[16:28:00] hi [17:08:43] youtube link? [17:10:00] ottomata, wrong channel [17:16:18] ottomata, https://www.youtube.com/watch?v=mB_oexqz8pA [17:17:14] ^ that's for the "Wikipedia as the Front Matter to All Research" brownbag, in 2hrs 45 mins. [17:19:45] quiddity, I think a few people are waiting for that. It went over the Wikidata and Wikisource mailinglists. [17:19:56] Yupyup :) [17:21:09] (I am, too!) [17:21:42] Yeah, me too :) [17:48:05] Jamesofur: ping :) [17:48:27] dcausse: gchat or email ;) [17:48:38] oh ok, feeling dumb :) [17:48:55] no worries :D [19:58:11] Wikipedia as the Front Matter to All Research: https://www.youtube.com/watch?v=mB_oexqz8pA [20:00:05] Does anyone know if this will be recorded? [20:00:08] the livestream [20:01:39] Gamaliel, I am pretty sure. There are older live streams in the youtube channel of WMF [20:02:02] o/ [20:03:53] has it started for anyone else? [20:04:02] hi halfak [20:04:08] Gamaliel, Not yet. [20:04:10] Hi abbey [20:04:36] Thanks. No hurry, just didn't want to be the only one staring at a black screen. [20:05:15] Looks like we just started. [20:05:18] we are starting [20:05:21] Stream should be up shortly [20:05:34] Stream is up [20:05:42] :) [20:05:57] o/ tarrow [20:06:03] FYI: just getting started. [20:10:13] Has it started? I don't have Flash on Fedora :( [20:10:21] yes [20:10:22] yep, it started. [20:10:41] tobias47n9e_: I think youtube works with HTML5 in firefox or similar [20:10:49] I don't recall installing flash on this machine. [20:12:16] guillom, I probably need H.264 or WebM package or something. Shuffles off to PostinstallerF ... [20:13:07] ah, government. figures. [20:13:08] lol @ "record" [20:13:16] "Academix" [20:13:17] yeah -I love it! [20:13:19] Brilliant. [20:13:49] sorry for the slightly choppy audio everyone [20:14:06] brendan_campbell, working OK for me. [20:14:10] it's getting choppy somewhere in the youtube encoding process, it's clean on our end [20:14:19] Oh... I'm on hangouts. [20:14:40] Hi everyone - if anyone has questions, please share them here, I will ask when the time comes. [20:14:43] Nevermind... hearing it nbow [20:14:52] Thanks abbey! [20:15:34] halfak: it seems to be improving now. it was never completely unintelligible, just slightly annoying [20:15:41] Want to add a link to the stream in the topic? [20:15:57] So we have a mod who can?\ [20:15:58] https://www.youtube.com/watch?v=mB_oexqz8pA [20:16:03] ^ Stream link [20:16:22] *Do [20:16:22] "The Wikipedia" [20:16:28] :) [20:16:36] I'm wondering if DOIs can be used by "regular people" e.g. on personal sites & blogs etc. [20:17:40] used by = used to refer to pages hosted on such sites [20:17:50] halfak: i'm recording the audio locally now just in case, i can replace the audio in the video with clean audio once this presentation is done if necessary [20:17:51] guillom: not really, because in the DOI scheme you have to register as a DOI handler [20:18:11] (they don't call them handlers, I forget what the exact term is) [20:18:13] brendan_campbell, cool That'll be good for the archives. Thanks. [20:18:32] only fifth? slackers. [20:18:32] guillom: but there are a bunch of way to get DOIs for various research products [20:18:39] It streams pretty good using VLC for me. [20:18:41] phoebe: ! Great to see you :) [20:18:52] And thanks for the response :) [20:18:57] Gamaliel: that list counts the major publishers, IIRC, so we're not slacking that much :) [20:19:35] * halfak wonders if we can get a DOI for each :m:Research: page. [20:19:46] you're right, getting beat by Science Direct is nothing to be ashamed of. [20:20:09] for those who don't know, sciencedirect & scopus are both Elsevier [20:20:20] Gamaliel, I'm still skeptical that we are really getting beat. [20:20:21] We have an issues with referrers [20:20:21] DarTar has been working on that. [20:20:30] I think Wikipedia is under-counted. [20:20:30] there's a schema called ARKs which I think individuals can get [20:20:55] halfak: mostly lots of cites that have dois don't actually have them in the reference, so yeah [20:20:59] * guillom bookmarks https://en.wikipedia.org/wiki/Archival_Resource_Key . [20:21:08] * halfak hisses at Elsevil [20:21:17] héhé [20:21:19] they killed toollabs' redis a few times with cocytus [20:22:08] http://wikipedia.labs.crossref.org/ [20:22:10] anyway, halfak, for meta pages we'd be a lot better off running our own PURL server :) [20:23:06] "outside the scholarly literature" I think what he means is outside traditional publishing :) [20:24:22] they should get low tech and point a webcam at the wall counter. [20:25:18] more interestingly re: the counting issue, a lot of stuff we cite (news, tech reports, webpages, etc) don't have DOIs... [20:25:32] probably the majority of stuff we cite, actually. So this is still a tiny fraction of our references [20:25:47] insert-old-wikipedia-logo-joke [20:25:57] Do Klingons use DOIs? [20:26:03] hahaha [20:26:17] guillom we should add DOIs to the logo versions, sheesh [20:26:24] :D [20:27:15] ooh [20:27:18] this is cool [20:28:16] Structured Citations FTW. [20:28:50] I think everybody agrees by now that we need 1 item for 1 doi? [20:28:52] want it. want it. now :p [20:29:11] that's my understanding of how it works, yes [20:29:22] oh you mean one wikidata item [20:29:24] Lets blow up Wikidata by 10^2 :) [20:29:25] http://labs.crossref.org/ [20:29:49] See also: http://librarybase.wmflabs.org [20:29:53] +1 for putting cites in Wikidata [20:29:56] * halfak claps [20:30:02] that's pretty redundant assuming we implement a permanent URL scheme in Wikidata, unless you imagine the whole thing moving elsewhere [20:30:05] SEND QUESTION to abbey [20:30:09] *s [20:32:25] front matter of all research <3 [20:32:30] [citation needed] can also be the entry point for new science [20:32:51] After the Credible Hulk, Credible Goats. [20:33:00] are goats the new paramecium? [20:33:19] madhuvishy gets halfak-points! [20:33:33] :D [20:33:35] madhuvishy, He might be speaking of the near-extinction of Alpine mountain goats? [20:33:56] * halfak looks for swalling [20:34:36] word on the wikis is that swalling is responsible for a much of Wikipedia's information about goats and sheep. [20:34:48] for sure re: sheep, not sure about goats [20:34:50] haha [20:34:56] :D [20:34:58] KNOWLEDGE ENGINE! [20:35:01] * YuviPanda runs away [20:35:08] YuviPanda: hahaha [20:35:11] phoebe, indeed. I might be misremembering [20:35:15] * YuviPanda freebases KNOWLEDGE ENGINE [20:35:24] halfak it seems likely, anyway [20:35:25] lol [20:35:38] you should call the elastic search thing knowledge engine [20:35:46] haha [20:36:03] for today, I'm going to call my left shoe 'knowledge engine' [20:36:33] can't stop laughing [20:36:38] YuviPanda: Call the other one "Strategy". [20:36:48] oooh [20:36:53] * YuviPanda walks on KE and Strategy [20:37:15] * madhuvishy puts an engagement hat on YuviPanda [20:37:18] guillom: so if i use the phrase 'the other shoe has dropped' it means we know about our strategy now? [20:37:46] is my wrist brace 'community'? I hate it and need it at the same time... [20:38:08] "users don't care about provenance" [20:38:10] hmm [20:38:38] * YuviPanda stops being OT [20:39:07] i love that almanac analogy. [20:39:11] btw, I know someone who consults and trusts an almanac now [20:40:04] and wikipedia doesn't actually count as a source [20:40:09] since it's just an intermediary [20:40:46] https://newrepublic.com/article/124425/dreamed-perfect-database [20:42:46] Fun fact: the French Wikipedia did start using a template-powered database for their citations: https://fr.wikipedia.org/wiki/Aide:Espace_r%C3%A9f%C3%A9rence [20:43:06] We should also create a Wikisource page for every DOI with a countdown that tells the time until is is public domain :) [20:43:22] guillom how has that worked out? [20:43:51] there's a few template-driven citation pages in en.wp but it's not a namespace [20:43:52] (Example: https://fr.wikipedia.org/wiki/R%C3%A9f%C3%A9rence:The_visual_display_of_quantitative_information_(Edward_R._Tufte) ) [20:43:53] Lydia_WMDE, Wikidata at the center! Yeah! [20:44:07] phoebe: I think it fell out of practice over time. I should look into it. [20:44:38] Not many edits in the RCs :/ https://fr.wikipedia.org/w/index.php?namespace=104&tagfilter=&title=Sp%C3%A9cial%3AModifications+r%C3%A9centes [20:45:07] I wish that I could describe all my papers on Wikidata and upload the PDFs to commons. [20:45:09] https://www.wikidata.org/wiki/Wikidata:WikiProject_Source_MetaData [20:45:13] Stupid publishing industry [20:45:28] I'm going to plug http://librarybase.wmflabs.org again. [20:46:06] halfak: it's better to consolidate papers with other papers -- there's a lot of value in building a collection. But have you double-checked all your publishing agreements? Posted the author version? [20:46:07] that CrofRef presentation is quite nice [20:46:14] "Parser Functions" in MediaWiki lingo [20:46:50] phoebe, I'm not lawyer enough to know if I can upload the author versions to commons. [20:46:58] is the CrossRef organization using Wikidata for data storing? [20:46:59] But it's clear that I can host on my website. [20:47:06] So I do so agressively. [20:47:26] halfak: honestly, that is better than commons from a library perspective. What's best of all is uploading them in a repository of some sort [20:47:44] But +1 for collections. I think we want to have individual items. One thing I want to find time for is proper open summaries on Meta. [20:47:45] I don't want scholarly papers in commons for a bunch of reasons [20:47:51] * YuviPanda becomes commons admin, deletes halfak's papers [20:47:53] unrelated to licensing of said papers [20:48:03] phoebe, commons != repository? [20:48:10] YuviPanda, :( ! [20:48:11] it's a repository of other stuff [20:48:12] :P [20:48:26] @hashar I will ask your question - it is first in line. [20:48:36] "Per Commons:NOTCOMMONSIMOLOL" [20:48:38] I want like things with like. Plus, we don't have the kind of OCR & metadata support that I'd want for a paper repository [20:48:48] Tweets/Second is the most important metric of our time :( [20:49:00] like downloading citations in lots of formats, linking in teh data, etc etc [20:49:00] phoebe, papers in commons, metadata in wikidata? [20:49:08] hashar: CrossRef was the first presentation. Dario's is related to references but not about CrossRef. [20:49:10] phoebe: +1 Commons has support problems [20:49:26] basically I want features for papers that are irrelevant for images, largely [20:49:28] Can't imagine why [20:49:30] oh so that is Dario speaking right now? [20:49:35] yes [20:49:51] * YuviPanda pats marktraceur [20:49:56] halfak: Papers on Wikisource is the plan isn't it. Only the pictures on commons? [20:50:19] phoebe, I think of Commons and Wikidata as data stores and other systems for indexing and presenting as using those data stores. [20:50:24] Jupyter notebooks on Wikiversity? [20:50:34] guillom: omg yes :D [20:50:43] (I'M JOKING DON'T DO IT :P) [20:50:46] guillom: it is fairly trivial to implement a notebook viewer as a mw extension btw [20:50:51] But regardless, I just want a place to put scholarly assets and add metadata about them. [20:51:10] halfak: Surely Wikisource is a data store, too...just for a different format [20:51:21] Yeah. Indeed. [20:51:24] Sort of a transformative data store [20:51:29] halfak - all your preprints are in arxiv, no? [20:51:32] push the assets to commons then reference them as item in Wikidata where you can add metadata? [20:51:34] 6. -> This is why we need Wikidata, No other thing can solve these wild queries. [20:51:44] isn't Wikidata supposed to eventually be used for Commons metadata? [20:51:49] cause the {{templates}} are getting old [20:51:50] phoebe, only discovered arxiv recently, but yes since I realized that I could :) [20:51:51] hashar: Wikibase is [20:52:02] harej! [20:52:04] hashar: WMDE needs to get on it though. No pressure Lydia_WMDE [20:52:05] hashar, yeah. That was my thought [20:52:12] james hare's project is librarybase.wmflabs.org [20:52:17] ^ [20:52:18] woo wmflabs :D [20:52:23] ^ [20:52:31] it has a new logo too [20:52:41] Although I was really hoping for a Credible-Hulk logo :p [20:52:55] guillom, what logo? [20:52:55] halfak: cool. I'd do that. [20:53:03] halfak: new logo for librarybase [20:53:09] * halfak loads it up [20:53:19] arxiv ids are also a permanent scheme, can be treated like dois etc in wikidata. [20:53:32] great Dario! [20:53:33] hashar: One of the initial irrational fears was that Wikidata would steal all of Commons's metadata and the Commons community would lose ownership of it, so Commons is going to be a Wikibase repository on its own...which makes sense, because file metadata is hardly of the same quality as the stuff on Wikidata now. [20:53:37] SEND YER QUESTIONS TO ABBEY! [20:53:38] my question was 'is that a goat'. answer was 'yes' [20:53:44] yay dario [20:53:49] yay dario [20:53:50] 70 Millions + probably on averag 3 authors. So 200 million items as a rough estimate? [20:54:19] my question is about getting all the stuff into Wikidata that doesn't currently have any permanent ids [20:54:19] My only question is "So when *do* we get citations in Wikidata" and I'm pretty sure I won't like the answer :p [20:54:47] Cleanest data import that wikidata ever got. [20:54:54] WOuld be from librarybase [20:55:20] gogogo Alex [20:55:21] Can we include the Wiki-related literature in the pilot project? p-p-p-pleaaase . [20:55:39] lolol re: web of science [20:56:02] guillom, sounds like a good place to go. [20:56:04] Could use referata. :) [20:56:30] http://wikipapers.referata.com/wiki/Main_Page [20:56:31] marktraceur: I guess it make sense to have commons have its own metadata system yeah. Then wiktionary will ask for another :-} [20:56:35] abbey: question for dario: how about using wikidata queries to identify uncited or poorly cited statements? [20:56:39] yes. Because at the moment I'm using citation templaters + Wikipapers template + Acawiki template and it's painful. [20:56:43] marktraceur: soon we will need a WikibaseFederation system [20:56:49] v.v [20:56:51] abbey: i.e. statements that aren't well cited currently in Wikipedia itself [20:56:57] Looks like wikipapers died in 2012 [20:57:02] But we can bring it back in a new form [20:57:44] got it Phoebe.. I have a few others collected to ask first [20:57:54] halfak: Nemo and JeanFred are still editing a bit. [20:58:12] <3 Nemo and JeanFred [20:58:16] abbey: can confirm about wikipedia being use as entry point. At least among teachers and university folks I know. They quick glance at wiki, look for papers then go dig for more related papers [20:58:22] OMG lets make a [research needed] tag for Wikipedia :) [20:58:47] tobias47n9e_: Then researchers will add a [funding needed] tag ;) [20:58:56] [data needed]? [20:59:00] [analysis needed]? [20:59:01] re: what geoff is saying of course we function like review articles... but it's uneven. In general though we follow same patterns of citing reviews, famous articles [20:59:02] hashar have your question too [20:59:20] tobias47n9e_ & guillom, we should have those on Meta' [20:59:27] s research namespace [20:59:28] but let's get real people: citations in wikipedia suck [20:59:35] (For clarity: all my questions were rhetorical. I don't have any serious questions.) [20:59:41] FYI: https://meta.wikimedia.org/wiki/Research:Index [20:59:48] we do not do a good job of surveying the literature or making the most comprehensive things in front of people [21:00:17] and what the "best citation" is depends a lot on your audience, anyway. [21:00:21] OT question: how far are we from the realization of the Signalling OAness project? What do we miss? [21:00:30] phoebe, we need to figure out how to make Wikipedia part of academic practicwe. [21:00:38] phoebe, Modern science has only had ~100 years. Give it some more centuries to perfectly describe the world. Wikipedia can't fix everything. [21:00:53] aubreymcfato: hi! and I don't think that's OT at all... they mentioned flagging OA [21:00:59] tobias47n9e_, but we should be able to clear discuss the state of science. [21:01:02] ALso, 100 years? [21:01:04] Boo [21:01:12] halfak: / tobias47n9e_: Once I get it off the ground, the Codex will include a gap analysis that sort-of serves the "Research needed" issue. [21:01:12] What's "Modern"? [21:01:18] phoebe: hi! :-) [21:01:25] halfak, OK lets start with Keppler :) [21:01:28] halfak: the thing is "the best cite" might not be academic. Usually isn't. The best cite might be a NY times article that everyone can get too [21:01:32] "the Codex"? [21:01:41] we know this instinctively as a community, which is why we cite the NYT all the dang time [21:01:43] halfak: My lit-review project [21:01:50] gotcha :) [21:02:02] * halfak needs to keep up with guillom's awesome projects better [21:02:16] not pasting the link here because it's empty at the moment! [21:02:41] Maybe paste anyway so I can watchlist [21:02:42] halfak: there's not much to keep up with right now. I'm just reading a lot! [21:02:43] :D [21:02:46] phoebe, I will ask your q next [21:02:52] https://meta.wikimedia.org/wiki/Research:Codex [21:03:07] * halfak watches [21:03:13] OT is that the 5th floor? [21:03:19] phoebe: yup [21:03:20] yes [21:04:00] guillom, Thanks. Watchlisted. [21:05:28] btw I think these discussions tend to make me despair about epistomology and whether we can ever actually cite/know anything [21:06:10] time will come! we are still at the stone age [21:07:06] phoebe, IMO it takes a community to come to know things. [21:07:12] Wikidata cites every data-point, that is better than the current state of most infoboxes. [21:07:21] We rely on others a lot more than anyone will really admit to know things. [21:07:30] wikidata makes it *possible* to cite every data point [21:07:39] phoebe: true :P [21:07:56] So if we can have an active community discussing and iterating on subject area, then that subject area can become much more known. [21:08:02] TrevorParscal: if you got a moment please take a look at https://phabricator.wikimedia.org/T119404#1838900 [21:08:13] abbey: yeah just a statement sorry [21:08:25] It would be great if we could substantially extend the walled-off academic communities via open knowledge projects like this. [21:08:27] * halfak is a dreamer [21:08:55] halfak I mean right back to the beginning though, like fundamental epistomology. How science works. That kind of thing. [21:09:05] which is a little out of scope for this conversation :) [21:09:26] any other questions I might have missed? [21:09:29] ask them now [21:09:31] but agree re: opening up the community [21:09:32] Oh. Well... Yeah. I'm out of my element in a phil. of science discussion. I'm much more useful for practical methodology. [21:09:57] abbey: thank the presenters from the IRC peanut gallery over here? :) [21:10:04] :D [21:10:06] yes I will. [21:10:14] thx [21:10:15] wow, finally something that halfak is out of his element in! [21:10:15] Best peanut galley in a while. :) [21:10:21] :P YuviPanda [21:10:35] * halfak continues his rocket surgery AI project [21:10:37] ;) [21:10:44] o/ [21:10:54] thanks everyone! [21:11:00] Great talk :) [21:11:02] yeah thanks ! [21:11:02] talks* [21:11:05] was nice [21:11:05] Thanks folks. [21:11:05] The conversation doesn't need to stop here. [21:11:13] Thanks all! [21:11:18] what would be fun [21:11:22] thanks [21:11:29] is to use Wikipedia / Wikidata to potentially link papers togethers [21:11:32] If you want to get involved in these initiatives and you're not already CC'd on emails and phab tasks like crazy, let me know and I'll loop you in. [21:11:39] link papers together? [21:11:51] ^ that [21:11:51] want [21:12:02] hashar like a who cites who type of deal? [21:12:05] Paper graph that is annotated by whether the citation is casual, negative or supporting. [21:12:11] <3 [21:12:14] i.e. [[article]] cites Paper A and Paper B, that might mean they are linked somehow [21:12:23] but I'd really like to know where the Signalling OAness project stopped: my understanding is that we are waiting for CrossRef to figure out some issues [21:12:43] so CrossRef on Paper A can indicate that Paper B might be of interested (because: boths are on [[Article]] ) [21:12:50] who-cites-who data is even more locked up than most bibliographic data, btw, because only three big players have really had the resources to do it: Web of Science, Scopus (Elsevier) and Google Scholar [21:12:58] you need a giant corpus and a bunch of resources [21:13:11] is it a problem of metadata about OAness? [21:13:17] and that cold work the other way around. If Paper A is linked on  [[Elephant]] and [[Water]], then maybe on the [[Elephant]] page we could hint about [[Water]] because both cites Paper A [21:14:04] halfak: I'd like to be looped in [21:14:08] hashar yes! and you can see papers that are highly, highly cited (like famous works, or super useful reviews) [21:14:15] so I could browse from article to paper back to article to another paper [21:14:18] across many wikipedia articles [21:14:28] and loose even more time on the net :D [21:14:58] does Wikidata has the CrossRef data being injected? [21:15:51] it sounds like they're talking about it? [21:16:14] I missed some of that [21:16:52] yeah I missed the first part [21:16:59] I have just found http://wikipedia.labs.crossref.org :D [21:17:17] so seems the data are out of there and some smart folks will figure out something to do with them [21:19:13] alright friends, good to see you all again [21:19:26] I'd best get back to work & then home, since I'm 3 hrs later now :) [21:20:17] we should all nerd out about library things again sometime! [21:21:29] phoebe, Hopefully! [21:22:31] hashar: The what-cites-what was in the model that Dario presented. [21:22:44] WikiPapers sort of does that with Semantic MediaWiki. [21:22:57] But it's laborious because it's all manual :/ [21:27:58] guillom: maybe Wikipapers community can be linked to Dario [21:28:31] The WikiPapers community is basically emrijp & Nemo_bis ;) [21:28:31] anyway it is great to see all that activity going on [21:28:41] oh [21:28:48] should be easy so :} [22:02:25] Hi [22:06:08] #fedora-meeting [22:19:16] #startmeeting holidays [22:19:17] Meeting started Fri Dec 4 22:19:17 2015 UTC and is due to finish in 60 minutes. The chair is northern1. Information about MeetBot at http://wiki.debian.org/MeetBot. [22:19:17] Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. [22:19:19] The meeting name has been set to 'holidays' [22:19:23] Hello [22:21:14] Hi [22:24:56] Hi [22:27:10] #endmeeting [22:27:11] Meeting ended Fri Dec 4 22:27:11 2015 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) [22:27:11] Minutes: https://tools.wmflabs.org/meetbot/wikimedia-office/2015/wikimedia-office.2015-12-04-22.19.html [22:27:11] Minutes (text): https://tools.wmflabs.org/meetbot/wikimedia-office/2015/wikimedia-office.2015-12-04-22.19.txt [22:27:11] Minutes (wiki): https://tools.wmflabs.org/meetbot/wikimedia-office/2015/wikimedia-office.2015-12-04-22.19.wiki [22:27:12] Log: https://tools.wmflabs.org/meetbot/wikimedia-office/2015/wikimedia-office.2015-12-04-22.19.log.html [22:28:23] kLklK [22:28:24] k [22:28:24] k [22:28:24] k [22:28:25] kk [22:28:25] k [22:28:25] k [22:28:25] k [22:28:26] kl [22:31:11] l#OS2UGBE [22:31:17] #OS2UGBE [22:31:51] !ops [22:31:58] yes? [22:32:00] oh [22:32:05] yes