[12:45:01] Hey, It's just been mentioned on Discord and I wonder if any of you guys have thoughts. We wondered if wikimedians were disproportionately autistic or ADHD (or other similar condition). I also think it'd be interesting to see what type of editor they fell into (content curator, technical, WikiGnome). [15:00:55] @RhinosF1 i'm not aware of any research. primarily because it would require surveying wikimedians about some pretty sensitive information and when people have suggested work similar to this, it generally has not been clear what the benefits are or that the data will be handled appropriately [15:00:56] e.g., see a recent example: https://meta.wikimedia.org/wiki/Research:The_Construction_and_Application_of_Personality_Profile_Based_on_User_Behavior_in_Wikipedia [15:01:48] isaacj: we said we didn’t know how comfortable people would be with it [15:02:30] yep [15:20:17] isaacj: we thought it would be interesting to know but wasn’t sure [15:20:42] We kind of expected it to be slightly disprportinate [15:28:39] isaacj: is it worth throwing something out to the research list? I agree with the there’s no clear benefit, more inquisitiveness. [15:41:42] you're welcome to. i'm certainly not an expert about this area but people might know of studies with different (but related) platforms. i do frankly lean towards discouraging some of this research but others might be aware of past work [15:43:07] Will do [15:49:21] hi everyone. [15:50:22] djellel and J-Mo: I asked mgerlach to capture his contributions to the Research Showcase as a goal for the coming 3 months. His focus will be on building the list of speakers for July-December with Djellel and inviting them. FYI only. [15:50:36] isaacj: sent - https://lists.wikimedia.org/pipermail/wiki-research-l/2020-April/007209.html [15:51:23] RhinosF1: :thumbs up: [15:51:41] :) [16:16:03] miriam, yt? [16:16:48] dsaez hola :) [16:18:46] hola! [16:19:16] miriam, do you know any service that can return a category or any kind of label for a given reference? [16:21:10] dsaez, that's a good question! we tried in the past with tiziano to get some topic category for webites in Wikipedia references, and we pretty much failed.. existing systems work very poorly. [16:21:58] dsaez we tried with Alexa Web services but it was missing 70% of the websites and also their category taxonomy was not useful for our purpose [16:23:01] which kind of categories do you need? this new article that assigns reliability scores to references was just published today: https://www.preprints.org/manuscript/202003.0460/v1 [16:23:58] if you need to classify books, OCLC classify API might be helpful: http://classify.oclc.org/classify2/ [16:24:55] miriam, got you, I'll have a look on that, thanks! [16:25:25] also mgerlach has worked on topic modeling for scientific articles, maybe he knows more on that front [16:25:59] dsaez: if you want I can share with you the results from the Alexa classification [16:54:05] miriam, there is an Alexa API? [16:59:22] dsaez yes [16:59:25] https://aws.amazon.com/awis/ [16:59:30] but you have to have an aws account [17:00:57] dsaez but for most enwiki references I have the results already [17:48:07] miriam, but I want to run it for covid related pages, those are new, you think you have them? [18:52:02] isaacj: we’ve had some great opinions on the thread! [19:02:36] RhinosF1: would agree -- glad you posted it. some research i was not aware of and now we'll hopefully have a better answer for if people raise the question in the future [19:04:28] isaacj: yeah [19:07:24] dsaez: do you need something clean you can immediately use or is it useful if I share with you a list of initiatives that are generating scores/labels for sources? [19:11:58] RhinosF1: re the stackoverflow survey, I didn't mean to suggest that we use it out of the box. some parts of that needs a redo if you are interested to think about running such a survey in the Wikimedia world for sure, as you say. :) [19:12:24] leila: yeah, still shows a lot [19:12:49] I think I took part in that [19:12:49] for sure. good luck! [19:12:58] niiice! [19:13:17] I’m still seeing what people have gathered before [19:13:45] If it’s worth proper research and is appropiate, it could happen. [19:14:37] makes sense. come chat with us about it whenever you want. (and we do have office hours on the 4th Wednesday of every month if you want to drop by for a deeper dive together.) [19:15:13] I’ll make sure it’s in the diary [19:15:29] I’d be happy to help put something together [21:45:40] leila, please share them, I'll have a look [22:30:46] dsaez: there are 8 of them. here you are [22:30:50] https://assets.ctfassets.net/tlowcqj4pb76/4lmUdUz36gQuOKO0UOwEU2/819497f46f25a9cfcbaa7d4b5db8e354/CredCoWebConf2018.pdf [22:30:58] https://disinformationindex.org/2019/05/measuring-up-designing-the-index/ [22:31:09] https://ifcncodeofprinciples.poynter.org/know-more/the-commitments-of-the-code-of-principles [22:31:17] https://jti-rsf.org/fileadmin/Redaktion/documents/JTIForPublicComments.pdf [22:31:24] https://www.newsguardtech.com/ratings/rating-process-criteria/ [22:31:38] https://api.newsq.net/ [22:31:47] https://www.trustmetrics.com/how-it-works-interior [22:31:55] https://thetrustproject.org/ [22:32:00] thx leila, I'll have look. [22:33:34] dsaez: sounds good. some of them may not have public free APIs. Whichever you want, I'm happy to send an email to and ask for a special rate or free access. I would expect they would like to collaborate and help us in this space. [22:52:10] dsaez: can you or miriam put a 15-min on my calendar to talk about internships? I have some updates and would like to share with you to make a decision together. [22:52:34] if it helps, I'm happy to do it before the virtual coffee time on Friday (tomorrow)