:06:11 :) 15:06:13 if nobody else has taken the channel 15:06:19 betherly ^ I was wrong before! 15:06:21 I booked it for 2 days 15:06:54 fine by me 15:07:22 anything else? 15:07:38 I don't know of anything 15:07:55 #topic urgent items 15:08:10 urgency is against the ethos of storyboard. 15:08:16 but saying that, any urgent items? 15:08:47 I got nothing 15:09:07 #topic in-progress work 15:09:53 I'm working on updating some of my old patches and things 15:10:35 I rarely get a big block of time to focus long enough to see anything complex through, so we'll see how far I get. 15:11:14 storyboard and gerrit can talk to each other now, that was my in progress thing 15:11:22 I'm looking at making comments editable, as the next part of the timeline rework adventure 15:11:27 :) 15:11:31 they already are editable, but not in the UI 15:11:36 patches still need to be merged for this to be puppeted but it is up on dev servers now 15:11:39 thanks zaro 15:11:58 yeah, I think it's a good first step 15:12:06 oh I had thought that ensuring comments were not editable was important to the timeline 15:12:11 fungi: can you confirm this 15:12:26 SotK: is there an audit trail for comment edits? 15:12:33 in launchpad a comment can not be edited so it is valuable as a preservation of history 15:12:43 persia: https://review.openstack.org/#/c/332208/ exists to provide that 15:12:54 it would disrupt the flow of conversation if someone could go back and edit a comment 15:12:54 comments have been editable via the api the whole time 15:13:05 the gerrit work is a good first step; I want it to update task statuses so hopefully can get to that next week 15:13:08 Then that should make it safe. 15:13:13 unless anyone else volunteers 15:13:56 okay I did not know the api allowed a comment to be edited 15:14:04 anteaya: i don't recall saying that comments shouldn't be editable, but if you know who did they would be the ones to ask about that concern 15:14:28 i know lp (and most other bug trackers) don't allow you to change comments after the fact 15:14:28 fungi: oh okay so as far are you are concerned, being able to edit a comment is fine? 15:14:40 i hadn't thought about it, to be honest 15:14:44 ah sorry 15:14:57 perhaps it was jeblair who said something about it 15:14:58 i just assumed they weren't editable since that's not typical of most bug trackers 15:15:38 it definitely alters some dynamics of teh system if people can go back and change what they said in the conversation (even if there is an audit trail to be able to find out they changed it, assuming you know to look) 15:16:03 From my perspective, it is important to know what was originally written, for all sorts of reasons, but that is not always the best presentation (especially when people comment angry or tired). 15:16:25 I agree with persia 15:16:39 i tend to view comments on a bug report as similar to a mailing list where people won't know to go back and reread what you said earlier to see if it changed, they tend to just assume conversations are append-only 15:16:50 For example, many forums have an edit feature, but make it clear that edit happened. 15:17:06 I'm in the append only camp 15:17:27 if a comment discussion is long and has a history I think the history needs to be preserved 15:17:39 folks get tired posting to mailing lists too 15:17:41 LP has only-admin-can-edit, which mostly means admin-deletes-iffensive-statements. 15:18:10 I'm not opposed to a storyboard admin being able to delete spam comments 15:18:44 I'm not in favour of someone posting something, kicking off a larger conversation, and then going back and changing the original comment 15:18:54 I still want an audit trail, but have no strong objection to ACLs for edit. 15:19:09 my plan was/is to make it clear that a comment has been edited, when the edit happens 15:19:19 and have all versions available to see in the UI 15:19:26 i'd mostly be concerned about confusion if people who were following the original conversation have a divergent recollection/understanding from people who come later and catch up reading the altered comments, and then start talking past each other because they don't realize they're working from two different versions of a discussion 15:20:16 Good point. 15:20:21 it might not be an issue for a story with a dozen comments, but when it has hundreds people won't necessarily know to go back and find the modified comments they'd previously read 15:20:40 or the audit trail 15:20:49 Especially if some people are participating.ostly be email notifications. 15:21:25 so you may make it obvious to people who are just now reading those comments that they've been changed, but the people who read them before won't necessarily be aware (perhaps active notification via e-mail solves for some of that?) 15:22:01 I can see this creating a lot of frustration 15:22:05 rather than clarity 15:22:20 * SotK finds being unable to go back and fix formatting/bad wording really annoying too 15:22:32 well we have a preview button 15:22:35 Active notification by email might just confuse more: out-of-order typo fixes might appear to be reiterations. 15:22:39 and the link to the markdown guide 15:23:29 The above notwithstanding, I think I like "comments are editable only by API users" least if all the options. 15:24:30 i find being unable to "fix" previous comments is a concern shared mostly by people who didn't spend decades relying mostly on usenet and mailing lists, where once you've said something it's out there forever. after a while you realize that your typos and errors are no worse than anyone else's, but you do hopefully think twice (and double-check) before you post something 15:25:25 Which is an increasing proportion of the population, sadly. 15:25:27 that said, if there are stakeholders in storyboard for whom being able to treat it more like a web forum with mutable commentary is a desirable use case, i see no reason not to make it a configurable option 15:25:49 yeah expecting our users to constantly re-read entire timelines is unrealistic I think 15:26:22 I don't see bug reports as a forum 15:26:30 so we could have storyboard.openstack.org configured not to allow non-admins to alter comments but someone else deploying a storyboard instance could enable mutable comments 15:26:50 a global config option perhaps 15:27:08 fungi: do you envision UI for the editing admins? 15:27:55 i don't recall if i ever said anything about editing comments, but if you asked me now, i would say i don't think they should be editable 15:28:02 persia: hadn't really thought about it. honestly for sb.o.o we would probably get away fine with a "hide comment" feature, or even just an api call 15:28:36 persia: in fact, if there was no feature for it at all, we would likely just use sql queries 15:28:51 in the (hopefully very rare) cases where it becomes necessary 15:29:37 jeblair: oh sorry if I mis-remembered but thanks for sharing your thoughts on the topic now 15:30:13 I believe the current state to be "comments can only be edited via API". This seems to be from long ago. I do not have a strong opinion about who should edit, but I want to +1 an audit trail patch, or -1 it in favour of admin-only on the API and no UI. 15:30:28 The current situation seems the least good. 15:30:49 I'm not sure its really worth the effort of making it configurable, if we don't want editable comments we may as well just remove the PUT endpoint for comments imo 15:30:49 can we disable comments can be edited via the api? 15:30:50 persia: ++ 15:31:13 SotK: I'm in favour of that direction 15:31:29 SotK: is removal easier than admin-only? 15:32:14 equal ease, less susceptible to accidentally being broken 15:33:00 If admins usually have DB access, then removal seems safer. 15:34:42 the nova bugs team were keen on having the ability to edit, iirc 15:34:46 * Zara looks for etherpad 15:34:58 https://etherpad.openstack.org/p/nova-bugs-team 15:35:08 might be a bit old by now 15:35:10 * SotK also wonders why editing comments is any worse than editing the description of a story 15:35:27 but that's how the task ended up listed in the first place, iirc 15:37:06 Zara: perhaps you and I can check with marcus from nova after the meeting 15:37:10 and get some clarity 15:37:17 if he is around this week 15:37:42 SotK: the description of a story isn't a conversation 15:38:02 like channel logs often I talk to you about something that I talked with someone else about yesterday 15:38:12 and link you to the channel logs so you can read the converstaion 15:38:13 but it is still the thing which causes the conversation, and the thing the conversation is about 15:38:41 if the channel logs change between my recollection and your reading of them, you and I don't have the same starting point for our converstaion 15:40:06 Zara: I see it on line 110: https://etherpad.openstack.org/p/nova-bugs-team 15:40:20 perhaps we can share the log of this meeting with marcus and then get his thoughts 15:40:36 SotK: I can't defend editing a story description 15:43:01 I'd really rather have an example implementation in review, then if people find it causes problems, we change it 15:43:10 I think that comment threads and references to previous conversations are two different kinds of discussion 15:43:32 otherwise we'll just keep discussing it. 15:43:57 anteaya: +1 for getting the thoughts of the nova bugs folk before making a decision 15:45:43 I'll continue approaching a solution that allows history and such to be displayed until we make that decision, since I've already done most of the work anyway 15:46:24 okay fair enough 15:46:28 thanks 15:47:40 np :) 15:47:58 worth noting (not saying lp feature parity is a requirement) but you can edit bug descriptions in lp 15:48:02 just not the comments after them 15:48:45 in a disagreement I would advocate lp feature parity would be the deciding factor 15:48:57 as the community already understands that behaviour 15:49:12 my general approach is "wait and see what sotk does, and then cross the 'this doesn't match what people want' bridge if I have to", because he tends to be very good at guessing. 15:49:50 okay once we have a reference impliemenation for folks to look at we can gather more info 15:50:50 okay, any other in-progress work? 15:51:00 that ended up being a discussion topic :) 15:51:34 xD 15:51:46 #topic Open Discussion 15:51:49 10 mins, go for it! 15:51:51 9 mins now 15:51:59 in other timeline news, there is now no pagination 15:52:04 oh yeah! 15:52:06 nice one 15:52:07 and there is permalinking to comments 15:52:25 yup 15:52:27 and I sent patches for improving filtering a little 15:52:42 ... whiiich I haven't got to yet 15:53:15 well I glanced at the ui and I had some comments to make but I'll look a bit closer first. 15:53:38 yay no pagination 15:53:46 and yay permalinking 15:57:40 are we just waiting for the clock to wind down? 15:57:49 I suspect so 15:58:09 i'm happy to answer any questions i missed in these last two minutes if you need 15:58:09 okay well thanks for the meeting, see you in the sprint I have to go do something now 15:58:19 back in a few 15:58:29 but yeah, see you all in the sprint! 15:58:31 hehe, sure 15:58:39 I don't think I have any questions right now 15:58:44 will probably remember afterward 15:58:57 and yeah, kept the meeting open in case someone wanted to discuss something 15:59:01 but I am not that someone 15:59:12 anyway, it's open discussion all day every day in #storyboard! 15:59:15 hope to see you all there! 15:59:24 and in #openstack-meeting for an extra day, apparently 15:59:25 :P 15:59:30 sorry about that again 15:59:38 will end this meeting now 15:59:41 #endmeeting