Journalist Clive Thompson has been writing about software and its effects on society for about 25 years now. “For the first 15 years, it was a matter of convincing people that software is important, and it’s going to change your life,” says Thompson, a contributing writer for the New York Times Magazine and a monthly columnist for Wired. He says it’s only in the last decade, with the rise of social media, ride-sharing apps, and the like, that he doesn’t have to convince anyone anymore.

Still, he realized, “the entire industry of software development is very mysterious to people.” Thus, he set about writing his insightful new book, Coders: The Making of a New Tribe and the Remaking of the World (Penguin Press). “I wanted to give the average person a glimpse into ‘Who are the people making all this software? Who is writing all this code?’ and what their hopes and dreams and visions and blind spots are, so that we could better understand the software world we live in right now.”

We recently spoke to Thompson about the joys of programming, the appeal of blockchain technology, and the prospect of the first coder president.

As a fellow writer, I particularly enjoyed the part of the book where you talk about how you found coding more satisfying than writing.
Here’s the thing about about writing: When you’re writing, you’re engaged in an act of persuasion. There is no proof that my column or my book is good, or that it quote-unquote works. The value of my writing is in the subjective eyes of other people. Whereas coding is much more binary, and a little more objective in terms of its value. If I write a little script or a program and I get it to work, and it’s doing something, you cannot come to me and say, “Yeah, that script’s not working.” It is. It is doing its thing. There is objective proof at the end. And that’s awfully pleasant. For a writer, who’s often so worried about the subjectivity of what we do, that’s really a blast. It also has a lot of these concrete wins. Like the moment that I go from something that I’m programming, and it’s not working, it’s not working, and suddenly it’s working, it’s just this rush of euphoria. And there’s nothing quite like that in writing.

So should I learn to code? All the trolls say that we journalists should.
When the denizens of 4chan were taunting journalists who’ve been laid off saying “Learn to code,” that was not a good-faith argument. So they can safely be ignored. They were not actually offering advice. They were attempting to tease and mock people who had lost their jobs, because they think of journalists as the enemy. So let’s put that off to the side.

But let’s ask in a good faith way: Should a journalist, or anyone, learn to code? To me, the answer is, it’s certainly fun to learn a little bit of it. There are three reasons why the average person should try to learn to code, in order of value. The first one is that it’s actually quite fun, because it’s a way of using your mind and puzzle-solving. It is as much fun as playing the guitar or watching TV or reading a book or going for a walk.

Get the BREAKERMAG newsletter, a weekly roundup of blockchain business and culture.

The third reason, the one that I think is not as valuable for journalists or writers or other people, is this idea that you’re going to get so good at coding that you’re going to change your career. That is certainly worth doing if it’s something you genuinely find intriguing and are super-driven to change your life that way. But it requires so much work and so much commitment that I think it’s only practical for people for whom it is a real drive.

There is a reason in the middle that no one talks about that I almost wish I’d talked about more in the book. It can be incredibly useful to learn a little bit of coding because it can make you enormously more effective in your regular job. If you learn a little bit of programming, you discover that you can automate dull tasks in your own work, or you can open up new possibilities, new skills, new things you can do. Just as a funny example, my book just came out, and I was finding myself, the way that authors rather pathetically do, visiting and refreshing my Amazon page to see “What’s my sales rank?” At some point, I realized, “This is crazy. I shouldn’t be doing this. It’s not psychologically healthy.” But because I know enough programming, I’ll just write a little bot that will go grab the contents of the page, scrape it, find the sales-rank stuff, strip it out, format it into a text message and text it to me, four times a day. And I won’t pay attention to the page at all, otherwise. It’s really delightful. It broke me of that horrible, stupid psychological habit.

What surprised you most about coders? What preconceptions of yours, if any, were smashed?
One thing that wasn’t so much a preconception as something I simply had no idea of is that coders are insanely able to deal with epic levels of frustration that would crush any other normal human being. It’s in the nature of writing software. If you get the slightest thing wrong, everything stops working. Amazon once was taken offline for like three hours on the East Coast because of a single misplaced character in a piece of routine maintenance by a coder. The U.S. once had to blow up a spaceship, a Mariner I, when it started veering off course because someone had misplaced a single character in code. So the level of precision necessary is incredibly high. And the amount of bugs and errors that you’re wrestling with when you’re coding is so common. I certainly knew that coders don’t sit there typing all day long with code flowing out of them. But I don’t think I understood how much of what they do is just dealing with the broken stuff.

When we were emailing prior to this interview, you mentioned that you weren’t able to include the reporting you did on blockchain and cryptocurrency. What kind of things did you learn?
The blockchain people are really interesting, for a couple reasons. One is that only very rarely do you come up with a whole new area of software development. The web was like that. Databases back in the day were sort of like that. It’s such a big area, and I think that’s part of the reason it attracts the kind of blue-sky thinkers, the idealists, who feel that corporate and commercial software development fields are so heavily colonized that they’re looking for a new area. And that is one of the things I found over and over again when I talked to blockchain people. They were the ones that were hungering for a new green field to plow.

The other thing that’s interesting is that the people that are interested in blockchain have a political culture that is a little distinct. People often say things like, “Tech people are so libertarian, particularly in Silicon Valley.” And that turns out not to be really true. When political scientists have done studies of tech folks in Silicon Valley—and there haven’t been many studies—they find that they’re actually slightly less in agreement with the basic statement of libertarian values than even garden-variety Democrats. They’re not libertarian at all, with the exception of being very, very anti-government when it comes to labor. They do not believe there should be any significant limits on how a company hires, fires, or dispatches its labor.

But blockchain really does have an extraordinarily high concentration of libertarians. They’re drawn to it partly because blockchain dovetails with a preexisting political belief of the danger of currency being controlled by governments, general suspicion of government overall, and a general sense that things would be a lot better if there were systems that allowed people to operate outside coercive government structures.

Since I started covering blockchain about a year ago, I’d never met so many libertarians in my entire life.
I’ve seen numbers that say that anywhere from 25 to 40 percent of people who are involved in blockchains would say, “Yes, I’m a straightforward libertarian.” And that is enormously higher than you would find in the programmer population at large.

Speaking of politics, it was recently revealed that Beto O’Rourke was a member of the hacking group Cult of the Dead Cow in the ’90s. Do you think America could benefit from a coder president?
Yeah, I think so. At this point in time, having a president who was familiar with code, and had maybe done a little bit of hacking, would be a terrific thing, because one would hope it would make them a little more aware of the complexities that lie within that area of human endeavor. One of the things you learn when you write about software is what’s known as Moravec’s paradox, which states that it’s the things that humans find easy that computers find hard; things that computers find hard, people find easy. I think that one of the useful things about having a coder president would simply be that they would be a little more aware of what is easy and what is hard in software, which is just helpful when you’re trying to make decisions about where priorities should be legislatively.

You dedicate parts of the book to women programmers, who were in fact pioneers in the field. Do you feel like things are getting better at all for women in programming?
There are a couple areas where they’re getting absolutely better, and a couple where I don’t think anything significant has yet changed. At the level of young women, there has been a great deal more interest in pursuing computer programming as a field. In the last five years, there has been a significant uptick in the percentage of women that are applying and getting into computer science programs in college. And then you have organizations like Girls Who Code and Black Girls Code encouraging girls to try it out.

Where things still need to happen is at the corporate level. You can graduate a lot of women, but they’re going to run into this buzzsaw of an industry that is still predominantly male and in many ways fixed in its ways and has so many really subtle forms of sexism. Sue Gardner interviewed 1,400 women [coders], and she found that they would arrive and they’d have a lot of fun, and they really enjoyed the challenges. And as they got more senior, the barriers started coming up, and they got a lot more resentment, a lot more pushback, a lot more denigration, and suspicion of their skills. And they said, “Screw it. I’m going to go somewhere else. I’m going to use my technical skills in some other area. I’ll work in IT in the medical industry; I’ll get paid a ton. But I’m not gonna sit here having to deal with neg hits from my peers inside this industry.”

That’s the part that most needs to change, and it’s harder to change, because it is culture, and culture changes slowly. It took decades for software to calcify from being the very open, meritocratic place that it was in the 1960s, where anyone could walk in and there were no preconceived notions of who looked and walked like a coder, to “Oh yeah, that’s guys.” And it’s going to take decades of work for it to shift in a different direction.

Prediction time. You talk a lot in the book about Facebook and Twitter and social media apps of that nature. What are the most promising apps of any sort that you’ve come across lately, the future household names?
I’m going to punt on that question, because I’m not really good at doing that. I don’t actually pay attention to it. I’m the type of journalist who, for my entire life, has been focused on a couple things—primarily the civic and social impact of software. So I’ve been awfully interested in “How is this affecting people?” but I’m not anyone who ever gave much thought to “Who is going to be the next rising star?”

Do you think that will still be talking about Facebook and Twitter in 10 years, or will they be replaced?
My suspicion is that within the next 10 years, yes, we we will probably still be dealing with some of the companies that have grown to be very big, because there is an awful lot of lock-in and network effect that those companies leverage. And they are now behaving exactly the way monopolists behave, doing everything they can—ranging from the way they run the service to the way they lobby for laws to be written—to entrench their status.

In fact, I’m not sure what would dislodge them right now, apart from a major technological shift to a different ecosystem that catches them unawares, in a way that the mobile phone turned out to be an unexpected gift for them. Twitter, Facebook, Instagram, and the other large social networks grew up at the exact instant that we went from visiting social media infrequently during the day to carrying it in our pockets all the time. And that turned out to be an enormous first-mover advantage gift. One could hypothesize that if a new ecosystem for how we interact with digital services, with software, were to emerge, it might be hard for them to adjust. And the entrants that did get a toehold might not want to sell out, not want to become a subsidiary of Facebook.

For a while, people were predicting that VR was essentially going to become a sufficiently new and weird interface that it was going to explode and create a new green field. And I think that’s why everyone was very excited by it. But it turns out that we’re not at the form factor yet where anyone really wants to hold something up to their face for very long. And maybe that changes, maybe something about the hardware changes in some way that we can’t predict.

Are you excited about the future of tech or cautiously optimistic?
I’m Canadian, I’m always cautiously optimistic. For all of the problems we have with big tech right now, A) You have public awareness about it and interest in it, which is great. B) Even the worker bees of Silicon Valley are becoming a little unsettled by the ethical and civic implications of their code. Like in the last year, we saw walkouts at Google, thousands of people; we saw a petition at Microsoft. In the case of Google, it was building AI for the military. In the case of Microsoft, it was building productivity tools for ICE. This is really, really new behavior. This is not something that the software developers of the ’80s and ’90s or even the aughts were routinely doing.

My journalistic technique tends to be to try to find the things that are interesting that we should move towards, while also identifying the things that are bad for society that we should move away from. I think a lot of people are doing a very good job of figuring out things we should move away from. But it’s always useful to say, “What are the things we should move towards?”

What kind of things should we be moving toward?
One is continuing to push on and accelerate. And this is part of what my book is trying to do—[foster] the public conversation about the ways in which software and large tech companies affect us. Frankly, I think it needs to start to be something people vote on, which takes us back to the idea of a coder president.

More people are being asked about their thoughts on issues like “Should we break up big tech?” Which is what Sen. Warren is talking about. Should we be concerned about about the monopolistic effects of large companies? Should we be concerned about the effects of DRM? Should we be pushing for far more R&D money for digital technologies, so that we have the kind of boom of the ’80s and ’90s that came out of the investments that happened in the ’50s, ’60s, and ’70s? There’s so many cool things that a more digitally aware public could make part of their demands on their public servants.

I love telling this one to libertarians: I’m like, “So you guys like AI, right? Deep learning is pretty cool. We’ve seen this turn overnight into a trillion dollar industry. Well, thank the Canadian government.” Because when three apostates of AI—Geoff Hinton, Yoshua Bengio, and Yann LeCun—were working on on neural nets and deep learning in the early aughts and no one thought it was worth a bucket of warm spit, one of the few places they could get funding and encouragement from was a Canadian government organization for tech that funded them. And this wasn’t millions of dollars—this was in the hundreds of thousands of dollars—and yet they got deep learning out of it. It is bonkers how a bit of government encouragement at the R&D level can push things forward in radically interesting new ways. There are these wonderful leverage points that the public could push for and make a priority for their politicians.

And on the technological side, I have a whole chapter devoted to cypherpunks, the folks who were working on code to try and liberate people. That was my favorite chapter to write because it’s all the weirdos who are trying to seize back autonomy and power from from big people. And some of that overlaps with blockchain, which is an attempt to create different paradigms for allowing software to operate online so it doesn’t have these highly centralized forces that we’ve seen can be such a problem with big tech companies. Let’s let’s throw more excitement, more entrepreneurs, and more research money at that, because that’s another thing that can open things up.

You just mentioned weirdos. Are coders weirder than the average person?
The one thing I’ve learned from being a journalist is that everyone is weird, man. You go talk to nurses, you go talk to the farmers, you go talk to the people who work in factories producing yarn by the ton, everyone is super-weird, particularly in the joy and obsessiveness that they bring to their work. Also, the reasons why they got into it are often super-weird. So coders are no weirder than the average person because the average person is already super-weird. But do they have things that make them weird in a unique way? Yes, absolutely. Someone on Twitter described coding as “telling rocks what to think.” There’s something about that that sort of gets into your soul, and it opens up creative possibilities, ways of thinking about the world that are very distinct and unique and breed types of weirdness that are delightful.

This interview has been edited and condensed. Photo courtesy of Clive Thompson.