Live data from Hacker News

The Risk of Discovery

paulgraham.com

111–120 of 344 posts

Re: The Risk of Discovery

#112

The core idea of this article is that high reward is usually accompanied by high risk. To exemplify this, pg gave a great example of Newton pursuing physics, alchemy and theology. However, this was just one example. I think the article could be greatly improved by giving, like, 50 examples (in an abbreviated form, like "Newton: physics, alchemy, theology"). With one example it's sorta easy to think, "eh that might ju…

It's a good example but I think the core of the article is that there is inherent success bias in the history of science and engineering. The scientists of the past that were working on dead end problems were not necessarily worse scientists, it could have been pure bad luck. But in hindsight it's difficult to tell those people from the incompetent.

Re: The Risk of Discovery

#113
post #37

Earlier quoted context omitted.

Judging that way each and every one of our actions (including me writing this comment) is a bet, because AFAIK we still live in a world where we do not have "complete information". From time to time I think about one of Mark Twain's short novels which I read as an adolescent, where a lady decides to take the left road instead of the one on the right (or vice-versa) when she reaches a road-junction, and because of tha…

Indeed! And how about that classic Robert Frost poem, "The Road Not Taken" [1] Edit: WOW, after reading the commentary on the poem, I never realized how wrong I was. The poem is a joke, he's saying it didn't really matter which path he took. All these years I took it so seriously, which apparently is a common mistake dating back to the very first time the poem was read in front of an audience. [1] https://www.poetryf…

There's an episode of The West Wing that brilliantly uses that misconception. Can't find it on youtube unfortunately.

Re: The Risk of Discovery

#114
post #99
post #82

Earlier quoted context omitted.

Capitalism don't require irrational behavior. What is needed are things like comparative advantage and economic profit[1]. Someone who has capital has different incentives than someone who whose major offering is labor or expertise. Newton's studies were motivated by his dislike of farming and need to revenge a bully. He got time for his most productive work because plague closed the University for two years. ---- [1…

Do you agree there is a difference in behavior between startups and big companies, though? In the way they are willing to invest into something? If you agree, then how can you say both act rationally? Why wouldn't rational people who have labor to offer, instead of joining a startup, go work for a big company? The pay is usually better I hear. Addendum: I am sure you can explain the difference in behavior by invoking…

You can explain it using prospect theory https://en.wikipedia.org/wiki/Prospect_theory

>If some entities are rational, they should perceive the world the same way and thus have the same preferences.

This is incorrect. Rational economic behavior is sequence of choices that gives the optimal level utility for the actor. Utility function is given and outside rationality.

Different people have different capabilities and resources available to them. Even if they had different utility functions they would behave differently for the reasons I described above.

Re: The Risk of Discovery

#115
post #59

It's amusing that he uses a modern chauvinism ("making a bet") in this way, while noting that some of Newton's "bets" were only "wasteful" (risky, bad, whatever) when viewed through the lense of modern knowledge. He has a good point, put poorly, and without good support in his chosen example.

What do you mean by calling the phrase "making a bet" a chauvinism? It just seems like a figure of speech. Gambling has existed for a very long time. (I don't mean to disagree or agree with you, just genuinely don't know what you mean, might end up going either way after you explain your cryptic (to me) remark)

In this context it imposes a modern view of "opportunity cost" that would simply not have occurred to Newton or any of his contemporaries. It's just too far back in history to really be useful as a support for his argument, especially given the convoluted and mixed approach used.

Re: The Risk of Discovery

#116
post #110
post #57

Earlier quoted context omitted.

Thank you for reading and commenting with good faith and a, as far as I can tell, correct historical frame in mind. As opposed to with some literalist naive and/or hostile, contemporary-navel-gazing nonsense that is completely missing the point of the essay. For more thinking in a similar direction, see Nassim Taleb's excerpt on the Lindy effect ( https://medium.com/incerto/an-expert-called-lindy-fdb30f146e... ), whe…

" If you hear advice from a grandmother or elders, odds are that it works at ninety percent. On the other hand, in part because of scientism and academic prostitution, in part because the world is hard, if you read anything by psychologists and behavioral scientists, odds are it works at less than ten percent " Some of my favorite books of all time were Taleb's, but I have to say, his writing is the perfect demonstra…

I agree 90% of his insights are worthless, and yet the real insights that remain make reading his books very worthwhile. If I could produce 1 page of genius for every 9 pages of trash I would be ecstatic! Most authors never produce any real lasting insights at all.

The value of contrarians is not in the things they're wrong about, it's in the things they're right about. I'm pretty sure that was also the point of pg's mini essay.

Re: The Risk of Discovery

#118

It's either ignorant or insulting to name theology in the same breath as alchemy. Theology is still an academic discipline and it doesn't even require a believe in the supernatural. Unless this is pg's attempt to insert himself into fight about "theology" vs. "religious studies".

How does theology not involve a belief in the supernatural?

Re: The Risk of Discovery

#119
post #48

I find it funny that capitalism requires what is essentially an irrational behavior. It seems to me that big public corporations, which avoid risk, and only buy startups once they have been established, are the rational actors. On the other hands, three guys in the garage doing a startup often do it not because they calculate expected profits, but because they want to do it for other than monetary reasons - typically…

About the people on the garage, I don't think you can say it is irrational if they get a big non-monetary benefit from doing it.

Nor if they do it for non-monetary benefit.

Re: The Risk of Discovery

#120
post #96

Wait, I thought alchemy and theology were pretty mainstream back then? How was it risky? Just because they turned out to be wrong, I don't think it was crazy for someone to be studying these things back then. Seems like a modern example would be someone studying something like string theory or dark matter and later we discover these things are wrong... but there is no reason to think that today.

The penalty for unsanctioned alchemy was hanging. His theological conclusions would have been considered heresy. The penalty for heresy was the loss of all property and status and, sometimes, death. He was indeed taking a serious risk in studying these things.

While I appreciate what you are saying, I think the type of risk pg was talking about was the risk of wasting his time/career, not the risk to his life. The more I think about it, I think a good modern example may turn out to be someone studying AI (specifically AGI). If it turns out to be possible, and you succeed, it would be huge. If it turns out to be not possible, you would have wasted your career. So maybe the lesson is to do more than one thing.
Post reply on HN