Hello @rebelethno , this is just to let you know that the final version of the POPREBEL dataset – the interviews, in three languages – are now published on Zenodo. It’s an impressive ethnographic dataset: 8,019 posts with 570,112 words, 12,189 annotations 156 informants, 1,509 codes,
The previous versions already have over 2,000 downloads (but that really means 400, since the whole dataset includes 5 files).
Awesome! But also shouldn’t Jitka and Djan and Jirka be listed as co-authors?
Indeed, well spotted. I added Jitka’s info from the other dataset, but I am missing information from @Jirka_Kocian and @Djan. I need:
@alberto Wonderful! Another significnat step forward. Alberto: I wrote to you inside of the Deliverable. Do you think it would be possible to prepare a few new (simple) visualisations in Tulip? I spent some time “eyeing” the whole document and it seems to me that we have some imballance in the “visualisation dentity” throught the text. For now, I am insetting what @Wojt prepared for me in Graphryder.
Sure. Hit me with a list.
Is this format OK? If that workds, say I will suggest 3-4 more, OK?
uh… no, those screenshots are unreadable. I would like the following information:
For example: “Polish corpus, ego network of Donald Tusk, d > 4”.
Remember that Graphryder and my Tulip scripts use slightly different definitions of d, so the graphs might not be identical.
OK, will need a few hours to prepare this, as some pre-Christmas chores are waiting… And need to think very carefully, so I do not waste your time. Thanks for your readines to do this!
But to begin:
Polish corpus, ego network of DDpolcor (political corruption), d > 7. The point is to see only three other nodes, Law and Justice Party, Civic Platform, and the Catholic Church
Discourse Admin Manual