mit is still a bubble

my writing professor says the Right Way to write stories is character-driven; you start with very detailed descriptions of the main characters you want to talk about, and if those characters are sufficiently compelling then their interactions will emerge organically and that naturally gives rise to your plot. we talk about how plot-first writing often leads to unmotivated character behavior, and how idea-first writing feels like it’s always bordering on propaganda. when i am in class this all makes perfect sense to me, but then immediately afterwards my friends point out that some of my favorite short stories, like the egg and story of your life, are very obviously idea-driven and plot-driven, respectively. these observations are not necessarily contradictory; for instance, it could be the case that character-driven writing produces the most consistently good short stories, but not the ones i like the most. regardless, this is confusing territory for me to navigate. i think this blog is a pretty even mix of all three kinds of writing though

in my computation structures class we are spending a few weeks programming in assembly language, specifically risc-v. some of my math friends simultaneously love puzzles and hate assembly, and i don’t really understand how this is possible, given that writing assembly is literally solving puzzles. so far i find it extremely fun, because it’s pretty cool how you can write any reasonable program using such a small set of instructions and registers, and the constraints involved force you to think harder about every line of code you write. some people tell me this means i have bad taste, because the whole point of modern languages is to think less about code details

i have been keeping up with all my classes and commitments and also successfully following performance engineering and cognitive augmentation, which is better than i was expecting. i’ve also been sleeping a lot by my standards, to the point where last saturday i got approximately zero sleep due to hackmit and was basically completely functional the day afterwards. so i guess the semester has been pretty good. but by far the most important thing for me to do right now (other than the obvious things, like staying healthy) is to get a better sense of what i care about and what problems i’m interested in working on, and i haven’t made much progress on these fronts. i don’t have significantly more knowledge about eg. the biotech or bci or crypto spaces than when i arrived on campus a month ago, and the list of companies i know i’d be happy working for after college has not gotten longer, which is a problem that i will try to remedy soon

speaking of hackmit, my team developed composing.studio which is pretty cool. most of the work was done by open-source packages (rustpad for collaborative editing, abcjs for text-to-note rendering and audio synthesis, monaco for syntax highlighting) and we just had to put them together in a nice way. i guess that’s ideal for a hackathon project, but people keep asking things like “how were you able to do this in one day” and the answer is very obviously that the hardest parts had already been done beforehand

i keep running into people who want to work in finance for non-altruistic reasons. i also keep running into stanford ea (effective altruism) people who are trying to help mit ea recruit new members as stanford has a strong ea community while mit does not. i also have friends who want to focus on impact but who think that ea has ineffective methods and ideology, so they are starting their own club which is simultaneously supportive of and adversarial to ea. there are also people like me, who were interested in joining said new club but decided not to because the club has very limited capacity which should go to other students

all of which is to say that if you talk to enough different kinds of people, it is impossible to ever tell what is happening on campus or who is the Most Correct about anything. as far as i’m aware, the only ways in which people ever hold stable beliefs here are by either having exceptional conviction in what they believe in, or by burying themselves in friend groups to protect themselves from the full diversity of thought on campus. i have already committed to trying to avoid the second option, which means i must embark on the first one

i’ve been meeting a lot of new people on campus, either via referrals or by sitting next to strangers in places like dining halls. many of them come from groups that i previously didn’t take seriously eg the business club, and they’re generally pretty cool. so far i think i’ve found everyone i’ve met to be interesting as long as they’re reasonably easy to talk to. i’m not saying people who are bad at conversations are boring; it’s just a lot harder for me to understand anything about them

it is difficult to comprehend how two people can meet the same new person, and one can think the new person is really cool while the other sees nothing of value in them, but this is the dynamic i keep encountering on a daily basis. i’m used to seeing this happen for organizations; for instance, i know a lot of people who think finance or frats or poker club are cool and a lot of other people who think they are utterly worthless, but it’s just jarring to see the same effect play out when the object in consideration is an actual person. of course i understand how this is possible — the same neural networks can be trained to learn anything, after all — but it makes me sad because i’d rather live in a world where everyone thinks everyone else is cool, you know? 

my vision is so foggy right now. it was this way freshman year too, but back then i was too near-sighted and too grounded in echo chambers to notice, whereas nowadays i can see the fog. i can see that everyone is giving me contradictory information about the content in my classes and the merit of ea and the value of other people and every other question worth asking, and i can see that the only solution here is to truly feel the things i claim to believe

i can also see that almost everyone is procrastinating on the question of what to do after college — and i mean this literally, as in, the number of college students i know who are spending large amounts of time, right now, doing things specifically targeted towards answering the question is in the single digits. a lot of people seem to think they’ll just figure it out over time by doing internships or taking classes (which in my opinion is equivalent to hoping that you can discover your favorite food via spoonfeeding), or that the question is not very urgent and a resolution can wait. maybe they’re right about that. but the reason i care so much about figuring this out now is that i agree with peter thiel and ben kuhn that the most probable scenario for people who procrastinate on figuring out what to do is that they slowly run out of time and end up feeling like they have no choice but to sell out, when in reality they had every opportunity to explore and simply never prioritized it enough. i don’t really feel like elaborating on this, but if you’re curious you should read the relevant sections of above links

it is difficult for me to recall the clarity i had back in alaska just a month ago, when nothing on campus makes any sense. it is difficult for me to remember which goals i think are important, when nobody seems to want to confront the primary question i am here to answer. i am handling the chaos better than any past version of me could, but it still does not feel good enough. sometimes mit feels completely absurd and i just wish i could see through it more clearly

5 thoughts on “mit is still a bubble

  1. you can literally post this on admissions with like a little editing for context

    how plot-first writing often leads to unmotivated character behavior » lewitt takes the exact opposite approach and says plot first, develop characters later, and then revise plot based on characters. this was what i took when i wrote my short stories for that class

    given that writing assembly is literally solving puzzles » “puzzles” is wide. there are word puzzles music puzzles logic puzzles id puzzles etc., thats just scratching the surface. low-level programming is okay, but i dislike relatively “toy” examples of how things are put together; to build anything of reasonable usefulness from assembly that hasn’t already been done before needs a lot of work. on the other hand, the ecosystem of libraries &c means that it’s a lot less effort to make something that’s actually *useful* cf hackmit project xp

    they had every opportunity to explore and simply never prioritized it enough » aS aN iNtErNaTiOnAl sTuDeNt, the options of things that you have a reasonable shot of getting are vanishingly tiny. like literally no intl i know is not in insert big company here. (something something maybe this is just an excuse for me not to think about it, but i think this is legitimate)

    the clarity i had back in alaska just a month ago… my vision is so foggy right now » one of the big reasons i never really bought into this postrat concept is because of this. i feel like you can make radical changes in your life and think really hard and whatever and then you feel good because youre mindful and all that stuff. cool! but then you go back to doing object-level things and those are just gone. like, what is the point, then?

    (hence why the mindfulness-adjacent things that work for me are smaller, more regular things; i told you about personal holidays which are honestly just an excuse to reflect over a period of time)

    Liked by 1 person

    1. you can literally post this on admissions with like a little editing for context » i have not made any posts because none of the new bloggers have been given the green light yet. it has nothing to do with post content 😛

      lewitt takes the exact opposite approach and says plot first, develop characters later, and then revise plot based on characters » fascinating 😮

      “puzzles” is wide… i dislike relatively “toy” examples of how things are put together; to build anything of reasonable usefulness from assembly that hasn’t already been done before needs a lot of work » yeah the point about puzzles being wide is fair and i had not considered that. but i don’t see how utility is relevant when talking about puzzles/assembly here; if you actually cared about usefulness i assume you wouldn’t be looking at puzzles in the first place?

      they had every opportunity to explore and simply never prioritized it enough » aS aN iNtErNaTiOnAl sTuDeNt, the options of things that you have a reasonable shot of getting are vanishingly tiny » i agree 😦 but i don’t think that’s what i mean by “explore”, and i also don’t think most people have this constraint 😛

      what i mean when i say nobody takes exploring seriously is that on a friday night you have options like a) party b) pset c) hang out with people d) read up on different ml problems and ai teams at various tech companies (for example, if you like ml) to figure out what you might like, and nobody treats “figure out what you want” with enough urgency to act on d)

      the clarity i had back in alaska just a month ago… my vision is so foggy right now » one of the big reasons i never really bought into this postrat concept is because of this » wait sorry, what postrat concept are you referring to? (to clarify, the fogginess is coming from the fact that everyone is saying different things and i haven’t sorted out which things to listen to yet, and i don’t think postrat has anything to do with it?)

      Liked by 1 person

      1. none of the new bloggers have been given the green light yet » oh ripperoni

        i don’t see how utility is relevant when talking about puzzles/assembly here » right oops mixed thing. i mean i dont enjoy doing assembly as a hobby because i dont find it that fun and its not “useful” enough for me to force myself to enjoy it?? or something???

        what postrat concept are you referring to » not quite sure how to articulate it but the general concept of like “wow i feel lost and therefore i should try to unlose myself” or something? which now that i write it down soudns stupid so maybe thats not it

        more like this being uncomfortable with not knowing what you’re doing far enough ahead or figuring things out far enough ahead or idk

        Liked by 1 person

  2. your comparison of writing assembly to solving puzzles is interesting. i’m don’t have strong feelings about assembly, nor am i an avid puzzle solver, but having done some amounts of both i would contest they are far from literally the same thing. perhaps the setting and purpose matter. puzzles are often written with enjoyment, interesting, and coolness in mind. i’m sure you could make assembly puzzles like them too, but most tasks you’d do with assembly in class or applications aren’t like that.

    Liked by 1 person

    1. hmm yeah its also possible i have not done enough assembly/puzzles. but i go into both with the mindset of “i will have fun solving the problem in front of me” and that is probably part of the reason why they end up being somewhat similar? idk

      Like

Leave a comment