[megameter]

I believe the most troubling thing about Crawford's path is simply in the inability to develop self-critique of his own philosophy. His thoughts on a subject seem to terminate in the thing of having a mathematical model of a topic, not what we get out of that model. It does not seem to matter if the model is inscrutable when presented within a system, or if the system degenerates into a single strategy. (I have a memory of playing "Balance of the Planet" and after struggling for some time, discovering that the model did not restrict my taxation of dirty energy. Therefore I could gain a nearly infinite budget to clean up the planet on turn 1 with no negative consequences beyond "people falling off roofs while installing solar panels." I'm not even kidding - for some reason roofing accidents are ranked up there with deforestation and carbon release as very important things to model about our impact on the planet.)

Plus, last I heard, he's still stuck on an evo-psychological model of society that is quite out of fashion these days, which doesn't exactly help matters.

Crawford's story is a good warning for anyone who embraces simulation as an "end in itself", rather than a medium, though. This was an idea in vogue with wargaming's golden era and is now carried forward by VR enthusiasts, among others.

[notahacker]

My first thought on reading him talking about equations to estimate numbers of fighters based on a few parameters is that there are entire subfields of academia devoted to debating regularities like this: most of economics, most of modern IR theory, substantial parts of psychology, sociology and politics. And the standard criticism (within the subfield, at least) isn't that there is something inherently wrong with expressing part of the world as a mathematical model, but that the mathematical model chosen is wrongly specified. The world isn't hostile to the concept of simulations, but it does have a horrible tendency to produce data that casts doubt on their accuracy.

[danielscrubs]

I’m always baffled by how little self proclaimed geniuses actually read current research papers.

It’s almost like they protect their pride by not challenging themselves, thereby being able to say they are knowledgeable with a conscience.

[TheAceOfHearts]

Do you have advice on how one can get more into reading interesting research papers in Computer Science and Software Engineering? I've previously tried subscribing to ACM but found that too many of their articles weren't of much interest or relevance to me. I definitely do feel like I've gotten lazy and that I don't challenge myself enough, but it's hard to break out of these patterns so I'd welcome any suggestions or recommendations which could help me grow.

I usually start reaching for research papers when I'm aware of a specific problem and I'm looking for the different approaches that have been taken to try and solve said problems.

Every few months I visit the NIST website and I browse through their catalogue to pick out interesting articles and publications to add to my queue.

[UncleMeat]

Most papers are available on author websites or arxiv. I'd check out the titles of papers in recent software engineering conferences (ICSE is a great starter) and read what you are interested in. For CS more broadly you'll need to start with a field and then find conferences since conferences are broken up by topic.

[danielscrubs]

I like to check the award section of “best thesis paper” of my local university for fun and then at work I have to read research papers in my area but then I just use a search engine to find it.

Start small and don’t take it too seriously! Little by little, just as learning a new language.

[amatic]

His article "Am I a genius" is also interesting: http://www.erasmatazz.com/personal/self/a-genius.html

> the main reason for this is that I’ve made no attempt to sell the idea. I simply wrote it up and put it on my website. I suppose that, were I to jump through the appropriate hoops, I could garner more interest for the idea. But that is beneath my pride; I am a thinker, not a salesman. I refuse to promote myself. I put the idea before the world and the world can take it or leave it. The world mostly leaves it.

I've heard some other people blaming a lack of "sales" for their ideas not spreading. As if you can not sell. Any sort of presentation is sales, if your product or idea is out there, it is selling itself.

Maybe the focusing on sales would, ironically, bring the understanding of what people find impressive in ideas, in games or products, and it might be completely different from what we thought before.

[notahacker]

The most important bit of doing any kind of complex sale is listening to others.

Reading the "most important" idea article that he made no attempt to sell is also illuminating. It's a mildly thought-provoking blog about how many fields have concepts of state and state change, and the concepts are interdependent and blur at the edges, leading to an unsupported conclusion that we think too much about data and inputs and not enough about CPUs. It would probably get a few upvotes and a few confused replies on LessWrong, but there's not really much for the computer scientists and creatives he clearly hopes will take notice to work with. Perhaps they might find a different version or some of his other ideas more valuable

[joshspankit]

Sales is morally bad: this idea that in order to sell, you must manipulate. Eg: “Used car salesmen” (said with disgust)

Money is evil

A bit to the side of the OP, but I feel like these two narratives have been particularly damaging to our planet’s long-term growth. To your point @amatic: If the author had let go of pride and at least explored the idea of effective communication with people as a way to spread ideas (aka sales), he may be far more objective in his self-appraisal and certainly gain perspective on how his ideas can be applied by others.

[partyboat1586]

>Plus, last I heard, he's still stuck on an evo-psychological model of society that is quite out of fashion these days, which doesn't exactly help matters.

>quite out of fashion

Yeah I don't get the impression this guy is into intellectual fads. If anything he's a contrarian and will deliberately take the opposing position to the mainstream.

Evo Psych is out because a bunch of idiots (and self proclaimed non intellectuals) use it to justify and explain everything and anything with stories and no evidence. Meanwhile real research continues but is sidelined due to this bad reputation.

Understandably contrarians see this happening and immediately take the position of Evo psych because it allows them to do a lot of hypothesis generation (their favourite pass-time) and because it flies in direct opposition to current political and academic movements.

[sago]

I agree, and definitely what I was hinting at.

I can understand why he enjoyed building mathematical models. And why reviews of "Balance of the Planet" praised that. But assuming that was the point, I think missed the point.

I think it would be like creating Demons Souls, loving creating the back story, getting lots of praise for it, and deciding that the best thing is an interactive fiction game with nothing but unlimited detail back story. Then wondering some years later why you didn't change the world, no one has gone further than you did, and all games seem shallow compared to your imagination. Fortunately Miyazaki went a different way.

Not a surprise that Will Wright and Sid Meier (who I know is very much a fan of complexity theory) went in a more successful direction.