[13:54:00] aloha :) [13:54:15] ciao dsaez - this may be of interest: https://news.iu.edu/stories/2019/09/iub/releases/12-botslayer-launch.html?fbclid=IwAR1xzze9N61EfUdTQfdgC4eJedQRIpC_GxrGkyTvzpdz7vHbJkWCVjnZ6a4 [13:56:29] cool, thx [15:02:40] o/ [17:42:59] dsaez: when you get a chance, could you point me to your code for the (flask?) app running the demo section recommendation app? [17:47:37] leila: outcome from today's meeting with New Readers is that there's a desire to run demographics surveys in Portuguese, Bahasa Indonesia, and Hindi and potentially volunteer translators to help with this. I'm going to do my best to support. Hindi and Portuguese at least the surveys are partially translated already [17:48:52] isaacj: do you want to include them as part of your next deployment in September? [17:49:10] if the volunteers are available in the next week or so, then yes i think that makes the most sense [17:49:43] isaacj: sounds good. Keep in mind that you can ask them for engineering resources for turning the survey on and off. [17:49:59] isaacj: that's a straightforward task and they can support, I assume, if they are interested in the results. [17:50:08] (it may take you longer to explain it, in which case, don't.;) [17:50:12] those languages are interesting to us too as they help expand the geographic diversity of the current surveys. and yes, that's a good point re engineering resources, especially if this continues to grow [17:50:18] and thanks for meeting with them and the update. [17:50:42] np - i should have done it earlier but better late than never [17:50:44] continues to grow ; ) [18:32:52] dsaez isaacj: do you know if someone in our team has given advice to the team that is working on IP masking? Mako has named me in https://meta.wikimedia.org/wiki/Talk:IP_Editing:_Privacy_Enhancement_and_Abuse_Mitigation#Consider_the_effect_on_researchers_using_Wikimedia_data/metadata , especially because he mentions the WMF research team and our potential involvement [18:33:35] Niharika: I just noticed you're here and you may be able to respond to the above, too. ^ [19:13:17] leila: Hey! I was just in a meeting with Mako and Kaylea. I have also talked with Aaron and Jonathan over email about this briefly. I would love to setup a meeting with you -- maybe once you have more feedback about the sockpuppet detection model? [19:14:12] We can talk about the general project and also potential sockpuppet detection tools we will be building. [19:32:50] isaacj : check this https://github.com/digitalTranshumant/wikiImagesDistance/blob/master/server.py [19:33:30] isaacj: and then this https://github.com/digitalTranshumant/wmf-interlanguage/tree/master/SectionRecommendation/app [19:33:45] and finally this: https://github.com/digitalTranshumant/wmf-interlanguage/blob/master/SectionRecommendation/app/test.py [19:33:58] isaacj , have a look and we can talk about that later or tomorrow. [19:34:20] leila, I don't know [20:06:11] thanks dsaez ! good idea [20:07:42] leila: I mentioned this to cwylo when I first heard about the proposal but did a poor job of following up on the formal talk page. it's a very good point by Mako (a lot of economics-esque research uses geolocated edits as well) and I've been meaning to follow up . also points to the growing importance of aggregated geo-editor datasets [21:31:11] dsaez: 'noted down. isaacj: yeah. it's fair that the assessment should include the cost for researchers. [21:32:27] Niharika: Sure. I'll leave a comment on the talk page then to acknowledge that I've ready it, I agree that assessing the impact on researchers make sense, and that we would be happy to chime in on that part of the assessment. [21:33:12] leila: Sounds good. :) [21:33:15] Niharika: re sockpuppet detection, I'm happy to talk at any point, we're basically iterating right now over the model and Srijan is going to refresh the model based on the latest feedback we have received from (one) checkuser [21:34:03] leila: Got it. Are you expecting to hear more responses? [21:40:24] Niharika: not immediately. we have told them not to spend more time on it until we iterate on the model and send fresh data based on the feedback. [21:41:04] Niharika: I expect the following couple of months be spent on back-and-forth with them to see how we can optimize the model based on their needs. [21:41:54] leila: Got it. I'll schedule a meeting with you to talk more about it. :) [21:42:05] Niharika: sounds good. : ) [21:44:51] leila: Done! Feel free to move it around if the time doesn't work for you. [22:01:21] Niharika: sure. thanks!