Populist discourses on sexuality (Poprebel subproject)

[alberto]

Hello @SZdenek, this is really exciting. We stand ready to help. Let’s see:

@SZdenek

(1) First, is it OK for you/Edgeriders to have posts/content that is not produced by individual people as it is other cases?

Yes, we can treat any text data, and audio and video too. However, it is important to keep different types of data separated from each other, which brings me to…

@SZdenek

(2) If yes then, secondly, is it possible to create a separate project, maybe within Wellbeing in Europe, but detached from Ethno POPREBEL?

Absolutely. With OpenEthno as it is now I would suggest creating a dedicated category, for visual separation, and adding to each topic a Discourse tag like #ethno-poprebel-media. However a reform of Open Ethno is under way, one that POPREBEL ethnographers suggested: we will now create a “project” database unit, and assign codes to a unique project. People will be able to reuse codes (including all the annotations made with those codes) in any project, but only by duplicating them. Which solution to use depends on the time scale of your idea. We will coordinate this with the devs.

@SZdenek

(3) Thirdly, should we create as many users to put the posts online as many media sources we have? (Similarly as it is done in Poprebel with the interviews were Mania and Jitka create a person and post the interview in their “name”?)

This depends on the nature of the data. If you perceive these data as originating from one source (“the media”), there is no need to differentiate. If you want to keep track of the identities of different sources, then yes, you would create an account on the platform for each source.

@SZdenek

We assume this will help to more easily create different SSNA visualisations by filtering based on the tags. Do we get it right?

It does not have to be tags. Here I used categories (“fora”) to encode languages. But yes, you could also use tags, it just needs a modification in the Python code. Graphryder would give you, by default, a graph with all co-occurrences in the corpus defined by #ethno-poprebel-media, and at the moment we have no way to sorting out the languages. But I can do that work with Tulip; also, I just asked Jan for some extra resources to support activities, one of which is tech support on Graphryder the way we have done it on OpenEthnographer. If that comes together, we might have some margin to extend the functionalities of GR.

@SZdenek

(5) Lastly, we will have definitely some codes that are already present in the Edgeryders, such as “activists”. But we might have it in different hierarchies.

Hierarchies are DAGs, so by construction any child can only have one parent. If you want to subvert the hierarchy, then yes, you will have to duplicate existing POPREBEL codes and assign them to different parents. But if that were to happen, it might mean that the herarchies are being stretched: my idea of hierarchy is the superset, not the arbitrary grouping. For example, dogs, beavers and giraffes can be children codes of mammals uncontroversially. The moment you start grouping second Punic war, War of the Two Roses and Napoleon's Egypt campaign under unjust wars you might be stretching it a bit. Also see above about re-using codes across different projects.