Dear Sam Altman

All Coherent and Fruitful Thought is Restricted

Version: 66149d0

Restricting speech leads to restricting ideas and therefore restricted innovation — the most successful societies have generally been the most open ones. Usually mainstream ideas are right and heterodox ideas are wrong, but the true and unpopular ideas are what drive the world forward. Also, smart people tend to have an allergic reaction to the restriction of ideas, and I’m now seeing many of the smartest people I know move elsewhere.


This is uncomfortable, but it’s possible we have to allow people to say disparaging things about gay people if we want them to be able to say novel things about physics. [1] Of course we can and should say that ideas are mistaken, but we can’t just call the person a heretic. We need to debate the actual idea.

E Pur Si Muove by Sam Altman

I want you to try a thought experiment. Simultaneously, I want you to try to have every possible thought that any person anywhere is capable of having.

Did it work?

I’m confident that it didn’t. Why? Because, your brain can’t simultaneously evaluate the universe of thoughts — time and space bind.

For a milder experiment, I want you to try simultaneously thinking about every bit of knowledge known to humankind.

Failed again, eh?

Okay, one last experiment. Try to simultaneously think about every facet of some small aspect of technological space — say, everything there is to know about Ruby on Rails.

Again, you couldn’t do it, right?

Take the Rails case. Happily, with enough effort and after enough accumulated and dissected mistakes (i.e. experience), you’ll acquire expertise. Less happily, even after all work, you’ll make mistakes.[1] But, hopefully, after all that work, you’ll have learned a critical lesson, the one that makes OSS so remarkably successful: it’s far easier to comprehend a good solution to a problem than it is to find one.

No matter how good you get at developing with Rails, if you’re stuck on something really challenging, it’s probably a good idea to Google around to see if anyone has worked on your problem before. If that fails, it’s best to ask around. Then and only then should you try to solve it yourself.[2]

Okay, so now say you’re in that last category: exploring new territory. If you asked around first, that space is not unconstrained. You have some information on where to go and some on where not to go. You may be traveling alone (unlikely), but the accretion of prior art still guides you. That’s actually how science, technology, culture, and society works. That’s how we route around our individual cognitive limitations. That’s how we as a species get to transcend — at least partially — time and space.

We collaborate.

In your blog post, you weren“’”t talking about technology. You were talking about a sociological aspect of American culture that has an effect on technology recruitment. That your post elicited outrage from a large body of people who study sociology, politics, and culture should provide you with some evidence that, perhaps, you are wrong. They were using their hard-won maps to a space you haven’t explored when screaming, “DANGER: THIS IS A REALLY BAD SOLUTION!”

Of course, Silicon Valley has a bias against sociology — and, in particular, sociologists. It’s a bit of a blind spot, really. But, imagine if a sociologist: 1) decided, “I’m going to learn rails;” 2) spent a few weeks doing so; then, 3) told you (a hypothetical expert) that your work on and using Rails is all bullshit. You’d probably think something along the lines of, “wow, what an asshole,” or “fuck, that’s not someone I’d want to work with.”

But, okay, fine, you’re better than most people, I guess. So, instead of reacting angrily, you suppress the range of human emotions that an adult exhibiting such childish arrogance would provoke. You take on the task of guiding the budding rails developer. What a task! Remember all that time you spent becoming an expert? Think how much labor it will take to course-correct them!

Then remember, most people don’t have lots of free time.

Now, consider your article and where you are in the Silicon Valley ecosystem. You are the CEO of the most visible startup incubator (or whatever you want to call it) in the country. When people — founders or otherwise — think ‘startups’, YCombinator is going to come up.

So, maybe it’s a real problem when you share your thoughts on something you don’t understand. Because, for better or worse, your position means you set boundaries. Similarly ignorant people — and, I mean that in the most innocent sense — will look at you, top of the food chain in their ecosystem, and translate that into assumed authority. You’re like a lighthouse compelling those people to smash themselves upon rocks that don’t affect you — or, more likely, through other people.

That’s why outrage ensued following your post. People use outrage as a negative social sanction because it works, to a degree. It helps limit harm by drawing a line and saying, “no, this is bad!” Sure, it would be better if someone could take you by the hand and calmly walk you through all the reasons you are wrong. But again, remember that arrogant rails newbie? Such effort takes a lot of time and it presupposes a receptivity that you haven’t demonstrated. And, in the meantime, quietly and politely debating you wouldn’t address the ongoing harm that you induce.

You are right on one part — culture acts as a filter upon collaboration. For some reason, you assume that the people who are harder to work with — those less capable of empathy or at least compassion — are the ones more likely to profitably explore the space of technological possibility. I’m not really sure how to address that. It’s pretty preposterous, especially when you consider how much of Silicon Valley operates in socio-technological space.

Then again, it may explain why Silicon Valley keeps fucking up the world. Technology changes the human environment. It’s important to ask, “should we do this?” That’s a hard thing to do because it often puts people at odds with the valorized Fuck-You-Money exit that they desperately want. It’s doubly hard when you don’t have the knowledge necessary to answer that question. And, it’s nearly impossible to do when you don’t even realize it’s a question that should be asked.

Your post expands the population of people working in technology — and, who choose to apply to YC, a firm that drives technological change — who wouldn’t even think to ask “should we do this?” questions.


[1] Actually, the never-zero probability of making a mistake is what makes any intellectual discipline satisfying, so it’s not too unhappily.

[2] At least if the problem is merely a means to an end you care about more.

[3] Originally published on Medium.