How to use the internet without turning into a zombie

Welcome to the internet! I see you’re new here; let me show you how things work around these parts. This is a pretty dangerous place - not in the exciting cyberpunk sense, it can’t kill you, but if you’re not careful it will turn you into a zombie and use you for fuel. Actually, the whole cyberpunk thing isn’t terrible as a metaphor: as on all frontiers, the internet features a few savvy people who exploit the system in ways that are increasingly illegible to the vast majority of their peers.

The most obvious way this happens is with companies that profit from picking low-hanging fruit (like Facebook and Uber). Then there’s the influencer/YouTuber class. I’m not really talking about either of these groups. Instead, I mean that there are some people who benefit in real life by essentially freeloading in environments where other people are monetized.

Dangers of the Internet

…And you will have a window in your head.
Not even your future will be a mystery
any more. Your mind will be punched in a card
and shut away in a little drawer.
When they want you to buy something
they will call you. When they want you
to die for profit they will let you know.
So, friends, every day do something
that won’t compute…
- Wendell Berry, Manifesto: The Mad Farmer Liberation Front

The standard form for this part: “the internet gave us cool things like email, but look at how much time teenagers spend on social media!” That one study where girls on Instagram have low self esteem gets mentioned, and also maybe something about political divisiveness. These problems are real, but there’s no more alpha in talking about them. I’d suggest that the problems you ususally hear talked about are symptomatic of a greater problem: zombification.

James C Scott has an excellent book, which I might review at some point, called Seeing Like a State. For now I’ll point you at Venkatesh Rao’s review if you want to learn more about the book, but, to oversimplify a book about the dangers of oversimplification, Seeing Like a State is about an idea called legibility. Legibility is how easy a system is to understand from the outside. To take an example from the book: for a very long time, people in England (and throughout Europe) had no fixed last names. A man named John might be referred to as “John Cooper” (because he made barrels), “John Short”, or “John Thomasson” (for obvious reasons) in different contexts. One person might have had a dozen surnames. However, this makes it very difficult for a government to take a census and figure out who they can tax or draft in a particular town. So the government tried to assign everyone fixed last names and take a census using those. Unsurprisingly, this was not popular, and census-takers were often stoned or run out of town.

This general form appears frequently in Seeing Like a State: a large organization (usually a state) finds that it needs to interact with a messy, organic reality, so it invents a legible abstraction to interact with instead. But the abstraction becomes reality: people still have last names like Cooper from when their ancestors’ names were frozen centuries ago. The naming example seems like a relatively harmless case to me, but this reification can have huge consequences, for better or for worse.

I mentioned zombification earlier. The zombie, obviously, is a metaphor for someone who has lost their individuality and become an avatar for consumerism or something. But another way to look at it is that a zombie is someone who has become perfectly legible. What a zombie wants, who they talk to, what they do and see and feel - these things can all be perfectly understood from the outside. And for internet zombies, they are all perfectly understood by the internet, because they are manufactured by the internet. Not all zombies are the same, you get to choose what kind of zombie you become. You can be a redpill zombie, a pro wrestling zombie, a fashion zombie, the choices seem endless. But they’re not, of course; I would think there are only a few hundred types.

A number of years ago, I was talking with an acquaintance who was surprised by the precautions I take to stay private online. I asked her if she noticed that things she looked at on social media, searched for, or emailed about tended to show up in ads. Isn’t it creepy how they’re stalking you, I asked? No, she said, she liked the convenience of getting ads that were relevant to what she wanted.

The most powerful entities on the internet have a vested interest in making you as legible as possible; they want you to be a type. The echo chambers everyone worries about drive engagement, but they also drive homogeneity not just of opinion but of identity, and it’s easier to write ads for a homogenous audience. Zombification, becoming a type (or collection of types) instead of a person, is the first significant danger of the internet.

The second danger is torpor. Even if you remain an individual on the internet, it’s very easy to disengage from the world because the internet is more compelling. I think of this as a raising of the activation energy you need to do something in real life. There’s an ideal level of activation energy, I think: some people already have lots of energy or very low activation energy and tend to overcommit themselves to low value-add stuff because they don’t explore enough alternatives. Other people, like me, tend to spend a lot of time thinking and reading and learning new things on the internet without getting a lot of benefit out of that because they never end up putting it into practice. If you’re the second type, you should probably bias against doing too much in-your-head type stuff, probably even to the point of being bored. This isn’t just an argument against spending a lot of time on the internet, but also against reading, TV, podcasts, etc.

Why Internet Then

The thing about the internet is that it’s basically a lottery: there are high potential rewards and a low chance of winning them. A lot of people know this, but think winning means becoming a tech founder or influencer. In fact, the most common way to win is by learning something really, really useful. For example, I learned how to code online by taking free classes, and it completely changed the path of my life. A friend of mine learned about modern dhamma online (by which I mean stuff like Mark Lippman’s meditation book or MCTB) and ended up getting heavily involved in the community, meditates daily, and has experienced some of the jhanas. People learn technical skills and discover ways to improve their mental, physical, and spiritual health online.

Once you figure out the risks and rewards, the rest is just math - you should spend time online if it has greater positive expected value than other ways of spending your time. I think it does. The mean probability of finding something really useful (and I think that most of the value does come from the rare, life-changingly useful stuff) on a given day is probably somewhere around 1 in 1000, so you would expect to find gold once every couple of years, if you’re looking for it every day. There are also lots of kind-of-useful things you find much more often (like a good recipe or something), but that’s less valuable.

What Do: General

I’d expect the 1/1000 number I gave above to vary by as much as an order of magnitude depending on the person and how they use the internet. In other words, some people might spend a decade on the internet and find nothing life-changingly useful, and other people seem to find really excellent stuff every few months. I think strategy accounts for a lot of this, so it’s worth finding one that works well. I’m into a lot of rationalist-adjacent stuff, so I’d probably suggest Astral Codex Ten and TPOT (stands for This Part Of Twitter and doesn’t really have a website, but @visakanv is well-connected to TPOT people and you can also look at who I’m following) as good starting points for finding high-value stuff. If you want to learn engineering stuff, Steve Brunton is a treasure.

What you want to avoid - or at least limit - is pure entertainment. Right now, the most popular YouTube channels by view count are all pure entertainment: WWE, a bunch of unboxing channels for kids, and a couple of music artists (if you limit yourself to US channels, that is; there are a bunch of very popular Indian channels too). I have a few friends who might object to the idea that listening to Taylor Swift is unlikely to be life-changing, but I’m going to stick by it - and she’s probably the most helpful in the top 50. When you do engage with pure entertainment, I think consciously developing good taste is helpful too. Sometimes you want to completely turn off your brain and see something totally mindless, but other times you can compromise between entertaining yourself and extracting value by finding entertaining content that’s genuinely useful, like a really excellent movie (if you want a recommendation, I think you can’t go wrong with a Studio Ghibli movie).

What Do: Specific

A lot of websites stay up by advertising to you. Some of them, like DuckDuckGo, are doing this in relatively responsible ways in order to provide you with an otherwise free service. Most of them want to extract all possible value from you regardless of the consequences. The items in this section are a starter guide from protecting yourself from the second type. I’d expect that most people know about most of the things on this list - but in the spirit of one-in-a-thousand chances: