[11:24:52] miriam: o/ [11:25:00] good afternoon leila :) [11:25:10] miriam: hi. hi. :) [11:25:14] miriam: how are you doing? [11:26:04] good, trying to improve the presentation for Boston :) I think it's too long to fit in 20 mins - as usual [11:26:13] how are you doing? [11:26:49] miriam: 'got it. I have to start working on the presentation, too. [11:27:29] miriam: I'm still processing emails and moving forward a few urgent items including prep for some interviews for today/tomorrow. [11:27:47] miriam: I ran into a wikimedia-l discussion that I thought you may want to be aware of. [11:27:51] * leila looks up the link [11:28:37] leila: ok! let me know if I can help for the presentation, my understanding is yours is going to be about research as a whole? [11:30:28] miriam: https://lists.wikimedia.org/pipermail/wikimedia-l/2019-June/092867.html , I /think/ refers to the analyses related to the missing citations. [11:31:29] miriam: I'm not sure how much we want to engage on the thread, given that the focus/point of it is not very clear to me, but if the statements made in this specific email have inaccuracies, it's worth to point them out on the thread to intervene the further spread of it. [11:32:13] miriam: One thing that immediately comes to my mind is that the analysis is not considering all languages. The other question I have is: what are the percentages of missing citations and how accurate these estimations are for articles such as biographies? [11:33:29] so leila this refers to a straight-forward analysis we did on the dumps, essentially counting how many articles have 0 tags (or similar indicators of references) [11:33:42] leila: so indeed 1/4 articles in english Wikipedia has 0 references [11:34:44] leila: we didn't mention anything about biographies, also because it's very hard to isolate biographies (e.g. with halfak's the topic model, they would be tagged as "Language and Literature") [11:35:13] miriam: in general there is a separate line of (counter-)argument to the issue of Wikipedia's reliability being discussed on the full thread which is: it really depends how you define reliability. reliability can be different than accuracy, and with some specific definitions of it, the fact that Wikipedia is being used broadly by may audiences is a clear signal (with all the issues that we know exist) that it's reliable. [11:35:13] miriam: (maybe the discussion between verifiability and reliability is mixed in the thread.) [11:35:13] miriam: ok. I'll leave it to you to decide if you want to interject on the thread or not. [11:35:35] miriam: right. [11:37:53] leila:this is the analysis they are referring to: https://meta.wikimedia.org/wiki/Research:Characterizing_Wikipedia_Citation_Usage/Second_Round_of_Analysis [11:39:02] miriam: thanks. do you think it's important to clarify anything from the study on the thread? [11:42:23] leila: probably that we can't quantify how many BLP are there! [11:45:16] miriam: ok. then that alone is perhaps not pointing out. it's a complicated thread to begin with and requires deep engagement if we are to enter. Let's see how it goes. :) [11:54:36] leila: I haven't read it in full [11:55:18] leila: just the last few email exchanges. So let me get deeper into that and see if/how we want to chime in [12:28:42] miriam: sounds good. After reading the last email on the thread, I'm fine with not engaging.