[03:32:10] halfak: Librarybase is on. I will want to consult with you at some point. [13:20:39] o/ [16:17:24] \o_ [21:08:31] hey halfak! I'm back from running pre-storm errands. Let me know when you want to mull over Dev Summit deets [21:08:43] +guillom [21:08:47] o/ J-Mo [21:09:08] What did ya'll have in mind? [21:09:39] I don't really know. Aside from the whole "incorporating research into product dev" trope, I'm not sure. [21:09:43] PAWS? [21:10:04] I'm happy interpreting the AI topic broadly. [21:10:09] There's also the APIs and Services topic [21:10:13] PAWS is a service [21:10:15] IMO [21:10:21] agree [21:10:32] Chase reached out re. Quarry and PAWS. [21:10:36] Let me find the thread [21:10:37] maybe "community research infrastructure" [21:10:50] that includes PAWS, Quarry, pageview API… [21:11:01] https://phabricator.wikimedia.org/T147593#2710467 [21:11:25] J-Mo, +1 I like that framing [21:11:29] brb. tech support for mother in law [21:11:33] lol [21:11:33] k [21:12:22] * guillom waves. [21:12:41] Hey guillom! Quick, let's talk about stuff while J-Mo's not looking :D [21:12:48] héhé [21:13:37] So, what I had in mind was revolving around two themes that I've been working on recently: documenting our knowledge, and researching systems. [21:13:44] the answer was "drag it to your desktop" in case anyone was curious [21:13:57] keep talking! [21:14:01] Both would help Product development, but not immediately [21:14:13] Which is why I wasn't sure they were appropriate for the summit. [21:14:54] guillom, I think the documentation thread is interesting. [21:15:00] what do you mean by "researching systems", guillom? [21:15:06] ^ good Q [21:15:11] (typing) [21:15:58] Documenting our knowledge is related to my efforts on the Research Codex; I thought having other researchers around, and product people, would be an opportunity to prioritize needs, inventorize what we know, and inventorize what we don't know [21:16:36] guillom, you know, it might be very valuable to have a sort of group hack session for 30 mins [21:17:01] Get everyone in a room. Point at codex. Have people fill in missing bits. [21:17:57] Probably not directly into the codex, but maybe in some preliminary space like an etherpad that you could merge later. [21:17:57] Researching systems is about what I'm working on these days with graffiti on my white boards :) I've created a skeleton at https://meta.wikimedia.org/wiki/User:Guillaume_(WMF)/Dynamics_of_Wikimedia_systems but in a nutshell, it's about applying the System dynamics methodology to trying to understand our Wikimedia systems. I'm playing with stuff like https://meta.wikimedia.org/wiki/File:2016-10-06_Wikimedia_system_dynamics_-_Early_draft. [21:17:57] jpg and maybe I could propose a session where people come in and critique my models to make them better. [21:18:46] Agreed, hence the focus on inventory of what we have and what we don't have (and an opportunity for product people to chime in, since they're consumers) [21:20:23] I have other ideas about documentation but they're longer term and not clear enough in my head yet; I'm drafting a business case on office wiki and will share it with doc-minded people in the next few weeks :) [21:20:28] guillom, for the system, I think it would be interesting to take advantage of tool devs context. Ask them "where do your tools fit into this system diagram -- or why don't they fit?" [21:21:02] I suspect there are some sorely under-supported parts of the system. [21:21:05] (Note that the current diagrams are also very incomplete and probably very wrong) [21:21:22] yeAH. halfak and I are both totes in to system theory, but I suspect that we'll need to provide a specifically pragmatic focus if we want to suck in more people and have a generative conversation. What would that be? [21:21:24] Totally fine. Applying reality to model is a good way to critique model [21:21:56] J-Mo, maybe this is a better way to think of the tools list [21:22:07] (with my tool devs framing [21:22:08] The "where do tools fit" (and also software features?) makes sense to me [21:22:14] yes, it's a great idea [21:22:15] agree [21:22:23] https://tools.wmflabs.org/hay/directory/ [21:22:27] Make a map of the sociotechnical ecosystem [21:22:37] Yeah. That sounds really cool [21:22:39] and by 'tools' we mean specifically things hosted on labs? [21:23:01] or just any adjuct tools that are not part of MW? [21:23:05] *adjunct [21:23:06] IMO, not limited to labs. Anything that people who call a tool [21:23:10] Even parts of MediaWiki [21:23:15] E.g. Special:RelatedChanges is a tool [21:23:20] mr. halfak [21:23:22] yeah, and Thanks, etc. [21:23:26] That DOCTOR HALFAK [21:23:28] mm. I see [21:23:30] oh, right. [21:23:31] dr. halfak [21:23:32] lol [21:23:58] Owing to the funding of Librarybase, I and my team are at your disposal. [21:24:01] or "Dr. Bossypants" [21:24:14] I accept Dr. Bossypants [21:24:17] Katherine poked fun at me the other day because I had halfak's and J-Mo's dissertations printed on my desk for reading. [21:24:34] OMG someone's going to read my dissertation :D [21:24:35] * J-Mo hopes guillom didn't notice all the typos [21:24:37] guillom: people didn't believe me when I said that I read J-Mo's dissertation [21:24:38] Ok, so that's one session on the sociotechnical map [21:24:46] J-Mo: I did :) [21:24:50] hare, in the middle of thinking about Dev summit sessions. Can talk librarybase later [21:24:52] dissertation reading party! [21:24:54] anyway [21:24:57] and maybe a session on inventory of knowledge and knowledge gaps? [21:25:12] I am proposing a session called "Tools for curating and organizing editing work: what has been done and where to go next" [21:25:14] guillom, yeah I agree that they are different focuses [21:25:20] I think that a role-of-tools-in-ecosystem topic is good, but still too broad. [21:25:23] If someone else is doing this, let me know so I can piggyback on their thing instead [21:25:26] J-Mo: I haven't read it yet; but I'll try to read past the typos :) [21:25:44] hare, +1 for that. Might want to get something like "cleanup templates" in the title so people have an anchor from which to start their critiques. [21:26:11] J-Mo: too broad because there's too much to map? [21:26:17] Well my specific objective is figuring out how to achieve synchronization between CollaborationKit, Wikipedia Requests, and Content Translation. [21:26:41] (Am I missing anything?) [21:26:47] too broad because people think of different types of tools, based on the work they do, and with a broad audience like this one it will be hard to get people focused [21:26:49] methinks [21:26:56] I hear you [21:27:10] J-Mo & guillom, could be OK depending on structure of the session [21:27:19] Maybe we could immediately break up into groups and try not to focus. [21:27:22] What if we break up by Labs vs. MediaWiki? [21:27:51] we should at least have a coherent problem statement, and I was thinking 'research tools' specifically before because it's relatively easy to frame that: "what kind of tools do different people who do different research things want, need, and already use?" [21:28:21] halfak. agree. what would you want to accomplish with such a session? [21:29:04] J-Mo, start with lose map of people/content flows in Wikipedia. Superimpose tools on map and observe empty/well-developed areas. [21:29:15] Use this to prioritize new work. [21:29:20] *loose [21:29:32] * halfak starts by losing the map [21:30:05] I do agree a map of research tools would be useful; it's just that I *also* want to have such a map for non-research things :) [21:30:07] hare might use such a map to get a sense for available tasks for routing [21:30:31] halfak, guillom, it sounds like an interesting session (scoped to 'all tools'). [21:30:50] We need a research/reflection loop on guillom's map [21:30:53] in your experience, is this a type/format of session that is common at DevSummit? [21:30:54] For the research tools to go [21:31:20] J-Mo, most sessions are (1) Product person talks a lot, (2) Short demo with discussion of demoed thing. [21:31:30] Quim said to be creative about the sessions, I imagine it's true for the format as well [21:31:33] I think this might be more productive than (1) and (2) [21:31:42] that's my problem: I was only at two last year: harej and Isarra's collabKit one, and volker's UX Standardization talk [21:32:00] halfak, I agree [21:32:10] I did a Product Person talks a lot for ORES last year. [21:32:53] This year, I'd rather do a "Let's imagine some tools using ORES" session where we write down product descriptions. [21:32:58] Or something like that. [21:33:20] What do you guys think about the idea of a "research room" where we can have research-focused presentations & workshops, and also hang out and maybe have impromptu, organic discussions when there's nothing planned? (if it's logistically possible) [21:33:26] So, anyway, this fits pretty solidly with the kind of thing I had in mind for sessions. [21:34:04] guillom, that sounds interesting to me. I'd really like to have a longer-form PAWS workshop [21:34:15] "Learn you some Research-fu" [21:34:15] (I don't want to prevent people from wandering to other sessions; I want to have a place for people to go when there's nothing really relevant) [21:35:37] I'd like to learn how PAWS works. [21:36:07] guillom: looks like they're planning an 'unconference' again: https://phabricator.wikimedia.org/T147955 (re: having a 'research room')_ [21:36:07] I think PAWS will be popular with the bot/research crowd. [21:36:08] halfak: So would it make sense to list all researchy things, and turn that into a "track" of sorts, with a room? [21:36:18] J-Mo: Hmm. [21:36:38] guillom, that might be a stretch. Let's bring that up to quim to see if the space would be reasonable for that. [21:36:48] The "topics" have already been solidified. [21:37:04] If there's an unconference planned, it can be part of that, I guess. [21:37:08] Then again, maybe we could just expand the AI topic to be "Research (and some AI stuff)" [21:37:28] AI is pretty firmly in the realm of "research" [21:37:29] Yeah, I don't want to disrupt any existing plans; just trying to make the most out of the event if I attend. [21:38:45] ok, I have a few minutes left before I need to go catch my bus. What would be the next steps? Make a list of ideas for workshops and discussions? [21:38:49] guillom, +1 to that. Just thinking of a political maneuver we might need to do. [21:39:04] 1. file phab tickets with a stub of the idea [21:39:27] 2. once we have a group of sessions together, let's talk to quim about having a dedicated room [21:40:31] Yes, the room is not a must-have, whereas the list is. [21:40:58] +1 [21:41:22] Well, this chat was useful to me. I hope I didn't prevent you from doing the brainstorming you mentioned ) [21:41:22] :) [21:41:41] This was the brainstorm! [21:41:41] Not at all. I think we're all better off :) [21:41:52] J-Mo's getting a real storm soon [21:42:02] two of 'em, in fact [21:42:25] I'm flying to the storm :/ [21:42:31] If they let us fly, even. [21:42:40] Vancouver and Seattle [21:43:01] Watch out for flying thorny tentacles if you need to go outside in the wind. I imagine all of those blackberry bushes creating a meat grinder in the storm. [21:43:04] oh that's right! I think the worst will be Saturday night. Things should be calming down by Sunday. When do you get in? [21:43:18] Saturday in Vancouver. [21:43:23] Hence the fun :D [21:43:44] blackberry bushes and flying Northwest Tree Octopi. [21:43:59] Alright, bus time. Thank you both again! [21:44:01] http://zapatopi.net/treeoctopus/ [21:44:22] lol