[15:40:19] Today is the day they plan to shut down one of the elevators at 149, from now until Mid May expect only one operational elevator at a time. [15:50:50] MID MAY? [15:50:58] what are they gonna do? :) [17:40:24] @Mark They are replacing pretty much every part of each elevator and renovating the interiors. [22:01:26] starting meeting E134 in just a sec [22:01:44] #startmeeting https://phabricator.wikimedia.org/E134 [22:01:45] Meeting started Wed Jan 13 22:01:44 2016 UTC and is due to finish in 60 minutes. The chair is robla. Information about MeetBot at http://wiki.debian.org/MeetBot. [22:01:45] Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. [22:01:45] The meeting name has been set to 'https___phabricator_wikimedia_org_e134' [22:01:48] o/ [22:01:59] \o [22:02:20] #topic Wikimedia meetings channel | Please note: Channel is logged and publicly posted (DO NOT REMOVE THIS NOTE) | Logs: http://bots.wmflabs.org/~wm-bot/logs/%23wikimedia-office/ | Topic: https://phabricator.wikimedia.org/E134 [22:02:38] we should really fix that bit in meetbot [22:02:46] (to keep the disclaimers) [22:02:59] yeah, good enough for now :-) [22:03:36] so...this is an informal retrospective on the Dev Summit [22:04:07] first off: anyone who wasn't there that wants to ask something about it? [22:04:15] * robla waits a sec [22:04:51] I wasn't there, and I have no idea went on. hopefully nothing scary was decided. :p [22:05:03] (notes are here: https://www.mediawiki.org/wiki/Wikimedia_Developer_Summit_2016 ) [22:05:20] YairRand: we decided all sorts of scary things! ;-) [22:05:22] were there any pain points that kept coming up, things that deserve allocation of resources? [22:05:47] for example I remember shadow namespaces were mentioned a couple of times [22:06:04] legoktm: can you speak to that one? ^ [22:06:23] YairRand: a lot of the conversations were of the "Problem definition" variety described here: https://www.mediawiki.org/wiki/Good_meetings#Taxonomy [22:07:05] I'm going to be working on shadow namespaces starting in April [22:07:39] (with official resourcing and stuff :)) [22:08:03] * robla just edited [[mw:Good meetings]] today as he was writing this comment: https://phabricator.wikimedia.org/T118473#1932535 [22:08:48] legoktm: cool! when was that decision made? [22:09:24] Yay, shadow namespaces! [22:09:52] There was quite a bit positive feedback for introducing dependency injection. I'll be working on that over the next couple of weeks, updating my rfc and code experiments. [22:10:44] Also, Editing (JamesF) seems determined to go ahead with multi content revisions. I'll be involved with the specs I expect, but no so much with implementation. [22:10:46] robla: I had talked about it with a few people beforehand, it's mostly confirmed as of mid-December [22:12:24] One thing I heard mentioned in a couple of sessions was a desire for a "Roadmap" for MediaWiki. I wsa never quite sure what people were looking for exactly with that. [22:13:09] I suppose a product vision of non-trivial feature enhancements [22:13:14] #link https://phabricator.wikimedia.org/T115762 <- shadow namespaces discussion at WikiDev '16 [22:14:09] One unplanned meeting we had at the summit was about working group structure [22:14:10] we have roadmaps for teams already [22:14:18] #link https://etherpad.wikimedia.org/p/WorkingGroups [22:14:35] TimStarling: true. the WMF is getting better at that all the time IMO [22:14:44] the roadmap discussion I think plays into the working group conversation a bit [22:14:56] o/ [22:15:35] \o [22:16:00] but you can't expect every last change to be in someone's roadmap [22:16:12] the only way to do that would be to reject changes on the basis that they weren't planned [22:16:24] the RFC process & working group discussion is about improving the way we make decisions / build consensus, so definitely related to building up a shared roadmap [22:16:32] robla: I didn't make it to that session. I have head the Rust model of governance mentioned by gwicke. I sounded interesting [22:16:58] *heard [22:17:09] https://github.com/rust-lang/rfcs/blob/master/text/1068-rust-governance.md [22:17:23] we plan to make the governance model discussion be the topic for next week [22:17:36] #link https://github.com/rust-lang/rfcs/blob/master/text/1068-rust-governance.md [22:18:43] http://www.ncameron.org/blog/my-thoughts-on-rust-in-2016/ has a summary of how the RFC process went last year: "In 2015, the Rust community: Created 331 RFCs, of which 161 were accepted and merged, 120 people submitted an RFC" [22:19:27] Next week's meeting: https://phabricator.wikimedia.org/E135 <- please put in reading material to read prior to that meeting in this Phab event [22:19:56] copied this over [22:22:15] so....the models that I'm aware of that we should study better: Rust, IETF, W3C, Python, WHAT-WG, any others? [22:23:54] * robla shifts in his chair in the uncomfortable silence :-) [22:24:49] yeah, there's not much engagement in this meeting [22:24:50] swift has a similar process [22:24:50] So... was there anything *missing* from the summit? [22:25:00] https://twitter.com/clattner_llvm/status/676472122437271552 [22:25:02] was there anything any of you would have wanted to discuss? [22:25:09] ...but we didn't have room for it? [22:25:52] quick question: how thorough are the notes? can I assume that if it's not in the notes, it wasn't discussed? [22:26:12] YairRand: I wouldn't go that far, but they are better than usual [22:26:31] pretty much every meeting had an assigned scribe [22:26:48] swift process description: https://github.com/apple/swift-evolution/blob/master/process.md [22:27:43] DanielK_WMDE: I guess for me I would have been excited to see more assigned owners/champions and action items but I felt that we covered a lot of ground for 2 days of solid meetings [22:27:50] YairRand: I'm very happy with the notes we took in the "Make code review not suck" session, which is where I experimented with a notetaking technique I've continued to use post summit. [22:28:10] And I think that we are trying to get better about following up. It will take practice [22:28:35] bd808++ [22:29:20] robla: perhaps debian, among others with http://dep.debian.net/ [22:30:38] jzerebecki: thanks, that's a good one for us to learn more from. are there any things Debian does you wish we did? [22:31:15] I heard from several participants that this dev summit was more constructive and less confrontational than past editions [22:32:58] gwicke: yeah, I'll take the blame for what happened at the Architecture Summit '14....I tried to bias everything toward "consensus meetings" as described here: https://www.mediawiki.org/wiki/Good_meetings#Taxonomy [22:33:45] robla: :) yes, but what quickly comes to my mind is probably not related to an rfc process and i'm not familiar with how the DEP process works in practice. [22:34:35] so maybe some of debian devs working for WMF can be asked to comment on that [22:35:08] good point [22:35:17] I agree, this years summit was much better than last years :) [22:35:48] robla: I think it's also a reflection of more shared ground on the overall direction [22:36:11] bd808: yea, I think focused problem solving discussions were pushed mostly to the unscheduled day, while the scheduled sessions here mostly intended to facility cross-cutting discussions of broader topic areas. [22:36:13] #idea jzerebecki> so maybe some of debian devs working for WMF can be asked to comment on [the Debian process] [22:36:52] robla: were any people blacklisted from attending this years summit? [22:37:26] jzerebecki: I'm not sure....I don't think so. [22:38:44] I also thought this year's summit was much better [22:38:55] The timing could have been better (not so close to new years), but I'm not sure what could've been done about that [22:39:26] timing was perfect for me [22:39:34] even if it was after the all staff instead of before, it would have been much better for me [22:39:36] #idea don't schedule a Dev Summit so close to when a major sporting event is also scheduled ;-) [22:39:56] two weeks later and I wouldn't've been able to attend [22:40:22] two weeks later an hotels will be umpty gazillion dollars [22:40:34] I only managed to find a few hours to read my email backlog and respond to a few things between christmas and dev summit [22:40:35] (and already fully booked) [22:41:56] I think an extra few days for pre-summit meetings would have improved the summit [22:43:16] I would have preferred more small meetings (or less overlap between them), even if to the detriment of the bug ones [22:43:34] the unconference meetings worked really well, the scheduled ones I went to not so much [22:43:35] the scheduling and venue stuff is perfect for attendees to comment on in the survey that rfarrand sent out [22:44:43] TimStarling: +1, the transition from holiday / slow period to dev summit was fairly abrupt [22:45:25] I believe she's closing off the survey very shortly, compiling the results next week, and planning on publishing the results of the survey shortly thereafter [22:47:07] #link https://phabricator.wikimedia.org/T123400 <- general WikiDev '16 retrospective task [22:51:32] tgr: what didn't go well about the scheduled sessions from your point of view? [22:52:55] they did not have a clear goal, did not fall into any of the categories in your schema and it wasn't clear by the end that anything was achieved [22:53:21] I only went to two, so totally not representative [22:54:01] tgr: mind, saying which ones? [22:55:03] in general I think large meetings have less throughput (number of people speaking at the same time per number of participants is smaller) so we should be careful to only do them when needed [22:55:04] * robla needs to start winding this down [22:55:31] they are good for consensus decisionmaking but not so much for the other meeting types [22:56:31] tgr: I think we need to get better at big meetings myself. link: http://www.3rd-force.org/meetingnetwork/readingroom/meetingguide_rapid_concensus.html [22:56:35] I actually think that making big decisions is hard in large meetings, unless there has been a lot of preparation & most participants are well familiar with the topic [22:56:46] jzerebecki: non-Wikimedia deployments, software engineering and code review [22:56:51] I understand it's frustrating and hard work, though [22:56:54] (yes, that's actually three) [22:57:33] gwicke: sure, the easiest is a single person making the decision :) [22:57:40] ok.....I think I'm going to end the official part of this meeting in 60 seconds [22:57:46] but you get more legitimacy with more involvement [22:57:59] there is much to be said for having time to reflect while homing in on a consensus [22:58:38] next week we plan to talk in more depth about the working area/working group process, and generally about governance [22:59:07] in person synchronous meetings are detrimental to reaching better consens [22:59:14] #link: https://phabricator.wikimedia.org/E135 <-next week [22:59:59] s/better/wider/ [22:59:59] tgr: I hope that we can handle more of this online, in an asynchronous fashion, with broader participation [22:59:59] #link https://phabricator.wikimedia.org/E135 <-next week [23:00:02] ok...really ending [23:00:09] #endmeeting [23:00:10] Meeting ended Wed Jan 13 23:00:09 2016 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) [23:00:10] Minutes: https://tools.wmflabs.org/meetbot/wikimedia-office/2016/wikimedia-office.2016-01-13-22.01.html [23:00:10] Minutes (text): https://tools.wmflabs.org/meetbot/wikimedia-office/2016/wikimedia-office.2016-01-13-22.01.txt [23:00:10] Minutes (wiki): https://tools.wmflabs.org/meetbot/wikimedia-office/2016/wikimedia-office.2016-01-13-22.01.wiki [23:00:11] Log: https://tools.wmflabs.org/meetbot/wikimedia-office/2016/wikimedia-office.2016-01-13-22.01.log.html [23:00:24] gwicke: no disagreement there but that hasn't worked out any better in the past [23:01:00] yeah, many RFCs don't see much discussion on the task itself [23:02:01] but, I think there are ways to encourage that