I.
- i finally started using an external keyboard and everything is so much easier now i feel dumb for not doing this much earlier. next step is external monitor i guess
- i’ve been playing with lunchclub.com and it’s been pretty interesting! would recommend trying; i’ve had some very good and very boring conversations so it’s definitely a lot of rng. there’ve been a few times where i got matched with someone significantly older than me who immediately assumed i was dumb or ignorant due to being a sophomore in college but also other times where i got matched with cool fellow undergrads and i always enjoy learning about what non-mit people are doing 🙂
- i’m back in the job search which is endless and soul-draining as usual 😦 theres always the amazon return offer which expires 11/1 but i’ve committed to declining that even if i’m still jobless in november, which feels a bit scary but is also objectively the Right Choice. i thought all this would be way easier now that i’m not a freshman, but it’s been more or less the same story eg. i got rejected by js after 5 hours of “onsite” interviews
- apparently an updated draft of my primes paper is getting published in the ima journal of applied math 😮 i’m pretty shocked by this since we (me + mentor) gave no indication about whether the refugee model we created was generalizable at all, even if it does work well in the scenario we tested (burundi ~2015). you hear a lot about how eg. the social science research world is flooded with overfitting and p-hacking, and although we didn’t do anything of that sort, there’s also no evidence in the paper that we didn’t? so uhh yay for publishing standards i guess
- also one of my essays from a few posts back is appearing in mit angles which is cool! i haven’t read many of the other essays featured but they’re probably interesting too, would recommend 🙂
- im getting very tired of all the memes about 2020; i don’t want to be that guy but for the most part they feel pretty counterproductive and are honestly getting a bit annoying. i have no words other than this tweet. please find a better coping mechanism than memeing the year tyty
II.
since a few weeks of school have passed, i can comment in greater detail about my classes
5.12 organic chemistry is.. for the most part easy to follow? i didn’t have any chem background so i crammed the 5.111 intro chem ocw videos for 2 days and that turned out to be enough in terms of chem background, though the psets are still hard. i’m really glad the prof is teaching it in a way which requires very little memorization (which is the main complaint i’ve heard about orgo from my non-mit friends) so this has been a pleasant surprise 🙂
6.009 fundamentals of programming is more or less what i’ve expected. i’m not learning too much about “how to python better”, but the labs are enjoyable to work on and also not too time-consuming, eg. the image processing lab had a “do whatever you want” section so i threw in some voronoi diagrams and ended up with my new facebook profile picture 😛
6.036 intro to ml is also basically what i expected. it does seem a bit more comprehensive than the coursera intro ml in both content and homework (eg. covering the perceptron which isn’t super important but is an ok motivating example, and this was absent in coursera) so that’s nice
6.849 geometric folding algorithms is tough in the sense that every week we work on open problems which i’ve made no progress on whatsoever, but the lecture videos are interesting and working on hard problems on calls is enjoyable even if i’m not contributing too much. if i do need to drop another class it’ll be this one though, even though it was the class i was most looking forward to going into the semester, just because i don’t think i have the time or the skill to do the content justice
18.821 project lab is kind of annoying.. probably my least favorite mit class so far? we have a bunch of assignments of the form “write mock papers / make mock video presentations” and these are not very fun. the “research problems” we work on also feel extremely superficial because they were specifically chosen so that we could make progress on them and have probably been reused for 15 iterations of the class or something, so it feels like a lot of wasted energy. i do think making good videos is a skill i’d enjoy learning, but the class focus is more on how to select content or talk about math in videos and doesn’t have much on videomaking or editing
21m.284 film music has been great! the first few weeks were a bit boring because there was lots of textbook reading on eg. film music terminology, but now that’s over with so we’re more or less just watching movies and talking about them. the prof seems to have chosen all the movies pretty carefully and there’s no way i’d be making time to watch movies if i weren’t in this class, so i appreciate it a lot. movies weve watched so far:
- catch me if you can– i thought this would be a typical heist/chase-type movie, but that turns out to be almost completely incorrect in pretty much the same way that the dark knight isn’t a superhero movie. instead the focus is on father-son relationships and youth/immaturity. would recommend 🙂
- pride and prejudice– i haven’t read the book but i’m told the movie captures it very well. maybe im dumb but i felt like the dialogue was somewhat hard to understand because it was often fast, quiet, and in old english lingo. other than that the film is very sweet and mainstream (like, literally the most predictable and happy ending possible) but it’s engaging and has interesting characters regardless and pretty nature shots; i think “Pride & Prejudice established the template for an infinity of romance novels, yet no subsequent love story has ever come close to equaling the delights of the original.” sums it up pretty well (from a review i read)
- glory– was good, but i knew the story of the 54th regiment already and have seen enough of the war + coming of age subgenre (yes there are racial themes here but i’d classify them under “coming of age”) so i was slightly bored by the plot. it was really touching though and probably good to see if you like civil war-era stories
- in the heat of the night– i liked this a lot! sort of surprised that this is maybe the first movie i’ve seen about the us south post-1950. it’s a pretty fresh story and a lot of extremely racist characters were portrayed in disgusting ways, which i guess means they did a good job? i feel like it’s still extremely relevant today which is both cool and very sad
III.
on friday night brian sent an email to the sparc alum mailing list advertising a living group in truchas, new mexico with mostly hacker-type people of which i am not one of. i already gave my parents 2 separate living group proposals for this fall, and both were rejected for not having rigorous enough covid protocols (specifically the lack of a quarantine period upon arrival). but this group was actually enforcing rules so it seemed worth a try and i scheduled a call with the organizers for the following afternoon
my initial three thoughts after the call were:
- you’re not cool enough to live with these people
- the average age here is 21 and you’re 19 and 2-year age gaps are incredibly significant until you’re more than ~23 years old
- you just gave a series of idiotic responses in what was essentially an admissions interview so stop worrying about it
so i was caught completely off guard when they emailed me back saying yes, they did have openings available and i had to commit by the next morning which is never a reasonable turnaround
IV.
every difficult choice in my life has the property that, even before i make the choice, i know i’ll regret whatever decision i come to. it’s not just a “the grass is greener on the other side” situation, it’s “the grass is greener on the other side and you chose the wrong side” or “the grass is greener on the other side because you chose this side” and each sentiment feels distinctively worse than the previous one. the regret fades of course, and usually doesn’t come back, but i never know how much i’m supposed to pay attention to it
the trouble with insisting on not describing yourself as a math person is that you don’t have a great answer when anyone does ask you to talk about yourself. you can practice, you can prepare a funny response, you can answer the question sideways, but none of those things come naturally and they probably won’t for at least a few more months or years. that was the trouble i ran into during our call, and it’s the trouble i still face when i ask myself how i’d fit in with these people
some days i feel like i can’t solve math problems or write code or read or exercise; the only thing i can do is write, just me and a blank document, churning thoughts into sentences. it’s the only thing i know i can always do, the only thing i’m sure will turn out alright. is that who i am now? because it feels oddly comforting and horrifying
anyway i made a pros and cons list for moving to new mexico. the pros were meeting new people and starting new projects and having increased motivation and accountability and not being at home, and the cons were potentially dying or having an emergency and being with people i might not like and no longer having any free time left for social stuff or personal work once i incorporated living group obligations into my schedule- i call this scenario “time death”. anyway all the cons were pretty bad and the pros weren’t really comparable so this list was not helpful at all
i’ve never really been sure where i stand on the introvert – extrovert spectrum. i’m not even entirely convinced these kinds of people actually exist, and when i take personality tests i always end up in the 45% – 55% range. i know that i don’t really need people to be happy, so why was i putting so much weight on the pros column? i also know i get infinite energy from being around people; that’s how i survived sleep deprivation at mit and summer camps. but i’m not sure if it’s people that energize me or just the idea of people, in the same way i can never be completely certain if i really care about people or just about their effects on myself. i was getting ready to reject going to new mexico and suddenly i was feeling immense motivation to make the most out of being at home, as if the act of turning down a cool opportunity would be enough to push me into being better. is it possible to draw fuel from a battery and then flip it around and also be powered by its absence?
at the same i feel like time death isn’t actually ever an option. time death just means you need to be better, and it never actually hits you because you always find a way to be better, whether that’s by dropping clubs and classes or giving up daily chess puzzles or something else entirely. and if you want to keep good grades or friends or anything else, that’s just another constraint on the manner in which you need to be better. time death exists in the sense that you can have too many priorities to satisfy all of them, but once you understand that and come to terms with the relative ordering of those priorities it’ll always work out. that’s what i’m counting on for now at least, and if i ever realize i was wrong i’ll be the first one to admit it
V.
3 weeks ago i found myself driving home after a run and my parents were in the back of the car. on the way back i realized this was my first time driving on a highway with both my parents, like, if i crashed now it would be game over, you know? so there was nothing to do but lock in on the road and take it a second at a time, to turn pedaling and steering into my entire world. in those 30 minutes i was more focused than i’d been in maybe six months. wouldnt it be nice to feel that way more often?
in some sense no decision is actually about the consequences of whatever situation you’re in. decisions are about what person you want to be and whether you’re moving in positive directions or not, and more than anything else right now i want to be the kind of person that gets things done. there are moments, hours, days of breathtaking clarity when i become that person, but it never lasts long enough
i’m still not sure why i made the choices i made. maybe it was the fallout from recruiting piling up. maybe it was the confirmation by my friends that spring 2021 would probably be virtual. maybe it was the way the first 3 weeks of the semester had gone by with hardly anything of note happening, other than me learning some orgo and watching some movies
maybe it was that you called me a superhuman at managing commitments and projects and i felt like i wasnt doing any of that at the moment, i was just going to class and doing homework wasting away. i’m sorry, i’m sorry for letting you down, i’m sorry i was never the person i thought i could be, i’m really really sorry but i promise i’ll keep trying until i learn how to be better and i will learn eventually, and i know that’s not good enough but it’s the only thing i can do so please forgive me, okay?
or maybe i just needed to see stars in the night sky for the first time in my life. i haven’t really done that yet, not more than five or ten or twenty anyway. maybe seeing stars wouldn’t even do anything for me and maybe after i saw them i’d still be the same person i’ve always been. but i’ll never know until i actually do it, and if i don’t try i’ll just keep wondering to myself, “what if you bothered to look at the stars once” for the rest of my life, you know?
maybe all you need is to go away and look up into the sky for a bit and try to be a little better than you were the day before, and if all that doesn’t work out then after a while you can come back to the life you’ve always had. if that life has left you behind by the time you return then maybe it’s not something you really want to keep around anyway, and maybe once you realize that you can stop regretting your choices because you’ll finally believe they were the right ones
i hope you can wait for me
started using an external keyboard » external keyboard has been one of the best purchases in my life last year. i also dont think the benefit of an external monitor is as big after you have a keyboard
playing with lunchclub.com » back when things were on campus there was https://connectmaven.com/about/connector/mit and they paid you $10 to have lunch with someone! i have never had a bad experience with connect
back in the job search » T_T T_T help i dont want to
haven’t read many of the other essays featured » you should recommend that people read my essay
about my classes » you are taking too many T_T
not learning too much about “how to python better” » i dont think i learned anything from 6009 other than “wow i can write a lisp parser”
perceptron which isn’t super important » i think perceptron is a pretty important conceptual handle. like the linear classifier just comes up frequently enough to be Important i think
i know i’ll regret whatever decision i come to » per https://mitadmissions.org/blogs/entry/to-the-exclusion-of-everything-else/ “Was it really a question of balance? Was it really a question of figuring out the proper ratio of academics to social life? Was it really just a matter of solving for the right proportion, as if making a decision could be as simple as those mixture problems, as if you could use algebra to determine the amount of poison with each choice?”
come to terms with the relative ordering of those priorities it’ll always work out » i did this the other day and realized that even if i enjoyed 6849 it was something that was lower in my priority list than other things i could not make time for so i dropped it and now i am marginally happier
LikeLiked by 1 person
“i also dont think the benefit of an external monitor is as big after you have a keyboard” wait isn’t it so you can do things like pset while watching movies and have better screenshare layouts (totally hypothetical examples)
i dont think im taking too many classes in particular 009 036 821 aren’t much new content and 284 is just movies? and 5.12 is hard but on pe/ne shrug
yeah i agree linear classification is important but like you get that from logistical regression too? the main takeaway from perceptron for me is how it’s guaranteed to finish under good conditions but these sorts of guarantees are usually rare/impractical in ml? correct me if im wrong
i still dk if im dropping 6849 oh well we’ll see how long recruiting drags on for maybe
LikeLiked by 2 people
logistic regression is indeed more stable, more theoretically grounded than perceptron, and has way less bad behavior. but also its way less elegant T_T
LikeLiked by 1 person
I think the pride and prejudice movie (2005) was better than the book and I normally don’t feel this way about movie adaptations 😛
LikeLiked by 1 person