[extropy-chat] Philosophical phylogeny

[Anders Sandberg]

Jay Dugger wrote:
> I eagerly await seeing more memetic phylogenic trees, Anders. Even
> more interesting: your methods and tools for generating these maps. If
> you've posted this already, where?

My current setup consists of simple files with rows consisting of
"name1" "name2"
"name3" "name4"
...
This read by a matlab script that puts the names into a set, and then
prints a file in the GML language with nodes and edges. The map layout is
then done with the wonderful program yEd, free for downloading and fully
functional. I only wish the clustering algorithm was documented.

Making the initial file and fiddling with the layout coloring and style is
what takes most time.

> I've not studied graph theory, so forgive an ignorant question. Does
> there exist a way to measure messiness?

Hmm, not that I know of. But one could compare to a random graph. If you
calculate the diameter and local clustering (the average number of
triangles each node is in), in a regular localized graph the clustering is
high and the diameter is high. In a small world graph diameter is small
but there is much clustering. In a random (messy) graph the clustering is
small and the diameter is small. Graphs with low clustering and high
diameter are probably tree-like (not messy?).

> Finally, for your transhumanist phylogeny do you plan to map an
> overlap between science fiction authors and non-fiction works? I think
> in particular of three examples. Bernal's influence on Stapledon;
> Danridge Cole's & G.K. O'Neill's influence on Zebrowski, Sterling,
> et.al.; and Scott Westerfeld's political visualization tools on your
> own similar work.

Yes, I don't think it makes sense to leave them out. And thanks for the tips!


-- 
Anders Sandberg,
Oxford Uehiro Centre for Practical Ethics
Philosophy Faculty of Oxford University