Summary
This page describes some personal thoughts and history regarding being stringent vs. laid-back with respect to one's work. Almost surely most of these points are unoriginal.
Contents
My history with flexibility of goals
From ages 13 to 22, I was somewhat of a perfectionist with respect to goals. This tendency was probably helped along by playing video games and by school, since in both of those cases, metrics for success are well defined, and there's a clear maximum possible level of task completion. I regarded it as obligatory to do everything that was assigned, although as mentioned here, I eventually learned that I wasn't able to apply the "finish everything" rule to reading assignments, since often skimming was necessary in order to complete readings on time. However, I could still apply the "finish everything" rule to all written homework assignments.
When I graduated from college in 2009 and began working at Microsoft, I had a bit of an existential crisis with regard to goal accomplishment. While I still had broad tasks at Microsoft, they weren't broken down into small pieces the way they are in school. I had the blessing and curse of being able to decide how to approach a project, what ideas to try, what timelines to shoot for, and so on. (The extent of this freedom varied from manager to manager.)
Freedom and responsibility to choose what to work on was already somewhat familiar from my free-time altruistic efforts. That said, in 2009, my feelings of existential angst became more pronounced in the altruistic realm too. Historically, I had tended to judge that certain projects looked particularly important, and then I worked on them; I felt that what I was doing was privileged in the sense of being more valuable than other possible actions. I had lots of uncertainty about how to make the world better, but I felt I could still do expected-value calculations and come up with optimal choices.
In 2009-2010, the feeling that I was doing what was optimal disintegrated and gave way to an uneasy realization that a lot of what I was doing amounted to random impulses based on arbitrary moral values and questionable epistemic convictions. Of course, I had always been a moral anti-realist, but until 2009, I at least believed that consciousness was an objective property that was potentially somewhat measurable. In other words, the only subjectivity in my moral views was the choice to be a (suffering-focused) utilitarian, after which there were no more free parameters. But once I realized that consciousness was also up to us to define, the scope of arbitrariness that I saw in my work significantly increased.
The following quote is from an email I sent to some friends on 27 June 2010. (I've made a few language tweaks to it.)
One of the biggest shifts in my thinking I've noticed in the past year or so has been to become increasingly less fastidious in my attention to minutiae and my insistence on doing things in one particular way that seems so obviously the "right" way. Instead, I've become more soft-edged and flexible, recognizing that a lot of these decisions about what matters most, what exact goals we set, and such are up to us to decide. A lot of the shift correlates with my change in stance on the hard problem of consciousness, from having a sense that there really is a consciousness "thing" out there, to being persuaded that what we consider conscious is fundamentally up to us.
The best way I can phrase this is to say that I've stopped "believing in magic." I used to view particular actions and ways of doing things as almost fundamentally important. Now I'm not so rigid and am willing to break self-imposed rules.
One difference is that I'm not doing math much anymore and am instead doing a lot of fuzzy, no-clear-right-answer work. Every day I confront the fact of having to decide exactly what I do and how I do it, without an obvious template or procedure to use. Even in my reducing-suffering activities, I tend to be okay with a broader array of options for how things are done, rather than insisting pedantically on doing a particular action that I see [perhaps illusorily] as being optimal.
As time went on, I became arguably even more tolerant of arbitrariness. For example, around 2013-2014, I allowed myself to have more than one utility function—one for "utilons" and another for "fuzzies"—even though this is hard to defend from a systematizing mindset.
The idea is similar to an earlier shift in my thinking (from around 2005) regarding grammar. I had previously assumed that correct grammar was very important and that there was something "in need of fixing" with bad grammar or misspellings. Later I came to realize that grammar is an arbitrary code that's subjective and changes with time. I still try to use proper grammar (1) for the sake of making my writings more credible and (2) as a form of mental exercise. But there's no deeper sense in which correct grammar is important. Moreover, breaking writing conventions when doing so aids clarity, such as by using logical quotation even when writing for Americans, is a choice open to me (except when I'm writing for someone else who demands adherence to different style guidelines).
In addition to apprehending our freedom to choose morality, I also increasingly recognize how ignorant we are in the face of gargantuan altruism-related uncertainties. I often feel like I'm just doing random things that seem plausibly to be net positive in terms of suffering reduction, with no idea whether they actually are net positive, or what kinds of long-run impacts they'll have. With this perspective, it's clear that the magical glow that I had previously seen around specific actions and altruistic causes was illusory.
One of my former teachers related the story of how, when he was in college, he came home over winter break and pontificated to relatives about how the world's problems could just be solved if X, Y, and Z. Looking back, he realized how ignorant he actually was and how much more complicated things are. Based on limited data, I've noticed a similar trend in myself and other effective altruists—that as one gets older, one's certainty about everything declines, and one becomes more tolerant of those who have different approaches for how to navigate life while blindfolded. Perhaps this trend is simply a function of how many times you've discovered that you were wrong and that the full picture is far more complex than you realized. While perhaps platitudinous, this quote from Socrates seems fitting: "I am wiser than this man, for neither of us appears to know anything great and good; but he fancies he knows something, although he knows nothing; whereas I, as I do not know anything, so I do not fancy I do."
How seriously to keep promises
During my existential angst of 2009-2010, I had doubts about commitments and rule-following. I thought something like, "If what I'm doing is pretty arbitrary rather than privileged, why should I adhere to the goals I've set rather than breaking them? I made them up in the first place, so they don't actually carry any authority." I had peeked behind the curtain and discovered that the Wizard of Oz was an old man pulling levers.
People adopt different strategies in response to this problem. For many, the answer is religion: "God is watching and wants me to keep my promises, and His commands imbue specialness to some actions over others." Some people appear to like rigid, fundamentalist religion because it tells them how to live their lives in a detailed way. Other people keep commitments using bosses at work, relationships, and social pressure—e.g., fear that other people will be disappointed if you don't follow through.
In my own case, I've discovered a certain mindset that I can get into in which keeping my promises feels like it has to be done, and contemplating breaking them amounts to treason. I don't know how I do this, but in the few cases where I actually bother to do it, it works pretty flawlessly. Sarah Constantin says: "I think the capacity to make a truly binding promise is a neat trick that more people should look into." This is basically a deontological or updateless mindset, in which calculations about whether it might be optimal to change course aren't allowed. For this reason, I'm extremely cautious about making self-promises and only do so for very short-term and relatively easy things. Throwing out your car steering wheel is helpful in the game of chicken, but trying to drive home without your steering wheel for more than short distances is liable to lead to a car crash.
Incidentally, this is the main reason I haven't taken the lifelong Giving What We Can pledge (the same pledge that inspired Sarah's comment); I cannot reasonably guarantee keeping such a pledge, nor am I confident that I should. And if I were to make less stringent pledges, this would weaken the power of my stronger pledges, because then there would be a possibility of wiggle room. I think the power of pledges comes from disallowing wiggle room, because disallowing wiggle room has the psychological effect of reducing temptations. If there's always a possibility of breaking the promise, then you're constantly wondering if you can get away with doing so "just this once". Constant temptation leads to distress, in a similar way as too much choice can. But if you commit in an absolute way, temptations don't arise, or are quickly quashed, because they're not even on the table as possibilities. I think some have commented on this point in the context of why veganism (or other rule-based diets, like lacto-vegetarianism) can be easier than flexitarianism for certain people.
There remains a question of how much to precommit to certain goals vs. how much to play life by ear. In my teenage years I was very much on the rigid-goals side of the spectrum, partly because I didn't have as much choice about what goals to set. Today I'm usually on the "make it up as we go along" side of the spectrum, where I follow my emotions rather than making specific plans. Still, I find value in both approaches in different contexts.
"Just a little bit" goals
As a kid, when I went sledding with other people, the other people got tired before I did. I would ask for "just one more" round of sledding before we finished for the day, and the other people agreed. After that round was over, I asked again for "just one more" round, and again the others agreed. I kept this up until these requests finally stopped working.
A similar strategy can sometimes be effective for setting goals. For example, rather than setting the goal of working for a full hour on some difficult task, you can set the goal of working on it for just 10 minutes, after which you're free to stop. Once you start working on the task, you often find that it's not as hard as it looked, and near the end of your first 10 minutes you might decide to do another 10 minutes, and then later another 10 minutes. You can keep chaining these tiny promises forward until you feel that it's actually time to stop.
Tiny promises like these are an interesting middle ground between fastidious and relaxed goals. You can make a tiny promise fastidiously to help overcome mental inertia, but because it's such a small commitment, there's little risk that you'll regret forcing yourself to adhere to the goal later on. This approach is relaxed in the sense that after you finish the tiny goal, you can then change plans to whatever suits your fancy.