search

tedawson / Networks-of-others

Materials for network analysis of Lives of Others, etc
0 stars 0 forks source link

Should have protocols (show where edges come from) #1

Open tedawson opened 7 years ago

tedawson commented 7 years ago

I think we should include some sort of text file showing when interactions occur in the film (justifying edges). e.g.:

4:08 Hempf speaks to Wiesler

Maybe this wouldn't need to include time, cause that would take forever... I have this for the first part written out by hand already, but need to come up with a consistent way to show relationships. Also, need to do the second half of the film...

If we decided to get fancy, we could write a script to pull from the protocols onto a csv of edges.

msheehan-DH commented 7 years ago

This might be a good question for Mickey and Bobby. What kind of data do we need to create and deliver to justify our graphs (and so others could recreate our graphs from the data)?

I have access to some student workers who might be able to do more of the (cough) grunt work. One computer science student is working to create OCR files from Fraktur (no guarantee this project will actually function). I can also get him to start cataloguing all these interactions. He could flag questionable instances for our review. I can look into that option...

On Aug 31, 2017, at 10:28, Ted Dawson notifications@github.com wrote:

I think we should include some sort of text file showing when interactions occur in the film (justifying edges). e.g.:

4:08 Hempf speaks to Wiesler

Maybe this wouldn't need to include time, cause that would take forever... I have this for the first part written out by hand already, but need to come up with a consistent way to show relationships. Also, need to do the second half of the film...

If we decided to get fancy, we could write a script to pull from the protocols onto a csv of edges.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tedawson/Networks-of-others/issues/1, or mute the thread https://github.com/notifications/unsubscribe-auth/AOqqOywWJ_OUnxppnAKWBtH11CisrOXYks5sdtEwgaJpZM4PJATQ.

tedawson commented 7 years ago

Agreed, @casadm and @bobbysmiley will probably be able to give some guidance on what data should be made available and how. I'll go ahead and invite them to collaborate on this repo.

Student workers doing grunt work sounds generally awesome... I think the actual cataloging of interactions from human readable format to CSVs can now be done by this critter I'm calling "Gephicate" (trying to decide whether to have the readme written in scatological language), but I'm sure we can find stuff for undergrads to do... possibly including watching the film and marking interactions, or at least proofing our work on that front.

Let me know how the Fraktur OCR thing goes. Sounds fun.