I recently attended Quora’s Open House, and got the opportunity to ask their CEO, Adam D’Angelo, a question about the company.

I asked him about their origin story, and how different the company’s present form was from what he and his cofounder envisioned. Adam talked about how they were at the trough of a wave that was about to crest when they started Quora. I thought that was a rather mature way of looking at it.

Quora, Reddit and Facebook are essentially web apps. Network effects played a considerable role in getting them off the ground. D’Angelo thinks the era of web apps is largely over - in that creators of new web products can rarely have the right place, right time advantage any longer. It is too saturated. And he had a point.

Peter Thiel talks about this too, in his fabulous book about startups and world domination Zero To One. Right in the book description,

The next Bill Gates will not build an operating system. The next Larry Page or Sergey Brin won’t make a search engine. And the next Mark Zuckerberg won’t create a social network.

With a caveat, of course - unless it is so ridiculously better than everything that has been built thus far that it’s a runaway success. Because innovation is often incremental, that is often a far fetched goal. An alternative would be to find a niche.

If a wave is peaking now, and you are not a part of it - you’re likely too late already. There’s this great tale about JFK’s father, Joe Kennedy, on similar lines. Joe Kennedy was apparently getting his shoes shined one fine day, and the shoeshine boy started giving him unsolicited stock tips. Kennedy decided the “stock market was too popular for its own good”, he recognized a bubble, in other words - and went on to sell most of his positions. Kennedy got “filthy rich” by subsequently trading options and betting against the market1.

Let’s look at popular interest in web development, and shadow that with Machine Learning and some other search terms.

search terms

This isn’t clear-as-day by any means, and a lot of factors contribute to these graphs. The intrigue surrounding AI is probably a factor for the rising search prevalence of Machine Learning. But one can make out peaks of some sort - people hearing of other people making money and trying to join in. Again, I realize this is rather wishy-washy, but there is some truth to it. Especially considering how we’re arguably closer than ever before to a state of perfect information in the economy.

Waves are useful when you want to make a quick buck, not necessarily a small amount of money, but quick money. That’s certainly a valid consideration, and drives decisions for a lot of people. If you’re into CS, picking a specialization that you think is on an upward trajectory would be a good bet from a financial perspective.

Now, a nearly foolproof strategy for career success would be simply to be the best at what you do. I like the idea of being a specialist-generalist, and there’s definitely value to that as well.

But if that sounds like way too much effort, pick something that you think is about to crest.