How to Start Google

Paradoxically, avoiding risks makes people more likely to take the riskier options.

This isn't an argument. Not everyone wants to stare at a computer for 15 hours a day. Maybe you do. Some people prefer in-person interaction and discussion, others don't.

Yes, in 1-to-1 comparisons, a YouTube video about a topic may be better than a university lecture in which you are mute and can't interact with the professor. But university isn't just mutely watching less qualified professors give lectures. Lectures are but a small part of the entire university experience, which also includes: learning how to work with others, discussion groups, office hours with the professor, managing your time to complete tasks, building social skills, dating, on and on. It's a complete package. The key element is that you are immersed in this environment and interact with the community of people around you. A university is not merely a place that stores information. When you attend a quality university, all of these things are a part of the learning experience. They are not separate.

To use myself as an example: I have a degree in philosophy from one of the higher-ranking programs in the world. The Internet and YouTube are full of lectures and podcasts and free books about philosophy. And yet – none of them comes even close to the experience of sitting in a room for 3 hours intensely discussing metaphysics, or logic, or Kant, or another complex philosophical topic with a small group of other students and professors. That doesn't happen on the Internet and no amount of innovative video formats is going to change it. When you directly interact with another human being face-to-face and discuss/critique ideas, (in my experience) you learn much better than just watching videos. There is a huge difference between someone that has merely read a lot of books on a topic, and someone that has been through a rigorous discussion and critique with qualified interlocutors.

Finally, on the networking element: if you look at most of the actual top performers in the tech industry, they all have one thing in common: they know each other. Learning how to interact with people – and not merely acquiring information – is a key element of actually being successful. And that's still learning. It's not some nebulous hand-wavy thing that you can brush away. Building the next Google, which is what the original essay was about, necessarily requires you to build and navigate complex human social structures. It's not something you can just watch in a video. Merely having information in your head isn't actually that useful if your goal is build a company.

Universities kind of double down on this equivalence in some ways, as they are the de facto credentialing system for the job market, although there's some evidence that this is for the first time in quite a while becoming less true in some sectors, it's still a huge part of the reality of how social mobility is gated.

So when Graham frames the role of "selective" universities in the success of various famous startups, he's totally right that they're excellent places to network. But his conception of their selectivity is pretty strange. It's pretty well-understood by now that a lot of mechanisms can get people into great schools via the wealth and influence of their parents rather than "smarts and determination". I once read someone who pointed out that bringing in a few "scholarship" kids who are less well off makes it seem like anyone could get in with the right grades and test scores, and the incentive is to have just enough of these to ambiguate who is there for what reasons. To launder the "wealthy, well-connected" with some "smart, determined" cred. It's telling therefore that he leaves out one of the most important parts of the networking value of universities: They're a great place to meet funders, and the top tier universities also have significant resources to facilitate student projects that may turn into companies, compared to other institutions

"Build a search engine so good all nerds on the early Internet abandon AltaVista and switch to it, no marketing required."?