Happy Tuesday. (Or Wednesday, I suppose, by the time this gets posted.) Tuesdays are always fun for me—I have classes on Tuesdays, which means my thinking time is uninterrupted by actual work. I spend a lot of time corralling stray thoughts when working (it is the rare occasion when I am fully sucked in—writing and occasionally coding can be beautiful moments). I also spend a lot of time corralling stray thoughts when I’m sitting in classes, but then I get to write them down in a Word document, and get prompted by everything that’s happening in class to write more things down, and then share all of that nonsense with you!
This post is going to be piecemeal, by section. There’s a lot of little things that I’ve been transcribing, along with many wonderful quotes from HPMoR, “Harry Potter and the Methods of Rationality” by Eliezer Yudkowsky that I’m reading. Warning that this post does contain spoilers for HPMoR. I can’t imagine any of you guys reading it on a request from me though, given that it’s something like 650,000 words. (That’s a lot of words. More than three times the length of the last Harry Potter book by J.K. Rowling, and that was an epic.)
Here we go!
How much useless knowledge can you have in a life? By useless, I mean things that you spent time and effort acquiring that are no longer at all relevant to what you do. I was thinking about this because one of my friends was bemoaning the fact that he didn’t take an advanced computer science class in undergrad that would have been useful for his current research. And I was like: “Dude! I analyzed neurons doing color vision in undergrad! At least you have a CS major— I haven’t even taken ‘Artificial Intelligence’!”
(I was arguing that a neuroscience background is less useful to research in computational cognitive science (where I am now) than a CS background, even if it was general CS. I had tremendous research opportunities during undergrad, they just aren’t particularly relevant to what I’m doing now.)
I think that a lot of people probably feel ill-prepared: it’s the rare person who knew exactly what they wanted to do in grad school, went and pursued that directly, and then ended up exactly where they thought they would in grad school. Some of us have more catch-up to do than others, but it’s funny to me to think about the might-have-beens when we were presumably making the best choices we could given our knowledge at the time.
Another of my friends was telling me a story about a Berkeley man who’d switched careers four times. He’d been a biochemist, a racecar driver, a jade miner, and I forget what he’s doing now. The friend who was describing this said that they wouldn’t be surprised if they themselves switched careers further in. That’s kind of terrifying to me, because I work on, like, 10-year plans, but it also feels antithetical because that’s a whole skillset that’s no longer being used. I wouldn’t be opposed to switching careers if something came up that I truly thought was better (…nothing comes close, at the moment) but in the abstract it’s always strange to consider that we’ve spent so much time and effort on things that are no longer of use to us.
It’s how life goes, though :). Life’s fun because of the uncertainty, and as long as we do the best we can with our current knowledge and follow our interests, I don’t think we can actually do better. I have so much to learn. And so much knowledge that shaped who I am, was probably not worth the amount of time I spent on it, but is an irrevocable part of how I’ve used my time and to some people one of the best uses of a flexible, changing life.
HPMoR: Hermione has been avoiding Harry. Harry has approached her.
I love this book so much. I love when things are spelled out so explicitly—that someone can say, “I want you to have a positive association with me” and then hand out chocolates to reinforce that. In real life, we probably create positive associations by smiling, by being warm—but it’s either implicit behavior, or you’re not allowed to talk about it if you’re doing it deliberately. I talk about strategies I’m using far more explicitly than most people, but it’s seemingly impossible to classify everything that people are doing. Take turn-taking, for example. In conversations, there are very specific rules for who has the lead in a conversation, and if someone has said something and you jump in with, “I think,” then pause for a while, everyone has to wait for you to finish what you were saying. I didn’t come up with that observation—there’s a substantial literature on turn-taking—but it’s just one of the many techniques that’s been described explicitly and suddenly you find people using it everywhere. There are so many rules that we can’t write them all down, and many of them are context- and situation-dependent, and I’ve always been fascinated by these things but I don’t quite know where I’m going with it. They’ve been observed before, they’ll be observed again, and it seems like each of us just has to learn them from birth just like everyone before us. Well, or stick them in wonderful books where people can appreciate them :).
HPMoR has had me wondering a lot about “cheating the system” ways of interacting with people, using knowledge of cognitive biases in humans to elicit a behavior you’d like to see. I think in today’s terms these are called “life hacks”, but a preliminary Google search of “social life hacks” doesn’t turn up a lot of the more interesting findings. HPMoR is also a lot about cheating the system in other ways—finding loopholes in society that won’t hurt anyone if exploited but are opportunities that aren’t pursued simply because they’re not normal. In one example, wizards can use their magic to create certain bonds between people, but it permanently drains their magic. Harry suggests that wizards on their deathbeds should create these bonds and sell them, and give the money as inheritance to their children. It’s a practical loophole that would be a service to the wizards’ children and the bond recipients, but it isn’t done in the magical world. There must be an equivalent number of these sorts of things in our world. What are they, and why aren’t they done?
Let’s consider an abstracted version of the wizarding money-making loophole, and why it wasn’t exploited. It’s an intuitive idea—there’s some behavior that you wouldn’t normally do because it’s long-term damaging, but if you’re only looking at a short-term situation then it would make sense to do it. (Note that the short-term situation doesn’t have to be dying; you could just be in some situation for a fixed length of time.) In HPMoR, our protagonist often argues that people often just don’t think of these sorts of things—it’s not that people are stupid for not doing these sorts of life tricks, it’s just that they’re unaware of them. But even if you do become aware of them, I can see how one wouldn’t execute these sorts of maneuvers because of social conventions.
Like, I can’t see any venture that involves making money being acceptable on one’s deathbed. It isn’t done. (Why isn’t it done? I don’t know—presumably money-making is just usually such a different situation that they aren’t commonly associated. The thing that’s really annoying me about this example, however, is that I’m not able to think of a real-world equivalent. Like, come on, there has to be a situation which one is going to earn money and do a service to others, that normally one wouldn’t do because it’s long-term harmful (plus socially unacceptable). All I’m coming up with are things like smoking and drugs and other pleasurable things that are long-term damaging, which one could certainly do on one’s deathbed (actually, this sort of works as a half-example) but doesn’t assist another party. Or even better, I wanted an example that wasn’t about dying, like giving something up when leaving a community or something that would help others but normally hurt you long-term in that community: knowledge, maybe, but I can’t think of something that you normally would not want to do, that would benefit you in the short-term AND others in the short-term. There have to be examples though. I’m probably being blinded by the “it isn’t proper to combine those two ideas” thing, which is exasperating!)
Relatedly, HPMoR also spends a lot of time picking at things like Quidditch, which doesn’t really make a lot of sense as a game. Specifically, making the Snitch worth so many points compared to the rest of the action means that people aren’t invested in the vast majority of the gameplay, and it’s not actually that exciting to watch someone else looking for something (the Snitch in this case.) Again, there should be many systems that don’t make sense in the real world. These ones, as least, are easy to find examples for—lots of games aren’t optimally interesting, lots of systems are broken in ways that it would be possible to fix, when countries have opt-out organ donation systems rather than opt-in organ donation systems a lot more organs get donated, all sort of examples. I think a lot of these will be listed on the LessWrong website, SlateStarCodex, TedTalks, Facebook posts from people upset about certain systems who are presenting solutions, numerous. In this case, however, the problem changes from one of identification and defeat of social pressure to fixing the problem and defeat of social pressure. And I haven’t gotten to the point of HPMoR where the author explains how to fix all of the world’s problems and convince people do things that are better. I don’t have the highest hopes—people seem to have been attempting to convince others to do things better for a long time, with political and social movements and such, and there does not seem to be a cure-all. That said, I’m verrry interested in this topic, specifically a route called “mechanism design”, which is roughly shaping the environment so that peoples’ natural strategies lead them to do what would make everyone happiest. My thoughts are very preliminary on topic at this point.
I was recently at a meeting (we’re switching back to real life :)) in which I had some notion of what was going on, but the details were outside my background and so were escaping me. Since I couldn’t really follow and evaluate the arguments people were making myself, I switched into meta mode, which is basically observing how people are reacting and how I was reacting. Several amusing incidents / observations followed.
The presenter brought up some four-letter acronym (let’s say BOYT), and then continued to discuss whatever “BOYT” was as the audience joined in with other “BOYT” arguments. Meanwhile, I’ve rapidly pulled out my laptop and am sitting on Google searching “BOYT”, “BOYT model”, “BOYT++”, “BOYT robotics” and getting “do you mean ‘boy’?” over and over again until the guy next to me finally takes pity, pulls out his phone, types in the spelled out name and gestures for my attention. I adore helpful people.
(I also very much enjoy when people read over my shoulder when I’m attempting to do smart things. I consider everything open on a screen to be fair game for knowledge collection, though I’ve realized that this is NOT a common assumption and is considered an invasion of privacy. I do act as though other people share my opinion, though, and rarely have non-audience-expected things up on my laptop in public spaces. Then again, I do seem to have a lower public-audience bar than most. Two ways this plays out.
One: I was formalizing a hypothesis today about why we might be able to remember what we’ve said to specific others so well, completely missing the point and trying to connect a different hypothesis about the inverse problem. And my friend Carson smiled very nicely and said: “Ah, you didn’t think of this because you’re an honest person, but the reason we can remember what we say so well to others is probably so we can lie.” And I like, oh, duh. From which I conclude that apparently being a very open person limits my perspective and hypothesis space. I mean, everyone’s constraining their hypothesis space in some way because of who they are—that’s why it’s important to talk to other people with different perspectives—but it’s nice to know where I have specific blind points. Also, this is a particularly complimentary blind point (and well-delivered), though I can think of a few of my nastier ones.
Two: I talk to other people about things other people do a lot. Like, it’s becoming problematic, how strong this instinct to share is :). I’ve developed methods to mitigate this need in the past—basically, just limit the blast zone, try to keep information to groups of people who aren’t connected. My problem, I think, is that a) I like telling people interesting things and people are one of my most interesting stimuli, b) I deal with emotional problems by downloading onto others and observing their reactions, c) I test a lot of my intellectual thoughts on others to get advice and feedback, d) I basically, and kind of fundamentally, believe that things aren’t useful unless they apply to a larger population. That’s why I’m utter crap at writing diaries or journals, but I’m happy to write a blog every week if I know that people will read it. And if I’m going to any effort to think thoughts, I pretty much want to share them if they’re interesting or personally evocative thoughts. I am absolutely sure that others don’t have this same problem of keeping secrets, and that I am spectacularly bad at this impulse control and the only reason I manage to keep anything quiet at all is because this need runs into my other strong don’t hurt people need. I’ve mostly considered this inability to keep secrets a good thing—means I’m honest, open, people call me “genuine”, means I can write—but this has got to be the major downfall, both on a personal level and personality fault.)
Returning from my tangent. One of the oddest things about not knowing what’s going on in an argument is that you become extremely aware of what other cues you’re using to figure out what’s valid. Apparently, my hind brain gives a lot of credit to whoever is making their voice sounds reasonable, whoever gets nods from other people, whoever everyone yields to, whoever talks fluidly, and whoever talks the most. I will listen much more closely to anyone who exhibits these traits, and ignore people who do not. I will listen less to awkward people, and people who do not make eye contact with me. I will listen more to people who seem to have demonstrated clear thinking, either in the past or in the number of responses they receive from others. In short, without having any idea of the actual content, I have so many biases about which people I’m listening to.
I don’t think I’m alone in this, but I also see examples of people who do better than this. I tend to dismiss arguments from people with whom I’ve disagreed in the past, without fully trying to comprehend their arguments. I will change this attitude if other people indicate through their engagement that the argument is valid. This situation isn’t just a me-problem, but it’s very strange to watch myself have these thoughts: oh, they weren’t clear last time, I’m not going to listen now. Shoot, wait, that other person who I respect just added their support to that argument, means I need to listen. Nah, the other-other person who I respect over there is going to summarize all of the discussion of the argument in a few minutes, I just need to hang on and listen when that happens.
This circuit of thoughts is a more minor problem when I understand the points being discussed. I’ll place lower weight on people whose arguments I haven’t understood in the past, but I’ll listen to everyone. This circuit of thoughts is a major problem when I don’t understand what’s going on. Basically, I judge everyone by my ability to have understood their past arguments, so if people weren’t clear in the past, I’ve just completely lost touch with you even though you could have been saying very valid points, just in a way that wasn’t crystal clear to someone who didn’t have the proper background.
It’s a pretty stupid way of doing things. I see why I do it—I’ve got narrow bandwidth for hard things that I don’t have the context for—but it’s deeply unfair and if I could focus better I wouldn’t have to use it at all. I want to be one of those people who listen to someone else’s badly-phrased ideas and hear the important message in it and then support the speaker and rephrase it for everyone else. I’m never that person (too affected by how others are behaving around me) but that’s what I should strive for.
I’ve recently been introduced to the term “steel man”, which is the opposite of “straw man” and means you argue against a stronger form of the argument that has been presented. In my mind, this means that you pick out the best thoughts from what someone has been saying to you, and then talk about those. Now that I have this descriptor, I can see examples of several people who use it very well. Prof. Griffiths will always take the best possible interpretation of every question you have, expanding even basic understanding questions with new knowledge that everyone can benefit from. I’ve been watching another graduate student do it pretty masterfully—he’s very encouraging, always takes the best possible form of the delivered argument, is flexible in his discussions, and just keeps reframing everything for you formally and then asking if that was what you meant. The postdoc who I’ve been working with also does it with me—he’s explained that everyone has good ideas, you just need to dig around until you find the best ones and then expand them a bit. I think that for everything I do there are some really great people who do things very well here, and emulation is hard, but I think the first step is just to notice.
(Ignore the last part about the zombie, that’s context-specific.) But I think this is THE skill I’m hoping to learn in grad school. Stop paying attention to cues, start evaluating all arguments, be confident enough and have enough bandwidth to listen to what people are saying. Well, except the believing that people are lying bit. I prefer to believe that everyone is always telling me the truth despite evidence to the contrary. I’m not rational in that way, but it just makes me much happier to believe that everyone’s telling me actually how they feel / what they believe. (There is, however, occasionally that odd moment when they then perform an action directly contradictory to what they said, and then I’m like, sigh, Monica…(and go ask my friend Tiffany :)).
I’m pretty sure my two singular goals in life are to be liked and not waste time. It is frankly amusing how much of my behavior can be derived from those two goals.
(Of course, the non-standard and still-in-progress ways that I’ve chosen to “make people like me”, and what I consider to be “not wasting time” add all the complexity back to the problem. It’s going to be so bizarre when we finally get what makes us human mapped into machinery. Which is gonna happen someday. I BELIEVE!)
From Hermione’s perspective:
This kind of writing makes me so happy. It just shows such a strong grasp of different characters. I always admire things that I’ve worked at and am still not very good at, and being able to understand two people in their own mindsets, not through my own, is something I value immensely in people. Maybe this can be accomplished if you have a lot of people in you, so you’re better at simulating what other people think. Maybe this can be accomplished if you are a more flexible person and readily incorporate traits from others. Maybe this can happen if you’re an intuitive person and can read other people really well. Well, I admire it, and here specifically I admire how both sentences are clear and yet display the voice of each of the characters.
Another thought I had about this paragraph is that elements of it can be weirdly reflective of my own writing. I don’t notice it so much on the blog, but I’ll read over notes I’ve written like “I totally dreamt about you last night, wishing you the best from both planes”, or “seems kind of derogatory—that’s not the right word—but like, informal, negating the impact.” People will understand and agree with these statements. What’s most jarring to me when I read them is the informality of the language coupled by the relative insight of the thought expressed. A college English teacher once told me I was holding myself by using simplistic language when trying to express complex ideas. She also told me she was glad I hadn’t been trained as an English major because I was able to express things in an accessible way. I know on my part that I’ve been thinking about being less slangy (using “like”, “dude”, really informal, immature words) for a long time on the blog, but that’s usually what appears. I can write formally—my goodness, just read some of my science papers, they’ve got passive tense and formalism nonsense all over the place—but sometimes I think this is still too uncontrolled. That said, I haven’t written a completely high-as-a-kite post in a while (ha. I’ve never been high, but those are the posts where I’m feeling buoyantly mood-swingy happy. Huh, hadn’t even realized I hadn’t done one of those recently!) so it’s not as bad as it has been.
I’ve been thinking recently that who we hang out with and what we like about others is actually pretty reflective of who we are. And what I hadn’t realized until recently was that I actually enjoy listening and asking questions a lot.
Some friends and I had dinner together recently (great fun, also amusing in that we all really enjoyed each others’ company but still wanted to do one-on-one meetings after) and one of them mentioned that I hadn’t been speaking a lot. And I thought: no, I’d been speaking plenty! At least, I felt completely engaged in the conversation and that I’d been listened to and that I’d been happy with my level of participation. And then I realized that maybe in sheer volume I do sometimes speak less. Not all the time—I can go on about things just like anyone, I can be loud—but I actually consider it just as enjoyable to have an equal level of engagement but not be saying quite as many words. And, surprisingly, this is kind of a new situation for me.
Usually I like talking. I like talking about myself, most people do, and when I’m actively listening it’s a turn-taking and attention-consuming thing before I know it’s going to be my chance to monologue for a while. I also find myself in conversations where I’d mostly be happy if people just left me alone and I could listen and take notes. This usually happens when someone’s delivering a lot of information to me and I feel like I need to go look it all up afterwards and need to absorb as much as possible without feeling any pressure to respond with good questions.
Then there’s the kind of conversation I’ve been having with Smitha, Falk, and Carson. Conversations in which I’m never completely sure where the ground is—they all know a lot, the knowledge bases aren’t the same—and I’m perfectly comfortable in that they’re all going to clearly and happily explain whatever I ask about. And that’s fun, because it means I get to suck down new information, get to interact with them as equals, get to ask them about overall frameworks and where the knowledge fits in, can contribute thoughts and insights as I have them, AND I don’t have to worry about directing the conversation. It’s fricking awesome.
I really appreciate being around people who like explaining. I think there’s a certain and limited type of person who is willing to and enjoys doing this, and has the knowledge base to do so. Moreover, everyone’s been listing to me what the advantages of explanation are, such as how many people learn when they’re explaining things, and that the process of explaining and talking generates new thoughts. Also, sometimes people like trying to communicate clearly as a goal, and sometimes people are using communication as a goal to improve their general processing speed (Smitha :P). Regardless, it’s great being in a learning mode throughout every conversation, I love prompting questions, and there’s always a nice back-and-forth. Finally, it’s been really special having people write comments down from my blog for discussion. My insights sometimes operate on a few hours time delay, and it’s been lovely to have people engage with me over what I’ve said here.
HPMoR draws the distinction between being ambitious (wanting to succeed) and having ambition (being goal-driven) more than once.
Here’s another quote about it from Harry’s perspective:
What I liked here was that “ambition” was called “the common endeavor” by a research scientist, and that Harry considers a lack of ambition as rather useless because it wouldn’t have a wide-spread impact.
When I read HPMoR, I commonly screenshot passages like this. It’s because I feel that I have always had ambition, but I’m not quite certain what the goal I’m trying to attain is. I have a pretty good idea of what those goals might be, and how I’m going to present them (I’m getting closer to that post), but the fact that I went after being ambitious before having an ambition is interesting to me. I feel like most people do this, in fact. We take the actions that are going to lead to some sort of success far before we figure out what success means to us.
Additionally, I liked the idea of “ambition” being “the common endeavor” to a scientist :). I’m not sure how to interpret it, but I like the following: everyone should have some goal they’re trying to achieve—it’s the common endeavor of the human race—and figuring out that goal is as part and parcel of that endeavor as achieving it, and moreover that the scientist’s goal of achieving new knowledge and learning is universal. I’m reading much too far into it, but it was something I was thinking about.
Finally, I have a strong sense that nothing I do is really worth it unless it affects more than me. That’s actually one of the traits I most appreciate about Harry—he’s so committed to whatever his ambitions are (Harry has ambitions, in this novel) and sometimes I read the quotes and think: that’s one of the most motivational ways I’ve ever heard what I feel framed.
Hmm :). I think that’s all for tonight, all—I started writing a paragraph but then lost the point partway through, which means it’s HIGH TIME for sleeping :). Much love to you all for reading, and sending you best wishes!