On 10/12/19 1:32 PM, Timon Walshe-Grey wrote:
I think your graphing software had a bit of a hiccup in the bottom right
corner. :P
I noticed that - it appears I just got unlucky, since it looks like
there's some randomness in the generation of the graph, but I didn't
realize that until after I sent my last email.
Seriously though, this is pretty interesting! You can kind of see which are the
most integral definitions (power, offices, switches, Agoran decisions, etc.,
plus the ones you've mentioned already) vs. which are more just implementing
other mechanics (Win by Paradox, birthday gifts, HAN/GHAN, ratification w/o
objection etc.)
Is there any chance you could publish your dataset? Would be fun to have a poke
around and see if there are any other insights available.
I'm working on a writing up a thesis for this. I'll publish my code
concurrently (not a pledge), at which point you could play with it all
you want. Some other analyses I've considered:
- Number of dependents by rule
- Number of dependencies by rule
- Circular dependencies
I'm sure there's a bunch of interesting stuff that I'm missing.
--
Jason Cobb
