Hi Dom --
Thanks for the link.
I read the editorial.
I was drawn to the mixing of blood between cheap specimen athletes, trying to figure out what again, about performance enhancing drugs? That's a polynomial exactly how? He stops short of explaining.
My strategy has been more to share, in the open source tradition, big pieces of what other countries are planning and implementing, plans to whip America's ass in the globalization game, unless big dummy decides he's entitled to enslave using brute force, which would require the demise of the USA as a value system, so with what army?
So it's not like Sputnik, where we just overnight launch this bleep bleep bleep thingy and everyone panics and tries to learn base 5.
No, I "leak" like the journalists like, talk about how the RSA is building a pipeline around personal avatar software, complete with on-line think tanks, very safe because sourced in and by the schools.
We're not talking exposing the kids to cybercreeps in some uncontrolled Internet environment. Telecasts, optical fiber... of course not every city gets to be Baghdad.
And still the Americans wallow, not lifting a finger to save themselves from their fate. Must be destiny or somethin', to have Big Dummy Textbooks (BDTs) around your necks, no A & B modules, no MITEs, and throw yerselves off a bridge, or whatever it is you're up to (I think the world would like to know some days -- looks idiotic, whatever it is).
Anyway, modeling is important. But the emphasis on closed form differential equations based in some axiomatic understanding ala e = m c c, just isn't the way to go in a lot of knowledge domains. You need a more neural network like approach, plus all these bit buckets with tweakable feedbacks like in Sims. It's almost like tuning a theramin (or a not-a-theramin), i.e. some analog musical instrument, but it's digital these days, maybe written in Python (more likely the engine is written in something faster, but with Python bindings for modelers wishing to use that language).
Here in Portland, there's a lot of emphasis on modeling stemming from Jay Forrester's studies at MIT (remember the Club of Rome?), more recently manifest in such affordable modeling apps as Stella. We've had statewide championships among modeling teams, showing up at OMSI to showcase their animated flows, of the prison population for example, which swells in proportion to our not teaching any modeling, or other math worth learning, in too many schools (yours for example?).
Kirby