[00:00:25] session finished, the notes are available here: https://etherpad.wikimedia.org/p/devsummit17-asynchronous-processing [00:01:25] Thanks volans ! [00:04:30] * yuvipanda will miss dev summit [00:04:45] Have fun and be productive everyone :) [00:05:12] Fun?! [00:05:16] IN OUR MEDIAWIKI?! [00:05:25] ...it's more likely than you think. [00:13:15] Next pile of sessions in 7 minutes. [00:18:09] Platforms are Products too! [00:18:09] NOTES: https://etherpad.wikimedia.org/p/devsummit17-platforms-are-products [00:18:09] Bringing "Enterprise MediaWiki" to the Wikimedia Foundation [00:18:09] STREAM: https://www.youtube.com/watch?v=Gj116oUF5-U [00:18:09] NOTES: https://etherpad.wikimedia.org/p/devsummit17-enterprise-mediawiki [00:18:10] Cross-wiki templates, translatable, reusable, manageable [00:18:12] STREAM: https://www.youtube.com/watch?v=IzG7unxMtMc [00:18:14] NOTES: https://etherpad.wikimedia.org/p/devsummit17-xwiki-templates [00:18:16] Support global preferences [00:18:19] NOTES: https://etherpad.wikimedia.org/p/devsummit17-support-global-preferences [00:18:45] All starting soon, probably. [00:24:01] CROSS-WIKI TEMPLATES IS HAPPENING. [00:24:17] Okay, so's the enterprise one too. [00:28:07] Wow, there's birds of paradise growing outside the window in the enterprise one. [00:28:51] If you all have questions regarding the Brining Enterprise MediaWiki to the Wikimedia Foundation let me know. [00:29:25] Many thanks to Isarra for sharing the links. [00:30:19] You're welcome. [00:30:37] enterprise mediawiki seems to be dropping a lot of frames. is that just me? [00:30:46] I'm getting the same. [00:30:56] Well, I was. [00:31:46] Also the audio quality is way worse than the templates one. In case that is remotely useful to anyone. [00:31:50] I'm seeing a "This event will begin in a few moments" on YouTube. [00:33:01] Try reloading? I have no idea. [00:35:04] he forgot composer [00:38:07] Developers. [00:38:08] Developers. [00:38:08] Developers. [00:38:09] Developers. [00:38:10] Developers. [00:38:34] Devlopers? [00:38:52] Developers. [00:39:16] Developers! [00:40:04] Developersss!!! [00:40:10] Developers! [00:44:36] Developers???? [00:44:47] Developers!!!! [00:51:52] Street sweepers! [00:52:13] (I hope you all don't hear that on the stream) [00:53:17] Not too bad. [00:53:31] SHOW ME THE MONEY [00:53:44] tgr_: I've created https://www.mediawiki.org/wiki/Developer_Wishlist as a draft. :-) [00:53:47] Ahahahah I heard it start up and go away in one and now I hear it in the other... XD [00:55:56] "Street sweepers" has been said so many times in this channel today... [00:57:20] Apparently there is also a gun that was called a street sweeper for a time... https://en.wikipedia.org/wiki/Armsel_Striker [00:58:24] yes [00:58:55] Sweep the streets. Sweep the wikis. The databases. The templates. The developers. [00:59:17] this is what i get reminded of when hearing the word streetsweeper! an old roommate told me he saw it sold at a US gun show [00:59:47] Horrible name for a gun >.> [01:03:26] "due to weather issue in SF the flight will have a minor delay. Boarding time sooner or later"... [01:03:29] Twitter has a picture of the unconference board, but it's low res. Any possibility of a high res version? [01:03:57] Tpt: >__< [01:04:14] yes... [01:04:30] Just because something is a hack doesn't mean it's bad, though... [01:04:55] rfarrand: here? [01:08:02] MediaWiki Foundation [01:08:19] Reedy: !HSMWF [01:08:21] ;-) [01:08:46] I'm sure that was discussed last year (and the year before) too ;) [01:09:18] Wikipemediawikiwiki foundation. You know you want it. [01:09:29] It needs someone to write a convincing argument for the WMF Board and have them approve it (and thus reverse a huge set of previous serious discussions). [01:09:37] Just talking about it does nothing. [01:11:08] Just curious, are any of the board here? [01:11:21] (in the room) [01:11:24] No. [01:19:16] matanya: yup, whats up? [01:21:37] in here for about 5 more min and then leaving. If you send me an email ill try to answer pretty quickly. If you are here already you are welcome to stop by WMF this evening [01:21:42] matanya ^^ [01:21:49] Why not mix local and global templates? We mix global and local images and it's greeeeaaaaaat. [01:26:49] Also I like how they did this entire discussion in front of a blank blue screen. And it was fine. [01:27:09] Especially how everyone was still standing slightly to the side of the screen, so we could still see the screen. [16:37:07] Good morning SFO and good later morning, afternoon and evening to other participants. We'll be kicking things off today in about an hour, with a Wikimedia Foundation Technology and Product Q&A. You can tune in here: https://www.youtube.com/watch?v=dnhECg-2UJs [16:44:35] hi Melody. [16:47:46] hi Elitre! [16:48:06] Notes for the first session will be here: https://etherpad.wikimedia.org/p/devsummit17-dev-and-product-q&a [17:06:39] It looks like notes for the first session will be here: https://etherpad.wikimedia.org/p/devsummit17-EasyTasks - I will update the program. [17:12:05] Melody, that looks like the pad for a different session? [17:12:40] You are right Elitre - sorry - got confused. Ignore that! [17:21:00] Hello all! We'll be getting started with the keynote in about 10 minutes. Here's everything you need to know. Please note that the streaming link has been updated. [17:21:00] STREAM: https://www.youtube.com/watch?v=f0JdtauJkKs [17:21:00] NOTES: https://etherpad.wikimedia.org/p/devsummit17-dev-and-product-q&a [17:21:00] PHAB TASK: https://phabricator.wikimedia.org/T153063 [17:32:41] The sound just got a lot less echo-y. Thank you! [17:32:45] (on the stream)' [17:34:43] Is there a designated notetaker for this session? [17:35:39] yes, there are 3. [17:35:58] good morning [17:36:28] Note-Taker(s): @ksmith, @Quiddity, @Aklapper [17:36:34] So the rule for asking questions today is, you're welcome to ask IF you are going to be involved with the answer after the Summit ends. Thanks! [17:36:37] Thanks! The notetaking document is here: https://etherpad.wikimedia.org/p/devsummit17-dev-and-product-q&a [17:37:39] brendan_campbell: are you doing the stream this morning? is it possible to move the camera a bit? can't really see the speakers on stage... [17:37:43] I don't know if this is logistically feasible, but since there are no slides, it would be great if the camera was closer to the stage and showing the the people talking, rather than the backs of the people sitting. [17:37:51] Jinx! [17:38:55] I actually hadn't volunteered for notes, but can try to help [17:39:15] meeple27: You have been volunteered! [17:39:29] voluntold [17:39:34] Right. [17:42:28] I mean, because of the in-front-of-the-stage choice, you probably had a better view of the speakers than those of us actually in the hall anyway. [17:42:34] guillom: unfortunately this is the closest i can get. [17:42:57] i didnt know we were streaming in this room until yesterday morning, so i have to make due the best i can. apologize for the inconvenience [17:43:21] unless anyone in this room has a 50 foot hdmi cable... [17:43:32] brendan_campbell: no problem, I'm appreciative of the stream regardless. Thank you! [17:43:54] Great audio brendan_campbell! [17:43:57] what guillom said [17:44:07] the audio is better than yesterday morning imo [17:44:10] thank you, i'm pulling audio off the mixer in the back (hence the view) [17:44:29] * aude waves [17:44:36] hey aude. [17:45:08] this session is also being professionally recorded, so that will be available at a later date [17:45:36] and the camera is only on the speakers there :) [17:46:17] who is speaking? victoria? [17:46:45] YEs [17:46:50] ok [17:48:37] aude: You can see notes here: https://etherpad.wikimedia.org/p/devsummit17-dev-and-product-q&a [17:48:54] Talking on the mailing list about long term plans would be a good start... [17:49:27] I could swear the street sweeper has gone past the venue more times in the last two days than is done in a year outside the office. [17:50:08] thanks [17:50:22] * aude sniffles with a cold [17:50:39] Does someone want to reword a question for me in a more intelligent manner than I am capable of this morning [17:50:47] * greg-g waves to aude [17:50:49] bawolff: that looks like a good comment for https://www.mediawiki.org/wiki/Talk:Wikimedia_Product . [17:50:55] Seddon: sure [17:51:01] and notes, for sure, my flight to SF would have been cancelled if i had planned to come (evil snow!) [17:51:21] Elitre: honestly its probably too snide to be constructive :) [17:52:07] * legoktm clap clap [17:52:11] bawolff: I know that Wes really welcomes feedback about communications, so he won't mind a gentle reminder if you think they are not using mailing lists enough :) [17:52:23] greg-g: Basically something along the lines of, how do we use and work with the network of affiliates other than WMDE to broaden and diversify the MW technical community [17:52:34] Seddon: sure [17:52:39] broaden, diverisfy and decentralise* [17:53:02] Seddon: that looks clear enough to me. it's being asked now though. [17:53:23] share the love [17:53:25] share the work :) [17:53:25] I mean it seems like we should start with communicating at all before worrying about diversifying the medium [17:53:31] :) [17:53:55] * Seddon shows <3 to greg-g [17:54:55] bawolff: then I would love to know from you about specific occasions where WMF failed at this :) (not necessarily here and now!) [17:55:42] Please remind people to introduce themselves when they talk (I think this is C Scott?) because we can't see them [17:55:49] this is cscott [17:56:07] thanks for the reminder, Guillom [17:56:14] * guillom hugs Elitre. [17:57:16] Seddon: good enough? [17:57:21] bawolff: Can you give some examples of "long term plans" that exist but aren't shared on "the" (wikitech?) mailing list? :-) [17:58:05] thanks James_F for rephrasing my question :) (also, for actually sharing your plans on a list! and wikimedia-l nonetheless.) [17:58:11] James_F: i would guess the problem is that there are no long term plans [17:58:33] or if there are, nobody has shared them with me :) [17:58:42] bawolff: Indeed. It's not about venue or communicating, it's about being less focussed on daily stuff and more on the long-term. [17:59:04] yeah DJ!! [17:59:24] I guess what im trying to say is it seems premature to discuss diversifying communicating longterm plans [17:59:32] * James_F nods. [17:59:46] But "focussing on a roadmap" means blocking code that diverges away from the roadmap, which means it's going to alienate people who don't agree with the direction. [17:59:59] more hackathons! [18:00:02] perhaps [18:00:03] roadmaps are hard [18:00:04] +1 [18:00:22] but people cant focus on things that dont exist [18:00:29] Indeed. [18:00:34] I think some teams are doing great at long term thinking. I think Parsing for example is sharing a lot of what they want to do and ask a lot of feedback, even about prioritizing what they're gonna do. Their page on mw.org speaks a bit about that (in addition to their sessions to events, Phab tasks etc.) [18:00:40] greg-g: yep :) danke [18:00:52] More hackathons and events such as BotAcademy(WMSE) [18:00:55] Elitre: Good point. [18:00:59] Elitre: indeed there is a lot of variation between teams [18:01:11] Seddon: any time [18:01:21] https://pbs.twimg.com/media/CBMW1tCU0AA_MOM.jpg [18:01:45] +1 to TheDJ [18:01:50] addshore: :) :) [18:02:43] thanks for the love, Cindy! [18:03:45] can i ask questions here? [18:03:49] aude: yes! [18:04:17] how does the architecture committee and RFC process fit in with the roadmap? what role do they have? [18:04:25] (if this hasn't been asked yet) [18:05:16] I don't see it among the most voted questions, no. [18:05:22] (i feel like the point of my question was missed, in that the final statement from victoria was about the foundation committing to support more community code) [18:05:24] aude: it was kiiinda asked yesterday during Daniel's "Platforms are Products too!" talk, but probably worth it again here [18:05:27] Anyone wants to ask it? [18:05:28] aude: want me to? [18:05:32] sure [18:05:33] As a volunteer, i would second what was said, but also point out that focusing on mediawiki is limiting the scope [18:05:44] aude: actually, which roadmap? [18:05:45] my point was that we need better support for community members/projects which *doesn't* involve hiring them or moving their code into production [18:05:51] greg-g: ok [18:06:04] mediawiki or wmf technical (if there is one) [18:06:07] kk [18:06:13] or quarterly goals [18:06:28] there's also the strategy process that katherine is leading [18:06:41] longer-term grants, perhaps, or dedicated support engineers. there should be some way we can support (for example) "enterprise mediawiki" or "semantic mediawiki" without moving their code into core. [18:06:44] there's also the Annual Plan... [18:06:50] yea, also annual plan [18:07:06] * aude also wonders if/how these things are coordinated [18:07:23] Looks like greg-g is the remoties' hero today. TY! [18:07:39] I work for all remoties :) [18:07:40] I dont know we should actually support smw [18:07:44] :) [18:07:56] beyond in how it interacts with core [18:07:59] e.g [18:08:04] aude, you will probably be asked to coordinate everything, since you asked about this :p [18:08:09] lol [18:08:25] If they want to upstream a change where it makes sense [18:08:36] Give things to do to chapters that don't have budget to hire someone to take care of communities, mmmmh. [18:08:51] who is asking the q right now? [18:08:51] Trizek: that doesn't work [18:08:57] that was chad [18:08:57] smw kind of does their own thing, though maybe that's a result of neglect [18:08:59] chad [18:09:01] but we shouldnt provide support for devs of every ext in the world [18:09:14] What?! Why not? [18:09:15] and not sure what they want, in terms of support [18:09:19] * legoktm hugs ostriches [18:09:21] matanya: that's exactly what I mean. [18:09:57] bravo ostriches [18:10:16] i was part of the FDC when wmde was working with wmf to get wikidata contract, if you make other orgs go through this, it will fail, for sure [18:10:28] bawolff: I think we should continue explicitly not supporting SMW. [18:11:33] (gabriel) [18:11:34] legoktm, Nemo_bis: You're welcome [18:12:44] What I keep hearing is that there are so many good stories that are worth telling to a wider world. (Which might also help with recruitment of new volunteers.) [18:13:00] matanya: It wasn't about "making" per se, it's about saying "We've got this project that the WMF doesn't have the resources for, can another org (chapter, thematic group, interested individuals applying for an IEG) take on the work?" [18:13:06] At least that's how I'd like to see it work :) [18:13:19] enabling :D [18:13:25] empowering? [18:13:32] ostriches: +1 [18:13:36] if we enable empowerment, we can leverage them [18:13:40] Enabling is a nice double-sided word. [18:13:40] bawolff: yeah, exactly. there should be a way we can *help* them without solving all their problems or upstreaming their code. [18:13:44] brion: something something synergy [18:13:47] :D [18:14:19] The independence also means we are UTTERLY SCREWED the moment something goes wrong. [18:14:35] Like we underestimate how long things will take, how much we need, etc etc... [18:14:38] Isarra: Ideally, things that get "outsourced" like that are self-contained, but useful [18:14:40] !hesaids(yner|trate)gy [18:14:41] (with regard to what hare is saying) [18:14:42] melodykramer +1 [18:15:05] bawolff: we don't really have a process for "you guys are doing good work and seem to have a large community, even though your product isn't aligned with our larger roadmap (which is moving in the wikidata direction)." [18:15:10] So yeah, we underestimated how long our development would take and now we're trying to finish everything even though we ran out of money months ago. Bloody demotivating. [18:15:11] 4 questions and 15 minutes left [18:15:29] ostriches: Well, yes. [18:15:37] Cscott: to what end though [18:15:46] Problem is, useful tools still often have some overall integration into everything else... [18:15:49] aude: smw doing their own thing is actually a strength, IMO. We need more communities who have the ability to accomplish their own goals, not depend on WMF to do things for them. [18:15:52] Like... they... agh. [18:16:03] i dont think anyone doubts smw is useful to some people and have large userbase [18:16:34] As long as nobody puts huge walls in the way of smw, they seem... fine. But that's just sort of considering general usage. Or whatever. [18:16:52] i guess i just dont know the smw devs or what they "want" from us [18:16:59] Isarra: Sure, integration, but not dependency on other teams. So like if Wikimedia Antarctica fails, it doesn't cause another group to fail as well [18:17:07] Oh, sure. [18:17:09] (btw, Wikimedia Antarctica should totally be a chapter :)) [18:17:11] Isarra: Is that short for Semantic MediaWiki? [18:17:21] Wikimedia Antarctica!! [18:17:23] bawolff: to continue to grow the non-WMF community. to have more users of our core codebase, more people familiar with the codebase, etc. some of those will contribute back to the core codebase, some may eventually move on to more WMF-aligned projects. [18:17:27] Smw=semantic mediawiki [18:17:29] rfarrand: sounds cold [18:17:32] bawolff: we need to have a "farm team", to use the baseball analogy. [18:17:43] We have more dependency than usual since our product is intended for wikimedia projects so it needs to go through security review at very least, but... eh. [18:17:48] * brion is baseball-illiterate [18:17:51] melodykramer: Yeah. [18:17:51] inb4 MediaWiki Farmers [18:17:54] lol [18:18:06] * anomie has a tendency to think SMW = Super Mario World [18:18:07] We should revive the farmers. [18:18:11] Snrk. [18:18:14] brion: "minor leagues". wait, that's also baseball. "NBA b-league"? What's the cricket version? [18:18:16] a farm team is a minor league team where your recruits practice and get experience and then are called up to the majors [18:18:21] cscott: that'll do :D [18:18:25] thx [18:18:58] idea is just that we need a community of folks doing stuff in the general vicinity of our codebase, without trying to force them all to align with the WMF roadmap. [18:18:59] That's how Code for America views it's fellows. They have a skillset, they're placed in a specific city for a year, and then they often take other roles within the digital gov ecosystem. [18:19:03] cscott: as it stands people switch from third party to wmf stuff [18:19:13] although it is rarer than it could be [18:19:32] bawolff: not everything should go in that direction [18:20:00] bawolff: right. and i'm arguing that one reason it is rare is that we don't have this large amorphous cloud of people doing interesting mediawiki things. [18:20:49] Matanya: i agree. People should do whatever interests them [18:21:08] cscott: Where can I see a list of projects that are taking place in the general vicinity of the codebase? I'd love to see what people are up to - not sure how. [18:21:23] cscott: so you want like a third party ext developer group that blogs a lot about what they do? [18:21:29] cscott: As opposed to having folks doing other things get randomly ploughed over? >.> [18:21:32] also, the WMF roadmap could be wrong? or poorly ordered. allowing folks to follow their own directions could end up in results which are ultimately more useful than rigid central planning. [18:21:36] Or would that still happen? [18:21:46] bawolff: yeah. [18:21:49] Isarra: yeah [18:21:59] one contact per developer :) [18:22:04] melodykramer: wikiapiry is probably a good place to start exploring other projects [18:22:27] smw easily meets my bar because they have grown to the point where they organize their own conferences & etc. they are obviously self-supporting and sustainable, even though not WMF-aligned. [18:22:42] brion: OCPD? [18:22:47] cscott: we could certainly use things like that and especially people writing tutorials [18:23:07] Have you seen this, bawolff https://whatcanidoformozilla.org/#!/progornoprog/proglang [18:23:11] i think some of the WMF support *might* include messaging about roadmap and trying to help 3rd parties find ways to align their codebases with WMF preferred direction [18:23:16] This is Moriel. [18:23:19] Our how to write an ext tutorial is pretty pathetic [18:23:32] but that should be helpful guidance, not coercion [18:23:39] + a million to moriel [18:23:51] Who is this? [18:23:52] docs docs docs [18:23:55] Moriel? [18:23:56] Isarra: moriel [18:23:59] aaah, Reedy is here! [18:24:12] now somebody's shouting out not sure who :D [18:24:15] The sad thing is our documentation is still soooo much better than anything else I've seen. [18:24:15] i don't actually know what the best way to support 3rd parties w/o absorbing them into WMF. but i want to encourage discussion in that direction. [18:24:19] EOM [18:24:23] Get people extra time to document, too! [18:24:24] yurik next [18:24:24] let's talk about documentation! ;) [18:24:27] They just need more time. [18:24:33] Sumana :( [18:24:33] bug 1! [18:24:38] They can document stuff as they go if they just aren't so ruched. [18:24:46] :'( [18:24:50] rushed [18:25:11] melodykramer: that site looks cool :P [18:25:24] brion: the shouting was Reedy ;) [18:25:28] why is mozilla using a hashbang url? thay should know better... [18:25:29] ah good :D [18:25:32] yuri's qn is one of the key questions i think [18:25:35] * cscott is an old man grumbling. [18:25:50] brion: addshore I don't disagree that we need to fix it. I just know it's not worked before, repeatedly [18:25:55] aude: was that good enough for you? [18:26:01] In the VE team we actually had a very good experience with a tech writer we hired, but only within certain parameters (document this library, its author will sit down with you for 10-20 hrs/wk to tell you about it and answer your questions) [18:26:02] the question, at least :) [18:26:12] I like victorias comment about inside vs outside [18:26:32] yeah, the oojs ui docs on mw.org were done by a tech writer hired specifically for it [18:26:36] That's how the oojs-ui class-level docs were written, with code samples that run live on the documentation dsite [18:26:37] Best question. [18:26:47] what did Reedy shout? [18:26:50] i think we need a tech writer specificly for introduvctory tutorials [18:26:52] legoktm: It doesn't work :P [18:26:54] YES. [18:26:56] I don't think Victoria really understood it [18:26:57] legoktm: "We tried that, it doesn't work". [18:27:02] Oh right and the mw.org pages too though I haven't read those [18:27:02] RoanKattouw: Yeah, that [18:27:06] to build new users [18:27:06] in response to "tech writer"? [18:27:09] Yup. [18:27:14] Autogenerated stuff from the code makes more sense [18:27:15] Get someone who is clueless to do tech writing. They can write as they learn. [18:27:18] Reedy: +1 [18:27:24] bawolff: "specificly for introduvctory tutorials" sounds good [18:27:30] Which, maybe someone releng/CI is the "fix" [18:27:36] If i knew how to spell [18:27:47] https://github.com/cfpb/readme_refresh_day the cfpb has a documentation refresh day. [18:27:49] i've heard mixed opinons about the mw.rg oojs ui docs, but most people thought they were very helpful [18:27:52] gerrit/ci doc is actually good [18:27:59] Our inline code comments are great [18:28:10] what did somebody just say off mic? [18:28:12] Yuri clarified that his question was about culture clash [18:28:13] addshore: I wrote a piece on how to write great tutorials for lay audiences: http://melodykramer.github.io/2015/03/04/how-to-write-a-great-tutorial [18:28:16] i think we've mostly managed to keep them up to date... except for th screenshots :P [18:28:20] ah, thanks halfak|Mobile [18:28:20] rfarrand: ^^^ [18:28:32] No problem. Knew someone would ask 😁 [18:28:40] docs > outdated docs > no docs [18:28:42] * addshore has generally heard people compain about all docs on mw.org and ask why we don't do docs better / have them next to the code (with example there etc) and have multiversion docs in an easy to use form [18:28:51] melodykramer: ++ good advice there [18:28:56] I'm not 100% sure I know what the lessons are from that, but I think they are probably at least 1) narrow scope is important and 2) you need to make a knowledgeable person's time available to them, and a lot at that [18:29:03] MatmaRex_mobile: i found it confusing to figure out all the options supported by a specific widget due to all the levels of subclassing [18:29:06] This is something I really enjoy writing. Happy to lead workshops or help out in this way. [18:29:21] addshore: if we had examples in the doc comments we could have multiversion docs built from source tree reliably [18:29:33] though embedding code examples in comments always feels weird [18:29:38] brion: maybe we should do that ;) [18:29:42] :D [18:29:48] Is there documentation outside of Wikimedia that you think is really spectacular? It would be interesting to list why they're good + what works. [18:29:51] brion: yeh, in comments can get very ugly, but in files in the repo at least ;) [18:29:59] Im happy victoria seems very pro outside devs [18:30:25] melodykramer: i know everybody hates it but honestly i love the php docs. :) [18:30:25] maybe we can follow this session up with written answers to the remaining questions or a tech talk later on [18:30:30] I will see what we can do [18:30:38] /docs/examples//example.md? ;) [18:30:38] it's well searchable, and the comments are sometimes quite helpful [18:30:55] meanwhile i hate hate hate apple's documentation for ios/macos ;) [18:31:00] 18:27 < Reedy> Which, maybe someone releng/CI is the "fix" [18:31:02] mainly because it's poorly linked [18:31:06] context? :) [18:31:08] and often very sparse [18:31:11] Thanks rfarrand. I think that's a good idea [18:31:16] This is a blog post from one of my former coworkers: https://jacobian.org/writing/great-documentation/ [18:31:38] I dislike our docs on mw.org and generally find out of date docs every time I look somethingup [18:32:01] greg-g: If the answer to docs is better autodocs, visibility of them etc... Rather than duping to mw.org It's gonna be relying on CI to get it published etc. Not saying your current team as is should be... It could be hire worthy [18:32:05] yeah keeping wiki docs in sync with source is hell [18:32:11] that's why i love doc comment generation [18:32:18] brion+1! [18:32:19] Reedy: ahhhhh, thanks :) [18:32:31] we should generate the mw docs on the $wgXXXX config variables from source somehow [18:32:38] Reedy: I just couldn't parse the backscroll effectively enough :) [18:32:38] of course anything that documents multi-class clusters can be harder that way [18:32:39] brion: cscott Reedy also +1 [18:32:48] the docs are great, but they are not in sync and there's nothing ensuring that new config vars get properly documented. [18:32:50] brion: I like this: http://opencomparison.readthedocs.io/en/latest/contributing.html and this: http://docs.fabfile.org/en/1.13/ [18:33:02] melodykramer: Let me find you a link but I quite like the beginner docs for Rust. I never ended up finding an excuse to program in it but I learned a lot about the language by reading the docs and liked the format. Specifically because it has examples like "if you write this reasonable-seeming code, it will be rejected for this somewhat-unintuitive reason [18:33:02] and here is why" [18:33:02] cscott: that should actually not be difficult, doc comments on the vars already exist. should be able to expand them and extract from there [18:33:03] Brion: we should make a third emmbed git into wikipage extension! [18:33:04] Can we do it from https://doc.wikimedia.org/mediawiki-core/master/php/ ? [18:33:13] :) [18:33:13] (The Rust compiler also gives you suggestions for fixes along with its error messages) [18:33:20] brion: yeah, but the existing mw docs are much larger than our current doc comments [18:33:20] That's really awesome RoanKattouw [18:33:23] greg-g: thanks :) [18:33:30] there might be a readability issue if DefaultSettings.php suddenly grows 3x [18:33:38] it's great to see this channel alive. [18:33:42] rust has a great community too, i hear they're good helping beginners [18:33:44] The problem --or one problem--isn't a lack of documentation. It's often too much, and conflicting, documentation of the same issue. People apparently don't like to kill others' pages, even when they are out of date or duplicative. So the thicket grows... [18:33:44] we might need to split up DefaultSettings.php into multiple files or somesuch to keep it manageable. [18:33:52] rfc is a way for people to participate [18:33:59] Also if you're super into documentation, I recommend this community: http://www.writethedocs.org/slack/ [18:34:02] jmatazzoni______: too much and conflicting...+1 to that [18:34:07] jmatazzoni______: yes, we'll need to prune the old pages :) [18:34:17] Being able to see what is relevant + what is current is key. [18:34:25] nothing worse than out of date docs [18:34:27] DELETE, we need to delete more [18:34:32] code and docs :) [18:34:40] these are more technical proposals (not product) [18:34:46] * addshore is a fan of pretty versioned docs like http://docs.guzzlephp.org/en/latest/ :P [18:34:50] often* [18:35:16] greg-g: similarly w/ 3rd party communities. ;) we need to DELETE default WMF support from them, but support them to maintain the code themselves (whatever that means, but aggressively, so they don't feel abandoned at all) [18:35:17] melodykramer: https://doc.rust-lang.org/book/ownership.html is the bit I liked [18:35:22] (and subsequent chapters) [18:35:35] there is image drag and drop in VE! [18:36:32] Id really like the extension infobox to be loaded from extension.json [18:36:48] melodykramer: Aligned with recent measurements in Design documentation to clearly mark outdated docs and move them :) [18:36:52] melodykramer: we should just fork the mozilla whatcanidoformozilla and create a draft verison of it for WM / MW and see what it looks / feels like [18:36:54] Almost all of our code works [18:37:03] addshore: That would be really cool. [18:37:14] bawolff: I'd really like to kill most of the content of MW.org and load it "live" from git. [18:37:18] some of it is shity and should be removed, but it works [18:37:19] * brion hmm, i may want to realign my priorities for 2017 to support roadmap & docs explicitly [18:37:23] James_F: +1 to that [18:37:30] we do have drag and drop... All depends on what you drag and where you drop... [18:37:32] yep [18:37:48] bawolff: That way we can version the documentation properly for the version you have, rather than in one document try to explain config for MW 1.12 all the way up to master. [18:37:50] James_F: hmm, didn't we have an OPW intern who wrote an extension to allow that? [18:37:54] contextual help in VE! /me hides [18:37:57] load stuff into MediaWiki from a Git repo [18:38:06] I'm also happy to be a guinea pig for existing documentation and work with someone to figure out what I can and can't understand (as a relatively new person here who can code.) [18:38:21] brion: https://phabricator.wikimedia.org/T155028 [18:38:26] melodykramer: I thought your entire team was up for that? :p [18:38:52] I think so! It's a particular passion of mine, but I'm sure there's a list of people who could help in this way! [18:38:56] Melodykramer: Lots of it is flat out missing. Try to write a mediaHandler for example [18:39:07] thanks everyone for chatting here. TTYL [18:39:15] yeah I don't even know what that means bawolff :P [18:39:34] \o/ [18:39:43] melodykramer: thumbnailing and metadata extraction for a new file format [18:40:57] bawolff: https://phabricator.wikimedia.org/T155029 [18:41:10] Two sessions coming up at 11AM PT [18:41:10] What should an AI do you for you? Building an AI Wishlist [18:41:10] STREAM: https://www.youtube.com/watch?v=j8ND7Uu4e_s [18:41:10] DISCUSS: https://webchat.freenode.net/?channels=#wikimedia-ai (NOTE: different IRC room) [18:41:10] NOTES: https://etherpad.wikimedia.org/p/devsummit17-ai-wishlist [18:41:10] Integrating MediaWiki (and other services) with dynamic configuration [18:41:10] STREAM: https://www.youtube.com/watch?v=XLIwFajicsY [18:41:11] DISCUSS: here (#wmhack) [18:41:11] NOTES: https://etherpad.wikimedia.org/p/devsummit17-integrating-mediawiki [18:44:01] * dbb waves from rainy Berkeley [18:53:21] bawolff: well, something like the extension matrix bot is imperfect but not too hard to make [18:54:57] .. added three followup comments here, just now https://etherpad.wikimedia.org/p/devsummit17-Labsdbs [18:56:06] Hey folks, the AI wishlist session backchannel is in #wikimedia-ai. You can join the stream here: https://www.youtube.com/watch?v=j8ND7Uu4e_s [18:56:13] The session will start in 4 minutes. [18:57:21] addshore: One more example: http://civicissues.codeforamerica.org/geeks/civicissues collects issues on GitHub labeled "help wanted" from across the civic tech movement. [18:58:00] melodykramer: ooooh, pretty [18:58:33] |I kind of feel like thats another one of our issues, we are a few steps behind in making our pages / tools / docs that are meant to attract people pretty enough for them to be attracted :P [18:58:48] chicken. egg. [19:00:00] https://usercontent.irccloud-cdn.com/file/jwgOONo0/ [19:00:00] There is one new unconference session that will also be livestreamed now. It's called MediaWiki History for Analytics. The stream is here: https://www.youtube.com/watch?v=qUAUSc2Nsh8 - I'm trying to find out more information about that session. [19:00:13] fast, cheap, pretty .. pick one [19:00:35] If anyone knows anything about that session, please let me know and I can update. [19:01:57] melodykramer: Where do I find the notetaking thingies for unconf sessions? Specifically I'm in the Multi-Content Revisions one [19:01:59] dbb: the joy is we can choose pretty for some things but not for others ;) [19:02:34] RoanKattouw: Is that an unconference session? I don't see it on the program. https://www.mediawiki.org/wiki/Wikimedia_Developer_Summit/2017/Program#Tuesday.2C_January_10th [19:02:45] halfak: can hear [19:02:53] Ah, found it [19:02:55] thanks for streaming [19:02:56] Looks like they've not been put up yet. [19:02:58] Looks like they just updated the schedule [19:03:10] * James_F refreshes. [19:03:22] Although, no notes link yet? [19:03:29] RoanKattouw: I just made this for you: https://etherpad.wikimedia.org/p/devsummit17-multi-content-revisions [19:03:35] Will update the schedule with links for the others! [19:03:43] perfect :D thx melodykramer [19:03:45] Yay thanks! [19:03:52] melodykramer: Thanks! [19:04:48] Live notes for Integrating MediaWiki (and other services) with dynamic configuration are here: https://etherpad.wikimedia.org/p/devsummit17-integrating-mediawiki [19:06:46] If you're attending an unconference session, there are now notes available in the program for each session: https://www.mediawiki.org/wiki/Wikimedia_Developer_Summit/2017/Program#Tuesday.2C_January_10th [19:18:43] is there a stream link for joe's session? [19:20:07] mutante: https://www.youtube.com/watch?v=XLIwFajicsY [19:20:49] Wikimania 2018 in capetown south africa [19:21:01] just announced [19:21:02] addshore: thank you [19:21:02] crazy... [19:21:10] (and awesome) [19:21:19] rfarrand: ooooooh [19:22:10] comments and updates welcoome on this -> http://wiki.osgeo.org/wiki/GitHostingSoftware#FOSS_dot-org_Git_Implementations [19:22:13] South Africa is happening? [19:22:14] Neeeaaat. [19:25:30] What exactly is 'dynamic configuration'? [19:25:49] * dbb guesses ..config based on changing inputs [19:26:01] like load balancing [19:26:32] Ah. [19:28:33] I guess I'll apply to go to wikimania next year :) [19:28:44] Isarra: https://phabricator.wikimedia.org/T149617 [19:29:13] So pleased we're going to have Wikimania for only the second ever time below the Equator. [19:29:43] mobrovac: Thanks. [19:29:51] I just got here, so I'm still catching up. >.< [19:33:59] Yay, we can continue our tour of apartheid countries! [19:34:33] Nemo_bis: Italy isn't that bad. [19:37:35] lol [19:43:14] Hi all - there have been several unconference sessions added to the conference program in the last hour for the next two sessions. Notes and streaming (when available) have also been added: https://www.mediawiki.org/wiki/Wikimedia_Developer_Summit/2017/Program#Tuesday.2C_January_10th [19:45:29] thanks melodykramer ! :) [19:45:38] Yes, thank you muchly. [19:46:44] Team effort: rfarrand, brendan_campbell, Srishti, Qgil... [19:47:15] Someone should get y'all a drink. [19:48:42] Isarra: don't worry ;) [19:49:12] Excellent. [19:50:12] Excited about Cape Town, precisely what I was hoping for [19:50:22] Having been there six months ago :) [19:59:01] melodykramer, bawolff, addshore, (re: earlier pointer to Mozilla's page) we have one, inspired by that. https://www.whatcanidoforwikimedia.org/ -- no changes for 11 months though. That's one problem with offwiki/non-embedded docs, easy to miss or forget about. [19:59:30] Didn't know that - thanks quiddity! [19:59:40] And harder to update, no? [20:00:04] well, it's a github repo, so as easy/hard as that. Easier than gerrit, for non-devs! [20:00:43] if there were a drinking game that required everyone to take a shot whenever someone dropped a neg on Java, this conference would be over quick [20:00:46] err, phab-code-repo, rather (now that gerrit has a web UI) [20:01:25] urandom: You mean, over in sys.lang.com.oracle.java.concept.temporal.Time.moment.Quick? ;-) [20:01:36] everyone drink! [20:06:56] Like people here need an excuse to drink. [20:09:09] Isarra: If I don't drink, I die of dehydration [20:09:31] Well, that too. [20:19:31] addshore: Note that our docs are versioned actually (but not very discoverable right now). This was one of my main concerns when we first started to make doc.wikimedia.org. We publish it by branch right now, and I'd like to do it for release tags as well. Especially the difference between master and current stable version is important to make sure the docs [20:19:31] are useful for developers not working on MediaWiki itself. [20:20:09] It was broken by https://phabricator.wikimedia.org/T150727 to naturally discover the peer directories though [20:28:35] Oh, come on, code review isn't livestreamed? >.> [20:29:17] Then again we all know code review is doomed, anyhow. [20:29:22] Dooomed doomed doomed. [20:29:45] Lol [20:31:12] victoria seemed to make possible noises about cr in first session [20:32:09] Anything promising? [20:32:14] Are you all supposed to be at lunch now? [20:32:24] Says it's lunchtime. [20:32:54] Just want to make sure I'm not missing anything with time zones! (I'm in a different one.) [20:33:17] Heh, same. Still close enough to lunchtime for me, though. [20:34:32] ...actually I'm sort of regretting eating this early. And it's later here. >.> [20:35:46] code review is part tech, part social skills imho [20:36:27] S/possible/positive [20:36:27] some groups do ultra-tiny code review almost constantly.. [20:37:16] What's ultra tiny? [20:38:00] ultra-tiny is a bit of an exaggeration, but basically the contents of one function, or single file refactoring .. like that level [20:38:25] Ah. [20:38:30] as opposed to real architectural change, or change of tools/libs.. which are big things [20:38:59] many small changes are always almost better than one large one, nobody likes merging large changes [20:39:06] That's a lot easier to review, but if you can't be guaranteed anyone will bother even with that, it's a lot harder to keep going in small patches... >.> [20:39:17] people will bother with it, if it's easy enough to review [20:39:20] Problem is, nobody seems to like merging anything a lot of the time. [20:39:35] You need to specifically track someone down and prod them with a stick, and that's if you start out knowing who they are... [20:39:43] .. some dev at Twitter talked about how they have their entire running system as once single code base, which changes live [20:39:50] side note [20:40:10] you need to fire&forget, upload and move to the next thing. then have positive surprises on some days, heh [20:40:21] avois the "tracking down on IRC part" [20:40:43] No, then other people's changes get merged in and yours becomes incompatible in the meantime. [20:40:43] big part is that people dont filter notifications though [20:41:00] if it's small enough it's rebase-able by just hitting the button in gerrit most of the times [20:41:07] And if you're doing a bunch of little things, they all depend on the previous one going in, of which there's no guarantee. [20:41:28] Not if theirs isn't little. [20:41:28] upload them without dependency unless you are forced to for code-reasons [20:41:42] so git reset between uploads [20:41:45] Someone moving the entire file out from under you makes rebasing a lot harder. [20:41:53] Which I've had happen. [20:41:57] It's fun. [20:41:59] yea, just trying to minimize how bad it is [20:42:00] Really fun. [20:42:01] well this is where its very wise to distinuish between CSS changes versus database drivers or network things [20:42:03] i know [20:42:42] make people check gerrit web ui by default more often, that means less IRC pings for all [20:42:54] and removing the need for realtime = good [20:44:17] Mhm. [20:49:50] Isarra: I hate file moves [20:50:08] They're a pain. [20:50:32] But I do that a lot too. Like when an extension starts out all in one file, but then gets bigger and bigger and there's like five classes all wedged in there... [20:50:42] Or particularly commonly, I want to use less. The file is css. Grr. [20:58:17] There are six upcoming sessions in 10 minutes. Two of them will be livestreamed. [20:58:17] Better code review processes for reviewers [20:58:17] NOTES: https://etherpad.wikimedia.org/p/devsummit17-code-review-processes [20:58:17] Algorithmic dangers and transparency -- Best practices [20:58:18] STREAM: https://www.youtube.com/watch?v=myB278_QthA [20:58:18] NOTES: https://etherpad.wikimedia.org/p/devsummit17-algorithmic-dangers [20:58:18] Redesigning Special:Search. By: Birgit Müller, Dan Garry [20:58:18] NOTES: https://etherpad.wikimedia.org/p/devsummit17-special-search [20:58:18] An API for monitoring dumps. [20:58:19] NOTES: https://etherpad.wikimedia.org/p/devsummit17-dump-monitoring [20:58:19] Unconference: Create UX/UI hub. By: Moushira [20:58:20] STREAM: https://www.youtube.com/watch?v=I62PdEtB6KE [21:10:33] Don't cross the streams. [21:13:52] link for Mou note taking? [21:14:37] NOTES: https://etherpad.wikimedia.org/p/devsummit17-ux-ui-hub [21:15:15] The stream hasn't started yet for that one. [21:15:21] Has the thing itself started, then? [21:15:23] okay, streams are started [21:15:38] thanks. [21:15:51] Thanks. [21:16:12] grammar fail [21:17:43] Are the street sweepers STILL going? [21:18:52] Yep [21:21:20] In Mou's session, is there a way the speakers could get closer to the mike? Not sure if possible, but hard to hear some of the participants. [21:21:30] Yeah, I can hardly hear anyone. >.< [21:21:44] Even Moushira was difficult. [21:21:49] will move the mic [21:23:57] The design process seems to vary a lot by who's doing it. Some teams just make something up and implement it on the fly, among other things. [21:24:13] moved the mic, moushira should be easier to hear now [21:24:23] Aye. [21:24:54] What's ideation? [21:25:10] Well if one doesnt have design resources, its hard to do anything other than make it up on the fly [21:26:06] https://en.wikipedia.org/wiki/Ideation_%28idea_generation%29 [21:26:07] Hey, it works. I'm just saying there doesn't seem to be one process anyway. [21:26:13] :) [21:26:13] Link for Mou's session: https://openideo.com/ [21:26:15] Well, sort of works. [21:26:32] Works better if there's a conversation that can happen about it, as with all of them. [21:26:55] Like if folks can just say whatever is needed, whatever isn't working, and, though it never happens in reality, whatever is good. >.> [21:27:14] Being able to have actual conversations is great, though. How do you do that? How do you talk to your users and whatnot? [21:27:27] ...I'm seriously asking. I have no idea. [21:28:02] greg-g: Now I'm more confused. >.> [21:28:45] Ideas are kind of useless, on their own. You need data, direction, problems to be solved and solutions to explore. But those aren't ideas... [21:29:21] Agh, I can't hear this person at all. [21:31:21] Is there anyone in the Mou session who can relay stuff? [21:31:55] Could someone say in Mou's session that this exists: http://www.knightfoundation.org/articles/knight-news-challenge-libraries-awards-16-million-support-innovative-ideas [21:32:05] And that process was based on the concepts mentioned, but for libraries. [21:32:09] (I was a judge for that competition.) [21:33:17] Isarra: Another example for that session: https://labs.usa.gov/#research-report - This was something that 18F did to analyze barriers that existed that prevented people from accessing gov't resources. All of the research was conducted in public libraries across the US. [21:33:48] Methodology: https://labs.usa.gov/files/FFD_Research_Methodology_v11.pdf Elitre ^^ two links/ideas for discussion. [21:34:23] are these about open ideation? [21:34:32] This one is: https://www.newschallenge.org/ [21:34:39] That was the library challenge from Knight [21:34:54] Neat. [21:34:54] on the OpenIDEO platform [21:34:58] looks like James beat me to it. [21:35:15] I pinged him in another channel because he was the only person I was sure who was actually there. [21:35:35] they used this to come up with new ideas for libraries, and it was extremely successful. [21:35:41] For the future, we need to somehow get dedicated relay people for these. As well as notetakers and all that. >.> [21:36:07] Thank you James. [21:36:07] thanks for links and suggestions. [21:38:05] Justification is really important. [21:38:08] For any change. >.> [21:38:44] That way there isn't a vs. in the first place. Instead it becomes something that can actually be discussed... [21:39:10] I REALLY WISH I COULD HEAR HAREJ. [21:39:36] Linting! [21:41:04] Communities don't like t ofight, they like to argue. [21:41:17] So argue back with actual arguments. They LIKE that. [21:42:18] ...treat people like gods damn people and they'll usually work with you. [21:42:42] * Isarra continues muttering impotently. [21:44:27] ...get the courage. Surprise people. They'll never see it coming. [21:46:14] I'm on a listserv that has many people who participate in open source projects where they discuss best practices and share ideas. It might be useful for people interested in this stuff: https://lists.freedesktop.org/mailman/listinfo/foundations [21:47:41] Actually, crap. If people keep insisting something is scary, it'll only make it scarier. But going up to the communities - any of them - needn't be scary at all. Just go somewhere, ask a question. Some people will respond. Some won't. Use whatever useful information you can get out of it. [21:48:00] And if anyone DOES attack you, that's on them. They're being dicks. It happens. [21:48:26] Don't take it personally, just disregard them. [21:49:10] And usually, people aren't dicks unless they're actually rather upset already. Usually they're bloody helpful. If you're going to them about a problem they've been dealing with all along, they may even get rather excited about it. [21:49:25] “Use whatever useful information you can get out of it.” -yes, yes, yes [21:49:38] * Volker_E can't agree more with Isarra [21:49:48] People got pretty excited about wikiproject X. WikiProject users knew what they were working with wasn't ideal. They knew they had problems. And we wanted to solve them. [21:50:19] Volker_E: Even when the information is just 'blah would be a better place to ask that'. [21:51:02] >.> [21:53:14] The top of this list is boring, look at that long tail we should attack: https://phabricator.wikimedia.org/P4732 [21:55:12] Oh hey, RSS... [21:57:04] Yeah I was looking at that one, heh [21:58:25] I abandoned a test change to project:sandbox :p [21:58:30] Yay, backlog-- [22:01:10] Resorted to group by dest_project_name and to put the long tail at the top [22:03:19] Argh, I wish I could hear. [22:03:34] We should cross-ref this list to archived/closed repos....probably a bunch could be a abandoned outright.... [22:03:38] Or had some way to even communicate back besides through harej... who I also can't hear. [22:03:52] Probably. >.> [22:04:13] Or need new maintainers... [22:05:11] The discussion for the UX/UI hub was smashed? [22:06:25] Glaargh. [22:08:00] > how to incorporate our larger audience [22:08:00] I suggest just talking to people. Like this. And on-wikis and mailing lists and whatever. Go pester folks. Wherever. [22:08:32] Not that I can actually suggest anything since there's no real connection between this channel and the session... [22:09:55] STOP FOCUSING SO MUCH ON IDEAS. Ideas aren't process. Get people involved in actual process, in problems, in what's needed, and how to determine what's needed... [22:10:31] ...I should probably just stop listening and back away. This is making me way too frustrated. [22:12:52] sorry about that, Isarra. [22:13:27] Eh. [22:13:32] I care too much. >.> [22:13:35] i've been copying some of your comments into the doc. [I'm sick, and multitasking] [22:13:44] <3 [22:13:52] I'm sorry for ranting so much. [22:13:57] thanks quiddity and everyone who took notes for the session. [22:14:06] Words are hard! [22:14:17] Isarra: it's ok. your comments belong to the doc :) [22:15:19] Yes, thank you calm sensible people. [22:23:36] Hello! There are five upcoming sessions at 2:30 PT. All session information can be found on the program. Two sessions are streaming for remote participants. [22:23:38] https://www.mediawiki.org/wiki/Wikimedia_Developer_Summit/2017/Program#Tuesday.2C_January_10th [22:24:19] And if you participated remotely in the DevSummit, we would really appreciate it if you could fill out this survey: https://docs.google.com/a/wikimedia.org/forms/d/e/1FAIpQLSeLavJQXW-Zo5AxNZHblbQI6hd0T4cFxr8KQYyZfD1Ly4DRMg/viewform [22:26:02] ...but I don't want to log into google. [22:27:41] oh. hmm. Let me check. [22:34:02] schema starting https://etherpad.wikimedia.org/p/devsummit17-scaling-database-schema [22:34:14] hi everyone [22:34:25] I'll be monitoring IRC for this scaling database schema talk [22:34:34] please ping me if you have a question you'd like me to repeat in the room [22:35:14] (: [22:36:50] Who ś the dancing guy in the norms one? [22:36:50] https://etherpad.wikimedia.org/p/devsummit17-developing-community-norms [22:36:59] I know I should know, but I forgot. [22:37:24] The Developing community norms for vital bots and tools is starting now at Hawthorne [22:37:33] THREE PEOPLE ARE WATCHING IT ONLINE, YOU KNOW. [22:37:33] Etherpad: https://etherpad.wikimedia.org/p/devsummit17-developing-community-norms [22:37:40] THERE ARE PEOPLE. [22:37:42] oh yay [22:37:43] HI. [22:37:49] Hi! [22:37:50] XD [22:37:53] lol hi :) [22:38:25] I'll monitor questions on irc for the community norms session and ask them [22:38:31] Thanks. [22:41:54] 'Expected' behaviour isn't necessarily what people hope for. Sometimes it's what people dread. >.> [22:42:08] But nitpicking words is also silly. [22:51:36] Expected behaviour is for all humans to die eventually [22:52:51] nihilism works on some levels.. anyway.. [22:53:02] .. just seeing this page https://commons.wikimedia.org/w/index.php?title=File:Labsdbs-_get_more_data,_faster.pdf&page=2 [22:53:11] And have their descendents depend on them for 70+ years after. [22:53:19] Apparently. [22:54:00] "I intend to live forever. So far, so good." -Steven Wright [22:54:12] I am not on your systems all the time.. this is a sophisticated context for the document.. in particular, the Category at the bottom reminds me of the very good use of "faceted search ready" tags the Debian system is using [22:55:05] * Isarra kicks youtube. [22:56:00] And if anyone suggests maybe I shouldn't be trying to use a starbucks for this and should get better internet, you should know I'm in a starbucks precisely because it IS better internet. >.> [22:56:13] ugh [22:57:04] sorry , knee-jerk reaction to a monitored "consumer" service [22:57:38] not intending to discuss now.. back to our regularly scheduled content [23:02:41] * Isarra kicks it. [23:06:43] You don't even necessarily need multiple maintainers or documentation if you at least have the source code... but it sure does help. [23:06:43] Why did they not require that in the first place? [23:06:56] newb question - how to get a dump of all physical locations in California ? [23:07:15] Isarra: We 'required' it but didn't have a system of enforcement [23:07:32] So folks just did whatever? >.< [23:07:33] context .. I maintain maps databases in Berkeley including Census, OSM, and detailed urban planning data [23:08:52] SPARQL only? .. my own search of "everything" locally after ingesting the corpus ? [23:09:40] that is better worded.. main wiki articles on physical locations in California [23:09:57] geohack stuff, whatever that might mean.. etc [23:10:46] a hectic and seemingly overworked response earlier was "wikidata ID" and not much else [23:14:55] context for the tool labs policies - https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Right_to_fork_policy and https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Abandoned_tool_policy [23:16:34] This is cool. [23:16:44] tool labs standards committee - https://wikitech.wikimedia.org/wiki/Help_talk:Tool_Labs/Tool_Labs_standards_committee [23:37:04] I don't know who's talking (labs one), but yes. Very important stuff. [23:37:10] Chase [23:37:14] and Bryan [23:37:19] I mean right now. [23:37:21] At this moment. [23:37:23] currently talking is Chase Pettet [23:37:29] Cool. No idea who that is. [23:37:46] Aside from apparently smart person. [23:37:49] he is the manager of the Labs ops team at the wmf [23:38:00] Ah. [23:38:07] Good. [23:42:29] And Byan is also awesome. [23:42:33] Bryuan [23:42:35] Byran [23:42:36] BRYAN. [23:42:39] :D [23:42:40] Bloody hell. [23:43:02] Isarra: Behave. [23:57:33] Hello! There are four sessions taking place at 4. One will be livestreamed. Information can be found on the Program landing page: https://www.mediawiki.org/wiki/Wikimedia_Developer_Summit/2017/Program#Tuesday.2C_January_10th