[Script Info]
Title: 
[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
Dialogue: 0,0:00:05.94,0:00:09.48,Default,,0000,0000,0000,,Hello everyone to the Data Quality panel.
Dialogue: 0,0:00:10.29,0:00:13.67,Default,,0000,0000,0000,,Data quality matters because\Nmore and more people out there
Dialogue: 0,0:00:13.67,0:00:19.29,Default,,0000,0000,0000,,rely on our data being in good shape,\Nso we're going to talk about data quality,
Dialogue: 0,0:00:20.03,0:00:26.00,Default,,0000,0000,0000,,and there will be four speakers\Nwho will give short introductions
Dialogue: 0,0:00:26.00,0:00:29.54,Default,,0000,0000,0000,,on topics related to data quality\Nand then we will have a Q and A.
Dialogue: 0,0:00:30.13,0:00:32.23,Default,,0000,0000,0000,,And the first one is Lucas.
Dialogue: 0,0:00:34.38,0:00:35.38,Default,,0000,0000,0000,,Thank you.
Dialogue: 0,0:00:35.90,0:00:39.90,Default,,0000,0000,0000,,Hi, I'm Lucas, and I'm going\Nto start with an overview
Dialogue: 0,0:00:39.90,0:00:43.81,Default,,0000,0000,0000,,of data quality tools\Nthat we already have on Wikidata
Dialogue: 0,0:00:43.81,0:00:46.11,Default,,0000,0000,0000,,and also some things\Nthat are coming up soon.
Dialogue: 0,0:00:46.93,0:00:50.62,Default,,0000,0000,0000,,And I've grouped them\Ninto some general themes
Dialogue: 0,0:00:50.62,0:00:53.76,Default,,0000,0000,0000,,of making errors more visible,\Nmaking problems actionable,
Dialogue: 0,0:00:53.76,0:00:56.32,Default,,0000,0000,0000,,getting more eyes on the data\Nso that people notice the problems,
Dialogue: 0,0:00:56.94,0:01:02.62,Default,,0000,0000,0000,,fix some common sources of errors,\Nmaintain the quality of the existing data
Dialogue: 0,0:01:02.62,0:01:03.97,Default,,0000,0000,0000,,and also human curation.
Dialogue: 0,0:01:05.06,0:01:09.87,Default,,0000,0000,0000,,And the ones that are currently available\Nstart with property constraints.
Dialogue: 0,0:01:10.39,0:01:12.42,Default,,0000,0000,0000,,So you've probably seen this\Nif you're on Wikidata.
Dialogue: 0,0:01:12.42,0:01:14.03,Default,,0000,0000,0000,,You can sometimes get these icons
Dialogue: 0,0:01:14.53,0:01:17.24,Default,,0000,0000,0000,,which check\Nthe internal consistency of the data.
Dialogue: 0,0:01:17.24,0:01:20.80,Default,,0000,0000,0000,,For example,\Nif one event follows the other,
Dialogue: 0,0:01:20.80,0:01:23.76,Default,,0000,0000,0000,,then the other event should\Nalso be followed by this one,
Dialogue: 0,0:01:23.76,0:01:27.16,Default,,0000,0000,0000,,which on the WikidataCon item\Nwas apparently missing.
Dialogue: 0,0:01:27.16,0:01:29.36,Default,,0000,0000,0000,,I'm not sure,\Nthis feature is a few days old.
Dialogue: 0,0:01:30.04,0:01:34.68,Default,,0000,0000,0000,,And there's also,\Nif this is too limited or simple for you,
Dialogue: 0,0:01:34.68,0:01:38.08,Default,,0000,0000,0000,,you can write any checks you want\Nusing the Query Service
Dialogue: 0,0:01:38.08,0:01:39.84,Default,,0000,0000,0000,,which is useful for\Nlots of things of course,
Dialogue: 0,0:01:39.84,0:01:44.54,Default,,0000,0000,0000,,but you can also use it\Nfor finding errors.
Dialogue: 0,0:01:44.54,0:01:46.97,Default,,0000,0000,0000,,Like if you've noticed\None occurrence of a mistake,
Dialogue: 0,0:01:46.98,0:01:49.71,Default,,0000,0000,0000,,then you can check\Nif there are other places
Dialogue: 0,0:01:49.71,0:01:51.96,Default,,0000,0000,0000,,where people have made\Na very similar error
Dialogue: 0,0:01:51.96,0:01:53.44,Default,,0000,0000,0000,,and find that with the Query Service.
Dialogue: 0,0:01:53.44,0:01:54.56,Default,,0000,0000,0000,,You can also combine the two
Dialogue: 0,0:01:54.56,0:01:57.87,Default,,0000,0000,0000,,and search for constraint violations\Nin the Query Service,
Dialogue: 0,0:01:57.88,0:02:01.24,Default,,0000,0000,0000,,for example,\Nonly the violations in some area
Dialogue: 0,0:02:01.24,0:02:03.76,Default,,0000,0000,0000,,or WikiProject that's relevant to you,
Dialogue: 0,0:02:03.76,0:02:06.83,Default,,0000,0000,0000,,although the results are currently\Nnot complete, sadly.
Dialogue: 0,0:02:08.42,0:02:09.88,Default,,0000,0000,0000,,There is revision scoring.
Dialogue: 0,0:02:10.69,0:02:12.67,Default,,0000,0000,0000,,That's... I think this is\Nfrom the recent changes
Dialogue: 0,0:02:12.67,0:02:16.22,Default,,0000,0000,0000,,you can also get it on your watch list\Nan automatic assessment
Dialogue: 0,0:02:16.22,0:02:20.25,Default,,0000,0000,0000,,of is this edit likely to be\Nin good faith or in bad faith
Dialogue: 0,0:02:20.25,0:02:22.31,Default,,0000,0000,0000,,and is it likely to be\Ndamaging or not damaging,
Dialogue: 0,0:02:22.31,0:02:24.20,Default,,0000,0000,0000,,I think those are the two dimensions.
Dialogue: 0,0:02:24.21,0:02:25.69,Default,,0000,0000,0000,,So you can, if you want,
Dialogue: 0,0:02:25.69,0:02:29.90,Default,,0000,0000,0000,,focus on just looking through\Nthe damaging but good faith edits.
Dialogue: 0,0:02:29.90,0:02:32.52,Default,,0000,0000,0000,,If you're feeling particularly\Nfriendly and welcoming
Dialogue: 0,0:02:32.52,0:02:37.12,Default,,0000,0000,0000,,you can tell these editors,\N"Thank you for your contribution,
Dialogue: 0,0:02:37.12,0:02:40.56,Default,,0000,0000,0000,,here's how you should have done it\Nbut thank you, still."
Dialogue: 0,0:02:40.56,0:02:42.19,Default,,0000,0000,0000,,And if you're not feeling that way,
Dialogue: 0,0:02:42.19,0:02:44.45,Default,,0000,0000,0000,,you can go through\Nthe bad faith, damaging edits,
Dialogue: 0,0:02:44.45,0:02:45.57,Default,,0000,0000,0000,,and revert the vandals.
Dialogue: 0,0:02:47.54,0:02:49.76,Default,,0000,0000,0000,,There's also, similar to that,\Nentity scoring.
Dialogue: 0,0:02:49.76,0:02:52.59,Default,,0000,0000,0000,,So instead of scoring an edit,\Nthe change that it made,
Dialogue: 0,0:02:52.59,0:02:53.90,Default,,0000,0000,0000,,you score the whole revision,
Dialogue: 0,0:02:53.90,0:02:56.48,Default,,0000,0000,0000,,and I think that is\Nthe same quality measure
Dialogue: 0,0:02:56.48,0:02:59.86,Default,,0000,0000,0000,,that Lydia mentions\Nat the beginning of the conference.
Dialogue: 0,0:03:00.37,0:03:04.57,Default,,0000,0000,0000,,That gives a user script up here\Nand gives you a score of like one to five,
Dialogue: 0,0:03:04.57,0:03:08.18,Default,,0000,0000,0000,,I think it was, of what the quality\Nof the current item is.
Dialogue: 0,0:03:10.04,0:03:15.53,Default,,0000,0000,0000,,The primary sources tool is for\Nany database that you want to import,
Dialogue: 0,0:03:15.53,0:03:18.36,Default,,0000,0000,0000,,but that's not high enough quality\Nto directly add to Wikidata,
Dialogue: 0,0:03:18.37,0:03:20.34,Default,,0000,0000,0000,,so you add it\Nto the primary sources tool instead,
Dialogue: 0,0:03:20.34,0:03:22.96,Default,,0000,0000,0000,,and then humans can decide
Dialogue: 0,0:03:22.96,0:03:26.02,Default,,0000,0000,0000,,should they add\Nthese individual statements or not.
Dialogue: 0,0:03:28.60,0:03:31.90,Default,,0000,0000,0000,,Showing coordinates as maps\Nis mainly a convenience feature
Dialogue: 0,0:03:31.90,0:03:33.59,Default,,0000,0000,0000,,but it's also useful for quality control.
Dialogue: 0,0:03:33.59,0:03:36.94,Default,,0000,0000,0000,,Like if you see this is supposed to be\Nthe office of Wikimedia Germany
Dialogue: 0,0:03:36.94,0:03:39.40,Default,,0000,0000,0000,,and if the coordinates\Nare somewhere in the Indian Ocean,
Dialogue: 0,0:03:39.40,0:03:41.53,Default,,0000,0000,0000,,then you know that\Nsomething is not right there
Dialogue: 0,0:03:41.53,0:03:44.79,Default,,0000,0000,0000,,and you can see it much more easily\Nthan if you just had the numbers.
Dialogue: 0,0:03:46.38,0:03:49.58,Default,,0000,0000,0000,,This is a gadget called\Nthe relative completeness indicator
Dialogue: 0,0:03:49.58,0:03:52.48,Default,,0000,0000,0000,,which shows you this little icon here
Dialogue: 0,0:03:53.01,0:03:55.65,Default,,0000,0000,0000,,telling you how complete\Nit thinks this item is
Dialogue: 0,0:03:55.65,0:03:57.61,Default,,0000,0000,0000,,and also which properties\Nare most likely missing,
Dialogue: 0,0:03:57.61,0:03:59.77,Default,,0000,0000,0000,,which is really useful\Nif you're editing an item
Dialogue: 0,0:03:59.77,0:04:03.17,Default,,0000,0000,0000,,and you're in an area\Nthat you're not very familiar with
Dialogue: 0,0:04:03.17,0:04:05.66,Default,,0000,0000,0000,,and you don't know what\Nthe right properties to use are,
Dialogue: 0,0:04:05.66,0:04:08.23,Default,,0000,0000,0000,,then this is a very useful gadget to have.
Dialogue: 0,0:04:09.60,0:04:11.40,Default,,0000,0000,0000,,And we have Shape Expressions.
Dialogue: 0,0:04:11.40,0:04:15.62,Default,,0000,0000,0000,,I think Andra or Jose\Nare going to talk more about those
Dialogue: 0,0:04:15.62,0:04:19.76,Default,,0000,0000,0000,,but basically, a very powerful way\Nof comparing the data you have
Dialogue: 0,0:04:19.76,0:04:20.76,Default,,0000,0000,0000,,against the schema,
Dialogue: 0,0:04:20.76,0:04:22.68,Default,,0000,0000,0000,,like what statement should\Ncertain entities have,
Dialogue: 0,0:04:22.68,0:04:25.68,Default,,0000,0000,0000,,what other entities should they link to\Nand what should those look like,
Dialogue: 0,0:04:26.23,0:04:29.37,Default,,0000,0000,0000,,and then you can find problems that way.
Dialogue: 0,0:04:30.37,0:04:32.36,Default,,0000,0000,0000,,I think... No there is still more.
Dialogue: 0,0:04:32.36,0:04:34.32,Default,,0000,0000,0000,,Integraality or property dashboard.
Dialogue: 0,0:04:34.32,0:04:36.77,Default,,0000,0000,0000,,It gives you a quick overview\Nof the data you already have.
Dialogue: 0,0:04:36.77,0:04:39.15,Default,,0000,0000,0000,,For example, this is from\Nthe WikiProject Red Pandas,
Dialogue: 0,0:04:39.66,0:04:41.68,Default,,0000,0000,0000,,and you can see that\Nwe have a sex or gender
Dialogue: 0,0:04:41.68,0:04:43.56,Default,,0000,0000,0000,,for almost all of the red pandas,
Dialogue: 0,0:04:43.56,0:04:46.85,Default,,0000,0000,0000,,the date of birth varies a lot\Nby which zoo they come from
Dialogue: 0,0:04:46.85,0:04:50.26,Default,,0000,0000,0000,,and we have almost\Nno dead pandas which is wonderful,
Dialogue: 0,0:04:51.44,0:04:52.60,Default,,0000,0000,0000,,because they're so cute.
Dialogue: 0,0:04:53.70,0:04:55.65,Default,,0000,0000,0000,,So this is also useful.
Dialogue: 0,0:04:56.38,0:04:59.18,Default,,0000,0000,0000,,There we go, OK,\Nnow for the things that are coming up.
Dialogue: 0,0:04:59.89,0:05:03.78,Default,,0000,0000,0000,,Wikidata Bridge, or also known,\Nformerly known as client editing,
Dialogue: 0,0:05:03.78,0:05:07.08,Default,,0000,0000,0000,,so editing Wikidata\Nfrom Wikipedia infoboxes
Dialogue: 0,0:05:07.68,0:05:11.72,Default,,0000,0000,0000,,which will on the one hand\Nget more eyes on the data
Dialogue: 0,0:05:11.72,0:05:13.44,Default,,0000,0000,0000,,because more people can see the data there
Dialogue: 0,0:05:13.44,0:05:18.84,Default,,0000,0000,0000,,and it will hopefully encourage\Nmore use of Wikidata in the Wikipedias
Dialogue: 0,0:05:18.84,0:05:20.92,Default,,0000,0000,0000,,and that means that more\Npeople can notice
Dialogue: 0,0:05:20.92,0:05:23.39,Default,,0000,0000,0000,,if, for example some data is outdated\Nand needs to be updated
Dialogue: 0,0:05:23.86,0:05:27.00,Default,,0000,0000,0000,,instead of if they would\Nonly see it on Wikidata itself.
Dialogue: 0,0:05:28.63,0:05:30.66,Default,,0000,0000,0000,,There is also tainted references.
Dialogue: 0,0:05:30.66,0:05:33.96,Default,,0000,0000,0000,,The idea here is that\Nif you edit a statement value,
Dialogue: 0,0:05:34.68,0:05:37.28,Default,,0000,0000,0000,,you might want to update\Nthe references as well,
Dialogue: 0,0:05:37.28,0:05:39.37,Default,,0000,0000,0000,,unless it was just a typo or something.
Dialogue: 0,0:05:39.90,0:05:43.66,Default,,0000,0000,0000,,And this tainted references\Ntells editors that
Dialogue: 0,0:05:43.66,0:05:49.76,Default,,0000,0000,0000,,and also that other editors\Nsee which other edits were made
Dialogue: 0,0:05:49.76,0:05:52.47,Default,,0000,0000,0000,,that edited a statement value\Nand didn't update a reference
Dialogue: 0,0:05:52.47,0:05:56.77,Default,,0000,0000,0000,,then you can clean up after that\Nand decide should that be...
Dialogue: 0,0:05:57.74,0:05:59.57,Default,,0000,0000,0000,,Do you need to do any thing more of that
Dialogue: 0,0:05:59.57,0:06:02.80,Default,,0000,0000,0000,,or is that actually fine and\Nyou don't need to update the reference.
Dialogue: 0,0:06:03.54,0:06:09.34,Default,,0000,0000,0000,,That's related to signed statements\Nwhich is coming from a concern, I think,
Dialogue: 0,0:06:09.34,0:06:12.36,Default,,0000,0000,0000,,that some data providers have that like...
Dialogue: 0,0:06:14.13,0:06:17.23,Default,,0000,0000,0000,,There's a statement that's referenced\Nthrough the UNESCO or something
Dialogue: 0,0:06:17.23,0:06:19.87,Default,,0000,0000,0000,,and then suddenly,\Nsomeone vandalizes the statement
Dialogue: 0,0:06:19.87,0:06:21.84,Default,,0000,0000,0000,,and they are worried\Nthat it will look like
Dialogue: 0,0:06:22.83,0:06:26.99,Default,,0000,0000,0000,,this organization, like UNESCO,\Nstill set this vandalism value
Dialogue: 0,0:06:26.99,0:06:28.71,Default,,0000,0000,0000,,and so, with signed statements,
Dialogue: 0,0:06:28.71,0:06:31.49,Default,,0000,0000,0000,,they can cryptographically\Nsign this reference
Dialogue: 0,0:06:31.49,0:06:33.56,Default,,0000,0000,0000,,and that doesn't prevent any edits to it,
Dialogue: 0,0:06:34.17,0:06:37.74,Default,,0000,0000,0000,,but at least, if someone\Nvandalizes the statement
Dialogue: 0,0:06:37.74,0:06:40.26,Default,,0000,0000,0000,,or edits it in any way,\Nthen the signature is no longer valid,
Dialogue: 0,0:06:40.26,0:06:43.40,Default,,0000,0000,0000,,and you can tell this is not exactly\Nwhat the organization said,
Dialogue: 0,0:06:43.40,0:06:47.06,Default,,0000,0000,0000,,and perhaps it's a good edit\Nand they should re-sign the new statement,
Dialogue: 0,0:06:47.06,0:06:49.85,Default,,0000,0000,0000,,but also perhaps it should be reverted.
Dialogue: 0,0:06:51.20,0:06:54.17,Default,,0000,0000,0000,,And also, this is going\Nto be very exciting, I think,
Dialogue: 0,0:06:54.17,0:06:56.85,Default,,0000,0000,0000,,Citoid is this amazing system\Nthey have on Wikipedia
Dialogue: 0,0:06:57.38,0:07:01.34,Default,,0000,0000,0000,,where you can paste a URL,\Nor an identifier, or an ISBN
Dialogue: 0,0:07:01.34,0:07:04.76,Default,,0000,0000,0000,,or Wikidata ID or basically\Nanything into the Visual Editor,
Dialogue: 0,0:07:05.26,0:07:08.24,Default,,0000,0000,0000,,and it spits out a reference\Nthat is nicely formatted
Dialogue: 0,0:07:08.24,0:07:11.05,Default,,0000,0000,0000,,and has all the data you want\Nand it's wonderful to use.
Dialogue: 0,0:07:11.05,0:07:14.34,Default,,0000,0000,0000,,And by comparison, on Wikidata,\Nif I want to add a reference
Dialogue: 0,0:07:14.34,0:07:18.80,Default,,0000,0000,0000,,I typically have to add a reference URL,\Ntitle, author name string,
Dialogue: 0,0:07:18.80,0:07:20.45,Default,,0000,0000,0000,,published in, publication date,
Dialogue: 0,0:07:20.45,0:07:25.14,Default,,0000,0000,0000,,retrieve dates,\Nat least those, and that's annoying,
Dialogue: 0,0:07:25.14,0:07:29.26,Default,,0000,0000,0000,,and integrating Citoid into Wikibase\Nwill hopefully help with that.
Dialogue: 0,0:07:30.24,0:07:33.60,Default,,0000,0000,0000,,And I think\Nthat's all the ones I had, yeah.
Dialogue: 0,0:07:33.60,0:07:36.40,Default,,0000,0000,0000,,So now, I'm going to pass to Cristina.
Dialogue: 0,0:07:37.79,0:07:42.34,Default,,0000,0000,0000,,(applause)
Dialogue: 0,0:07:43.78,0:07:45.47,Default,,0000,0000,0000,,Hi, I'm Cristina.
Dialogue: 0,0:07:45.47,0:07:47.67,Default,,0000,0000,0000,,I'm a research scientist\Nfrom the University of Zürich,
Dialogue: 0,0:07:47.67,0:07:51.42,Default,,0000,0000,0000,,and I'm also an active member\Nof the Swiss Community.
Dialogue: 0,0:07:52.70,0:07:57.90,Default,,0000,0000,0000,,When Claudia Müller-Birn\Nand I submitted this to the WikidataCon,
Dialogue: 0,0:07:57.90,0:08:00.41,Default,,0000,0000,0000,,what we wanted to do\Nis continue our discussion
Dialogue: 0,0:08:00.41,0:08:02.42,Default,,0000,0000,0000,,that we started\Nin the beginning of the year
Dialogue: 0,0:08:02.42,0:08:07.44,Default,,0000,0000,0000,,with a workshop on data quality\Nand also some sessions in Wikimania.
Dialogue: 0,0:08:07.44,0:08:10.54,Default,,0000,0000,0000,,So the goal of this talk\Nis basically to bring some thoughts
Dialogue: 0,0:08:10.54,0:08:14.43,Default,,0000,0000,0000,,that we have been collecting\Nfrom the community and ourselves
Dialogue: 0,0:08:14.43,0:08:16.56,Default,,0000,0000,0000,,and continue discussion.
Dialogue: 0,0:08:16.56,0:08:20.06,Default,,0000,0000,0000,,So what we would like is to continue\Ninteracting a lot with you.
Dialogue: 0,0:08:21.56,0:08:23.37,Default,,0000,0000,0000,,So what we think is very important
Dialogue: 0,0:08:23.37,0:08:27.58,Default,,0000,0000,0000,,is that we continuously ask\Nall types of users in the community
Dialogue: 0,0:08:27.58,0:08:32.24,Default,,0000,0000,0000,,about what they really need,\Nwhat problems they have with data quality,
Dialogue: 0,0:08:32.24,0:08:35.00,Default,,0000,0000,0000,,not only editors\Nbut also the people who are coding,
Dialogue: 0,0:08:35.00,0:08:36.24,Default,,0000,0000,0000,,or consuming the data,
Dialogue: 0,0:08:36.24,0:08:39.49,Default,,0000,0000,0000,,and also researchers who are\Nactually using all the edit history
Dialogue: 0,0:08:39.49,0:08:40.80,Default,,0000,0000,0000,,to analyze what is happening.
Dialogue: 0,0:08:42.37,0:08:48.43,Default,,0000,0000,0000,,So we did a review of around 80 tools\Nthat are existing in Wikidata
Dialogue: 0,0:08:48.43,0:08:52.38,Default,,0000,0000,0000,,and we aligned them to the different\Ndata quality dimensions.
Dialogue: 0,0:08:52.38,0:08:54.36,Default,,0000,0000,0000,,And what we saw was that actually,
Dialogue: 0,0:08:54.36,0:08:57.68,Default,,0000,0000,0000,,many of them were looking at,\Nmonitoring completeness,
Dialogue: 0,0:08:57.68,0:09:02.82,Default,,0000,0000,0000,,but actually... and also some of them\Nare also enabling interlinking.
Dialogue: 0,0:09:02.82,0:09:08.44,Default,,0000,0000,0000,,But there is a big need for tools\Nthat are looking into diversity,
Dialogue: 0,0:09:08.44,0:09:12.82,Default,,0000,0000,0000,,which is one of the things\Nthat we actually can have in Wikidata,
Dialogue: 0,0:09:12.82,0:09:15.96,Default,,0000,0000,0000,,especially\Nthis design principle of Wikidata
Dialogue: 0,0:09:15.96,0:09:17.90,Default,,0000,0000,0000,,where we can have plurality
Dialogue: 0,0:09:17.90,0:09:20.31,Default,,0000,0000,0000,,and different statements\Nwith different values
Dialogue: 0,0:09:21.03,0:09:22.24,Default,,0000,0000,0000,,coming from different sources.
Dialogue: 0,0:09:22.24,0:09:24.92,Default,,0000,0000,0000,,Because it's a secondary source,\Nwe don't have really tools
Dialogue: 0,0:09:24.92,0:09:27.75,Default,,0000,0000,0000,,that actually tell us how many\Nplural statements there are,
Dialogue: 0,0:09:27.75,0:09:30.89,Default,,0000,0000,0000,,and how many we can improve and how,
Dialogue: 0,0:09:30.89,0:09:32.83,Default,,0000,0000,0000,,and we also don't know really
Dialogue: 0,0:09:32.83,0:09:35.54,Default,,0000,0000,0000,,what are all the reasons\Nfor plurality that we can have.
Dialogue: 0,0:09:36.49,0:09:39.20,Default,,0000,0000,0000,,So from these community meetings,
Dialogue: 0,0:09:39.20,0:09:43.08,Default,,0000,0000,0000,,what we discussed was the challenges\Nthat still need attention.
Dialogue: 0,0:09:43.08,0:09:47.25,Default,,0000,0000,0000,,For example, that having\Nall these crowdsourcing communities
Dialogue: 0,0:09:47.25,0:09:49.61,Default,,0000,0000,0000,,is very good because different people\Nattack different parts
Dialogue: 0,0:09:49.61,0:09:51.83,Default,,0000,0000,0000,,of the data or the graph,
Dialogue: 0,0:09:51.83,0:09:54.62,Default,,0000,0000,0000,,and we also have\Ndifferent background knowledge
Dialogue: 0,0:09:54.62,0:09:59.16,Default,,0000,0000,0000,,but actually, it's very difficult to align\Neverything in something homogeneous
Dialogue: 0,0:09:59.16,0:10:04.92,Default,,0000,0000,0000,,because different people are using\Ndifferent properties in different ways
Dialogue: 0,0:10:04.92,0:10:08.40,Default,,0000,0000,0000,,and they are also expecting\Ndifferent things from entity descriptions.
Dialogue: 0,0:10:09.00,0:10:12.72,Default,,0000,0000,0000,,People also said that\Nthey also need more tools
Dialogue: 0,0:10:12.72,0:10:16.00,Default,,0000,0000,0000,,that give a better overview\Nof the global status of things.
Dialogue: 0,0:10:16.00,0:10:20.73,Default,,0000,0000,0000,,So what entities are missing\Nin terms of completeness,
Dialogue: 0,0:10:20.73,0:10:26.12,Default,,0000,0000,0000,,but also like what are people\Nworking on right now most of the time,
Dialogue: 0,0:10:26.12,0:10:30.52,Default,,0000,0000,0000,,and they also mention many times\Na tighter collaboration
Dialogue: 0,0:10:30.52,0:10:33.31,Default,,0000,0000,0000,,across not only languages\Nbut the WikiProjects
Dialogue: 0,0:10:33.31,0:10:35.57,Default,,0000,0000,0000,,and the different Wikimedia platforms.
Dialogue: 0,0:10:35.57,0:10:38.86,Default,,0000,0000,0000,,And we published\Nall the transcribed comments
Dialogue: 0,0:10:38.86,0:10:42.96,Default,,0000,0000,0000,,from all these discussions\Nin those links here in the Etherpads
Dialogue: 0,0:10:42.96,0:10:46.16,Default,,0000,0000,0000,,and also in the wiki page of Wikimania.
Dialogue: 0,0:10:46.16,0:10:48.48,Default,,0000,0000,0000,,Some solutions that appeared actually
Dialogue: 0,0:10:48.48,0:10:53.00,Default,,0000,0000,0000,,were going into the direction\Nof sharing more the best practices
Dialogue: 0,0:10:53.00,0:10:55.76,Default,,0000,0000,0000,,that are being developed\Nin different WikiProjects,
Dialogue: 0,0:10:55.76,0:11:01.24,Default,,0000,0000,0000,,but also people want tools\Nthat help organize work in teams
Dialogue: 0,0:11:01.24,0:11:03.84,Default,,0000,0000,0000,,or at least understanding\Nwho is working on that,
Dialogue: 0,0:11:03.84,0:11:07.82,Default,,0000,0000,0000,,and they were also mentioning\Nthat they want more showcases
Dialogue: 0,0:11:07.82,0:11:12.02,Default,,0000,0000,0000,,and more templates that help them\Ncreate things in a better way.
Dialogue: 0,0:11:12.95,0:11:15.16,Default,,0000,0000,0000,,And from the contact that we have
Dialogue: 0,0:11:15.16,0:11:18.72,Default,,0000,0000,0000,,with Open Governmental Data Organizations,
Dialogue: 0,0:11:18.72,0:11:20.07,Default,,0000,0000,0000,,and in particularly,
Dialogue: 0,0:11:20.07,0:11:23.10,Default,,0000,0000,0000,,I am in contact with the canton\Nand the city of Zürich,
Dialogue: 0,0:11:23.10,0:11:26.21,Default,,0000,0000,0000,,they are very interested\Nin working with Wikidata
Dialogue: 0,0:11:26.21,0:11:29.90,Default,,0000,0000,0000,,because they want their data\Nto be accessible for everyone
Dialogue: 0,0:11:29.90,0:11:33.68,Default,,0000,0000,0000,,in the place where people go\Nand consult or access data.
Dialogue: 0,0:11:33.68,0:11:36.55,Default,,0000,0000,0000,,So for them, something that\Nwould be really interesting
Dialogue: 0,0:11:36.55,0:11:38.60,Default,,0000,0000,0000,,is to have some kind of quality indicators
Dialogue: 0,0:11:38.60,0:11:41.08,Default,,0000,0000,0000,,both in the wiki,\Nwhich is already happening,
Dialogue: 0,0:11:41.08,0:11:42.80,Default,,0000,0000,0000,,but also in SPARQL results,
Dialogue: 0,0:11:42.80,0:11:46.07,Default,,0000,0000,0000,,to know whether they can trust\Nor not that data from the community.
Dialogue: 0,0:11:46.07,0:11:48.23,Default,,0000,0000,0000,,And then, they also want to know
Dialogue: 0,0:11:48.23,0:11:51.42,Default,,0000,0000,0000,,what parts of their own data sets\Nare useful for Wikidata
Dialogue: 0,0:11:51.42,0:11:56.04,Default,,0000,0000,0000,,and they would love to have a tool that\Ncan help them assess that automatically.
Dialogue: 0,0:11:56.04,0:11:59.07,Default,,0000,0000,0000,,They also need\Nsome kind of methodology or tool
Dialogue: 0,0:11:59.07,0:12:03.89,Default,,0000,0000,0000,,that helps them decide whether\Nthey should import or link their data
Dialogue: 0,0:12:03.89,0:12:04.89,Default,,0000,0000,0000,,because in some cases,
Dialogue: 0,0:12:04.90,0:12:07.14,Default,,0000,0000,0000,,they also have their own\Nlinked open data sets,
Dialogue: 0,0:12:07.14,0:12:09.75,Default,,0000,0000,0000,,so they don't know whether\Nto just ingest the data
Dialogue: 0,0:12:09.75,0:12:13.42,Default,,0000,0000,0000,,or to keep on creating links\Nfrom the data sets to Wikidata
Dialogue: 0,0:12:13.42,0:12:14.42,Default,,0000,0000,0000,,and the other way around.
Dialogue: 0,0:12:14.95,0:12:20.04,Default,,0000,0000,0000,,And they also want to know where\Ntheir websites are referred in Wikidata.
Dialogue: 0,0:12:20.04,0:12:23.36,Default,,0000,0000,0000,,And when they run such a query\Nin the query service,
Dialogue: 0,0:12:23.36,0:12:24.85,Default,,0000,0000,0000,,they often get timeouts,
Dialogue: 0,0:12:24.85,0:12:28.18,Default,,0000,0000,0000,,so maybe we should\Nreally create more tools
Dialogue: 0,0:12:28.18,0:12:32.24,Default,,0000,0000,0000,,that help them get these answers\Nfor their questions.
Dialogue: 0,0:12:33.15,0:12:36.21,Default,,0000,0000,0000,,And, besides that,
Dialogue: 0,0:12:36.21,0:12:39.36,Default,,0000,0000,0000,,we wiki researchers also sometimes
Dialogue: 0,0:12:39.36,0:12:42.02,Default,,0000,0000,0000,,lack some information\Nin the edit summaries.
Dialogue: 0,0:12:42.02,0:12:44.95,Default,,0000,0000,0000,,So I remember that when\Nwe were doing some work
Dialogue: 0,0:12:44.95,0:12:48.92,Default,,0000,0000,0000,,to understand\Nthe different behavior of editors
Dialogue: 0,0:12:48.92,0:12:53.40,Default,,0000,0000,0000,,with tools or bots\Nor anonymous users and so on,
Dialogue: 0,0:12:53.40,0:12:56.15,Default,,0000,0000,0000,,we were really lacking, for example,
Dialogue: 0,0:12:56.15,0:13:01.11,Default,,0000,0000,0000,,a standard way of tracing\Nthat tools were being used.
Dialogue: 0,0:13:01.11,0:13:03.15,Default,,0000,0000,0000,,And there are some tools\Nthat are already doing that
Dialogue: 0,0:13:03.16,0:13:05.23,Default,,0000,0000,0000,,like PetScan and many others,
Dialogue: 0,0:13:05.23,0:13:07.72,Default,,0000,0000,0000,,but maybe we should in the community
Dialogue: 0,0:13:07.72,0:13:13.53,Default,,0000,0000,0000,,discuss more about how to record these\Nfor fine-grained provenance.
Dialogue: 0,0:13:14.17,0:13:15.32,Default,,0000,0000,0000,,And further on,
Dialogue: 0,0:13:15.32,0:13:20.80,Default,,0000,0000,0000,,we think that we need to think\Nof more concrete data quality dimensions
Dialogue: 0,0:13:20.80,0:13:24.96,Default,,0000,0000,0000,,that are related to link data\Nbut not all the types of data,
Dialogue: 0,0:13:24.96,0:13:30.72,Default,,0000,0000,0000,,so we worked on some measures\Nto access actually the information gain
Dialogue: 0,0:13:30.72,0:13:33.88,Default,,0000,0000,0000,,enabled by the links,\Nand what we mean by that
Dialogue: 0,0:13:33.88,0:13:36.68,Default,,0000,0000,0000,,is that when we link\NWikidata to other data sets,
Dialogue: 0,0:13:36.68,0:13:38.20,Default,,0000,0000,0000,,we should also be thinking
Dialogue: 0,0:13:38.20,0:13:41.92,Default,,0000,0000,0000,,how much the entities are actually\Ngaining in the classification,
Dialogue: 0,0:13:41.92,0:13:45.60,Default,,0000,0000,0000,,also in the description\Nbut also in the vocabularies they use.
Dialogue: 0,0:13:45.60,0:13:51.04,Default,,0000,0000,0000,,So just to give a very simple\Nexample of what I mean with this
Dialogue: 0,0:13:51.04,0:13:54.27,Default,,0000,0000,0000,,is we can think of--\Nin this case, would be Wikidata
Dialogue: 0,0:13:54.27,0:13:57.77,Default,,0000,0000,0000,,or the external data center\Nthat is linking to Wikidata,
Dialogue: 0,0:13:57.77,0:14:00.49,Default,,0000,0000,0000,,we have the entity for a person\Nthat is called Natasha Noy,
Dialogue: 0,0:14:00.49,0:14:02.60,Default,,0000,0000,0000,,we have the affiliation and other things,
Dialogue: 0,0:14:02.60,0:14:05.24,Default,,0000,0000,0000,,and then we say OK,\Nwe link to an external place,
Dialogue: 0,0:14:05.24,0:14:08.92,Default,,0000,0000,0000,,and that entity also has that name,\Nbut we actually have the same value.
Dialogue: 0,0:14:08.92,0:14:12.89,Default,,0000,0000,0000,,So what it would be better is that we link\Nto something that has a different name,
Dialogue: 0,0:14:12.89,0:14:16.88,Default,,0000,0000,0000,,that is still valid because this person\Nhas two ways of writing the name,
Dialogue: 0,0:14:16.88,0:14:19.71,Default,,0000,0000,0000,,and also other information\Nthat we don't have in Wikidata
Dialogue: 0,0:14:19.72,0:14:21.76,Default,,0000,0000,0000,,or that we don't have\Nin the other data set.
Dialogue: 0,0:14:22.39,0:14:24.65,Default,,0000,0000,0000,,But also, what is even better
Dialogue: 0,0:14:24.65,0:14:27.77,Default,,0000,0000,0000,,is that we are actually\Nlooking in the target data set
Dialogue: 0,0:14:27.77,0:14:31.39,Default,,0000,0000,0000,,that they also have new ways\Nof classifying the information.
Dialogue: 0,0:14:31.39,0:14:35.35,Default,,0000,0000,0000,,So not only is this a person,\Nbut in the other data set,
Dialogue: 0,0:14:35.36,0:14:39.52,Default,,0000,0000,0000,,they also say it's a female\Nor anything else that they classify with.
Dialogue: 0,0:14:39.53,0:14:43.40,Default,,0000,0000,0000,,And if in the other data set,\Nthey are using many other vocabularies
Dialogue: 0,0:14:43.40,0:14:46.59,Default,,0000,0000,0000,,that is also helping in their whole\Ninformation retrieval thing.
Dialogue: 0,0:14:47.37,0:14:51.23,Default,,0000,0000,0000,,So with that, I also would like to say
Dialogue: 0,0:14:51.23,0:14:55.81,Default,,0000,0000,0000,,that we think that we can\Nshowcase federated queries better
Dialogue: 0,0:14:55.81,0:15:00.45,Default,,0000,0000,0000,,because when we look at the query log\Nprovided by Malyshev et al.,
Dialogue: 0,0:15:01.28,0:15:04.30,Default,,0000,0000,0000,,we see actually that\Nfrom the organic queries,
Dialogue: 0,0:15:04.30,0:15:06.92,Default,,0000,0000,0000,,we have only very few federated queries.
Dialogue: 0,0:15:06.92,0:15:12.80,Default,,0000,0000,0000,,And actually, federation is one\Nof the key advantages of having link data,
Dialogue: 0,0:15:12.80,0:15:16.90,Default,,0000,0000,0000,,so maybe the community\Nor the people using Wikidata
Dialogue: 0,0:15:16.90,0:15:18.90,Default,,0000,0000,0000,,also need more examples on this.
Dialogue: 0,0:15:18.90,0:15:22.67,Default,,0000,0000,0000,,And if we look at the list\Nof endpoints that are being used,
Dialogue: 0,0:15:22.67,0:15:25.40,Default,,0000,0000,0000,,this is not a complete list\Nand we have many more.
Dialogue: 0,0:15:25.40,0:15:30.48,Default,,0000,0000,0000,,Of course, this data was analyzed\Nfrom queries until March 2018,
Dialogue: 0,0:15:30.48,0:15:34.81,Default,,0000,0000,0000,,but we should look into the list\Nof federated endpoints that we have
Dialogue: 0,0:15:34.81,0:15:37.05,Default,,0000,0000,0000,,and see whether\Nwe are really using them or not.
Dialogue: 0,0:15:37.81,0:15:40.44,Default,,0000,0000,0000,,So two questions that\NI have for the audience
Dialogue: 0,0:15:40.44,0:15:43.00,Default,,0000,0000,0000,,that maybe we can use\Nafterwards for the discussion are:
Dialogue: 0,0:15:43.00,0:15:46.00,Default,,0000,0000,0000,,what data quality problems\Nshould be addressed in your opinion,
Dialogue: 0,0:15:46.00,0:15:47.41,Default,,0000,0000,0000,,because of the needs that you have,
Dialogue: 0,0:15:47.41,0:15:50.40,Default,,0000,0000,0000,,but also, where do you need\Nmore automation
Dialogue: 0,0:15:50.40,0:15:52.94,Default,,0000,0000,0000,,to help you with editing or patrolling.
Dialogue: 0,0:15:53.87,0:15:55.15,Default,,0000,0000,0000,,That's all, thank you very much.
Dialogue: 0,0:15:55.78,0:15:57.53,Default,,0000,0000,0000,,(applause)
Dialogue: 0,0:16:06.03,0:16:08.60,Default,,0000,0000,0000,,(Jose Emilio Labra) OK,\Nso what I'm going to talk about
Dialogue: 0,0:16:08.60,0:16:14.72,Default,,0000,0000,0000,,is some tools that we were developing\Nrelated with Shape Expressions.
Dialogue: 0,0:16:15.54,0:16:19.37,Default,,0000,0000,0000,,So this is what I want to talk...\NI am Jose Emilio Labra,
Dialogue: 0,0:16:19.37,0:16:23.22,Default,,0000,0000,0000,,but this has... all these tools\Nhave been done by different people,
Dialogue: 0,0:16:23.92,0:16:28.48,Default,,0000,0000,0000,,mainly related with W3C ShEx,\NShape Expressions Community Group.
Dialogue: 0,0:16:28.48,0:16:29.48,Default,,0000,0000,0000,,ShEx Community Group.
Dialogue: 0,0:16:30.14,0:16:36.08,Default,,0000,0000,0000,,So the first tool that I want to mention\Nis RDFShape, this is a general tool,
Dialogue: 0,0:16:36.08,0:16:40.68,Default,,0000,0000,0000,,because Shape Expressions\Nis not only for Wikidata,
Dialogue: 0,0:16:40.68,0:16:44.17,Default,,0000,0000,0000,,Shape Expressions is a language\Nto validate RDF in general.
Dialogue: 0,0:16:44.17,0:16:47.57,Default,,0000,0000,0000,,So this tool was developed mainly by me
Dialogue: 0,0:16:47.57,0:16:50.88,Default,,0000,0000,0000,,and it's a tool\Nto validate RDF in general.
Dialogue: 0,0:16:50.88,0:16:55.14,Default,,0000,0000,0000,,So if you want to learn about RDF\Nor you want to validate RDF
Dialogue: 0,0:16:55.14,0:16:58.62,Default,,0000,0000,0000,,or SPARQL endpoints not only in Wikidata,
Dialogue: 0,0:16:58.62,0:17:00.89,Default,,0000,0000,0000,,my advice is that you can use this tool.
Dialogue: 0,0:17:00.89,0:17:03.26,Default,,0000,0000,0000,,Also for teaching.
Dialogue: 0,0:17:03.26,0:17:05.64,Default,,0000,0000,0000,,I am a teacher in the university
Dialogue: 0,0:17:05.64,0:17:09.15,Default,,0000,0000,0000,,and I use it in my semantic web course\Nto teach RDF.
Dialogue: 0,0:17:09.16,0:17:12.12,Default,,0000,0000,0000,,So if you want to learn RDF,\NI think it's a good tool.
Dialogue: 0,0:17:13.03,0:17:17.60,Default,,0000,0000,0000,,For example, this is just a visualization\Nof an RDF graph with the tool.
Dialogue: 0,0:17:18.59,0:17:22.64,Default,,0000,0000,0000,,But before coming here, in the last month,
Dialogue: 0,0:17:22.64,0:17:28.44,Default,,0000,0000,0000,,I started a fork of rdfshape specifically\Nfor Wikidata, because I thought...
Dialogue: 0,0:17:28.44,0:17:33.08,Default,,0000,0000,0000,,It's called WikiShape, and yesterday,\NI presented it as a present for Wikidata.
Dialogue: 0,0:17:33.08,0:17:34.44,Default,,0000,0000,0000,,So what I took is...
Dialogue: 0,0:17:34.44,0:17:39.90,Default,,0000,0000,0000,,What I did is to remove all the stuff\Nthat was not related with Wikidata
Dialogue: 0,0:17:39.90,0:17:44.80,Default,,0000,0000,0000,,and to put several things, hard-coded,\Nfor example, the Wikidata SPARQL endpoint,
Dialogue: 0,0:17:44.80,0:17:49.04,Default,,0000,0000,0000,,but now, someone asked me\Nif I could do it also for Wikibase.
Dialogue: 0,0:17:49.04,0:17:52.00,Default,,0000,0000,0000,,And it is very easy\Nto do it for Wikibase also.
Dialogue: 0,0:17:52.76,0:17:56.28,Default,,0000,0000,0000,,So this tool, WikiShape, is quite new.
Dialogue: 0,0:17:57.02,0:17:59.84,Default,,0000,0000,0000,,I think it works, most of the features,
Dialogue: 0,0:17:59.84,0:18:02.47,Default,,0000,0000,0000,,but there are some features\Nthat maybe don't work,
Dialogue: 0,0:18:02.47,0:18:06.28,Default,,0000,0000,0000,,and if you try it and you want\Nto improve it, please tell me.
Dialogue: 0,0:18:06.28,0:18:12.68,Default,,0000,0000,0000,,So this is [inaudible] captures,\Nbut I think I can even try so let's try.
Dialogue: 0,0:18:15.38,0:18:16.94,Default,,0000,0000,0000,,So let's see if it works.
Dialogue: 0,0:18:16.95,0:18:20.07,Default,,0000,0000,0000,,First, I have to go out of the...
Dialogue: 0,0:18:22.45,0:18:23.45,Default,,0000,0000,0000,,Here.
Dialogue: 0,0:18:24.23,0:18:28.32,Default,,0000,0000,0000,,Alright, yeah. So this is the tool here.
Dialogue: 0,0:18:28.32,0:18:29.84,Default,,0000,0000,0000,,Things that you can do with the tool,
Dialogue: 0,0:18:29.84,0:18:35.28,Default,,0000,0000,0000,,for example, is that you can\Ncheck schemas, entity schemas.
Dialogue: 0,0:18:35.28,0:18:38.61,Default,,0000,0000,0000,,You know that there is\Na new namespace which is "E whatever,"
Dialogue: 0,0:18:38.61,0:18:44.80,Default,,0000,0000,0000,,so here, if you start for example,\Nwrite for example "human"...
Dialogue: 0,0:18:44.81,0:18:48.81,Default,,0000,0000,0000,,As you are writing,\Nits autocomplete allows you to check,
Dialogue: 0,0:18:48.81,0:18:52.00,Default,,0000,0000,0000,,for example,\Nthis is the Shape Expressions of a human,
Dialogue: 0,0:18:52.79,0:18:55.94,Default,,0000,0000,0000,,and this is the Shape Expressions here.
Dialogue: 0,0:18:55.94,0:18:59.84,Default,,0000,0000,0000,,And as you can see,\Nthis editor has syntax highlighting,
Dialogue: 0,0:18:59.84,0:19:04.56,Default,,0000,0000,0000,,this is... well,\Nmaybe it's very small, the screen.
Dialogue: 0,0:19:05.68,0:19:07.59,Default,,0000,0000,0000,,I can try to do it bigger.
Dialogue: 0,0:19:09.19,0:19:10.97,Default,,0000,0000,0000,,Maybe you see it better now.
Dialogue: 0,0:19:10.97,0:19:14.24,Default,,0000,0000,0000,,So... and this is the editor\Nwith syntax highlighting and also has...
Dialogue: 0,0:19:14.24,0:19:17.85,Default,,0000,0000,0000,,I mean, this editor\Ncomes from the same source code
Dialogue: 0,0:19:17.85,0:19:19.64,Default,,0000,0000,0000,,as the Wikidata query service.
Dialogue: 0,0:19:19.64,0:19:23.96,Default,,0000,0000,0000,,So for example,\Nif you hover with the mouse here,
Dialogue: 0,0:19:23.96,0:19:27.96,Default,,0000,0000,0000,,it shows you the labels\Nof the different properties.
Dialogue: 0,0:19:27.96,0:19:31.30,Default,,0000,0000,0000,,So I think it's very helpful because now,
Dialogue: 0,0:19:32.59,0:19:38.60,Default,,0000,0000,0000,,the entity schemas that is\Nin the Wikidata is just a plain text idea,
Dialogue: 0,0:19:38.60,0:19:42.49,Default,,0000,0000,0000,,and I think this editor is much better\Nbecause it has autocomplete
Dialogue: 0,0:19:42.49,0:19:43.74,Default,,0000,0000,0000,,and it also has...
Dialogue: 0,0:19:43.74,0:19:48.24,Default,,0000,0000,0000,,I mean, if you, for example,\Nwanted to add a constraint,
Dialogue: 0,0:19:48.24,0:19:51.57,Default,,0000,0000,0000,,you say "wdt:"
Dialogue: 0,0:19:51.57,0:19:56.88,Default,,0000,0000,0000,,You start writing "author"\Nand then you click {\i1}Ctrl+Space{\i0}
Dialogue: 0,0:19:56.88,0:19:58.92,Default,,0000,0000,0000,,and it suggests the different things.
Dialogue: 0,0:19:58.92,0:20:02.39,Default,,0000,0000,0000,,So this is similar\Nto the Wikidata query service
Dialogue: 0,0:20:02.39,0:20:06.44,Default,,0000,0000,0000,,but specifically for Shape Expressions
Dialogue: 0,0:20:06.44,0:20:11.98,Default,,0000,0000,0000,,because my feeling is that\Ncreating Shape Expressions
Dialogue: 0,0:20:11.98,0:20:15.84,Default,,0000,0000,0000,,is not more difficult\Nthan writing SPARQL queries.
Dialogue: 0,0:20:15.84,0:20:21.26,Default,,0000,0000,0000,,So some people think\Nthat it's at the same level,
Dialogue: 0,0:20:22.28,0:20:26.30,Default,,0000,0000,0000,,It's probably easier, I think,\Nbecause Shape Expressions was,
Dialogue: 0,0:20:26.30,0:20:31.24,Default,,0000,0000,0000,,when we designed it,\Nwe were doing it to be easier to work.
Dialogue: 0,0:20:31.24,0:20:35.00,Default,,0000,0000,0000,,OK, so this is one of the first things,\Nthat you have this editor
Dialogue: 0,0:20:35.00,0:20:36.62,Default,,0000,0000,0000,,for Shape Expressions.
Dialogue: 0,0:20:37.37,0:20:41.47,Default,,0000,0000,0000,,And then you also have the possibility,\Nfor example, to visualize.
Dialogue: 0,0:20:41.47,0:20:44.80,Default,,0000,0000,0000,,If you have a Shape Expression,\Nuse for example...
Dialogue: 0,0:20:44.80,0:20:49.39,Default,,0000,0000,0000,,I think, "written work" is\Na nice Shape Expression
Dialogue: 0,0:20:49.39,0:20:53.30,Default,,0000,0000,0000,,because it has some relationships\Nbetween different things.
Dialogue: 0,0:20:54.82,0:20:58.16,Default,,0000,0000,0000,,And this is the UML visualization\Nof written work.
Dialogue: 0,0:20:58.16,0:21:02.09,Default,,0000,0000,0000,,In a UML, this is easy to see\Nthe different properties.
Dialogue: 0,0:21:02.79,0:21:06.79,Default,,0000,0000,0000,,When you do this, I realized\Nwhen I tried with several people,
Dialogue: 0,0:21:06.80,0:21:09.22,Default,,0000,0000,0000,,they find some mistakes\Nin their Shape Expressions
Dialogue: 0,0:21:09.22,0:21:12.99,Default,,0000,0000,0000,,because it's easy to detect which are\Nthe missing properties or whatever.
Dialogue: 0,0:21:13.59,0:21:15.77,Default,,0000,0000,0000,,Then there is another possibility here
Dialogue: 0,0:21:15.77,0:21:19.52,Default,,0000,0000,0000,,is that you can also validate,\NI think I have it here, the validation.
Dialogue: 0,0:21:20.50,0:21:25.28,Default,,0000,0000,0000,,I think I had it in some label,\Nmaybe I closed it.
Dialogue: 0,0:21:26.27,0:21:30.99,Default,,0000,0000,0000,,OK, but you can, for example,\Nyou can click here, {\i1}Validate entities.{\i0}
Dialogue: 0,0:21:32.31,0:21:34.23,Default,,0000,0000,0000,,You, for example,
Dialogue: 0,0:21:35.40,0:21:41.92,Default,,0000,0000,0000,,"q42" with "e42" which is author.
Dialogue: 0,0:21:42.82,0:21:46.18,Default,,0000,0000,0000,,With "human,"\NI think we can do it with "human."
Dialogue: 0,0:21:49.05,0:21:50.05,Default,,0000,0000,0000,,And then it's...
Dialogue: 0,0:21:50.69,0:21:56.36,Default,,0000,0000,0000,,And it's taking a little while to do it\Nbecause this is doing the SPARQL queries
Dialogue: 0,0:21:56.36,0:21:59.13,Default,,0000,0000,0000,,and now, for example,\Nit's failing by the network but...
Dialogue: 0,0:21:59.66,0:22:01.58,Default,,0000,0000,0000,,So you can try it.
Dialogue: 0,0:22:02.76,0:22:07.03,Default,,0000,0000,0000,,OK, so let's go continue\Nwith the presentation, with other tools.
Dialogue: 0,0:22:07.03,0:22:12.35,Default,,0000,0000,0000,,So my advice is that if you want to try it\Nand you want any feedback let me know.
Dialogue: 0,0:22:13.13,0:22:15.54,Default,,0000,0000,0000,,So to continue with the presentation...
Dialogue: 0,0:22:18.92,0:22:20.23,Default,,0000,0000,0000,,So this is WikiShape.
Dialogue: 0,0:22:23.80,0:22:26.51,Default,,0000,0000,0000,,Then, I already said this,
Dialogue: 0,0:22:27.68,0:22:34.16,Default,,0000,0000,0000,,the Shape Expressions Editor\Nis an independent project in GitHub.
Dialogue: 0,0:22:35.60,0:22:37.47,Default,,0000,0000,0000,,You can use it in your own project.
Dialogue: 0,0:22:37.47,0:22:41.04,Default,,0000,0000,0000,,If you want to do\Na Shape Expressions tool,
Dialogue: 0,0:22:41.04,0:22:45.64,Default,,0000,0000,0000,,you can just embed it\Nin any other project,
Dialogue: 0,0:22:45.64,0:22:48.24,Default,,0000,0000,0000,,so this is in GitHub and you can use it.
Dialogue: 0,0:22:48.87,0:22:51.97,Default,,0000,0000,0000,,Then the same author,\Nit's one of my students,
Dialogue: 0,0:22:52.68,0:22:55.70,Default,,0000,0000,0000,,he also created\Nan editor for Shape Expressions,
Dialogue: 0,0:22:55.70,0:22:57.80,Default,,0000,0000,0000,,also inspired by\Nthe Wikidata query service
Dialogue: 0,0:22:57.80,0:23:00.68,Default,,0000,0000,0000,,where, in a column,
Dialogue: 0,0:23:00.68,0:23:05.10,Default,,0000,0000,0000,,you have this more visual editor\Nof SPARQL queries
Dialogue: 0,0:23:05.10,0:23:07.14,Default,,0000,0000,0000,,where you can put this kind of things.
Dialogue: 0,0:23:07.14,0:23:09.12,Default,,0000,0000,0000,,So this is a screen capture.
Dialogue: 0,0:23:09.12,0:23:12.66,Default,,0000,0000,0000,,You can see that\Nthat's the Shape Expressions in text
Dialogue: 0,0:23:12.66,0:23:17.82,Default,,0000,0000,0000,,but this is a form-based Shape Expressions\Nwhere it would probably take a bit longer
Dialogue: 0,0:23:18.60,0:23:23.40,Default,,0000,0000,0000,,where you can put the different rows\Non the different fields.
Dialogue: 0,0:23:23.40,0:23:25.80,Default,,0000,0000,0000,,OK, then there is ShExEr.
Dialogue: 0,0:23:26.88,0:23:31.88,Default,,0000,0000,0000,,We have... it's done by one PhD student\Nat the University of Oviedo
Dialogue: 0,0:23:31.88,0:23:34.08,Default,,0000,0000,0000,,and he's here, so you can present ShExEr.
Dialogue: 0,0:23:38.15,0:23:40.02,Default,,0000,0000,0000,,(Danny) Hello, I am Danny Fernández,
Dialogue: 0,0:23:40.02,0:23:43.80,Default,,0000,0000,0000,,I am a PhD student in University of Oviedo\Nworking with Labra.
Dialogue: 0,0:23:44.71,0:23:47.72,Default,,0000,0000,0000,,Since we are running out of time,\Nlet's make these quickly,
Dialogue: 0,0:23:47.73,0:23:52.64,Default,,0000,0000,0000,,so let's not go for any actual demo,\Nbut just print some screenshots.
Dialogue: 0,0:23:52.64,0:23:57.90,Default,,0000,0000,0000,,OK, so the usual way to work with\NShape Expressions or any shape language
Dialogue: 0,0:23:57.90,0:23:59.52,Default,,0000,0000,0000,,is that you have a domain expert
Dialogue: 0,0:23:59.52,0:24:02.31,Default,,0000,0000,0000,,that defines a priori\Nhow the graph should look like
Dialogue: 0,0:24:02.31,0:24:03.56,Default,,0000,0000,0000,,define some structures,
Dialogue: 0,0:24:03.56,0:24:06.98,Default,,0000,0000,0000,,and then you use these structures\Nto validate the actual data against it.
Dialogue: 0,0:24:08.12,0:24:11.64,Default,,0000,0000,0000,,This tool, which is as well as the ones\Nthat Labra has been presenting,
Dialogue: 0,0:24:11.64,0:24:14.44,Default,,0000,0000,0000,,this is a general purpose tool\Nfor any RDF source,
Dialogue: 0,0:24:14.44,0:24:17.38,Default,,0000,0000,0000,,is designed to do the other way around.
Dialogue: 0,0:24:17.38,0:24:18.76,Default,,0000,0000,0000,,You already have some data,
Dialogue: 0,0:24:18.76,0:24:23.16,Default,,0000,0000,0000,,you select what nodes\Nyou want to get the shape about
Dialogue: 0,0:24:23.16,0:24:26.72,Default,,0000,0000,0000,,and then you automatically\Nextract or infer the shape.
Dialogue: 0,0:24:26.72,0:24:29.79,Default,,0000,0000,0000,,So even if this is a general purpose tool,
Dialogue: 0,0:24:29.79,0:24:34.06,Default,,0000,0000,0000,,what we did for this WikidataCon\Nis these fancy button
Dialogue: 0,0:24:34.88,0:24:37.08,Default,,0000,0000,0000,,that if you click it,\Nessentially what happens
Dialogue: 0,0:24:37.08,0:24:42.08,Default,,0000,0000,0000,,is that there are\Nso many configurations params
Dialogue: 0,0:24:42.08,0:24:46.25,Default,,0000,0000,0000,,and it configures it to work\Nagainst the Wikidata endpoint
Dialogue: 0,0:24:46.25,0:24:47.97,Default,,0000,0000,0000,,and it will end soon, sorry.
Dialogue: 0,0:24:48.73,0:24:52.88,Default,,0000,0000,0000,,So, once you press this button\Nwhat you get is essentially this.
Dialogue: 0,0:24:52.88,0:24:55.13,Default,,0000,0000,0000,,After having selected what kind of notes,
Dialogue: 0,0:24:55.13,0:24:59.36,Default,,0000,0000,0000,,what kind of instances of our class,\Nwhatever you are looking for,
Dialogue: 0,0:24:59.36,0:25:01.32,Default,,0000,0000,0000,,you get an automatic schema.
Dialogue: 0,0:25:02.32,0:25:07.11,Default,,0000,0000,0000,,All the constraints are sorted\Nby how many modes actually conform to it,
Dialogue: 0,0:25:07.11,0:25:09.77,Default,,0000,0000,0000,,you can filter the less common ones, etc.
Dialogue: 0,0:25:09.77,0:25:12.13,Default,,0000,0000,0000,,So there is a poster downstairs\Nabout this stuff
Dialogue: 0,0:25:12.13,0:25:14.60,Default,,0000,0000,0000,,and well,\NI will be downstairs and upstairs
Dialogue: 0,0:25:14.60,0:25:16.45,Default,,0000,0000,0000,,and all over the place all day,
Dialogue: 0,0:25:16.46,0:25:19.08,Default,,0000,0000,0000,,so if you have any further\Ninterest in this tool,
Dialogue: 0,0:25:19.08,0:25:21.48,Default,,0000,0000,0000,,just speak to me during this journey.
Dialogue: 0,0:25:21.48,0:25:24.62,Default,,0000,0000,0000,,And now, I'll give back\Nthe micro to Labra, thank you.
Dialogue: 0,0:25:24.62,0:25:29.26,Default,,0000,0000,0000,,(applause)
Dialogue: 0,0:25:29.81,0:25:32.58,Default,,0000,0000,0000,,(Jose) So let's continue\Nwith the other tools.
Dialogue: 0,0:25:32.58,0:25:34.98,Default,,0000,0000,0000,,The other tool is the ShapeDesigner.
Dialogue: 0,0:25:34.98,0:25:37.24,Default,,0000,0000,0000,,Andra, do you want to do\Nthe ShapeDesigner now
Dialogue: 0,0:25:37.24,0:25:39.29,Default,,0000,0000,0000,,or maybe later or in the workshop?
Dialogue: 0,0:25:39.29,0:25:40.60,Default,,0000,0000,0000,,There is a workshop...
Dialogue: 0,0:25:40.60,0:25:44.44,Default,,0000,0000,0000,,This afternoon, there is a workshop\Nspecifically for Shape Expressions, and...
Dialogue: 0,0:25:45.26,0:25:47.94,Default,,0000,0000,0000,,The idea is that was going to be\Nmore hands on,
Dialogue: 0,0:25:47.94,0:25:52.32,Default,,0000,0000,0000,,and if you want to practice\Nsome ShEx, you can do it there.
Dialogue: 0,0:25:52.88,0:25:55.72,Default,,0000,0000,0000,,This tool is ShEx...\Nand there is Eric here,
Dialogue: 0,0:25:55.72,0:25:56.89,Default,,0000,0000,0000,,so you can present it.
Dialogue: 0,0:25:57.97,0:26:00.69,Default,,0000,0000,0000,,(Eric) So just super quick,\Nthe thing that I want to say
Dialogue: 0,0:26:00.69,0:26:05.71,Default,,0000,0000,0000,,is that you've probably\Nalready seen the ShEx interface
Dialogue: 0,0:26:05.71,0:26:07.60,Default,,0000,0000,0000,,that's tailored for Wikidata.
Dialogue: 0,0:26:07.60,0:26:12.93,Default,,0000,0000,0000,,That's effectively stripped down\Nand tailored specifically for Wikidata
Dialogue: 0,0:26:12.93,0:26:17.94,Default,,0000,0000,0000,,because the generic one has more features\Nbut it turns out I thought I'd mention it
Dialogue: 0,0:26:17.94,0:26:19.98,Default,,0000,0000,0000,,because one of those features\Nis particularly useful
Dialogue: 0,0:26:19.98,0:26:23.20,Default,,0000,0000,0000,,for debugging Wikidata schemas,
Dialogue: 0,0:26:23.20,0:26:29.22,Default,,0000,0000,0000,,which is if you go\Nand you select the slurp mode,
Dialogue: 0,0:26:29.22,0:26:31.44,Default,,0000,0000,0000,,what it does is it says\Nwhile I'm validating,
Dialogue: 0,0:26:31.44,0:26:34.69,Default,,0000,0000,0000,,I want to pull all the the triples down\Nand that means
Dialogue: 0,0:26:34.70,0:26:36.27,Default,,0000,0000,0000,,if I get a bunch of failures,
Dialogue: 0,0:26:36.28,0:26:39.59,Default,,0000,0000,0000,,I can go through and start looking\Nat those failures and saying,
Dialogue: 0,0:26:39.59,0:26:41.80,Default,,0000,0000,0000,,OK, what are the triples\Nthat are in here,
Dialogue: 0,0:26:41.80,0:26:44.12,Default,,0000,0000,0000,,sorry, I apologize,\Nthe triples are down there,
Dialogue: 0,0:26:44.12,0:26:45.65,Default,,0000,0000,0000,,this is just a log of what went by.
Dialogue: 0,0:26:46.33,0:26:49.18,Default,,0000,0000,0000,,And then you can just sit there\Nand fiddle with it in real time
Dialogue: 0,0:26:49.18,0:26:51.03,Default,,0000,0000,0000,,like you play with something\Nand it changes.
Dialogue: 0,0:26:51.03,0:26:54.16,Default,,0000,0000,0000,,So it's a quicker version\Nfor doing all that stuff.
Dialogue: 0,0:26:55.36,0:26:56.48,Default,,0000,0000,0000,,This is a ShExC form,
Dialogue: 0,0:26:56.48,0:26:59.46,Default,,0000,0000,0000,,this is something [Joachim] had suggested
Dialogue: 0,0:27:00.04,0:27:04.63,Default,,0000,0000,0000,,could be useful for populating\NWikidata documents
Dialogue: 0,0:27:04.63,0:27:07.34,Default,,0000,0000,0000,,based on a Shape Expression\Nfor that that document.
Dialogue: 0,0:27:08.10,0:27:11.68,Default,,0000,0000,0000,,This is not tailored for Wikidata,
Dialogue: 0,0:27:11.68,0:27:14.08,Default,,0000,0000,0000,,but this is just to say\Nthat you can have a schema
Dialogue: 0,0:27:14.08,0:27:15.40,Default,,0000,0000,0000,,and you can have some annotations
Dialogue: 0,0:27:15.40,0:27:17.52,Default,,0000,0000,0000,,to say specifically how I want\Nthat schema rendered
Dialogue: 0,0:27:17.52,0:27:19.03,Default,,0000,0000,0000,,and then it just builds a form,
Dialogue: 0,0:27:19.03,0:27:21.19,Default,,0000,0000,0000,,and if you've got data,\Nit can even populate the form.
Dialogue: 0,0:27:24.52,0:27:26.16,Default,,0000,0000,0000,,PyShEx [inaudible].
Dialogue: 0,0:27:28.02,0:27:31.08,Default,,0000,0000,0000,,(Jose) I think this is the last one.
Dialogue: 0,0:27:31.82,0:27:34.08,Default,,0000,0000,0000,,Yes, so the last one is PyShEx.
Dialogue: 0,0:27:34.68,0:27:38.15,Default,,0000,0000,0000,,PyShEx is a Python implementation\Nof Shape Expressions,
Dialogue: 0,0:27:39.19,0:27:42.68,Default,,0000,0000,0000,,you can play also with Jupyter Notebooks\Nif you want those kind of things.
Dialogue: 0,0:27:42.68,0:27:44.43,Default,,0000,0000,0000,,OK, so that's all for this.
Dialogue: 0,0:27:44.43,0:27:47.17,Default,,0000,0000,0000,,(applause)
Dialogue: 0,0:27:52.92,0:27:57.07,Default,,0000,0000,0000,,(Andra) So I'm going to talk about\Na specific project that I'm involved in
Dialogue: 0,0:27:57.07,0:27:58.07,Default,,0000,0000,0000,,called Gene Wiki,
Dialogue: 0,0:27:58.08,0:28:04.60,Default,,0000,0000,0000,,and where we are also\Ndealing with quality issues.
Dialogue: 0,0:28:04.60,0:28:06.68,Default,,0000,0000,0000,,But before going into the quality,
Dialogue: 0,0:28:06.68,0:28:09.23,Default,,0000,0000,0000,,maybe a quick introduction\Nabout what Gene Wiki is,
Dialogue: 0,0:28:09.86,0:28:15.18,Default,,0000,0000,0000,,and we just released a pre-print\Nof a paper that we recently have written
Dialogue: 0,0:28:15.18,0:28:18.16,Default,,0000,0000,0000,,that explains the details of the project.
Dialogue: 0,0:28:19.82,0:28:23.84,Default,,0000,0000,0000,,I see people taking pictures,\Nbut basically, what Gene Wiki does,
Dialogue: 0,0:28:23.85,0:28:28.03,Default,,0000,0000,0000,,it's trying to get biomedical data,\Npublic data into Wikidata,
Dialogue: 0,0:28:28.03,0:28:32.20,Default,,0000,0000,0000,,and we follow a specific pattern\Nto get that data into Wikidata.
Dialogue: 0,0:28:33.13,0:28:36.81,Default,,0000,0000,0000,,So when we have a new repository\Nor a new data set
Dialogue: 0,0:28:36.81,0:28:39.60,Default,,0000,0000,0000,,that is eligible\Nto be included into Wikidata,
Dialogue: 0,0:28:39.60,0:28:41.29,Default,,0000,0000,0000,,the first step is community engagement.
Dialogue: 0,0:28:41.29,0:28:43.78,Default,,0000,0000,0000,,It is not necessary\Ndirectly to a Wikidata community
Dialogue: 0,0:28:43.78,0:28:46.12,Default,,0000,0000,0000,,but a local research community,
Dialogue: 0,0:28:46.12,0:28:50.29,Default,,0000,0000,0000,,and we meet in person\Nor online or on any platform
Dialogue: 0,0:28:50.29,0:28:52.88,Default,,0000,0000,0000,,and try to come up with a data model
Dialogue: 0,0:28:52.88,0:28:56.20,Default,,0000,0000,0000,,that bridges their data\Nwith the Wikidata model.
Dialogue: 0,0:28:56.20,0:28:59.94,Default,,0000,0000,0000,,So here I have a picture of a workshop\Nthat happened here last year
Dialogue: 0,0:28:59.94,0:29:02.66,Default,,0000,0000,0000,,which was trying to look\Nat a specific data set
Dialogue: 0,0:29:02.66,0:29:05.28,Default,,0000,0000,0000,,and, well, you see a lot of discussions,
Dialogue: 0,0:29:05.28,0:29:09.78,Default,,0000,0000,0000,,then aligning it with schema.org\Nand other ontologies that are out there.
Dialogue: 0,0:29:10.32,0:29:15.51,Default,,0000,0000,0000,,And then, at the end of the first step,\Nwe have a whiteboard drawing of the schema
Dialogue: 0,0:29:15.51,0:29:17.34,Default,,0000,0000,0000,,that we want to implement in Wikidata.
Dialogue: 0,0:29:17.34,0:29:20.44,Default,,0000,0000,0000,,What you see over there,\Nthis is just plain,
Dialogue: 0,0:29:20.44,0:29:21.77,Default,,0000,0000,0000,,we have it in the back there
Dialogue: 0,0:29:21.77,0:29:25.24,Default,,0000,0000,0000,,so we can make some schemas\Nwithin this panel today even.
Dialogue: 0,0:29:26.56,0:29:28.40,Default,,0000,0000,0000,,So once we have the schema in place,
Dialogue: 0,0:29:28.40,0:29:31.32,Default,,0000,0000,0000,,the next thing is try to make\Nthat schema machine readable
Dialogue: 0,0:29:32.36,0:29:36.84,Default,,0000,0000,0000,,because you want to have actionable models\Nto bridge the data that you're bringing in
Dialogue: 0,0:29:36.84,0:29:39.69,Default,,0000,0000,0000,,from any biomedical database\Ninto Wikidata.
Dialogue: 0,0:29:40.39,0:29:45.18,Default,,0000,0000,0000,,And here we are applying\NShape Expressions.
Dialogue: 0,0:29:46.47,0:29:52.52,Default,,0000,0000,0000,,And we use that because\NShape Expressions allow you to test
Dialogue: 0,0:29:52.52,0:29:57.04,Default,,0000,0000,0000,,whether the data set\Nis actually-- no, to first see
Dialogue: 0,0:29:57.04,0:30:01.78,Default,,0000,0000,0000,,of already existing data in Wikidata\Nfollows the same data model
Dialogue: 0,0:30:01.78,0:30:04.72,Default,,0000,0000,0000,,that was achieved in the previous process.
Dialogue: 0,0:30:04.72,0:30:06.64,Default,,0000,0000,0000,,So then with the Shape Expression\Nwe can check:
Dialogue: 0,0:30:06.64,0:30:10.93,Default,,0000,0000,0000,,OK the data that are on this topic\Nin Wikidata, does it need some cleaning up
Dialogue: 0,0:30:10.93,0:30:15.01,Default,,0000,0000,0000,,or do we need to adapt our model\Nto the Wikidata model or vice versa.
Dialogue: 0,0:30:15.94,0:30:19.87,Default,,0000,0000,0000,,Once that is in place\Nand we start writing bots,
Dialogue: 0,0:30:20.67,0:30:23.80,Default,,0000,0000,0000,,and bots are seeding the information
Dialogue: 0,0:30:23.80,0:30:27.31,Default,,0000,0000,0000,,that is in the primary sources\Ninto Wikidata.
Dialogue: 0,0:30:27.85,0:30:29.30,Default,,0000,0000,0000,,And when the bots are ready,
Dialogue: 0,0:30:29.30,0:30:33.00,Default,,0000,0000,0000,,we write these bots\Nwith a platform called--
Dialogue: 0,0:30:33.00,0:30:36.20,Default,,0000,0000,0000,,with a Python library\Ncalled Wikidata Integrator
Dialogue: 0,0:30:36.20,0:30:38.17,Default,,0000,0000,0000,,that came out of our project.
Dialogue: 0,0:30:38.70,0:30:42.92,Default,,0000,0000,0000,,And once we have our bots,\Nwe use a platform called Jenkins
Dialogue: 0,0:30:42.92,0:30:44.54,Default,,0000,0000,0000,,for continuous integration.
Dialogue: 0,0:30:44.54,0:30:45.76,Default,,0000,0000,0000,,And with Jenkins,
Dialogue: 0,0:30:45.76,0:30:51.16,Default,,0000,0000,0000,,we continuously update\Nthe primary sources with Wikidata.
Dialogue: 0,0:30:52.18,0:30:55.89,Default,,0000,0000,0000,,And this is a diagram for the paper\NI previously mentioned.
Dialogue: 0,0:30:55.89,0:30:57.24,Default,,0000,0000,0000,,This is our current landscape.
Dialogue: 0,0:30:57.24,0:31:02.06,Default,,0000,0000,0000,,So every orange box out there\Nis a primary resource on drugs,
Dialogue: 0,0:31:02.06,0:31:07.83,Default,,0000,0000,0000,,proteins, genes, diseases,\Nchemical compounds with interaction,
Dialogue: 0,0:31:07.83,0:31:10.87,Default,,0000,0000,0000,,and this model is too small to read now
Dialogue: 0,0:31:10.87,0:31:17.47,Default,,0000,0000,0000,,but this is the database,\Nthe sources that we manage in Wikidata
Dialogue: 0,0:31:17.47,0:31:20.56,Default,,0000,0000,0000,,and bridge with the primary sources.
Dialogue: 0,0:31:20.56,0:31:22.36,Default,,0000,0000,0000,,Here is such a workflow.
Dialogue: 0,0:31:22.87,0:31:25.31,Default,,0000,0000,0000,,So one of our partners\Nis the Disease Ontology
Dialogue: 0,0:31:25.31,0:31:27.67,Default,,0000,0000,0000,,the Disease Ontology is a CC0 ontology,
Dialogue: 0,0:31:28.18,0:31:31.99,Default,,0000,0000,0000,,and the CC0 Ontology\Nhas a curation cycle on its own,
Dialogue: 0,0:31:32.76,0:31:35.74,Default,,0000,0000,0000,,and they just continuously\Nupdate the Disease Ontology
Dialogue: 0,0:31:35.74,0:31:39.69,Default,,0000,0000,0000,,to reflect the disease space\Nor the interpretation of diseases.
Dialogue: 0,0:31:40.34,0:31:44.36,Default,,0000,0000,0000,,And there is the Wikidata\Ncuration cycle also on diseases
Dialogue: 0,0:31:44.36,0:31:49.84,Default,,0000,0000,0000,,where the Wikidata community constantly\Nmonitors what's going on on Wikidata.
Dialogue: 0,0:31:50.41,0:31:51.60,Default,,0000,0000,0000,,And then we have two roles,
Dialogue: 0,0:31:51.60,0:31:55.48,Default,,0000,0000,0000,,we call them colloquially\Nthe gatekeeper curator,
Dialogue: 0,0:31:56.01,0:31:59.56,Default,,0000,0000,0000,,and this was me\Nand a colleague five years ago
Dialogue: 0,0:31:59.56,0:32:03.41,Default,,0000,0000,0000,,where we just sit on our computers\Nand we monitor Wikipedia and Wikidata,
Dialogue: 0,0:32:03.42,0:32:08.60,Default,,0000,0000,0000,,and if there is an issue that was\Nreported back to the primary community,
Dialogue: 0,0:32:08.60,0:32:11.76,Default,,0000,0000,0000,,the primary resources, they looked\Nat the implementation and decided:
Dialogue: 0,0:32:11.76,0:32:14.24,Default,,0000,0000,0000,,OK, do we do we trust the Wikidata input?
Dialogue: 0,0:32:14.85,0:32:18.56,Default,,0000,0000,0000,,Yes--then it's considered,\Nit goes into the cycle,
Dialogue: 0,0:32:18.56,0:32:22.69,Default,,0000,0000,0000,,and the next iteration\Nis part of the Disease Ontology
Dialogue: 0,0:32:22.69,0:32:25.41,Default,,0000,0000,0000,,and fed back into Wikidata.
Dialogue: 0,0:32:27.42,0:32:31.48,Default,,0000,0000,0000,,We're doing the same for WikiPathways.
Dialogue: 0,0:32:31.48,0:32:36.60,Default,,0000,0000,0000,,WikiPathways is a MediaWiki-inspired\Npathway and pathway repository.
Dialogue: 0,0:32:36.60,0:32:40.90,Default,,0000,0000,0000,,Same story, there are different\Npathway resources on Wikidata already.
Dialogue: 0,0:32:41.46,0:32:44.71,Default,,0000,0000,0000,,There might be conflicts\Nbetween those pathway resources
Dialogue: 0,0:32:44.72,0:32:46.70,Default,,0000,0000,0000,,and these conflicts are reported back
Dialogue: 0,0:32:46.70,0:32:49.52,Default,,0000,0000,0000,,by the gatekeeper curators\Nto that community,
Dialogue: 0,0:32:49.52,0:32:53.72,Default,,0000,0000,0000,,and you maintain\Nthe individual curation cycles.
Dialogue: 0,0:32:53.72,0:32:57.07,Default,,0000,0000,0000,,But if you remember the previous cycle,
Dialogue: 0,0:32:57.07,0:33:03.04,Default,,0000,0000,0000,,here I mentioned\Nonly two cycles, two resources,
Dialogue: 0,0:33:03.57,0:33:06.30,Default,,0000,0000,0000,,we have to do that\Nfor every single resource that we have
Dialogue: 0,0:33:06.30,0:33:08.06,Default,,0000,0000,0000,,and we have to manage what's going on
Dialogue: 0,0:33:08.06,0:33:09.18,Default,,0000,0000,0000,,because when I say curation,
Dialogue: 0,0:33:09.18,0:33:11.38,Default,,0000,0000,0000,,I really mean going\Nto the Wikipedia top pages,
Dialogue: 0,0:33:11.38,0:33:14.54,Default,,0000,0000,0000,,going into the Wikidata top pages\Nand trying to do that.
Dialogue: 0,0:33:14.54,0:33:19.32,Default,,0000,0000,0000,,That doesn't scale for\Nthe two gatekeeper curators we had.
Dialogue: 0,0:33:19.86,0:33:22.78,Default,,0000,0000,0000,,So when I was in a conference in 2016
Dialogue: 0,0:33:22.78,0:33:26.93,Default,,0000,0000,0000,,where Eric gave a presentation\Non Shape Expressions,
Dialogue: 0,0:33:26.93,0:33:29.28,Default,,0000,0000,0000,,I jumped on the bandwagon and said OK,
Dialogue: 0,0:33:29.28,0:33:34.24,Default,,0000,0000,0000,,Shape Expressions can help us\Ndetect what differences in Wikidata
Dialogue: 0,0:33:34.24,0:33:41.16,Default,,0000,0000,0000,,and so that allows the gatekeepers to have\Nsome more efficient reporting to report.
Dialogue: 0,0:33:42.28,0:33:46.02,Default,,0000,0000,0000,,So this year,\NI was delighted by the schema entity
Dialogue: 0,0:33:46.02,0:33:50.76,Default,,0000,0000,0000,,because now, we can store\Nthose entity schemas on Wikidata,
Dialogue: 0,0:33:50.76,0:33:53.18,Default,,0000,0000,0000,,on Wikidata itself,\Nwhereas before, it was on GitHub,
Dialogue: 0,0:33:53.86,0:33:56.82,Default,,0000,0000,0000,,and this aligns\Nwith the Wikidata interface,
Dialogue: 0,0:33:56.82,0:33:59.35,Default,,0000,0000,0000,,so you have things\Nlike document discussions
Dialogue: 0,0:33:59.35,0:34:00.76,Default,,0000,0000,0000,,but you also have revisions.
Dialogue: 0,0:34:00.76,0:34:05.26,Default,,0000,0000,0000,,So you can leverage the top pages\Nand the revisions in Wikidata
Dialogue: 0,0:34:05.26,0:34:12.26,Default,,0000,0000,0000,,to use that to discuss\Nabout what is in Wikidata
Dialogue: 0,0:34:12.26,0:34:14.06,Default,,0000,0000,0000,,and what are in the primary resources.
Dialogue: 0,0:34:14.97,0:34:19.69,Default,,0000,0000,0000,,So this what Eric just presented,\Nthis is already quite a benefit.
Dialogue: 0,0:34:19.69,0:34:24.34,Default,,0000,0000,0000,,So here, we made up a Shape Expression\Nfor the human gene,
Dialogue: 0,0:34:24.34,0:34:30.22,Default,,0000,0000,0000,,and then we ran it through simple ShEx,\Nand as you can see,
Dialogue: 0,0:34:30.22,0:34:32.43,Default,,0000,0000,0000,,we just got already ni--
Dialogue: 0,0:34:32.43,0:34:34.64,Default,,0000,0000,0000,,There is one issue\Nthat needs to be monitored
Dialogue: 0,0:34:34.64,0:34:37.32,Default,,0000,0000,0000,,which there is an item\Nthat doesn't fit that schema,
Dialogue: 0,0:34:37.32,0:34:43.14,Default,,0000,0000,0000,,and then you can sort of already\Ncreate schema entities curation reports
Dialogue: 0,0:34:43.14,0:34:46.24,Default,,0000,0000,0000,,based on... and send that\Nto the different curation reports.
Dialogue: 0,0:34:48.06,0:34:52.79,Default,,0000,0000,0000,,But the ShEx.js a built interface,
Dialogue: 0,0:34:52.79,0:34:55.86,Default,,0000,0000,0000,,and if I can show back here,\NI only do ten,
Dialogue: 0,0:34:55.86,0:35:00.36,Default,,0000,0000,0000,,but we have tens of thousands,\Nand so that again doesn't scale.
Dialogue: 0,0:35:00.36,0:35:04.65,Default,,0000,0000,0000,,So the Wikidata Integrator now\Nsupports ShEx support as well,
Dialogue: 0,0:35:05.17,0:35:07.43,Default,,0000,0000,0000,,and then we can just loop item loops
Dialogue: 0,0:35:07.43,0:35:11.49,Default,,0000,0000,0000,,where we say yes-no,\Nyes-no, true-false, true-false.
Dialogue: 0,0:35:11.50,0:35:12.50,Default,,0000,0000,0000,,So again,
Dialogue: 0,0:35:13.06,0:35:16.51,Default,,0000,0000,0000,,increasing a bit of the efficiency\Nof dealing with the reports.
Dialogue: 0,0:35:17.26,0:35:22.66,Default,,0000,0000,0000,,But now, recently, that builds\Non the Wikidata Query Service,
Dialogue: 0,0:35:23.18,0:35:24.100,Default,,0000,0000,0000,,and well, we recently have been throttling
Dialogue: 0,0:35:24.100,0:35:26.56,Default,,0000,0000,0000,,so again, that doesn't scale.
Dialogue: 0,0:35:26.56,0:35:31.39,Default,,0000,0000,0000,,So it's still an ongoing process,\Nhow to deal with models on Wikidata.
Dialogue: 0,0:35:32.20,0:35:36.68,Default,,0000,0000,0000,,And so again,\NShEx is not only intimidating
Dialogue: 0,0:35:36.68,0:35:40.36,Default,,0000,0000,0000,,but also the scale is just\Ntoo big to deal with.
Dialogue: 0,0:35:41.07,0:35:46.08,Default,,0000,0000,0000,,So I started working, this is my first\Nproof of concept or exercise
Dialogue: 0,0:35:46.08,0:35:47.68,Default,,0000,0000,0000,,where I used a tool called yED,
Dialogue: 0,0:35:48.18,0:35:52.59,Default,,0000,0000,0000,,and I started to draw\Nthose Shape Expressions and because...
Dialogue: 0,0:35:52.59,0:35:58.10,Default,,0000,0000,0000,,and then regenerate this schema
Dialogue: 0,0:35:58.10,0:36:01.28,Default,,0000,0000,0000,,into this adjacent format\Nof the Shape Expressions,
Dialogue: 0,0:36:01.28,0:36:04.52,Default,,0000,0000,0000,,so that would open up already\Nto the audience
Dialogue: 0,0:36:04.52,0:36:07.43,Default,,0000,0000,0000,,that are intimidated\Nby the Shape Expressions languages.
Dialogue: 0,0:36:07.96,0:36:12.31,Default,,0000,0000,0000,,But actually, there is a problem\Nwith those visual descriptions
Dialogue: 0,0:36:12.31,0:36:18.23,Default,,0000,0000,0000,,because this is also a schema\Nthat was actually drawn in yEd by someone.
Dialogue: 0,0:36:18.23,0:36:23.84,Default,,0000,0000,0000,,And here is another one\Nwhich is beautiful.
Dialogue: 0,0:36:23.84,0:36:29.41,Default,,0000,0000,0000,,I would love to have this on my wall,\Nbut it is still not interoperable.
Dialogue: 0,0:36:30.28,0:36:32.13,Default,,0000,0000,0000,,So I want to end my talk with,
Dialogue: 0,0:36:32.13,0:36:35.73,Default,,0000,0000,0000,,and the first time, I've been\Nstealing this slide, using this slide.
Dialogue: 0,0:36:35.73,0:36:37.59,Default,,0000,0000,0000,,It's an honor to have him in the audience
Dialogue: 0,0:36:37.60,0:36:39.42,Default,,0000,0000,0000,,and I really like this:
Dialogue: 0,0:36:39.42,0:36:42.36,Default,,0000,0000,0000,,"People think RDF is a pain\Nbecause it's complicated.
Dialogue: 0,0:36:42.36,0:36:43.98,Default,,0000,0000,0000,,The truth is even worse, it's so simple,
Dialogue: 0,0:36:45.58,0:36:48.13,Default,,0000,0000,0000,,because you have to work\Nwith real-world data problems
Dialogue: 0,0:36:48.13,0:36:50.03,Default,,0000,0000,0000,,that are horribly complicated.
Dialogue: 0,0:36:50.03,0:36:51.45,Default,,0000,0000,0000,,While you can avoid RDF,
Dialogue: 0,0:36:51.45,0:36:55.76,Default,,0000,0000,0000,,it is harder to avoid complicated data\Nand complicated computer problems."
Dialogue: 0,0:36:55.76,0:36:59.54,Default,,0000,0000,0000,,This is about RDF, but I think\Nthis so applies to modeling as well.
Dialogue: 0,0:37:00.11,0:37:02.77,Default,,0000,0000,0000,,So my point of discussion\Nis should we really...
Dialogue: 0,0:37:03.39,0:37:05.88,Default,,0000,0000,0000,,How do we get modeling going?
Dialogue: 0,0:37:05.88,0:37:10.83,Default,,0000,0000,0000,,Should we discuss ShEx\Nor visual models or...
Dialogue: 0,0:37:11.43,0:37:13.27,Default,,0000,0000,0000,,How do we continue?
Dialogue: 0,0:37:13.47,0:37:14.84,Default,,0000,0000,0000,,Thank you very much for your time.
Dialogue: 0,0:37:15.10,0:37:17.79,Default,,0000,0000,0000,,(applause)
Dialogue: 0,0:37:20.00,0:37:21.19,Default,,0000,0000,0000,,(Lydia) Thank you so much.
Dialogue: 0,0:37:21.69,0:37:24.00,Default,,0000,0000,0000,,Would you come to the front
Dialogue: 0,0:37:24.00,0:37:27.74,Default,,0000,0000,0000,,so that we can open\Nthe questions from the audience.
Dialogue: 0,0:37:28.61,0:37:30.20,Default,,0000,0000,0000,,Are there questions?
Dialogue: 0,0:37:31.51,0:37:32.51,Default,,0000,0000,0000,,Yes.
Dialogue: 0,0:37:34.25,0:37:36.89,Default,,0000,0000,0000,,And I think, for the camera, we need to...
Dialogue: 0,0:37:38.84,0:37:40.97,Default,,0000,0000,0000,,(Lydia laughing) Yeah.
Dialogue: 0,0:37:43.09,0:37:46.27,Default,,0000,0000,0000,,(man3) So a question\Nfor Cristina, I think.
Dialogue: 0,0:37:47.37,0:37:51.64,Default,,0000,0000,0000,,So you mentioned exactly\Nthe term "information gain"
Dialogue: 0,0:37:51.64,0:37:53.69,Default,,0000,0000,0000,,from linking with other systems.
Dialogue: 0,0:37:53.69,0:37:55.62,Default,,0000,0000,0000,,There is an information theoretic measure
Dialogue: 0,0:37:55.62,0:37:58.00,Default,,0000,0000,0000,,using statistic and probability\Ncalled information gain.
Dialogue: 0,0:37:58.00,0:37:59.54,Default,,0000,0000,0000,,Do you have the same...
Dialogue: 0,0:37:59.54,0:38:01.74,Default,,0000,0000,0000,,I mean did you mean exactly that measure,
Dialogue: 0,0:38:01.74,0:38:04.17,Default,,0000,0000,0000,,the information gain\Nfrom the probability theory
Dialogue: 0,0:38:04.17,0:38:05.24,Default,,0000,0000,0000,,from information theory
Dialogue: 0,0:38:05.24,0:38:09.02,Default,,0000,0000,0000,,or just use this conceptual thing\Nto measure information gain some way?
Dialogue: 0,0:38:09.02,0:38:13.02,Default,,0000,0000,0000,,No, so we actually defined\Nand implemented measures
Dialogue: 0,0:38:13.70,0:38:20.16,Default,,0000,0000,0000,,that are using the Shannon entropy,\Nso it's meant as that.
Dialogue: 0,0:38:20.16,0:38:22.70,Default,,0000,0000,0000,,I didn't want to go into\Ndetails of the concrete formulas...
Dialogue: 0,0:38:22.70,0:38:24.98,Default,,0000,0000,0000,,(man3) No, no, of course,\Nthat's why I asked the question.
Dialogue: 0,0:38:24.98,0:38:26.70,Default,,0000,0000,0000,,- (Cristina) But yeah...\N- (man3) Thank you.
Dialogue: 0,0:38:33.09,0:38:35.05,Default,,0000,0000,0000,,(man4) Make more\Nof a comment than a question.
Dialogue: 0,0:38:35.05,0:38:36.24,Default,,0000,0000,0000,,(Lydia) Go for it.
Dialogue: 0,0:38:36.24,0:38:39.84,Default,,0000,0000,0000,,(man4) So there's been\Na lot of focus at the item level
Dialogue: 0,0:38:39.84,0:38:42.55,Default,,0000,0000,0000,,about quality and completeness,
Dialogue: 0,0:38:42.55,0:38:47.37,Default,,0000,0000,0000,,one of the things that concerns me is that\Nwe're not applying the same to hierarchies
Dialogue: 0,0:38:47.37,0:38:51.48,Default,,0000,0000,0000,,and I think we have an issue\Nis that our hierarchy often isn't good.
Dialogue: 0,0:38:51.48,0:38:53.46,Default,,0000,0000,0000,,We're seeing\Nthis is going to be a real problem
Dialogue: 0,0:38:53.46,0:38:55.77,Default,,0000,0000,0000,,with Commons searching and other things.
Dialogue: 0,0:38:56.77,0:39:00.60,Default,,0000,0000,0000,,One of the abilities that we can do\Nis to import external--
Dialogue: 0,0:39:00.60,0:39:04.84,Default,,0000,0000,0000,,The way that external thesauruses\Nstructure their hierarchies,
Dialogue: 0,0:39:04.84,0:39:10.29,Default,,0000,0000,0000,,using the P4900\Nbroader concept qualifier.
Dialogue: 0,0:39:11.04,0:39:16.17,Default,,0000,0000,0000,,But what I think would be really helpful\Nwould be much better tools for doing that
Dialogue: 0,0:39:16.17,0:39:21.21,Default,,0000,0000,0000,,so that you can import an\Nexternal... thesaurus's hierarchy
Dialogue: 0,0:39:21.21,0:39:24.11,Default,,0000,0000,0000,,map that onto our Wikidata items.
Dialogue: 0,0:39:24.11,0:39:28.20,Default,,0000,0000,0000,,Once it's in place\Nwith those P4900 qualifiers,
Dialogue: 0,0:39:28.20,0:39:31.49,Default,,0000,0000,0000,,you can actually do some\Nquite good querying through SPARQL
Dialogue: 0,0:39:32.49,0:39:37.53,Default,,0000,0000,0000,,to see where our hierarchy\Ndiverges from that external hierarchy.
Dialogue: 0,0:39:37.53,0:39:41.35,Default,,0000,0000,0000,,For instance, [Paula Morma],\Nuser PKM, you may know,
Dialogue: 0,0:39:41.35,0:39:43.53,Default,,0000,0000,0000,,does a lot of work on fashion.
Dialogue: 0,0:39:43.53,0:39:50.52,Default,,0000,0000,0000,,So we use that to pull in the Europeana\NFashion Thesaurus's hierarchy
Dialogue: 0,0:39:50.52,0:39:53.81,Default,,0000,0000,0000,,and the Getty AAT\Nfashion thesaurus hierarchy,
Dialogue: 0,0:39:53.81,0:39:57.96,Default,,0000,0000,0000,,and then see where the gaps\Nwere in our higher level items,
Dialogue: 0,0:39:57.96,0:40:00.51,Default,,0000,0000,0000,,which is a real problem for us\Nbecause often,
Dialogue: 0,0:40:00.51,0:40:04.36,Default,,0000,0000,0000,,these are things that only exist\Nas disambiguation pages on Wikipedia,
Dialogue: 0,0:40:04.36,0:40:09.27,Default,,0000,0000,0000,,so we have a lot of higher level items\Nin our hierarchies missing
Dialogue: 0,0:40:09.27,0:40:14.48,Default,,0000,0000,0000,,and this is something that we must address\Nin terms of quality and completeness,
Dialogue: 0,0:40:14.48,0:40:15.97,Default,,0000,0000,0000,,but what would really help
Dialogue: 0,0:40:16.64,0:40:20.87,Default,,0000,0000,0000,,would be better tools than\Nthe jungle of pull scripts that I wrote...
Dialogue: 0,0:40:20.87,0:40:26.01,Default,,0000,0000,0000,,If somebody could put that\Ninto a PAWS notebook in Python
Dialogue: 0,0:40:26.56,0:40:31.97,Default,,0000,0000,0000,,to be able to take an external thesaurus,\Ntake its hierarchy,
Dialogue: 0,0:40:31.97,0:40:34.60,Default,,0000,0000,0000,,which may well be available\Nas linked data or may not,
Dialogue: 0,0:40:35.38,0:40:40.58,Default,,0000,0000,0000,,to then put those into\Nquick statements to put in P4900 values.
Dialogue: 0,0:40:41.16,0:40:42.16,Default,,0000,0000,0000,,And then later,
Dialogue: 0,0:40:42.17,0:40:44.53,Default,,0000,0000,0000,,when our representation\Ngets more complete,
Dialogue: 0,0:40:44.53,0:40:49.69,Default,,0000,0000,0000,,to update those P4900s\Nbecause as our representation gets dated,
Dialogue: 0,0:40:49.69,0:40:51.59,Default,,0000,0000,0000,,becomes more dense,
Dialogue: 0,0:40:51.59,0:40:55.38,Default,,0000,0000,0000,,the values of those qualifiers\Nneed to change
Dialogue: 0,0:40:56.23,0:40:59.53,Default,,0000,0000,0000,,to represent that we've got more\Nof their hierarchy in our system.
Dialogue: 0,0:40:59.53,0:41:03.73,Default,,0000,0000,0000,,If somebody could do that,\NI think that would be very helpful,
Dialogue: 0,0:41:03.73,0:41:07.12,Default,,0000,0000,0000,,and we do need to also\Nlook at other approaches
Dialogue: 0,0:41:07.12,0:41:10.76,Default,,0000,0000,0000,,to improve quality and completeness\Nat the hierarchy level
Dialogue: 0,0:41:10.76,0:41:12.38,Default,,0000,0000,0000,,not just at the item level.
Dialogue: 0,0:41:13.31,0:41:14.84,Default,,0000,0000,0000,,(Andra) Can I add to that?
Dialogue: 0,0:41:16.36,0:41:19.90,Default,,0000,0000,0000,,Yes, and we actually do that,
Dialogue: 0,0:41:19.91,0:41:23.55,Default,,0000,0000,0000,,and I can recommend looking at\Nthe Shape Expression that Finn made
Dialogue: 0,0:41:23.55,0:41:27.33,Default,,0000,0000,0000,,with the lexical data\Nwhere he creates Shape Expressions
Dialogue: 0,0:41:27.33,0:41:29.64,Default,,0000,0000,0000,,and then build on authorship expressions
Dialogue: 0,0:41:29.64,0:41:32.53,Default,,0000,0000,0000,,so you have this concept\Nof linked Shape Expressions in Wikidata,
Dialogue: 0,0:41:32.53,0:41:35.00,Default,,0000,0000,0000,,and specifically, the use case,\Nif I understand correctly,
Dialogue: 0,0:41:35.01,0:41:37.18,Default,,0000,0000,0000,,is exactly what we are doing in Gene Wiki.
Dialogue: 0,0:41:37.18,0:41:40.84,Default,,0000,0000,0000,,So you have the Disease Ontology\Nwhich is put into Wikidata
Dialogue: 0,0:41:40.84,0:41:44.68,Default,,0000,0000,0000,,and then disease data comes in\Nand we apply the Shape Expressions
Dialogue: 0,0:41:44.68,0:41:47.25,Default,,0000,0000,0000,,to see if that fits with this thesaurus.
Dialogue: 0,0:41:47.25,0:41:50.92,Default,,0000,0000,0000,,And there are other thesauruses or other\Nontologies for controlled vocabularies
Dialogue: 0,0:41:50.92,0:41:52.56,Default,,0000,0000,0000,,that still need to go into Wikidata,
Dialogue: 0,0:41:52.56,0:41:55.40,Default,,0000,0000,0000,,and that's exactly why\NShape Expression is so interesting
Dialogue: 0,0:41:55.40,0:41:57.96,Default,,0000,0000,0000,,because you can have a Shape Expression\Nfor the Disease Ontology,
Dialogue: 0,0:41:57.96,0:41:59.64,Default,,0000,0000,0000,,you can have a Shape Expression for MeSH,
Dialogue: 0,0:41:59.64,0:42:01.76,Default,,0000,0000,0000,,you can say: OK,\Nnow I want to check the quality.
Dialogue: 0,0:42:01.76,0:42:04.06,Default,,0000,0000,0000,,Because you also have\Nin Wikidata the context
Dialogue: 0,0:42:04.06,0:42:09.57,Default,,0000,0000,0000,,of when you have a controlled vocabulary,\Nyou say the quality is according to this,
Dialogue: 0,0:42:09.57,0:42:11.64,Default,,0000,0000,0000,,but you might have\Na disagreeing community.
Dialogue: 0,0:42:11.64,0:42:16.08,Default,,0000,0000,0000,,So the tooling is indeed in place\Nbut now is indeed to create those models
Dialogue: 0,0:42:16.08,0:42:18.14,Default,,0000,0000,0000,,and apply them\Non the different use cases.
Dialogue: 0,0:42:18.81,0:42:20.92,Default,,0000,0000,0000,,(man4) The ShapeExpression's very useful
Dialogue: 0,0:42:20.92,0:42:25.93,Default,,0000,0000,0000,,once you have the external ontology\Nmapped into Wikidata,
Dialogue: 0,0:42:25.93,0:42:29.47,Default,,0000,0000,0000,,but my problem is that\Nit's getting to that stage,
Dialogue: 0,0:42:29.48,0:42:34.88,Default,,0000,0000,0000,,it's working out how much of the\Nexternal ontology isn't yet in Wikidata
Dialogue: 0,0:42:34.88,0:42:36.26,Default,,0000,0000,0000,,and where the gaps are,
Dialogue: 0,0:42:36.26,0:42:40.66,Default,,0000,0000,0000,,and that's where I think that\Nhaving much more robust tools
Dialogue: 0,0:42:40.66,0:42:44.29,Default,,0000,0000,0000,,to see what's missing\Nfrom external ontologies
Dialogue: 0,0:42:44.29,0:42:45.54,Default,,0000,0000,0000,,would be very helpful.
Dialogue: 0,0:42:47.68,0:42:49.06,Default,,0000,0000,0000,,The biggest problem there
Dialogue: 0,0:42:49.06,0:42:51.20,Default,,0000,0000,0000,,is not so much tooling\Nbut more licensing.
Dialogue: 0,0:42:51.80,0:42:55.25,Default,,0000,0000,0000,,So getting the ontologies\Ninto Wikidata is actually a piece of cake
Dialogue: 0,0:42:55.25,0:42:59.30,Default,,0000,0000,0000,,but most of the ontologies have,\Nhow can I say that politely,
Dialogue: 0,0:42:59.96,0:43:03.26,Default,,0000,0000,0000,,restrictive licensing,\Nso they are not compatible with Wikidata.
Dialogue: 0,0:43:04.07,0:43:06.68,Default,,0000,0000,0000,,(man4) There's a huge number\Nof public sector thesauruses
Dialogue: 0,0:43:06.68,0:43:08.21,Default,,0000,0000,0000,,in cultural fields.
Dialogue: 0,0:43:08.21,0:43:10.85,Default,,0000,0000,0000,,- (Andra) Then we need to talk.\N- (man4) Not a problem.
Dialogue: 0,0:43:10.85,0:43:12.38,Default,,0000,0000,0000,,(Andra) Then we need to talk.
Dialogue: 0,0:43:13.62,0:43:19.19,Default,,0000,0000,0000,,(man5) Just... the comment I want to make\Nis actually answer to James,
Dialogue: 0,0:43:19.19,0:43:22.40,Default,,0000,0000,0000,,so the thing is that\Nhierarchies make graphs,
Dialogue: 0,0:43:22.37,0:43:24.04,Default,,0000,0000,0000,,and when you want to...
Dialogue: 0,0:43:24.58,0:43:28.89,Default,,0000,0000,0000,,I want to basically talk about...\Na common problem in hierarchies
Dialogue: 0,0:43:28.89,0:43:30.82,Default,,0000,0000,0000,,is circle hierarchies,
Dialogue: 0,0:43:30.82,0:43:33.80,Default,,0000,0000,0000,,so they come back to each other\Nwhen there's a problem,
Dialogue: 0,0:43:33.80,0:43:35.92,Default,,0000,0000,0000,,which you should not\Nhave that in hierarchies.
Dialogue: 0,0:43:37.02,0:43:41.30,Default,,0000,0000,0000,,This, funnily enough,\Nhappens in categories in Wikipedia a lot
Dialogue: 0,0:43:41.30,0:43:42.99,Default,,0000,0000,0000,,we have a lot of circles in categories,
Dialogue: 0,0:43:43.90,0:43:46.61,Default,,0000,0000,0000,,but the good news is that this is...
Dialogue: 0,0:43:47.71,0:43:51.58,Default,,0000,0000,0000,,Technically, it's a PMP complete problem,\Nso you cannot find this,
Dialogue: 0,0:43:51.58,0:43:53.41,Default,,0000,0000,0000,,and easily if you built a graph of that,
Dialogue: 0,0:43:54.47,0:43:57.05,Default,,0000,0000,0000,,but there are lots of ways\Nthat have been developed
Dialogue: 0,0:43:57.05,0:44:00.62,Default,,0000,0000,0000,,to find problems\Nin these hierarchy graphs.
Dialogue: 0,0:44:00.62,0:44:04.86,Default,,0000,0000,0000,,Like there is a paper\Ncalled {\i1}Finding Cycles{\i0}...
Dialogue: 0,0:44:04.86,0:44:07.96,Default,,0000,0000,0000,,{\i1}Breaking Cycles in Noisy Hierarchies,{\i0}
Dialogue: 0,0:44:07.96,0:44:12.67,Default,,0000,0000,0000,,and it's been used to help\Ncategorization of English Wikipedia.
Dialogue: 0,0:44:12.67,0:44:17.14,Default,,0000,0000,0000,,You can just take this\Nand apply these hierarchies in Wikidata,
Dialogue: 0,0:44:17.14,0:44:19.54,Default,,0000,0000,0000,,and then you can find\Nthings that are problematic
Dialogue: 0,0:44:19.54,0:44:22.48,Default,,0000,0000,0000,,and just remove the ones\Nthat are causing issues
Dialogue: 0,0:44:22.48,0:44:24.59,Default,,0000,0000,0000,,and find the issues, actually.
Dialogue: 0,0:44:24.59,0:44:26.96,Default,,0000,0000,0000,,So this is just an idea, just so you...
Dialogue: 0,0:44:28.78,0:44:29.93,Default,,0000,0000,0000,,(man4) That's all very well
Dialogue: 0,0:44:29.93,0:44:34.40,Default,,0000,0000,0000,,but I think you're underestimating\Nthe number of bad subclass relations
Dialogue: 0,0:44:34.40,0:44:35.40,Default,,0000,0000,0000,,that we have.
Dialogue: 0,0:44:35.40,0:44:39.68,Default,,0000,0000,0000,,It's like having a city\Nin completely the wrong country,
Dialogue: 0,0:44:40.25,0:44:44.87,Default,,0000,0000,0000,,and there are tools for geography\Nto identify that,
Dialogue: 0,0:44:44.88,0:44:49.20,Default,,0000,0000,0000,,and we need to have\Nmuch better tools in hierarchies
Dialogue: 0,0:44:49.20,0:44:53.48,Default,,0000,0000,0000,,to identify where the equivalent\Nof the item for the country
Dialogue: 0,0:44:53.48,0:44:57.67,Default,,0000,0000,0000,,is missing entirely,\Nor where it's actually been subclassed
Dialogue: 0,0:44:57.67,0:45:01.80,Default,,0000,0000,0000,,to something that isn't meaning\Nsomething completely different.
Dialogue: 0,0:45:02.80,0:45:07.16,Default,,0000,0000,0000,,(Lydia) Yeah, I think\Nyou're getting to something
Dialogue: 0,0:45:07.17,0:45:12.02,Default,,0000,0000,0000,,that me and my team keeps hearing\Nfrom people who reuse our data
Dialogue: 0,0:45:12.02,0:45:13.99,Default,,0000,0000,0000,,quite a bit as well, right,
Dialogue: 0,0:45:15.00,0:45:16.64,Default,,0000,0000,0000,,Individual data point might be great
Dialogue: 0,0:45:16.64,0:45:20.16,Default,,0000,0000,0000,,but if you have to look\Nat the ontology and so on,
Dialogue: 0,0:45:20.16,0:45:21.86,Default,,0000,0000,0000,,then it gets very...
Dialogue: 0,0:45:22.39,0:45:26.44,Default,,0000,0000,0000,,And I think one of the big problems\Nwhy this is happening
Dialogue: 0,0:45:26.44,0:45:30.74,Default,,0000,0000,0000,,is that a lot of editing on Wikidata
Dialogue: 0,0:45:30.74,0:45:34.54,Default,,0000,0000,0000,,happens on the basis\Nof an individual item, right,
Dialogue: 0,0:45:34.54,0:45:36.20,Default,,0000,0000,0000,,you make an edit on that item,
Dialogue: 0,0:45:37.65,0:45:42.08,Default,,0000,0000,0000,,without realizing that this\Nmight have very global consequences
Dialogue: 0,0:45:42.08,0:45:44.24,Default,,0000,0000,0000,,on the rest of the graph, for example.
Dialogue: 0,0:45:44.24,0:45:50.04,Default,,0000,0000,0000,,And if people have ideas around\Nhow to make this more visible,
Dialogue: 0,0:45:50.04,0:45:53.18,Default,,0000,0000,0000,,the consequences\Nof an individual local edit,
Dialogue: 0,0:45:54.00,0:45:56.54,Default,,0000,0000,0000,,I think that would be worth exploring,
Dialogue: 0,0:45:57.55,0:46:01.58,Default,,0000,0000,0000,,to show people better\Nwhat the consequence of their edit
Dialogue: 0,0:46:01.58,0:46:03.43,Default,,0000,0000,0000,,that they might do in very good faith,
Dialogue: 0,0:46:04.48,0:46:05.48,Default,,0000,0000,0000,,what that is.
Dialogue: 0,0:46:06.94,0:46:12.24,Default,,0000,0000,0000,,Whoa! OK, let's start with, yeah, you,\Nthen you, then you, then you.
Dialogue: 0,0:46:12.24,0:46:13.92,Default,,0000,0000,0000,,(man5) Well, after the discussion,
Dialogue: 0,0:46:13.92,0:46:18.26,Default,,0000,0000,0000,,just to express my agreement\Nwith what James was saying.
Dialogue: 0,0:46:18.26,0:46:22.47,Default,,0000,0000,0000,,So essentially, it seems\Nthe most dangerous thing is the hierarchy,
Dialogue: 0,0:46:22.47,0:46:23.91,Default,,0000,0000,0000,,not the hierarchy, but generally
Dialogue: 0,0:46:23.91,0:46:28.02,Default,,0000,0000,0000,,the semantics of the subclass relations\Nseen in Wikidata, right.
Dialogue: 0,0:46:28.02,0:46:32.56,Default,,0000,0000,0000,,So I've been studying languages recently,\Njust for the purposes of this conference,
Dialogue: 0,0:46:32.56,0:46:35.26,Default,,0000,0000,0000,,and for example, you find plenty of cases
Dialogue: 0,0:46:35.26,0:46:39.46,Default,,0000,0000,0000,,where a language is a part of\Nand subclass of the same thing, OK.
Dialogue: 0,0:46:39.46,0:46:43.58,Default,,0000,0000,0000,,So you know, you can say\Nwe have a flexible ontology.
Dialogue: 0,0:46:43.58,0:46:46.26,Default,,0000,0000,0000,,Wikidata gives you freedom\Nto express that, sometimes.
Dialogue: 0,0:46:46.26,0:46:47.26,Default,,0000,0000,0000,,Because, for example,
Dialogue: 0,0:46:47.26,0:46:50.72,Default,,0000,0000,0000,,that ontology of languages\Nis also politically complicated, right?
Dialogue: 0,0:46:50.72,0:46:55.04,Default,,0000,0000,0000,,It is even good to be in a position\Nto express a level of uncertainty.
Dialogue: 0,0:46:55.04,0:46:57.98,Default,,0000,0000,0000,,But imagine anyone who wants\Nto do machine reading from that.
Dialogue: 0,0:46:57.98,0:46:59.47,Default,,0000,0000,0000,,So that's really problematic.
Dialogue: 0,0:46:59.47,0:47:00.47,Default,,0000,0000,0000,,And then again,
Dialogue: 0,0:47:00.47,0:47:03.69,Default,,0000,0000,0000,,I don't think that ontology\Nwas ever imported from somewhere,
Dialogue: 0,0:47:03.69,0:47:05.49,Default,,0000,0000,0000,,that's something which is originally ours.
Dialogue: 0,0:47:05.49,0:47:08.32,Default,,0000,0000,0000,,It's harvested from Wikipedia\Nin the very beginning I will say.
Dialogue: 0,0:47:08.32,0:47:11.32,Default,,0000,0000,0000,,So I wonder...\Nthis Shape Expressions thing is great,
Dialogue: 0,0:47:11.32,0:47:15.58,Default,,0000,0000,0000,,and also validating and fixing,\Nif you like, the Wikidata ontology
Dialogue: 0,0:47:15.58,0:47:18.19,Default,,0000,0000,0000,,by external resources, beautiful idea.
Dialogue: 0,0:47:19.03,0:47:20.03,Default,,0000,0000,0000,,In the end,
Dialogue: 0,0:47:20.03,0:47:25.44,Default,,0000,0000,0000,,will we end by reflecting\Nthe external ontologies in Wikidata?
Dialogue: 0,0:47:25.44,0:47:28.65,Default,,0000,0000,0000,,And also, what we do with\Nthe core part of our ontology
Dialogue: 0,0:47:28.65,0:47:30.64,Default,,0000,0000,0000,,which is never harvested\Nfrom external resources,
Dialogue: 0,0:47:30.64,0:47:31.98,Default,,0000,0000,0000,,how do we go and fix that?
Dialogue: 0,0:47:31.98,0:47:35.28,Default,,0000,0000,0000,,And I really think that\Nthat will be a problem on its own.
Dialogue: 0,0:47:35.28,0:47:39.01,Default,,0000,0000,0000,,We will have to focus on that\Nindependently of the idea
Dialogue: 0,0:47:39.01,0:47:41.05,Default,,0000,0000,0000,,of validating ontology\Nwith something external.
Dialogue: 0,0:47:49.35,0:47:53.38,Default,,0000,0000,0000,,(man6) OK, and constrains\Nand shapes are very impressive
Dialogue: 0,0:47:53.38,0:47:54.50,Default,,0000,0000,0000,,what we can do with it,
Dialogue: 0,0:47:55.20,0:47:58.48,Default,,0000,0000,0000,,but the main point is not\Nbeing really made clear--
Dialogue: 0,0:47:58.48,0:48:03.23,Default,,0000,0000,0000,,it's because now we can make more explicit\Nwhat we expect from the data.
Dialogue: 0,0:48:03.23,0:48:06.89,Default,,0000,0000,0000,,Before, each one has to write\Nits own tools and scripts
Dialogue: 0,0:48:06.89,0:48:10.60,Default,,0000,0000,0000,,and so it's more visible\Nand we can discuss about it.
Dialogue: 0,0:48:10.60,0:48:13.64,Default,,0000,0000,0000,,But because it's not about\Nwhat's wrong or right,
Dialogue: 0,0:48:13.64,0:48:15.87,Default,,0000,0000,0000,,it's about an expectation,
Dialogue: 0,0:48:15.87,0:48:18.10,Default,,0000,0000,0000,,and you will have different\Nexpectations and discussions
Dialogue: 0,0:48:18.11,0:48:20.74,Default,,0000,0000,0000,,about how we want\Nto model things in Wikidata,
Dialogue: 0,0:48:21.25,0:48:23.10,Default,,0000,0000,0000,,and this...
Dialogue: 0,0:48:23.10,0:48:26.28,Default,,0000,0000,0000,,The current state is just\None step in the direction
Dialogue: 0,0:48:26.28,0:48:28.04,Default,,0000,0000,0000,,because now you need
Dialogue: 0,0:48:28.04,0:48:31.04,Default,,0000,0000,0000,,very much technical expertise\Nto get into this,
Dialogue: 0,0:48:31.04,0:48:35.72,Default,,0000,0000,0000,,and we need better ways\Nto visualize this constraint,
Dialogue: 0,0:48:35.72,0:48:39.100,Default,,0000,0000,0000,,to transform it maybe in natural language\Nso people can better understand,
Dialogue: 0,0:48:40.94,0:48:43.77,Default,,0000,0000,0000,,but it's less about what's wrong or right.
Dialogue: 0,0:48:44.92,0:48:45.92,Default,,0000,0000,0000,,(Lydia) Yeah.
Dialogue: 0,0:48:50.99,0:48:53.89,Default,,0000,0000,0000,,(man7) So for quality issues,\NI just want to echo it like...
Dialogue: 0,0:48:53.89,0:48:57.01,Default,,0000,0000,0000,,I've definitely found a lot of the issues\NI've encountered have been
Dialogue: 0,0:48:58.84,0:49:02.33,Default,,0000,0000,0000,,differences in opinion\Nbetween {\i1}instance of{\i0} versus {\i1}subclass{\i0}.
Dialogue: 0,0:49:02.33,0:49:05.96,Default,,0000,0000,0000,,I would say errors in those situations
Dialogue: 0,0:49:05.96,0:49:11.52,Default,,0000,0000,0000,,and trying to find those\Nhas been a very time-consuming process.
Dialogue: 0,0:49:11.52,0:49:14.84,Default,,0000,0000,0000,,What I've found is like:\N"Oh, if I find very high-impression items
Dialogue: 0,0:49:14.84,0:49:16.05,Default,,0000,0000,0000,,that are something...
Dialogue: 0,0:49:16.05,0:49:21.63,Default,,0000,0000,0000,,and then use all the subclass instances\Nto find all derived statements of this,"
Dialogue: 0,0:49:21.63,0:49:26.22,Default,,0000,0000,0000,,this is a very useful way\Nof looking for these errors.
Dialogue: 0,0:49:26.22,0:49:28.07,Default,,0000,0000,0000,,But I was curious if Shape Expressions,
Dialogue: 0,0:49:29.84,0:49:31.58,Default,,0000,0000,0000,,if there is...
Dialogue: 0,0:49:31.58,0:49:36.93,Default,,0000,0000,0000,,If this can be used as a tool\Nto help resolve those issues but, yeah...
Dialogue: 0,0:49:40.51,0:49:42.56,Default,,0000,0000,0000,,(man8) If it has a structural footprint...
Dialogue: 0,0:49:45.91,0:49:49.31,Default,,0000,0000,0000,,If it has a structural footprint\Nthat you can...that's sort of falsifiable,
Dialogue: 0,0:49:49.31,0:49:51.19,Default,,0000,0000,0000,,you can look at that\Nand say well, that's wrong,
Dialogue: 0,0:49:51.19,0:49:52.67,Default,,0000,0000,0000,,then yeah, you can do that.
Dialogue: 0,0:49:52.67,0:49:56.92,Default,,0000,0000,0000,,But if it's just sort of\Ntrying to map it to real-world objects,
Dialogue: 0,0:49:56.92,0:49:59.08,Default,,0000,0000,0000,,then you're just going to need\Nlots and lots of brains.
Dialogue: 0,0:50:05.77,0:50:08.63,Default,,0000,0000,0000,,(man9) Hi, Pablo Mendes\Nfrom Apple Siri Knowledge.
Dialogue: 0,0:50:09.15,0:50:12.77,Default,,0000,0000,0000,,We're here to find out how to help\Nthe project and the community
Dialogue: 0,0:50:12.77,0:50:15.64,Default,,0000,0000,0000,,but Cristina made the mistake\Nof asking what we want.
Dialogue: 0,0:50:16.47,0:50:20.05,Default,,0000,0000,0000,,(laughing) So I think\None thing I'd like to see
Dialogue: 0,0:50:20.96,0:50:23.52,Default,,0000,0000,0000,,is a lot around verifiability
Dialogue: 0,0:50:23.52,0:50:26.37,Default,,0000,0000,0000,,which is one of the core tenets\Nof the project in the community,
Dialogue: 0,0:50:27.06,0:50:28.59,Default,,0000,0000,0000,,and trustworthiness.
Dialogue: 0,0:50:28.59,0:50:32.41,Default,,0000,0000,0000,,Not every statement is the same,\Nsome of them are heavily disputed,
Dialogue: 0,0:50:32.41,0:50:33.65,Default,,0000,0000,0000,,some of them are easy to guess,
Dialogue: 0,0:50:33.65,0:50:35.54,Default,,0000,0000,0000,,like somebody's\Ndate of birth can be verified,
Dialogue: 0,0:50:36.07,0:50:39.08,Default,,0000,0000,0000,,as you saw today in the Keynote,\Ngender issues are a lot more complicated.
Dialogue: 0,0:50:40.20,0:50:42.13,Default,,0000,0000,0000,,Can you discuss a little bit what you know
Dialogue: 0,0:50:42.13,0:50:47.27,Default,,0000,0000,0000,,in this area of data quality around\Ntrustworthiness and verifiability?
Dialogue: 0,0:50:55.44,0:50:58.14,Default,,0000,0000,0000,,If there isn't a lot,\NI'd love to see a lot more. (laughs)
Dialogue: 0,0:51:00.65,0:51:01.65,Default,,0000,0000,0000,,(Lydia) Yeah.
Dialogue: 0,0:51:03.31,0:51:06.55,Default,,0000,0000,0000,,Apparently, we don't have\Na lot to say on that. (laughs)
Dialogue: 0,0:51:08.02,0:51:12.30,Default,,0000,0000,0000,,(Andra) I think we can do a lot,\Nbut I had a discussion with you yesterday.
Dialogue: 0,0:51:12.30,0:51:15.77,Default,,0000,0000,0000,,My favorite example I learned yesterday\Nthat's already deprecated
Dialogue: 0,0:51:15.77,0:51:20.28,Default,,0000,0000,0000,,is if you go to the Q2, which is earth,
Dialogue: 0,0:51:20.28,0:51:23.34,Default,,0000,0000,0000,,there is statement\Nthat claims that the earth is flat.
Dialogue: 0,0:51:24.18,0:51:26.06,Default,,0000,0000,0000,,And I love that example
Dialogue: 0,0:51:26.06,0:51:28.39,Default,,0000,0000,0000,,because there is a community\Nout there that claims that
Dialogue: 0,0:51:28.39,0:51:30.42,Default,,0000,0000,0000,,and they have verifiable resources.
Dialogue: 0,0:51:30.42,0:51:32.25,Default,,0000,0000,0000,,So I think it's a genuine case,
Dialogue: 0,0:51:32.26,0:51:34.64,Default,,0000,0000,0000,,it shouldn't be deprecated,\Nit should be in Wikidata.
Dialogue: 0,0:51:34.64,0:51:40.38,Default,,0000,0000,0000,,And I think Shape Expressions\Ncan be really instrumental there,
Dialogue: 0,0:51:40.39,0:51:41.83,Default,,0000,0000,0000,,because what you can say,
Dialogue: 0,0:51:41.83,0:51:44.86,Default,,0000,0000,0000,,OK, I'm really interested\Nin this use case,
Dialogue: 0,0:51:44.86,0:51:47.13,Default,,0000,0000,0000,,or this is a use case where you disagree,
Dialogue: 0,0:51:47.13,0:51:51.06,Default,,0000,0000,0000,,but there can also be a use case\Nwhere you say OK, I'm interested.
Dialogue: 0,0:51:51.06,0:51:53.45,Default,,0000,0000,0000,,So there is this example you say,\NI have glucose.
Dialogue: 0,0:51:53.45,0:51:55.84,Default,,0000,0000,0000,,And glucose when you're a biologist,
Dialogue: 0,0:51:55.84,0:52:00.18,Default,,0000,0000,0000,,you don't care for the chemical\Nconstraints of the glucose molecule,
Dialogue: 0,0:52:00.18,0:52:03.20,Default,,0000,0000,0000,,you just... everything glucose\Nis the same.
Dialogue: 0,0:52:03.20,0:52:05.97,Default,,0000,0000,0000,,But if you're a chemist,\Nyou cringe when you hear that,
Dialogue: 0,0:52:05.97,0:52:08.19,Default,,0000,0000,0000,,you have 200 something...
Dialogue: 0,0:52:08.19,0:52:10.44,Default,,0000,0000,0000,,So then you can have\Nmultiple Shape Expressions,
Dialogue: 0,0:52:10.44,0:52:12.72,Default,,0000,0000,0000,,OK, I'm coming in with...\NI'm at a chemist view,
Dialogue: 0,0:52:12.72,0:52:13.89,Default,,0000,0000,0000,,I'm applying that.
Dialogue: 0,0:52:13.89,0:52:16.69,Default,,0000,0000,0000,,And then you say\NI'm from a biological use case,
Dialogue: 0,0:52:16.69,0:52:18.52,Default,,0000,0000,0000,,I'm applying that Shape Expression.
Dialogue: 0,0:52:18.52,0:52:20.36,Default,,0000,0000,0000,,And then when you want to collaborate,
Dialogue: 0,0:52:20.36,0:52:22.78,Default,,0000,0000,0000,,yes, well you should talk\Nto Eric about ShEx maps.
Dialogue: 0,0:52:23.91,0:52:28.87,Default,,0000,0000,0000,,And so...\Nbut this journey is just starting.
Dialogue: 0,0:52:28.87,0:52:32.24,Default,,0000,0000,0000,,But I personally I believe\Nthat it's quite instrumental in that area.
Dialogue: 0,0:52:34.29,0:52:35.54,Default,,0000,0000,0000,,(Lydia) OK. Over there.
Dialogue: 0,0:52:37.95,0:52:39.17,Default,,0000,0000,0000,,(laughs)
Dialogue: 0,0:52:40.60,0:52:46.04,Default,,0000,0000,0000,,(woman2) I had several ideas\Nfrom some points in the discussions,
Dialogue: 0,0:52:46.04,0:52:50.90,Default,,0000,0000,0000,,so I will try not to lose...\NI had three ideas so...
Dialogue: 0,0:52:52.39,0:52:55.20,Default,,0000,0000,0000,,Based on what James said a while ago,
Dialogue: 0,0:52:55.20,0:52:59.00,Default,,0000,0000,0000,,we have a very, very big problem\Non Wikidata since the beginning
Dialogue: 0,0:52:59.00,0:53:01.57,Default,,0000,0000,0000,,for the upper ontology.
Dialogue: 0,0:53:02.36,0:53:05.34,Default,,0000,0000,0000,,We talked about that\Ntwo years ago at WikidataCon,
Dialogue: 0,0:53:05.34,0:53:07.43,Default,,0000,0000,0000,,and we talked about that at Wikimania.
Dialogue: 0,0:53:07.43,0:53:09.82,Default,,0000,0000,0000,,Well, always we have a Wikidata meeting
Dialogue: 0,0:53:09.82,0:53:11.66,Default,,0000,0000,0000,,we are talking about that,
Dialogue: 0,0:53:11.66,0:53:15.78,Default,,0000,0000,0000,,because it's a very big problem\Nat a very very eye level
Dialogue: 0,0:53:15.78,0:53:23.12,Default,,0000,0000,0000,,what entity is, with what work is,\Nwhat genre is, art,
Dialogue: 0,0:53:23.12,0:53:25.46,Default,,0000,0000,0000,,are really the biggest concept.
Dialogue: 0,0:53:26.20,0:53:33.12,Default,,0000,0000,0000,,And that's actually\Na very weak point on global ontology
Dialogue: 0,0:53:33.12,0:53:37.45,Default,,0000,0000,0000,,because people try to clean up regularly
Dialogue: 0,0:53:38.02,0:53:41.05,Default,,0000,0000,0000,,and broke everything down the line,
Dialogue: 0,0:53:42.52,0:53:48.65,Default,,0000,0000,0000,,because yes, I think some of you\Nmay remember the guy who in good faith
Dialogue: 0,0:53:48.65,0:53:51.78,Default,,0000,0000,0000,,broke absolutely all cities in the world.
Dialogue: 0,0:53:51.78,0:53:57.54,Default,,0000,0000,0000,,We were not geographical items anymore,\Nso violation constraints everywhere.
Dialogue: 0,0:53:58.72,0:54:00.28,Default,,0000,0000,0000,,And it was in good faith
Dialogue: 0,0:54:00.28,0:54:03.62,Default,,0000,0000,0000,,because he was really\Ncorrecting a mistake in an item,
Dialogue: 0,0:54:04.17,0:54:05.73,Default,,0000,0000,0000,,but everything broke down.
Dialogue: 0,0:54:06.35,0:54:09.37,Default,,0000,0000,0000,,And I'm not sure how we can solve that
Dialogue: 0,0:54:10.22,0:54:15.71,Default,,0000,0000,0000,,because there is actually\Nno external institution we could just copy
Dialogue: 0,0:54:15.71,0:54:18.49,Default,,0000,0000,0000,,because everyone is working on...
Dialogue: 0,0:54:19.15,0:54:22.04,Default,,0000,0000,0000,,Well, if I am performing art database,
Dialogue: 0,0:54:22.04,0:54:24.60,Default,,0000,0000,0000,,I will just go\Nat the performing art label,
Dialogue: 0,0:54:24.60,0:54:29.36,Default,,0000,0000,0000,,or I won't go to the philosophical concept\Nof what an entity is,
Dialogue: 0,0:54:29.36,0:54:31.20,Default,,0000,0000,0000,,and that's actually...
Dialogue: 0,0:54:31.20,0:54:34.56,Default,,0000,0000,0000,,I don't know any database\Nwhich is working at this level,
Dialogue: 0,0:54:34.56,0:54:36.83,Default,,0000,0000,0000,,but that's the weakest point of Wikidata.
Dialogue: 0,0:54:37.94,0:54:40.81,Default,,0000,0000,0000,,And probably,\Nwhen we are talking about data quality,
Dialogue: 0,0:54:40.81,0:54:44.03,Default,,0000,0000,0000,,that's actually a big part of it, so...
Dialogue: 0,0:54:44.03,0:54:48.57,Default,,0000,0000,0000,,And I think it's the same\Nwe have stated in...
Dialogue: 0,0:54:48.57,0:54:50.45,Default,,0000,0000,0000,,Oh, I am sorry, I am changing the subject,
Dialogue: 0,0:54:51.40,0:54:55.77,Default,,0000,0000,0000,,but we have stated\Nin different sessions about qualities,
Dialogue: 0,0:54:55.77,0:54:59.40,Default,,0000,0000,0000,,which is actually some of us\Nare doing good modeling job,
Dialogue: 0,0:54:59.40,0:55:01.24,Default,,0000,0000,0000,,are doing ShEx,\Nare doing things like that.
Dialogue: 0,0:55:01.97,0:55:07.66,Default,,0000,0000,0000,,People don't see it on Wikidata,\Nthey don't see the ShEx,
Dialogue: 0,0:55:07.66,0:55:10.39,Default,,0000,0000,0000,,they don't see the WikiProject\Non the discussion page,
Dialogue: 0,0:55:10.39,0:55:11.39,Default,,0000,0000,0000,,and sometimes,
Dialogue: 0,0:55:11.39,0:55:14.96,Default,,0000,0000,0000,,they don't even see\Nthe talk pages of properties,
Dialogue: 0,0:55:14.96,0:55:19.63,Default,,0000,0000,0000,,which is explicitly stating,\Na), this property is used for that.
Dialogue: 0,0:55:19.63,0:55:23.89,Default,,0000,0000,0000,,Like last week,\NI added constraints to a property.
Dialogue: 0,0:55:23.89,0:55:26.32,Default,,0000,0000,0000,,The constraint was explicitly written
Dialogue: 0,0:55:26.32,0:55:28.69,Default,,0000,0000,0000,,in the discussion\Nof the creation of the property.
Dialogue: 0,0:55:28.69,0:55:34.55,Default,,0000,0000,0000,,I just created the technical part\Nof adding the constraint, and someone:
Dialogue: 0,0:55:34.55,0:55:37.18,Default,,0000,0000,0000,,"What! You broke down all my edits!"
Dialogue: 0,0:55:37.18,0:55:41.54,Default,,0000,0000,0000,,And he was using the property\Nwrongly for the last two years.
Dialogue: 0,0:55:41.54,0:55:46.87,Default,,0000,0000,0000,,And the property was actually very clear,\Nbut there were no warnings and everything,
Dialogue: 0,0:55:46.87,0:55:49.92,Default,,0000,0000,0000,,and so, it's the same at the Pink Pony\Nwe said at Wikimania
Dialogue: 0,0:55:49.92,0:55:54.72,Default,,0000,0000,0000,,to make WikiProject more visible\Nor to make ShEx more visible, but...
Dialogue: 0,0:55:54.72,0:55:56.92,Default,,0000,0000,0000,,And that's what Cristina said.
Dialogue: 0,0:55:56.92,0:56:02.37,Default,,0000,0000,0000,,We have a visibility problem\Nof what the existing solutions are.
Dialogue: 0,0:56:02.37,0:56:04.24,Default,,0000,0000,0000,,And at this session,
Dialogue: 0,0:56:04.24,0:56:06.86,Default,,0000,0000,0000,,we are all talking about\Nhow to create more ShEx,
Dialogue: 0,0:56:06.86,0:56:10.73,Default,,0000,0000,0000,,or to facilitate the jobs\Nof the people who are doing the cleanup.
Dialogue: 0,0:56:11.60,0:56:15.84,Default,,0000,0000,0000,,But we are cleaning up\Nsince the first day of Wikidata,
Dialogue: 0,0:56:15.84,0:56:20.92,Default,,0000,0000,0000,,and globally, we are losing,\Nand we are losing because, well,
Dialogue: 0,0:56:20.92,0:56:22.96,Default,,0000,0000,0000,,if I know names are complicated
Dialogue: 0,0:56:22.96,0:56:26.16,Default,,0000,0000,0000,,but I am the only one\Ndoing the cleaning up job,
Dialogue: 0,0:56:26.66,0:56:29.67,Default,,0000,0000,0000,,the guy who added\NLatin script name
Dialogue: 0,0:56:29.67,0:56:31.58,Default,,0000,0000,0000,,to all Chinese researcher,
Dialogue: 0,0:56:32.09,0:56:35.62,Default,,0000,0000,0000,,I will take months to clean that\Nand I can't do it alone,
Dialogue: 0,0:56:35.62,0:56:38.78,Default,,0000,0000,0000,,and he did one massive batch.
Dialogue: 0,0:56:38.78,0:56:40.24,Default,,0000,0000,0000,,So we really need...
Dialogue: 0,0:56:40.24,0:56:44.16,Default,,0000,0000,0000,,we have a visibility problem\Nmore than a tool problem, I think,
Dialogue: 0,0:56:44.16,0:56:45.73,Default,,0000,0000,0000,,because we have many tools.
Dialogue: 0,0:56:45.73,0:56:50.26,Default,,0000,0000,0000,,(Lydia) Right, so unfortunately,\NI've got shown a sign, (laughs),
Dialogue: 0,0:56:50.26,0:56:52.12,Default,,0000,0000,0000,,so we need to wrap this up.
Dialogue: 0,0:56:52.12,0:56:53.56,Default,,0000,0000,0000,,Thank you so much for your comments,
Dialogue: 0,0:56:53.56,0:56:56.61,Default,,0000,0000,0000,,I hope you will continue discussing\Nduring the rest of the day,
Dialogue: 0,0:56:56.61,0:56:57.84,Default,,0000,0000,0000,,and thanks for your input.
Dialogue: 0,0:56:58.36,0:56:59.94,Default,,0000,0000,0000,,(applause)