I recently read a piece on the world’s most powerful computer being developed. How so many people can work on something for so many years without knowing whether anything will even come of it, a friend I’d shared it with reacted.
How can so many people work on something for so long without knowing if anything will come of it? It’s not too hard, I think. For one, they get paid well for it, so the bottom of Maslow’s pyramid is more than taken care of. For another, even in such long gestation projects, you can see incremental progress along the way in the form of milestones, which offer encouragement and keep morale high. And, of course, a lot of people are genuinely interested in what they’re working on; to be paid very well to work on what interests you, without being henpecked to meet some artificial deadlines every other month, I imagine, is a dream very few people are able to realize.
It’s also a net positive for society overall, I think, that some people do work on such long-term, uncertain bets. If no one took such risks to worked on highly doubtful projects, we wouldn’t have many of the things we do today. As a result, even though most such bets don’t pan out and there’s nothing to ‘show’ for all the effort and years people spend, on the whole, the human race comes out on top, even if most of the individuals involved mightn’t.
It may be that to do great work, you also have to waste a lot of time. In many different areas, reward is proportionate to risk. If that rule holds here, then the way to find paths that lead to truly great work is to be willing to expend a lot of effort on things that turn out to be every bit as unpromising as they seem… Newton’s case, at least, suggests that the risk/reward rule holds here. He’s famous for one particular obsession of his that turned out to be unprecedentedly fruitful: using math to describe the world. But he had two other obsessions, alchemy and theology, that seem to have been complete wastes of time. He ended up net ahead. His bet on what we now call physics paid off so well that it more than compensated for the other two… Here’s an even more alarming idea: might one make all bad bets? It probably happens quite often. But we don’t know how often, because these people don’t become famous.
Paul Graham, Bus Ticket Theory of Genius, Risk of Discovery
Might one make all bad bets? Simple probabilities would indicate that that’s the most likely outcome. If you have even a one percent chance of succeeding, and over the course of a working lifetime you manage to make ten such bets, you still have a ninety percent chance of missing all of them. With a five percent chance of success there’s a sixty percent chance of missing all of them, and at ten percent, a one-third chance. Of course, in the case of something like quantum computing or Newton’s bet, the odds are way less than a percent.
Even if this choice is beneficial for humanity overall, however, that doesn’t mean one is bound to go for it. The only pertinent analysis that really matters is at the individual’s level, from their point of view.
The expected value of working on a big, uncertain, long term bet, versus working on a smaller, less far off, more certain one, is comparable. In the former case the payoff is extraordinarily high but the certainty smaller, in the latter, the payoff, while still high, much lesser, and the certainty, while still low, much higher.
It is, I think, like aiming for a local maximum as opposed to a global one. A top college, and then a high paying or prestigious job, for example, which, though it might seem like a global maximum to those who dream of it, a tremendous achievement, to many, is often a local maximum for those in it – something you achieve quickly, and then, so to say, peak – very early. On paper, you seem very impressive, all crests and no troughs, only highs, no lows, no real failure, except a hypothetical one. “It is dangerous to venture. And why? Because one may lose. But not to venture is shrewd. And yet, by not venturing, it is so dreadfully easy to lose that which it would be difficult to lose in even the most venturesome venture, and in any case never so easily, so completely as if it were nothing …one’s self.“
In a global maximum, on the other hand, I suppose, you don’t peak, for a very long time, perhaps ever. And a global maximum presumably requires not getting stuck or stopping at a local one, so you don’t look very impressive to anyone for a very long time, until, and if, you ever reach your global high, in which case you suddenly do. In the most likely scenario, however, you probably miss the peak you thought existed, in which case, those who went for the local maxima now appear to have been much smarter.
Of course, like any expected value calculation, neither is universally ‘better’; one must make their own calculations. One always romanticizes what one lacks. Those at a local crest often conjure imaginary global ones; those struggling and far from their intended global high sometimes ruing the local ones they spurned along the way. Listening to someone who’d struggled with their own endeavour for the better part of a decade before folding it up – to ‘go back to where they started from’, more or less, showed me this more clearly than I could’ve ever intellectualized myself.
I suppose a local maximum is less ‘ambitious’ because it’s easier, and hence, one often hears the subtle condemnation that at least those who strive for global highs do not live with the regret of what might’ve been. “by not venturing, it is so dreadfully easy to lose that which it would be difficult to lose in even the most venturesome venture…one’s self. I wonder, though, if it is so straightforward, if this attempt at hypothetical potential regret minimization doesn’t itself skew a person towards a choice they’d not otherwise make.
Like with any other pendulum, this one too swings from one extreme to another – in the past, perhaps, the overemphasis on safer local maximums over risky global ones. The glorification, today, I think, is more of the other side – praying at the altar of hypothetical global maximums, and the slick warnings to shun local ones, lest they trap you.
None of which is to revert to the prior extreme of sheltering exclusively in local maxima and spurning the seductions of global ones. A global maxima, though, I think, needs to come from within, as perhaps Amjad Masad did with Replit, rather than be pursued for the sake of it just because it is a global maximum, a pursuit that I’d guess rarely meets with success.
Not everyone need have a global maximum lurking inside them somewhere, however. A large chunk of humanity, I’d guess, maybe the majority, might not have any particular grand goal they want to achieve; a local maximum, perhaps, is good enough for them. Worshipping global maxima, though, means the choice between the two kinds of maxima is no longer a balanced one – even a person with no concrete global maximum in mind might well feel urged to pursue a one for the sake of it. And when a choice is made with unbalanced mind, the expected value from it usually turns out to be negative.
When you value only grand, global maxima, perhaps you forget, as Alain de Botton points out, that there’s nothing wrong with an ordinary life, that it’s more than good enough. And you devalue what you actually have, which is no longer good enough, because of what you don’t, which is all what you believe matters. “Thus when the ambitious man, whose slogan was “Either Caesar or nothing”, does not become Caesar, he is in despair over it. But this signifies something else, namely, that precisely because he did not become Caesar he now cannot bear to be himself. Consequently he is not in despair over the fact that he did not become Caesar, but he is in despair over himself for the fact that he did not become Caesar.”