Semester review

Well, it’s been a heck of a semester! Following Peter Hurford’s excellent example, I’ve decided to start writing up reviews of how my life has been going every so often, for a couple reasons:

• It forces me to actually sit down and think about how my life has been going. It’s easy to get absorbed in minutiae if I don’t do this, and I want to make sure that I spend an appropriate amount of time in strategic-mode.
• It lets other people learn from my successes and mistakes (ideally). I find that insight into how other people’s lives actually work is a useful antidote to the massive selection bias that comes from being stuck inside Ben Kuhn all the time, so hopefully I can provide some to others.
• It exposes me to more feedback from the rest of the world. (Theoretically. If anyone actually makes it to the bottom.)

So without further ado, here’s what I’ve been doing for the past semester:

Jumping through academic hoops

To channel the spirit of Robin Hanson, “college is not about learning things; college is about costly signalling.”

Of course, if you can minimize the amount of time you spend on costly signalling, you can get a good amount of learning done, too. And in this department I’m doing quite well: my semester grades are all on target, but more importantly, I was able to overcome my perfectionism. For instance:

• I paid more attention to how long I was spending on problem sets, and occasionally punted on problems that were too tedious.

• I realized that Harvard’s distribution requirements were a joke (about which more later—I’m feeling a rant coming on), so I optimized for making my general education course minimally attention-draining.

• I began considering the possibility of using advanced placement to skip ahead one or more semesters, in case I decide that the main benefit of college at this point is signalling.

(Note: I’m not actually that cynical about college! In addition to being a massive signal-fest, it does provide a lot of actual value.)

Learning things

The reason that I split this off from succeeding academically (rather than filing them both under “taking classes”) is that I feel I’m doing significantly worse at learning things. For the first two years of Harvard, I could feel myself becoming more technically proficient at math and computer science: I got noticeably better at solving problems, figuring out algorithms, getting mathematical intuition, etc. This process seems to have slowed—perhaps I’m hitting diminishing returns?

On the other hand, I haven’t been able to make progress in finding good humanities or social sciences classes, either. I came to Harvard having already taken a fairly strong humanities curriculum, while my first experience with college humanities courses was through the General Education program—hence the classes I took were for the most part superficial and full of students who didn’t really want to be there, and generally disappointing.

On reflection, I think it was wrong to extrapolate from Gen Eds that I wouldn’t be able to get very much out of any humanities class. But I now only have three semesters left, which means I don’t really have time to flail around looking for good courses.¹ I’m not sure of a good solution to this: in particular, it seems like what courses are “good” varies widely from person to person, based on previous experience, how their brain works, etc. So it takes a fair amount of work to verify people’s recommendations. I plan to try harder at this in future semesters, but I’m not confident of success.

On the social science side, I think I have a better handle on things; I know of at least two courses that I’ll very likely get things out of (advanced microeconomics and econometrics). However, I’m still on the lookout for other courses that might teach me noticeable amounts.

I’m in the process of talking to Jonah and Vipul of Cognito Mentoring about course choice. Another idea that came to me while talking to them is that, if it’s really the case that humanities and social science just aren’t universities’ comparative advantage currently—lectures just don’t give very much value added compared to reading blogs, books, etc.—then maybe I should be focusing on things where college classes definitely are a value-add, namely, procedural knowledge. (For example, procedural subjects I think are plausibly high-value for me include public speaking, improv theater, drawing, hardware, and creative writing.) I don’t have strong thoughts yet what would be more valuable—I have some more thinking to do here.

Writing

I decided to step up my writing activity this semester, even going so far as to set up a Beeminder goal for blogging twice a week (as is probably obvious by now, I stopped during finals—I had 30 pages of other material to write!). My goals were to spread effective altruist ideas, get better at coming up with new non-trivial interesting ideas, and improve my writing.

At meme-spreading I was quite successful. I wrote an op-ed on effective altruism in the Harvard Crimson that got mentioned by the philosophy professor of one of my friends at a different university. Additionally, I suspect (though I don’t have hard evidence) that posting my EA-related entries on Facebook was a significant factor in getting some of my friends interested in effective altruism.

At coming up with new, non-trivial, interesting ideas, I did less well. I managed two out of three fairly often, but I wouldn’t count any of my posts this semester as all three. I came closest with A Critique of Effective Altruism, but I’d still categorize most of that post as derivative—only the synthesis was novel, and perhaps the coining of the phrase “pretending to actually try”, which has seen some discussion. (Of course, it was useful on other metrics, as I’ll go into below.)

At improving my writing, I mostly failed. Granted, this was the least important of my three goals, but I still thought that writing more would at least somewhat help. However, it seems I’d have to write a lot more for that to work. If I wanted to double down on this goal next semester, I would focus on more editing, reading, and being adventurous with format and subject matter. However, I don’t think that’s worth the time investment; my writing does the job already.

I also got a few unforseen benefits by writing a lot. It’s a great source of status and social reinforcement. A Critique of Effective Altruism started some good conversation about effective altruism (on Less Wrong and elsewhere). And it allows me to non-hypocritically encourage my friends to write more, and I enjoy reading my friends’ writing.

Going forward, I’d like to write about non-effective-altruism subjects more frequently; I would prefer to have a more diverse readership, and would rather not become a “single-issue person” (to the extent that I’m not already). Also, my experiment with “mini-posts” (writing more and shorter posts for the week) seemed tentatively like a good idea, but I never followed up; I’d like to experiment more with that and other formats.

Overall, I worry that I’m biased in my assessment of the usefulness of writing—it’s public and has fast feedback, so it provides disproportionately much social reinforcement. For instance, in hindsight, the critique of effective altruism was clearly influenced by which criticisms I anticipated would resonate with others (i.e. would get the most social reinforcement), in addition to (instead of?) which criticisms were the most damning. I’m still looking for a good solution to this; candidates include stopping posting to Facebook, or posting at a reduced frequency, or keeping my writing to myself altogether; I’m not sure which I favor yet.

Co-running Harvard Effective Altruism

A major plan for me and my co-president John was to take Harvard Effective Altruism (formerly Harvard High-Impact Philanthropy) from a few people hanging out to a Real Student Organization. We ran the first iteration of our new Philanthropy Fellowship program, in which we got a bunch of people to apply to attend a series of awesome speakers and do a fundraiser at the end of the semester. The fundraiser is now in progress, and succeeded admirably in getting new members interested (enrollment is up something like 500%) and raising our profile on campus.

I was less successful in preparing for next semester’s iteration of the fellowship. I’m going to have to scramble a bit to get our speakers lined up over winter break, though I don’t think that’ll be a large issue.

My main mission for the future is to prepare the group for our first real leadership turnover. To that end, I’m going to try to step down as co-president ASAP so we can start the succession. Secondarily, I want us to do more than just the Philanthropy Fellowship next semester: we now have enough engaged members that we can start doing side projects as well.

Making friends (and influencing people?)

I’m pretty content with my social group, so previously I didn’t put much thought into expanding it. However, I recently decided to make this a prioriy because I realized that getting to know lots of awesome people is actually one of the things that’s hardest in the outside world relative to college. I made a lot of progress on that this semester, but mostly through my roommate’s doing, not my own.

My attempts to maintain/build existing relationships were also more successful than in previous semesters, largely because I got over my ridiculous aversion to asking people if they wanted to hang out without having an excuse. However, as the semester progressed, because I was getting busier (and because it is absolutely impossible to get ahold of Harvard students), I got worse at proactively spending time with people I wouldn’t otherwise see.

These both strike me as high-return areas to work on. I suspect I won’t be able to find very many other communities as varied and interesting as Andrew’s weekly dinners, but doing so would be quite valuable if I could. On the other hand, I see pretty clear low-hanging fruit for improving the ways that I maintain my existing friendships, which I hope to pick over the coming semester.

Self-improving

This semester was also my first chance to put some CFAR material into practice. As I predicted, the benefits were pretty large. Of particular interest:

• I rearranged how I use email and Facebook, which had a surprisingly large effect on the amount of work I’m able to get done in a day.
• I designed and started using a to-do system and Beeminder, and am vastly more on top of things as a result.
• I started doing more things that made me uncomfortable, including advertising HEA more aggressively and cold-emailing people more (these were the two biggest wins).

I’ll probably write more about these at some point. Of course, some of my attempts didn’t go so well: in CFAR jargon, I’d particularly like to work more on the offline habit training and againstness units this coming semester. (For those who haven’t been to a CFAR workshop, I’m leaving this jargon here as a handle to myself—I’ll explain what these are once I can explain how they’re useful to me, to avoid making things too silly.)

Being in a long-distance relationship

Over the summer I started dating Ruthie, who promptly moved to the Bay Area. I had a lot of trepidation about whether I’d be able to carry on a long-distance relationship while still spending enough time on schoolwork, HEA, writing, etc. But, it turns out, by providing a massive incentive for me to be intentional about how I structure my time, it’s actually made me more on top of things. Some of my most productive work this semester has happend en route to SFO (after all, there’s no email or Internet to distract me…)

Maintaining Harvard Class

Unlike most of my semester plans, maintaining Harvard Class was pretty much a pure failure. At the beginning of the semester, Billy and I had almost finished an awesome rewrite; three months later, we’re essentially no further. The only work I did on it was the bare minimum required to keep the data up-to-date (and barely even that). In the mean time, the other course catalog seems to have leapfrogged us and is gaining back users (despite their blindingly obvious inferiority). At this point, I’m not feeling very motivated to continue to work. I think I now have higher-return projects, and if I’m going to be coding, I’d rather it not be yet more web-dev. That said, I might pick it up over winter or summer break, to get some closure and because it would make a good portfolio piece.

Deciding what to do with my life

Like any self-respecting college student, I’m also in the process of searching my soul to figure out my life’s calling. After writing up my previous plans, I started doing a career advice case study with Paul Christiano of 80,000 Hours, but it’s now on hold until I see how my winternship at GiveWell goes. I also started devising plans for the summer and for next year, although they’re still in the very early stages. Overall I don’t think this plan needs any adjustments.

Relaxing

Although I did (surprisingly) get to relax this semester, I realized recently that I’ve gotten a lot worse at it. Most of the things I tried to do (playing various video games, browsing the Internet, watching a movie) ended up not being particularly enjoyable. I hypothesize that I’m confusing my “wanting” and “liking” neural circuitry, and if I’m more mindful during relaxation activities of whether I actually enjoy them, I’ll have a better sense of how to relax effectively. In the mean time, I now know that “wanting to do X to relax” isn’t particularly indicative of X being relaxing, so I should try more different things and pay attention to how they work.