Open ethno data: towards a new Masters of Network?

Notes:

  • April
  • Secound data set from DLab

What to analyse and why?

  • 2 layers: Organisation + Hackathon tracks.
    • Hackathon Tracks are there to give people and idea: Usual format: You can either join or propose a track. Only requirement is you do a projects that is going to be represented around a graph. EX: “I want to form a team around ‘Can we see weather the view of technology is more dystopian or utopian?’” You just have to have a question, and then use the methodology to move from here to there.

NEEDS:

  • We need to define summaries of what the conversations are about, broadly.
  • We need to have people propose tracks.
  • We need to prepare the data.
  • We need to facilitate and connect people and groups running up to the hackathon.
  • We need to support people with specialised expertise to solve issues.
  • We need to facilitate the hackathon.

Hackathon Facilitation:

Internal Research Question:

  • How should the “service for academic research" GraphRyder look like? (RezNet)

Wednesday kickoff event (April 28th) and Friday for final presentation (April 30th) instead?

1 Like

Of course you are right. Had not adjusted that. Edited now :slight_smile:

The synchronous timeslots are rather uncomfortable for @johncoate timezone wise, so it would in fact probably be good if @Leonie or so could maybe take on a bit of the moderation/hosting role as @amelia suggested if you find you need that.

Also, Maria will be in her maternity leave then. What is it I am being signed up for? My NGI contract this year is very small in terms of hours (about 40 minutes per weekday), so please factor that in.

MCing Day 1, essentially.

I would say this takes precedence.

ok.

This is around the time I will have my PhD thesis defence. Once I receive the official date, I can let you know when and how I can help with this!

3 Likes

Draft post for the new MoN here. It’s called 5 because it comes after 4, but I am pretty sure it will actually be the 6th.

  1. Venice (intro)
  2. Rome (policy)
  3. Rome (online communities)
  4. Bordeaux (ethnography)
  5. Brussels (Networks of Care)
  6. Online (Networks of meaning)

Oh darn I see this far too late, and the invitation too.
January is crazy here, I apologize!

Hello all, reviving this topic as we move towards MoN.

We now have a date, a schedule, a minisite. People have started signing up (thread).

What we need:

  • Confirmation of the Gather space. @melancon, did you get around to set it up? If not, would it not be safer to drop it (since it would be a new tool for most of us), and default to Zoom rooms (the devil we know)?
  • More ethnographers! @amelia, I recall you saying that would not be a problem. What did you have in mind? Also, what are your thoughts about involving our own ethnographers? FWIW, I would recommend that Edgeryders ethnographers would get their hands dirty playing with graphs.
  • @Jan and @Richard, we have been discussing POPREBEL people participating too, perhaps in the visualization challenge or perhaps with its own challenge. What are your thoughts? Still interested?

Do people need a 30 mins catch-up meeting?

(Ping also @johncoate, who is going to be doing community management).

I think this could be useful for the POPREBEL ethnographers. @Maniamana, @jitka.kralova and @Djan, would this be of interest?

2 Likes

Yes, we would love it if you three would join – and it would really help you with building the visualisations of your data that you’d like to see. If you can make it a priority, please do.

@alberto, are you free during our POPREBEL ethno bi-weekly tomorrow? If you could come introduce the MoN concept and explain this year’s hackathon to the team, that would be really helpful!

Involving our own POPREBEL ethnography team was exactly what I had in mind when I said this, because given where we are in the project this will be really useful for their research. :slight_smile:

@amelia and @Richard sounds interesting :slight_smile:

which data exactly would @Maniamana, @jitka.kralova and myself be visualizing? Wouldn’t the team coding our interviews be visualizing the data? Thanks

I am already quite well acquainted with SNA and SSNA, and really deep into ethnographic fieldwork now, so not sure whether I will have time to play around for fun with a data-set that is still a work-in-progress;) Also, is this event like an ER workshop for all, or a part of the study’s methodology?

Both. It is a chance to work closely in a two-and-a-half days sprint, in interdisciplinary teams with at least one ethnographer and one network/data scientist. The idea is to use it to (1) get a better, more hand-on knowledge of the raw data and the process that produces the graphs from them, and (2) harden the methodology. Example expected results of 2 are: more informative approach to visualizations; and rules of thumb to assign interpretations in terms of semantic to algorithmically detectable network configurations.

And it is open to all, on the odd chance that some smart person no one knew walks in and gives us a brilliant idea.

Your call. But POPREBEL data are here: https://zenodo.org/record/4562451

When it comes to POPREBEL I am not sure if there is enough data yet to draw any significant conclusions, so perhaps such workshop should be taking place later in time, when we have more concise data, so our work doesn’t go in vain? But if it is just to be an exercise for learning purposes, not really relevant to the study itself, that’s a whole different story, especially as we are not sure who will be on the data science/network science side in those collabs, what are their skills, what do they know about the project, which I guess really influences the quality of the outcome, whether it might be of any use at all.
I would like to hear more on this idea, especially as I thought there was a plan to have a more regular network analysis workshop aimed for the POPREBEL team members that do not know the method well. Perhaps, if you are able to join us tomorrow, that would be great :blush: :blush:

1 Like

The data are not that small. You will be pressed quite hard to find annotated corpora of this size.

ethno dataset posts authors annotations codes
NGI Exchange 3,935 311 5,551 1,096
POPREBEL 2,206 313 5,679 1,448
OPENCARE 3,676 270 5,769 1,609

But whether it is quality data, that’s a whole different story - at least when it comes to non-English fieldworks, which are crucial to this part of the project.

1 Like