This interview is part of the “A Peek behind the Curtain” interview series.
Gregory Lewis is a DPhil Scholar at the Future of Humanity Institute, where he investigates long-run impacts and potential catastrophic risk from advancing biotechnology. Previously, he was an academic clinical fellow in public health medicine and before that a junior doctor. He holds a master’s in public health and a medical degree, both from Cambridge University.
In this interview, Greg and I discuss his productivity style, experience with forecasting, and advice for careers in biosecurity.
Note: This interview is from early 2021 and some parts may no longer accurately represent the guest's views. The transcript has been edited for clarity and readability.
——————————————————————————————
Lynette: What is rest versus rejuvenation for you?
Greg: I'm not sure I'm very good at this, but I think I luck out in the sense of maybe having some degree of resilience. I'm not sure what it is, but I find that I'm not that good at looking after myself, but I seem to do fine. I think other people have to spend a lot more effort to look after themselves properly and seem to have more challenges despite that. So I got lucky in that respect.
I have some scattered thoughts around this, which might just be more like a survey of my own shortcomings. One is, I have a sense of junk food dopamine. Things like video games, watching YouTube, whatever else, that maybe loosely correlates to what I think I understand about rest and rejuvenation. If I've been working over two weekends in a row or something, or working flat out on a project for ages, I do get a sense of like, "Well, I just want to be a rat in the Skinner box, pumping myself of dopamine." That occasionally happens. It's usually fairly short periods. Usually, a weekend is the archetypal one. I feel like I've done this thing now, powered through something mildly aversive, so I’m going to reward myself by staying in my room, playing computer games, and buying takeaway for many hours on end. These are not activities my better self would particularly endorse, but I think they probably reconcile occasionally as a cost of doing business.
I think one thing I would like to do, which I would like to develop further, is to have more of a more organized routine, which I can interleave variously wholesome activities, like reading books, doing exercise, going on walks in a more regular way. I have done that over periods in the past. What typically happens is I get a key project which need to be done really urgently, or I just get obsessed by work. Then it falls fall by the wayside and it takes a while to get back in the saddle.
Probably what I should do is following Tara's lead and evaluating whether in fact I'm benefiting a lot from doing this exercise, or I should just accept myself as someone who tends to work in bursts, and does things somewhat sporadically. It's more idealized, I guess, to schedule my entire day to be most effective, but I've not got there yet.
Another thing related to that is, I typically think I should improve more on taking dedicated time off. What typically happens is I work all days in a given week, for example, just like work less on weekends and more during the week. It's like smearing my work time generally across days and hours.
I feel like maybe I should at least try a bit more concentration and see if having more dedicated time could serve me better. I've done that occasionally. It seems to have worked a little bit, but hasn't proven resilient or sustainable? I'm not sure what the right word is. When something big comes along, I’m just like “it’s rubbish, just focus on getting this done.” Maybe I should accept that, or just develop more self-discipline in terms of how I manage my own time.
That's what I do for self-care as it is. Despite that, and as terrible as that sounds, I enjoy like a pretty uneventful and happy mental life. I'm not sure if that's what does it, but it seems to work okay anyway.
Lynette: Got it. You were defining the rest as dopamine relaxation stuff. Is reading and doing active stuff your rejuvenation version?
Greg: I feel the other way around, actually. I read this stuff on LessWrong about like rest days and recovery days, and recovery was more, in my case, like I'm burned out. It's like, "Oh, man, I just can't do anything besides doing this." It's more recovery style, I guess. Not necessarily I feel like I have do and I couldn't do more if I wanted to, but like, "Oh man, I feel like I deserve it," or something.
Where rest seems more like wholesome activities. Like, “I could read a good classic literature,” or “I'm going to do some creative writing,” or some other pastime, which doesn't feel quite as nihilistic as “I'm going to upgrade the numbers on my virtual avatar in this computer game.”
[Break – Call dropped]
Lynette: Did you ever have doubts about whether you were good enough to pursue your career?
Greg: I've never really had doubts about whether I was good enough. I've always been surprised at the prevalence of imposter syndrome in so many people I admire so greatly. It does make me sometimes wonder whether, in fact, I am indeed the impostor because I’m the only one who doesn’t think I am – and everyone else feels like that sometimes, which I'm sure is an exaggeration. Also, to be honest, I'm more prone to think too much of myself rather than too little. Arrogance is quite helpful to guard against costly self-doubt. Obviously, the cost for anyone who wants to talk to me is not trivial, but whatever.
Also, it’s not something I hugely fear either way. I feel like maybe there’s some underlying sense that you could be stealing a job from someone better suited if you're not good enough. I never really felt that. If there are people out there who think they could do my job better and do lots of jobs better than me, they should definitely get in touch. But I’ve really found myself-- If I start to review my CV and what I've done so far, it compares reasonably well or favorably to other folks doing similar sorts of positions in the EA community. I don't feel like I'm out of place on the objective merits.
I guess the other worry would be, of course, whether I'm better than average when it comes to people who could be doing this job, or if, in fact, I'm just way below the level which this job needs because of how important the work it's doing is. I guess that might be true in a sense. It's definitely the case that I want a much smarter person than me doing the work I'm doing, but that can't be helped. It feels like working the best based on who I am versus whatever various counterparts of mine --who don't exist and who'd be better than me -- would do instead.
Similarly, though I guess maybe my ego would disagree, I also feel I wouldn't be hugely perturbed if I couldn't do some massively important thing or some potentially massively important thing. Like, back in the day, the career I could aim for was to be the EAs’ dedicated family doctor so I could give more stimulants.
I think if it turns out – if I wake up one morning, I get like Nick Bostrom and all my colleagues telling me, "Greg, well, thanks for what you've done, but we've unfortunately come to the realization that you are so far below the bar of this work, that it's really better that you just quit and do something else." Well, it would be surprising, I think, and disappointing, but I wouldn’t feel like it’s a big disaster for how my life would go. I enjoyed being a doctor, for example. I wouldn't mind doing that again if there was some reasonable way that would contribute to the human project, etc. etc.
I'm not naturally disposed to have these sorts of concerns and so far as I can think it through, I didn't feel-- at least for me -- there's great cause to.
Lynette: For you personally, do you find that enjoying what you do is correlated with or necessary for doing it well?
Greg: I think correlated with, but not necessary. If you got me to do something like transcribe a document from one format to another format and update the citations manually one by one, which I occasionally have to do. Thankfully rarely. That's not a very enjoyable topic and I sort of find myself reluctant to work. I probably won’t do it very well for that sort of reason. There are some tasks I know I dislike intensely, and happily there's a few I enjoy a lot. Like commenting on an interesting google doc, for example, is something I naturally like doing. I'm probably better at that than trying to schedule something where it's very operational and I'm not naturally well suited to the sort of work perhaps.
A lot of what I worry about and what I’m working on is stopping other people from being malicious or stupid, like misusing biotechnology. It's a little bit weird to say you really enjoy this line of work because there’s a sense where the fact your job is necessary in the first place reflects various unflattering things about the world we live in and the human condition as a whole. So I'm not sure I want to say I really enjoy the work I do, but I find it fulfilling – that’s a I guess with better word -- and interesting as well.
Lynette: Yeah, maybe there's something there also about it being work that feels easier to motivate yourself to do. Do you have exceptions where you have to force yourself to do your work?
Greg: Not in the typical case. For my job, I can do a lot of different things in my work. Even if I really dislike doing X, I can sort of start working on Y instead. It's not the case that for everything in my work, I’m blocked from doing all of it. I think over time I've gotten less guilty of the thought of, “oh, I find X aversive but I need to do it, so I'm going to avoid it for as long as -- or longer -- than it should be possible to do.” I know what happens next and it’s usually not good. I’ve managed to get better over time at avoiding that.
Prototypical ways I get blocked, one is when I have a problem, maybe how do I structure what I'm going to say about something or how do I puzzle out this modeling problem or something like that. It's a little bit frustrating if you’re stuck there for a couple of days, but you have a sense like, "Oh, I’ll probably get through it. It's interesting problem to think about anyway.” I keep it running through my head while I do other stuff. I usually find that after a while I have some idea and then I start making progress again.
The other blocker is if something is really tedious or maybe I messed up something so I have to fess up and try to correct it. Things like that, garden variety aversive things. I guess over time I've sort of gotten better at just attacking these things and just doing them. Obviously this makes my job easier. I think I've gotten better over time at doing what needs to be done that's not easy or particularly enjoyable, and finding out again and again that it's all better if I tackle it head-on rather than trying to put it off for as long as humanly possible.
Lynette: Sounds good. You want to say more about learning to tackle it head-on instead of put it off?
Greg: Yes, I'm not sure if I have a crisp story. It's just a sense of—Similar to being a recovering introvert who always thought going to a party was a terrible thing to do yet always, every time I went, both enjoying it and then on reflection going, "Wow. I definitely make the right call there." Regretting it when I decide not to do so. I guess over time I've learned that-- at least for me, generally, the sense of aversion usually fades after I actually start doing the aversive task. It's not fun, but it's okay. A sense of relief after you've done it and gotten it off your plate. It seems like worth doing rather than maybe like spending a week or two weeks or two months going, "Oh man, I need to do this. It's even worse since I haven't done it for this long." Getting stuck in that sort of cycle.
That doesn’t mean I'm immune to it. It still happens to me occasionally, but at least compared to how it used to be, I'm better about it, which is a blessing.
Lynette: Yes. It sounds like you need to force yourself through the initial acclimation process sometimes. Have you ever gone long-term needing to force yourself to do something and had that be a sustainable part of your routine?
Greg: I think it hasn't really happened, so hasn't really been a problem. I think if it were the case that there was some project X or some responsibility I really hate doing but I always force myself to do it, like eating vegetables every minute of every day or something like that. I guess that would probably be a problem. I guess I would be like “Is there some way I can not do this or change what's going on?” It feels not hugely sustainable – it feels costly to sustain. Maybe too costly to be sustaining.
I think my pattern of work at the moment is more that there are these aversive things I need to be doing, but it’s never the case that there’s a monolith of an aversion which just keeps being there for a long period of time.
It’s like, if you're working on a project -- having the idea was great, and writing the initial Google Doc was really interesting, getting comments or replies was also really good. I may have to format it all in this particular thing for a journal. It’s a pain in the ass and I don’t really want to do it. That's the aversive bit, but it’s one slice of a larger bit of work.
Interviewer Do you think that following the kinds of things you do find enjoyable and motivating has shaped your career path?
Greg: Maybe to an extent. But I think it’s been a fairly small one. Although it's terribly cliché, I have -- at least as far as I can tell, unless there’s self-deception -- been prioritizing career choices based on global impact and doing the most good and stuff, with fairly little attention paid to whether I actually enjoy doing the work. That may have been fairly reckless, but I’ve never had a very crisp sense that there was something I really disliked doing, or at least, not in one of the candidate job options for me. I generally enjoyed working as a practicing doctor for the couple of years I did it.
[Break – Call dropped]
Lynette: What is your process for deciding which ideas are worth pursuing and which ideas you should let go of?
Greg: So when it comes to research, I probably front-load a lot of my decision making about whether to pursue certain things, so maybe it's worth distinguishing like initiation verses continuation.
In respect for the first, probably quite a lot of it relies on these like fairly brief snap judgments. It's like a set of categories, like, “oh, this should be on my shortlist, it’s really good,” or “this is an emergency, needs to be done for whatever reason,” and then almost like a really bad project, and it's like, “there's some stuff that’s pretty meh, that I’m indifferent towards” and so which like, “oh, it seems like nice to do, but it doesn't seem like a very high priority.” These are fairly broad classes, but they seem to work.
Then also I tend to mostly work on whatever really needs to be done or would be really great to do. I don't really trust my ability to forecast those to have a ranked list in order, so then maybe the projects you’re picking depends on exogenous things like, “oh, I once had a call a policy-maker who mentioned this topic like, should we be giving out vaccines really early even before the phase three trial data comes out?” Then I can start maybe looking at that as a potential avenue for research.
Also, I am certainly interested in one topic vs another one like, oh, this thing on health service at patients like COVID-19, I probably wouldn't prioritize it very highly. It seems potentially interesting, but it's not really the sort of field I'm very excited about pursuing right now, whereas this thing on vaccine timelines or vaccine licensure seems way more interesting to dive into. So it feels like prioritization then becomes a lot more haphazard and based on a mix of sentiment and exogenous things which happen, being led maybe too much by the circumstances.
I typically find I'm more prone to the error of quitting things too early versus sinking a lot of sunk costs into things that aren’t very valuable. I think it’s partly owed to the fact for typically the things I work on are not massively large most of the time, so it's a sense of like, “oh, if you're already halfway through then continuing is maybe two weeks of your time” or maybe less than that, in fact.
Maybe it's worth just like sticking the course because there's still probably some benefit. It will be good value of information. So it often feels it's better to do that. I find myself more often regretting things I didn't follow through on rather than things I kept going with that I shouldn't have. The former is much more common than the latter, so I tend to do an initial exploration perhaps to just vet it. If it does, then I try and commit pretty firmly to seeing it through to some (hopefully planned out in advanced) end product, like a white paper, or a paper, or whatever.
I still have a fairly large number of half-finished products on my back burner, but I try to aim in that direction even if I err from it. Likewise, I find I tend to work better with fewer projects rather than more of them. Ideally one at a time, which is not usually very feasible.
Instead, I try a one-in-one-out policy, with a set number of very large projects. I try to not add another one to the list until I've taken one off it, which again, doesn't always survive first contact, but it's better moving in that direction. Another thing I try and do is obviously, some of Cal Newport's like deep works thing. There's also a lot of, especially with my current job, management and some collaborations on communication stuff.
A lot of things are fairly bitty but I have to try and keep up with, so I try to schedule those towards the afternoon. I try and fill my mornings, hopefully, with bigger advances on a more substantial piece of work. Again, this is all not always happening, but aiming in that direction seems to fare better than not doing so, even if I'm not always wholly successful in my actions matching my beliefs. Who'd have thought?
Lynette: Yes. Do you have any explicit process or idea for your gut intuition for deciding which of these projects rise up to the top of the list when you're doing that first pass over them?
Greg: Yes. I think it feels at the moment like fairly haphazard and intuitive. It could be improved. Maybe one leaf I can pick out of Tara’s book (who you interviewed), would be maybe more forecasting or even knowing the right questions to ask yourself, so I sometimes find my approach doesn't go as well as I hoped it would.
I often go, "If only I'd asked ask myself that question at the start, maybe I would have realized it's a risk going ahead." and maybe that would cause me either to try and mitigate it or just to pick something else instead. One example, somewhat recently, was I was trying to work to a very, very tight deadline towards getting this particular piece of research done. Also on vaccines. My entire life revolves around vaccines at the moment.
I was mulling over whether to collaborate in this group. I thought, "Oh, it'll be good to collaborate with them because they had different skill sets I didn't have and they could add more cache of their background." It didn't work as well as I'd hoped perhaps because they were not operating the tempo I had hoped to operating at, so the deadline kept slipping.
Maybe if I'd have thought “what's the base rate of this sort of thing happening?” at the start, I'd have realized, "Oh, this seems actually quite common and quite likely for many good reasons." These people are often very busy and have ten other things to work on, and you may not be top their priority lists, so maybe it’s top of your priorities but not theirs and so on and so forth.
Maybe if I'd thought about it more carefully, I could have maybe planned ahead what to do differently, or indeed, just tried something else. Where there is no actual way I can get a project with these consideration done in this timescale, so this should be a non-starter and I should do something else instead. But alas. Maybe I'll learn my lesson the second time around, but I didn't figure it out at first sight.
Lynette: What have you done to deliberately improve your ability to do this forecasting with practical applications?
Greg: I think there’s a variety of skills involved in forecasting. I do a fair amount of it on the Good Judgment Open and things. That's one skill which I think one can drill by practicing, which is what I did.
But the metacognitive skills, like knowing to ask yourself to make a forecast, are not trained by having questions you can answer. There, I've begun to try and make some steps on this. I had a fairly large Google doc of notes I make to myself to try and remind myself. I sometimes have a question like that, or prompts I try and make and then try and learn so that, “Oh, if I'm trying to do X, maybe I should think about Y and Z,” and then Y and Z might well become like fairly explicit forecast which I may not write down but have in my head and then use to inform action.
Like, “The timeline to completion here is more than six months, so you have to start now.” Or “My best guess to time of completion is six months, but the tail risk of it taking more than N months is X% and that seems like a bit too high for a specific good idea to start right now,” and various things of that nature. It’s occasionally happened when planning out my Ph.D., like when I'll complete it, and then seeing if I can finally defeat Hofstadter's law by explicitly forecasting the degrees which I was set to fall for Hofstadter's law. We shall see if that works. That's the attempt at least.
Lynette: Okay. It sounds like your research is somewhat different from the normal academic, ivory tower description, where you're trying to do much more applied things. What does that look like, or how do you think about your research in terms of actually accomplishing things with it or influencing decisions?
Greg: Yes, it's the worst thing. That's probably a fairly recent development. Pre-pandemic or a year or so ago, it was maybe much more standard ivory tower, so I would try to really understand a problem and disentangle it. I was starting to take a slightly more practical mindset then. Maybe my field is also one where there’s less massive Knightian uncertainty around it. There’s a sense like, “oh, you can see how bio could be a problem and you can see how it could help.” My understanding is I'm ignorant on AI. I often see very different views on how AI is a problem and how to fix it across practitioners. Whereas in bio, you might disagree about how big some problems are versus others, but there's more of a general consensus of what the shape of the problem looks like.
It's got especially more applied now because COVID-19, although costly in all other ways, is probably, and hopefully will be, the biggest policy window of any of our careers in terms of trying to make things better going forward. You want to start shipping stuff and probably shipping a lot of stuff rather than lots of conceptual work, for example.
In that vein, it feels like there's a lot more problem-solvy, bottom-up, opportunistic sort of things, much more policy focus. I can give an example. There's going to be less in the way of trying to understand the very broadest contours of risk landscape. Some of us do work in that, but it's mostly because you think you know the answer you're trying to persuade others to believe the same thing.
For example, I'm working on a paper trying to say we should be still very concerned about deliberate misuse of bio, especially by states. I'm pretty sure I'm right about this, but I'm trying to articulate it. It's not really because I think it's going to be a really interesting conversation to have where I could find out I'm wrong about this, but more because I think our viewership basically believe what I do and I'm trying to persuade them. It's much more didactic, or proselytizing perhaps, than lots of research in this area.
The other difference is that because it's very proselytizing, the problems are more almost like engineering or mechanism design things. Like, "Oh, we're fairly sure that, for example, these things will help. We aren't sure which are the best ones. We think all of these are things we want to have on our list of things we want to be doing."
For example, wouldn't it be nice if you could develop vaccines, new pathogens in a matter of weeks rather than a matter of months? How could that happen? When you start looking at that problem and go, “well, there's various bottlenecks to how fast you can do this. Which ones do we have ideas on how you could help?” It's like, "Oh, there's a couple we could work on."
Human challenge trials have been talked about a lot, mostly thanks to folks over at 1Day Sooner. Is there a way to make those more widely accepted? Is there a useful contribution I can make in my own work to push that forward? It feels maybe there’s less top-down prioritization or theorizing or strategizing. More this bottom-up, execution-based thing of getting stuff done, which-- Again, how good I am at that is very uncertain, but that's what I feel I should be aiming myself towards.
Lynette: How do you decide when and who to defer to for other's opinions versus relying on your own intuition or understanding?
Greg: Maybe there's two categories. One is daily life such as it is versus my work. In day-to-day, what I try and do is -- given that I wrote this very, very long piece on modesty, I do practice what I preach. Like, is it good to have a higher minimum wage, picking on a political interesting topical topic. I'll probably discount my impressions because I think they're pretty poorly informed, therefore fairly worthless. The question of course then is what to replace it with. How you cultivate the right or the most accurate consensus to defer to is tricky. I'm not sure if there are any cheat codes on how to do it really well. One thing I found, which has helped is happening upon like certain groups or communities who seem to have better than average accuracy here.
EA and rationalists probably count as one, but also super-forecasters. I think in some areas where some degree of background knowledge, I can have adjacent areas which aren't really where I work. Like for in and around public health, who tends to have more reasonable opinions and who tends to have better arguments. I tend to follow these people, but it still is a little bit opaque. It's definitely not a solved problem, especially not by me.
There is also that question about what I do in my work. The difference is that I tend to play my impressions more when it comes to doing work. The reason is that I think-- even if I have a view which diverges a lot from established wisdom or consensus -- it may still be worth me exploring that and trying to figure out more because I'd rather find out I'm wrong. Less likely but still possibly, I might find some consideration which shifts the overall consensus, which seems like a valuable activity.
I think this is very similar to what Scott Summer is talking about when he talks about “my workaround role as a macro-economist is I'm going to talk about my own particular theory in economics which I think I think we should be following and argue for that.” But if I was made president of the world, I wouldn't be following my pet theory, I'd be following the consensus view. I'd like to have a similar attitude here.
Also, in bio, there's often fewer people around. So, there's less people to defer to. In some topics there's not many people working on them, so maybe not anyone to defer to even if I wanted to.
There's also a sense where there's not very stark disagreement in many areas. When I chat to my colleagues and people who work in the area, it’s often just more “I would rate consideration X more highly than you would. You'd rate Y more highly than I would.” But the general picture is fairly similar between us both. So, whether or not I differ to consensus or not wouldn't change a lot of my day-to-day life.
But in fact what I would do is, I would definitely take myself as the academic peer and not superior to colleagues of mine with similar degrees of experience, like Cassidy Nelson who works with me in a very similar job title. I don't think I know better than her more than half the time what's the right thing to do in a given dilemma.
Lynette: What about when setting your own goals? How do you decide which things are most important and most worth prioritizing?
Greg: Yes, it's a mix. I write yearly plans. I have this broad outline of things I want to be aiming towards. My underlying meta-cognition here is I still don't trust my intuitions to get this broad picture stuff right without assistance. I trust it more with low-level stuff. I have this broad outline yearly. I occasionally keep track of how I'm doing with respect to those objectives, but generally not have lots and lots of careful evaluation and prioritization of the things I'm trying to do. Like, I'm trying to work out more. I try a few things rather than carefully tracking how many minutes I spent doing this thing versus this thing. Then I see if I settle on something good. If I don't, then that's a case of taking stock and doing it again.
Lynette: Okay. Do you incorporate others' opinions, either via looking at what other people are prioritizing or seeking feedback when setting high-level goals?
Greg: In terms of my personal life, generally not directly. I don't have a peer review buddy who I send my year plan to or anything like that. There are people I admire, in a variety of ways. Often I get inspiration from what they're doing and try to copy the bits which are good, but there’s less of a sense of direct peer-review of “how should Greg run his personal life?” Maybe that would be good, actually. Maybe I could do with a manager. Who knows?
Anyway, personal life aside. When it comes to professional activity, there is much more of consensus-building activity because bio in particular is a fairly delicate area. So I really don’t want to go off and actually do something stupid. Also, a sense where bio would benefit a lot from people who know a lot about a subject weighing in like, "Oh, I'm more excited about this thing than this other thing. Maybe I should aim more in this direction than this other direction." Discussions like that I find quite helpful. That's usually done by the magic of Google docs, where I write out, “these are my thoughts, this my idea”, and we all comment and then it's hopefully much better.
Also, a thing I’ve found vaguely helpful (and extremely nerdy), I sometimes treat myself like an RPG character playing myself. Often what happens is -- It's pretty well satirized by Rob Wiblin, who talks about, “I've planned out my perfect day. From eight to nine I meditate, from nine to ten I gratitude journal.” You get all these ostensibly wholesome activities and then 15 minutes actually doing work. There's this way you over plan. Then not only that, but if you fall off the wagon, you go, "Oh, it's all a complete disaster. I've failed everything." I think “well, screw that,” It's like you spend six months planning your year because you keep messing up the plan the first time round.
I've begun to go away from this like, "Oh, try different experiments on myself and see which ones work better. Can I prod myself into doing exercise this way versus this other way? Can I tweak my diet in this sense? Does this work? Try for a couple of weeks to see how it goes. If it doesn't work out, then try something else." If you try X-Y round back to Z again, then maybe think about, "Well, is this actually that important to you? Because it seems like nothing's really working. Maybe you should just accept it.” I think it's slightly better than, "I'm going to transform my life attitudes towards these things." But maybe that's just me.
Lynette: Okay. Maybe related to that, one of the questions I was curious to ask you is how do you get work done without a routine? Though maybe that just applies to you?
Greg: It varies. I think I have had routines I've maintained like a fair amount of time, and in terms of sustaining productivity that’s generally where I've faired best in terms of doing work. I'd regularly wake up at time X. I would do exercise for this amount of time. It was not perfectly but reasonably well scheduled. In terms of sustainable activity, that felt pretty good.
The challenge, especially with COVID, is I have exogenous emergencies or just like obsessions. “This needs to be done, preferably by yesterday but if not, sooner” sort of things. Where you feel you could spend an hour on your exercise bike, or you could just get this done. But if that turns out to take several days or two weeks, then you're no longer following a routine anymore.
The other thing is sometimes I get -- maybe obsessed is not the right word, but I've got this project I'm doing and I feel like, "Oh, I just know all the things need doing. I'm just knocking them all out. It's like going really well." I want to keep just doing that and it feels like it's a good trade to make. Just sacrifice all routine to reap the benefits of this flow state or whatever you want to call it. That occasionally happens.
What I haven't really mastered is how to try and combine the two. Bracketing out the COVID firefighting stuff, which is one of those things. If we ignore that, occasionally I’ve said, "Oh, just do this really good burst of activity." Balancing that against having a reasonably sustainable routine, I haven't quite figured out how to do that very well yet.
So it seems, in answer to your question, I am probably typically more productive with a routine, so I try and keep a routine. But occasionally I find that I just hit this flow state and I just want to keep working for several hours or several days on this topic. I just keep on going, and I haven't worked out how to balance out the two or whether I should double-down on one strategy versus the other one. Currently, it’s a bit of a mix and therefore, to some degree, a bit of a mess.
Lynette: Okay. How do you focus on one project at a time without feeling scattered and having all the other things you could be doing sitting on your mind?
Greg: When I get hit by inspiration, that's not a problem. Obviously, that's not all the time, unfortunately for me. If it's a sense like, "Oh, my gosh I have ten different things I could do, which one do I do?" I often find that arbitrarily picking one and focusing on it for the next two hours, seems to be a good way of avoiding getting too scattered.
There's this optimizer's curse trap of like, "Oh, is this most important thing right now? Should I be ordering these tasks?" It feels, at least to me, that the overhead of trying to find this optimal order is sufficiently unpleasant and takes enough time, that you are better just doing something really dumb. Like, just picking one and trying to do it. Both because it may have been more efficient, but also -- if you are feeling scattered and like, "Oh, man, what I do now? I could do this, I could do this." -- maybe a sense of success like, "I've done something—Hey, I've done something today. I've done this thing. I've cleared that off my plate. That's great." Might then get you out of a rut so to speak. At least I’ve found that, on occasion.
Lynette: It also sounds like you have already done the first pass of prioritization. You're picking from a handful of things you think all have good potential, and so the difference in prioritization between them might be less or harder to determine. Does that sound right?
Greg: Sometimes a large part of this is stuff I've already committed to in various ways. It's like, "Oh man, I have to catch up on this thing and this thing and this other thing," and then it's even easier because I have a sense that it's probably good not to drop it, to try and avoid dropping stuff. It tends to go better in the long run. I think there's that side of it as well.
Then another reason against this careful prioritization is that you hope these things are mostly on a par. So, I often find I'm picking the easiest one like, "Hey man, this thing's really hard. Let's pick an easy topic just like do for a bit and then start working forward from there."
Lynette: What helps you get started in the morning, getting down to this deep work instead of procrastinating or doing easier things?
Greg: I think what I generally prefer is to plan out what I'm going to do the night before or have a rough sense of “the thing I really want to do today is X,” where X might be like finish off this complicated table for my paper or something, and then try and do that thing first. Then if I accomplish it, it's great. It's like I’ve already won today even if I completely mess up the rest of it. That seems to go well. Otherwise, things like taking an early morning walk or exercise can also help if I’m not immediately stimulated to do things. So doing easy things first helps more when I’m feeling scattered, rather than my default.
Lynette: All right, okay.
Greg: Again, I’m far from perfect at actually following through on this.
Lynette: Sure.
Greg: I occasionally open up Facebook first thing on my browser rather than Google docs, for example. But it seems to get better when I try and push in that direction even imperfectly, so voila.
Lynette: Taking a bit bigger step back, what are the failure modes that you watch out for in your work? The things that would stop you from having generated useful output at the end of the day, either productivity things or higher-level strategy things?
Greg: Yes, good. What do I actually--? Do I look out or anything? [silence] I feel like I don’t have huge amounts on insight here. [silence] It definitely isn't because failure is completely alien to me. I have no idea what could possibly go wrong with my things I do. I'm always repeating what things I've talked about before, so I guess one way of thinking things go wrong if you don't push from time-- Maybe we just don't really think about-- I'm not huge on the lookout for failure modes of my work.
One thing I tend to think about is “Could my project backfire?” which is more a bio-specific thing. That’s more careful reflection, taking stock, and re-evaluating. In terms of if I know x is a good thing to do, what will stop x from happening? I guess I've changed them mostly but I'm not usually worried about it like, it's almost not following through or not being hugely productive for a while or something like that.
Lynette: What would be the things that would cause you to be unproductive for a while? What's the biggest failure modes there?
Gregory: Getting stuck, obviously, is one thing. Having an aversive task and not doing anything else either or doing other stuff half-heartedly. That's happened to me more than a few times. I think it's happening less as time goes on.
The risk to productivity is not so much like this week where you just do nothing (unless I'm planning to. I occasionally take a break for a while), but more of a sense of chronic occasional losses or maybe not so occasional losses here and there which add up.
One thing I keep a little bit of a lookout on -- Although it hasn’t ever really been a problem for me and hopefully won't ever be in the future, burnout is infamously common. It’s talked about a lot and rightly so. I never really felt anything approximate to it, but I appreciate that given my job and so on, there are some risk factors. So I try and keep some lookout for that. I know from secondhand, this is very good to avoid.
Gregory: Similarly, any chronic mental health challenge. Again, I happily haven't had that, but people tell me these really suck.To be honest, I think I typically err in the direction of not working myself hard enough rather than working myself too hard.
Lynette: What's a skill you've spent deliberate effort developing that's paid off a lot?
Gregory: Yes, I think I'll definitely rate forecasting as a skill which I have deliberately done. I often did it when I wanted to take some break from some problem, like interleave it with something else.
It paid off in a few ways. One way it paid off was that some of the subject matter of geopolitics is relevant to my day job. A lot of my work is sort of international relations and politics adjacent, so developing better knowledge of how the world works has been helpful in informing work I do day-to-day.
I think the skills for forecasting have been helpful. It's a little bit hard for me to tell how much practice improved me at it, and how much I was just innately or whatever, but I think it probably has helped in terms of better work on calibration and things like that. Forecasting is an omniskill. It has applications in various things, not least in planning out one's projects if I used it better, so I think it comes in handy in a variety of places.
It's also occasionally come in handy when people ask me for advice, especially with COVID-19 like, "Oh, will we be able to hold this conference? Will it be canceled?"
It's probably also helped in terms of self-knowledge in that I'm pretty good at this, so that's handy to know that I can offer advice in terms of at least probabilistic forecasting. It seems handy to know for how much I should back my own judgments on certain matters.
In terms of my ego, which doesn't really need much more inflating, I got appointed super forecaster earlier this year, and that's a nice thing to brag about yourself.
Lynette: Sounds good.
Gregory: I'm not trying to say everyone should be doing it. It depends on what role you're doing or what things you want to improve about yourself obviously. For me at least, I found it helpful and it was, rarely for me, something I deliberately tried. I think a lot of the other things I developed some skill in have been developed, not entirely accidentally, but somewhat indirectly.
I'm allegedly fairly good at writing and I also review others' writing suggesting how it could be improved. That skill wasn't deliberately cultivated or trained. It was developed by spending a lot of my time writing stuff and reading stuff and commenting on stuff all over the internet. A lot of my misspent youth and things like that. That paid off, but it wasn't a deliberate strategy on my part.
Lynette: Sure. How important do you think it is to find your niche? In your career it sounds you did a lot of things that paid off eventually but weren't originally strategically planned. I'm curious if you think this is common? Do you need a lot of exploration or is there a “get to the skill immediately” strategy that people could apply if they were trying?
Gregory: I consider a lot of my career extremely serendipitous. Maybe I should have imposter syndrome for this reason. Thanks to some weird case of causal arrows linking up, it's turned out pretty well. For example, going to medical school is a very large investment of your time. For most EA roles, it's not hugely relevant for what we want to do thereafter.
However, for bio, it's actually quite handy to have that credential. Now maybe not handy enough. You don’t want 18-year-olds going to medical school for this purpose. But as someone who already went to medical school, it's partly a sunk cost but the asset has some value, more than it would be typically expected to. I think a lot of those things have paid off for me, thanks to good luck.
I think that means I probably wouldn't recommend this as a strategy for other people. I think there's a lot of returns to reasoned deliberation here rather than following your heart or your interests.
I think some good forward planning is helpful. I think it's a little bit tricky in some areas, like bio where I think the typical career path for people who work in this area are people who started working in other disciplines and got interested in solving this mid-career and moved into it. I think this does imply there should be shortcuts to someone who wants to target this directly rather than spend two decades doing something else and move into it. It does give you some background knowledge but it just doesn't feel very efficient. However, it does mean there isn’t a very clear path of what you would do because no one else has done it that way before. It's a challenge of giving people career advice.
Firstly, the area is pretty poorly mapped. Secondly, the people already there happened onto the right place. So, giving a direct strategy is hard. However, I think trying to do this is probably superior in most cases to being led by interests and exploring a lot of things. If one needs to do that, that's totally fine but I don't think it should be lionized as a virtue.
Lynette: Okay. If you're comfortable sharing, what are one or two of the biggest struggles you've had in getting to where you are now? Things that seemed huge and insurmountable at the time?
Greg: Yes. I may have a really crap answer, I'm afraid. This is chapter n of n+1 of Greg living a charmed life. I never really felt like I had insurmountable struggles basically at any point of my life, regarding almost anything.
What's the closest I've gone? I'm just trying to think of-- I don't know if I'm really at risk of going off the rails. I had to resit a couple exams in Medical school because I messed them up, but it never felt like an insurmountable problem to being a doctor, and it obviously wasn't. Yes, I generally fell through a lot of stuff with a fairly easy run of it.
Lynette: Do you have strategies that you find useful for getting value out of interviews or talking to experts, that could be leveraged for more efficiently getting to a useful part in your career?
Greg: Yes, that's a good question. I'm not sure I have a good answer because I'm often on the receiving end, so people are trying to get me to give them good information. Hopefully, I succeed occasionally.
Lynette: Anything they ask that helps this along?
Greg: I was going to say, that's the natural thing to ask next. I often have, not stock answer that's I don't really care what they're saying, I just going to give them the same response either away, but I have a sense of-- Especially when it comes to what to do with your career, I have a fairly typical thing I say, which is along the lines of, "This area does reward experience and credentials, and so often building up relevant career capital is a way to go, whether that's in grad school or professional service. If there's a really promising immediate year opportunity, which most people won't have because they seldom exist, they are often better served going into this path and trying to build up stock for later."
Any advice I'd typically give? In terms of how to best interrogate that advice-- What did I do when I was an 18-year-old asking these questions? Maybe one thing is trying to get a sense of either calibration or confidence of recommendations. I think my impression is that EA land has gone wrong in the past along the lines of miscommunications like, "Oh, org x says y." While y is meant as a tentative suggestion, we take that as a cast-iron recommendation and then make lots of decisions. Which then, because it was tentative, actually turned out to be mistaken, and we're a bit screwed over by that. That's obviously not just the responsibility of the receiver, but also the transmitter. One way of avoiding this, is trying to interrogate—“Have you examples of this? Are there counter-examples? Can you give a credence on this?” I guess those things would help.
Lynette: This was lovely. Thank you for talking. I just really enjoyed it.
Greg: I'm glad. Yes, I feel like I just don't have this very good insight into my special sauce if I have any. I'm not sure if anyone in fact should look up to me as a role model for productivity. I definitely don't feel I am. I feel like I just have lucky traits and have got lucky in various ways. Ways which are not very translatable. Like-- just read really fast. It's really helpful if you can read fast. Yes, great, but what if I can't? It's always something like that, right? I did pretty well at school, but I'm not sure I can give advice on how to do well at school. I guess I'm quite intelligent in a variety of ways and that really helps, but that feels unhelpful. Despite that, maybe there are some pearls in there you can unearth. If you can, thank heavens.
Lynette: No, it was good. I think just hearing what other people do is still interesting and insightful in a way that is often missed. Have a great day.
Greg: What a pleasure. Have a great day, and catch you soon. All the very best.
——————————————————————————————
Enjoying the interview? Subscribe to Lynette’s newsletter to get more posts delivered to you.