Saturday, December 26, 2020

Alan Watts

Alan Watts died way before the Internet was a thing, but this didn't stop his talks on "philosophy" being perpetually popular on online media platforms. I never really spent time perusing them but whenever I come across them they never fail to impress. Quite hypnotizing really. I suspect that the answers he found to Important Questions of Life are pretty similar to what I found myself (albeit I probably disagree with him on some minor points).

And after all these years, I've started to believe that, even though truth is inherently subjective, there are models of truth that lead to a labyrinth of never-ending questions (the stuff that PhD in Philosophy theses are made of), and there are models that are (relatively) clear, concise and satisfactory -- and the latter ones lead to better answers for Important Questions of Life... and the answers do generally converge, perhaps because the number solutions decrease as you optimize for some attributes.

Yet the funny thing is, there's actually no "objective" meaning to finding the solutions to these Questions. Any notion that these solutions help one solve practical problems in real life are probably misguided, although sometimes it might help you in making slightly better decisions -- but so would pretty much any arbitrary/complicated belief system that has withstood Darwinistic trials.

So in the end it is just like solving riddles. The journey is more important than the destination. There's no medal or achievement for solving them, and it's even meaningless to discuss them among people who have already solved them. There's occasionally some fun in hinting the way towards the solution for those who haven't got it yet, but in retrospect even that might be considered an unnecessary spoiler.

----------


Sunday, December 20, 2020

Lies, Damn Lies...

Statistics

Statistics tip: Always try to get data that's good enough that you don't need to do statistics on it - Randall (xkcd2400)

This rings so true. When data points to an obvious conclusion, you don't need fancy analysis to arrive at some hypothetical correlation that may or may not be real. Perhaps the emphasis on statistical tools and methods within the sciences and social sciences in recent decades are a symptom that real progress is hard to come by, and data had to be marinated with statistical sauce to be made to work.

Statistics is also heavily utilized in "big data" as well -- even disregarding the "little data" applications and  puffs-and-smoke buzzwords, it may be true that even "real" big data is somewhat doomed from the start -- sure, you may crunch the numbers and discover that a 2.4% increase in user engagement can be achieved by tweaking some feature, and that often translates well into bottom line numbers... but the crucial question is unanswered -- does it make a better product? Does the user actually benefit?

And thus the old adage: Lies, Damn Lies, ... and Statistics.

Saturday, December 19, 2020

Queen's Gambit

It felt like the Cinderella tropes crossed with Hollywood obsessions.

Production and acting was great, but a good story should at least be somewhat believable. Drugs and alcohol and these highly "rational" intellectual pursuits don't really mix. I had a hard time believing anyone, regardless of gender or race, would be able to perform competitively by habitually binging on drugs and alcohol that are well known to dull the senses. Perhaps in the more artistic pursuits one would have an advantage if they could see things differently with an altered perception, but chess is objective and requires utmost mental clarity.

We all need aspirations to get by, and having a strong, talented heroine competing in a game dominated by men is a wonderful story if told right. Yet believability is subjective. Knowing the "vibe" of the kind of people that win national championships in intellectual games give off, prevents me from enjoy the story as most other people would. There's just not enough passion, not enough drive for self-improvement, not enough emphasis on self-discipline, and too much pettiness, too much plebbery. At a world championship level, every player is blessed with all the talent that is humanly possible -- it is a statistical impossibility that anybody, regardless of gender or race, could rely on mostly talent and barge in and win while being distracted by serious personal issues on the side. You compete by giving up less important things in your life to focus on doing what it takes make it to the very top. And the sad truth is that, the privileged requires fewer sacrifices. Hollywood made a valiant attempt at balancing our protagonist's passion, character and talent with her emotional struggles to make things relatable with the average viewer, but in retrospect given the ecosystem there, there's probably no way it could have succeeded in writing anything remotely believable (to me at least).

Cinderella is a more believable story if it had a fairy than if it had not. But the audience wanted a fairytale without a fairy-godmother nor a prince charming. The modern woman is too independent for that, they say. And so, after experimenting with drugs, alcohol and sex, the girl in dirty rags suddenly became Queen on her own, without even really half-struggling, just because.

The funny thing is, while women and men do of course pull off some incredible feats in real life by winning some sort of lottery (often genetic), the rules of fiction is different. Truth is stranger than fiction, they say, and I think they're right. You need a fairy in a fairytale. A story of some person winning some kind of lottery, then living happily ever after, makes for a horrible story because it's "unrealistic", even though this happens to many people on a daily basis.