Let me reiterate and I am willing to die on this hill: not all fields of study can be called science, nor do all fields merit a PhD.
The purpose of science is to glean out some fundamental truth about the universe. And the bar for what is considered true is set very very high.
1. It must explain existing phenomena
2. Its findings must be replicable.
3. It must make specific, testable, falsifiable predictions.
And the specific, testable, falsifiable parts are extremely important. If your field doesn’t do this, you are not doing science. Period. You are presenting an opinion.
A lot of humanities papers seem to do just that. Present opinion. Yes, arguably there is some structure to it. There are some axioms. And there is some internal logical consistency. But in the end it is you and your supervisor’s personal opinion.
Now this kind of work is indeed necessary. But that is not academics. That is not research. You don’t need to dole out PhDs for someone presenting their personal opinion using made up jargon and purposefully dense language.
There are brilliant authors, poets, and essayists out there. And they are presenting their opinions in the open marketplace of ideas, where those ideas either fall or stand on their own. Their contributions to the world are as important as any scientist. Just please don’t dilute the meaning of the word “science”.
Our academic institutions were built to be places of truth seeking. Let's restore of some of that.
What Isreastine needs is a God Emperor who will enforce a terrible peace for 3500 years while turning the whole place into lush green plains.
On this Golden Path, all specialness will be lost and the people there will be transformed into virtuous non-retards.
A story like the one in Perfect Days is really only possible to tell when it's set somewhere nice like Tokyo. It's only really possible to do what the movie is showing when there aren't random assholes who will fuck you up just for the sake of it.
Think about how vague and imprecise even the most specific (non-technical) term in any human language is. Mathematics, on the other hand, is built on maximally pure, strict, logical relationships -- how could it not be a good model for what is essentially cause and effect?
People who think that mathematics is "unreasonably effective" at describing reality need to reflect on what they think constitutes an "effective description" of something. "Mathematics is an extremely good descriptor of reality" -- as opposed to what?
There is nothing in the world that can be meaningfully called both purely psychology and science.
a) There's basically no predictive theory other than obvious statistical correlations
b) All experiments that could elucidate interesting improvements to the theory are forbidden
@ianmiles Yes, this whole "black samurai in feudal Japan" is forced and stupid, but this is a bit unfair, imo. If you think this samurai character should be Japanese because he's in Japan, it shouldn't be a problem that he ends up killing mostly Japanese people.
Do you guys understand how scary of a combo physicist + programmer is. Masters of the universe fr. Additionally, everyone who is in this category is confirmed hot. Without exception.
@mhQBWiDQV5kZ Quick Google search says the national dropout rate is already at about that level, but I think it would make a big difference in how high schoolers perceive the institution if they know that the 30% chance is likely to be due mostly to skill issues
@mhQBWiDQV5kZ Seems like you'd have a hard time finding instructors who know and appreciate a subject well enough to make a truly rigorous course on it and don't want to do research. And good luck getting enough students who know there's a 30% chance they won't walk away with a degree.
Overall I'm liking Haskell a lot more than OCaml for a great many reasons, but (restricting to functional languages), if there's one thing I think OCaml really got right, it's the role that modules play in the hierarchy of all things.
OCaml functors, then, are a great idea. They just fall flat because you can only use them on the module level, rather than on the per-function level. Just give me OCaml with bounded polymorphism on the per-function level, list comprehensions, and fucking line comments.