Fashionable Problems

I love Joseph Campbell but his advice to his students to “follow their bliss” may not always be optimal.

I read the AI book “Mind Inside Matter” in the late 1970s, and even though I have done a ton of non-AI architecture and software development, I have also been able to work on AI problems like knowledge representation, expert system, NLP, neural networks, and deep learning starting in 1982. I definitely followed my bliss, but I have never been world class in my profession, but I have enjoyed myself.

The platitude is directed toward those living a life they feel is negative, but who slog through for whatever reason: fear of change, obligation to other people, etc.

Taken as feel-good, head-in-the-sand optimism, of course it's facile advice.

But imo there is genuine wisdom in it, and it's this: You are much more likely to do something well, and to continue to do it until you reach a high-level of expertise, if you don't have to force yourself to do it. People who naturally love working out are more likely to be fit. That's what the heuristic is.

Actually, I think I know the reason. pg writes to maximize brevity and directness. He's always interested in the shortest logical path from A to B. When language is optimized that way, it has a force that can sound like an implicit claim to authority. So his essays land that way with some readers, and then they react with a sort of protest: who is this guy and why does he get to act like an authority? But he isn't—what he's doing is stripping out everything extraneous, and that includes the soft touches that normally soothe the reader—"this is what I think", "your mileage may vary", that kind of thing. He's not stripping them out because he wants to provoke or thinks he's an expert or anything like that. Rather, it's a matter of taste. He likes to make things minimal. The same impulse is behind Bel, or the design of HN's front page. This is someone who might spend weeks getting rid of a single line of code if he was convinced the program could be shorter.

Before I met pg I thought it was impossible for someone to talk the way that he writes. But he does. He's always going for the shortest way to put things. What's clearer in spoken language than in writing is why he does this. It's for pleasure. In person it's infectious, because he clearly does it for its own sake and for fun. It's also essentially the thought process of an essayist: someone on the hunt for just the turn of phrase that makes a point in just the right way, and then moves on.

> If that is the case why do you need others expertise to vet it before posting?

They might think of something you missed or find parts that are unclear. But I don't think he's looking for experts so much as good readers.

Aside: This SMBC shall be long etched in my memory: "All the best work has been done over here! [...] The funding is here too!"

---

Here's a teaser challenge for someone who has access to HN comments data for analytics: How often does an early comment come to dominate the discussion, rather than a better comment which comes later? Assume that there is some IID stochastic process which generates comments of varying quality (play with your distribution of choice), and plot what the time distribution of "best" comments looks like under that model -- compared to the time distribution of most upvoted/discussed comments on HN (can try different composite metrics here too).

---

Provocative meta question for PG: Are (venture-funded) startups the "fashionable" idea that is over-subscribed? How would one know? (suppose we restrict the discussion to Silicon Valley)