Jimmy Wales on Systems and Incentives (Ep. 109)

The Wikipedia founder explains how the online encyclopedia will maintain objectivity in polarizing times.

Jimmy Wales used to joke that choosing to build Wikipedia on a non-profit, non-advertising model was either the best or worst decision he ever made — but he doesn’t joke about that anymore. “If you think about advertising-driven social media…it’s driven them in many cases to prioritize agitation and argumentation in a negative sense over education and learning and thoughtfulness.” In his now ceremonial role, Jimmy spends a lot of time thinking about how to structure incentives so that the Wikipedia community stays aligned on values and focused on building an ever-improving encyclopedia.

Jimmy joined Tyler to discuss what happens when content moderation goes wrong, why certain articles are inherently biased, the threat that repealing section 230 poses to Wikipedia, whether he believes in Conquest’s Law, the difference between “paid editing” and “paid advocacy editing,” how Wikipedia handles alternative accounts, the right to be forgotten, his unusual education in Huntsville, Alabama, why Ayn Rand is under- and over-rated, the continual struggle to balance good rules and procedures against impenetrable bureaucracy, how Wikipedia is responding to mobile use, his attempt to build a non-toxic social media platform, and more.

Watch the full conversation

Recorded October 21st, 2020

Read the full transcript

TYLER COWEN: Hello, everyone, and welcome back to Conversations with Tyler. Today I’m talking with Jimmy Wales of Wikipedia fame. Jimmy, welcome.

JIMMY WALES: Thank you. It’s very good to be here.

COWEN: Let’s say there’s a benevolent donor of either money or time. Where in the Wikipedia system is the highest marginal return for that person?

WALES: Well, I think it depends on that person because if you’re looking to make a marginal impact, if you have unique knowledge on some fairly obscure topic that no one else has, then you may be the best person or the only person in the world who can actually make a difference in that area.

On the other hand, if you want to be the 11 millionth person who’s keeping an eye on the Donald Trump article, your contribution, while important, is probably going to be, marginally speaking, a little bit less, so I think it just depends.

Certainly, what we see, oftentimes, is that people who have the largest impact for the amount of time they’re spending are people who are working in smaller languages in the developing world. If you’ve got some knowledge there, you can have a huge impact just by helping to get the community going, building the community, and so forth. It depends on the skill set.

COWEN: Even for a fairly fixed historical topic — take Shakespeare’s play Hamlet — this year, there have already been at least 35 edits to that entry. What exactly, at the margin, is being done there? Why does that happen?

WALES: Well, it’s very interesting. I haven’t reviewed that particular article, but oftentimes, things that are relatively stable — what we see are slight improvements to wording, making something that seems a little bit confusing read a little more smoothly, adding some obscure fact.

Actually, one of the things that’s very interesting — when we first got started, we took a look at and we imported a handful of articles from the 1911 edition of Britannica, which was in the public domain. It was no longer under copyright so we were allowed to use it.

We imported some articles, and you would think, “Okay, since 1911, what have we really learned about Julius Caesar?” Well, it turns out, a lot. Even on very old topics like Hamlet, scholarship has carried on, scholarship continues, new things are learned, new perspectives are generated. For things like Hamlet, the impact of Hamlet on the world, of course, is still a thing that happens in the contemporary world.

It’s fun to think about, if we look at the Wikipedia entries from 10 years ago, what has changed in the last 10 years on some topic that you think would have been stable for the last 10 years? It would be quite interesting to actually study that — and varied.

COWEN: Do you think of Wikipedia first and foremost as an enlightenment project because it’s obviously teaching people? A conservative project — it’s conserving the best knowledge of the past? Or is it also a radical project — that it plays a role in mobilizing social movements?

WALES: A little bit of all three, I guess, although probably more of the enlightenment project, which in its own way is quite radical in today’s culture. Certainly, when we look at the tone and the style of Wikipedia, which continues to be quite dry, quite matter of fact — the Wikipedia style — and contrast that with a lot of what’s going on in our media culture — things like Twitter and so forth — where you do tend to have a much more outrageous style.

In some ways, I think that it’s quite a radical project to say, “Okay, hold on, everybody. Why don’t we slow down?” If we’re going to talk about something like — Obamacare is my favorite example — are we for it or against it? Actually, before we know if we’re for it or against it, maybe shouldn’t we actually find out what it is? Shouldn’t we just read a neutral, clear description as best as we can find to understand it before we come to some conclusion?”

That, in today’s culture, may be a little bit radical, even though in another sense, obviously, it’s quite old fashioned.

COWEN: If we put aside your official Wikipedia role and just ask you, “Jimmy Wales, as an individual human being, are you an exopedian, a metapedian, or a mesopedian?”

WALES: [laughs] You’ll have to define the three for me.

COWEN: An exopedian is one who emphasizes the encyclopedia-like nature of Wikipedia and the knowledge-building component of the project. A metapedian is more of a focus on the social mechanisms, the processes, the rules, the community. A mesopedian is, supposedly, they tell me, a kind of moderate. Which of those is the greatest attraction to you?

WALES: I must be the moderate in the sense that —

COWEN: There are Wikipedia pages on all of these terms, by the way.

WALES: It’s amazing, it’s amazing. I’m a moderate on that. In fact, there was an interesting discussion today that I was involved with, about — forgotten her name, but she’s running for the Senate in Iowa. She’s the Democratic candidate [Theresa Greenfield], and Wikipedia doesn’t have a page about her.

Somebody said this to me on Twitter: “Why doesn’t Wikipedia have a page on this politician who’s the front runner, at the moment, to be elected to the Senate in Iowa?” I said, “Gee, I don’t know. Let me go check.” What I found was this enormously convoluted, very, very Wikipedia discussion about the rules and the process, and there was a deletion a while back because she was not notable when she first announced her candidacy because she was not someone who we could find any information about.

However, she has become notable in the last several months, including national media coverage. But the whole thing got caught in this web of complicated Wikipedia-ness. I cited one of our oldest rules, the ignore-all-rules rule, to say if some rule is preventing you from making Wikipedia better, then ignore that rule.

COWEN: Does that rule apply to itself?

WALES: [laughs] It does, yes. In fact, it’s probably the most frequently ignored rule of all. Interestingly, it was a situation where, in today’s discussion, you would say I’m the first one — I care most about “let’s make the encyclopedia be encyclopedic and good quality and neutral” and all of that, and the rules are secondary to that.

At the same time, I think the social processes and the rules and thinking about how do we design them is actually, in the long run, the important part and the hard part because when I look at a situation like what we had today, I say, “Gee, this is all hopeless. We got this one badly wrong.” I’m like, “Okay, but what exactly should we change?” Because these rules are not completely random. There are reasons for them.

As somebody said, it’s like the garage band which gets nominated for deletion and requests reinstatement 15 times till people are sick of it because they’re no longer notable. It’s actually the reason some of the procedures are in place. It’s just to give us a mechanism for dealing with those things. But in some cases, those procedures go wrong. We have to really think about, “Okay, what went wrong? How do we make that better?”

It’s a very different kind of model, as compared to moderation at Twitter or YouTube or somewhere like that, where it’s behind the scenes, invisible. This is an open dialogue. It’s a chewing on ideas and so on. It’s very, very public, although often impenetrable in the sense that it’s a lot of words.

COWEN: Which kinds of topics do you think Wikipedia is not so very well geared to cover instructively?

WALES: There are cases where Wikipedia is biased, not because of any intention of people to be biased, but just because no one cares about that thing except fans of that thing. I always give the example — if you look at our articles on Japanese anime, they tend to be quite positive because nobody actually cares about Japanese anime except for people who love Japanese anime.

It’s a binary thing. There aren’t people who — unlike, say, famous politicians, where people do take sides. If you don’t like Japanese anime, you probably don’t watch it, and you don’t write about it in Wikipedia. That’s fairly harmless. It’s Japanese anime, and if the articles are not sufficiently critical, that’s probably not a huge issue.

There are certain kinds of disputes where the amount of energy that we put into trying to keep the articles good quality is probably far out of proportion to the results, although the results are quite good. I don’t mean that the results aren’t good.

If you look at something like George W. Bush, and you think about, here’s this article, and it’s this long, and it’s very detailed. Then you say, “How many human hours were spent hammering away at making this what it is today?” It’s an enormous amount of time spent doing it. You probably could have done it in a smaller amount of time if you just sat down four or five thoughtful people and said, “Let’s work on this.”

On the other hand, there’s something interesting and something valid about this open, public process where everything is open to challenge and a further debate and a further debate.

COWEN: Do you think it’s the case that Wikipedia shifts the relative social balance away from philosophy and toward history? If I want to read about the life of Napoleon, Wikipedia is clearly a very good place to go. If I want to understand better, what is beauty, it’s less obvious to me that I should start with Wikipedia. Aren’t you making history more central to our thinking? It doesn’t have to be bad, but do you think that’s true?

WALES: It’s interesting because if you wanted to understand the concept of beauty as understood in philosophy, Wikipedia would be a good place to go, but what you would be reading about, in a sense, is the history of philosophy. Therefore, I’m agreeing with what you said.

COWEN: It’s history again.

WALES: Yeah, yeah. It’s history again. There’s an old meme, and it is largely true, that if you go to Wikipedia and you go to any entry, once you get past the pronunciation and you click the first link of everything, then it takes only a few clicks, and you end up at philosophy.

If I went to your biography in Wikipedia, it probably takes me four or five clicks. By just clicking on the first link of each thing that I come to, then you end up at philosophy. That’s just simply because of the genus-species nature of a definition or an explanation of something. I’m not sure that means philosophy is central in the sense you’re talking about.

COWEN: Now, as you know, Wikipedia is open. It’s free. It doesn’t have ads. It’s a dream of the early tech utopians. Why is it the only surviving dream of that kind that has persisted?

WALES: Well, it’s an interesting thing, and I’m not sure it’s the only, but it’s certainly the most famous and the largest. There’s a few different reasons. I’ve been on a real kick lately to talk a lot about how business model drives outcomes, sometimes in surprising ways. Or they’re not that surprising after the fact, but they were maybe not anticipated.

For Wikipedia, it isn’t so much that it’s a nonprofit, although that probably has a certain impact as well, but the fact that Wikipedia does not have an advertising business model means that the Wikimedia Foundation, the organization behind Wikipedia, really faces a different set of incentives. Even as a nonprofit, you have to think in a businesslike way, and you have to say, “Well, how are we going to pay the bills? And how do we achieve our mission?”

If you think about advertising-driven social media, the real incentive is to show you as many ads as possible, so they really obsess over time on-site, and it’s driven them to create addictive products. It’s driven them, in many cases, to prioritize agitation and argumentation, in a negative sense, over education and learning and thoughtfulness.

If you went to Twitter, and you found out the three things you wanted to know, and you left, they wouldn’t get a chance to show you very many ads. But if you go to Twitter, and you see 18 people who are wrong and you’ve got to [laughs] straighten them out, you might stay for hours doing this very addictive hobby.

When we think about things at Wikipedia — for example, we could probably increase engagement if we use some of the very basic machine learning techniques to start showing people random promotional links to other things than Wikipedia and then have the machine learn over time how to show you links that are more interesting so that you end up staying on the site longer.

Now, it might turn out that that’s completely normal and thoughtful, in fact, if you go to a well-known economist, that it turns out that the way to keep you on the site longer is to show you other concepts of economics and economic theory. But it might turn out, and probably would turn out, the best thing to do is, when you go to look up Tyler Cowen, to show you on the sidebar links to Kim Kardashian, Donald Trump, whatever the hot topic of the day is and so on, which is not really what you want from an encyclopedia.

When we think about that, our incentive structure at Wikipedia is not to optimize time on-site. It’s to say, look, every now and then, normally at the end of the year, we say, “Hey, would you donate some money?” Nobody has to donate. The only reason people do donate — and this is what donors tell us — is they think, “This is meaningful. This is important to my life. This should live. This should exist.”

We got a year between the time we ask you for money to become a part of your life and make it meaningful enough for you to say, “I want to click, and I want to contribute to this.” That’s just a completely different set of incentives in front of us in terms of, how do we optimize that?

COWEN: How contingent do you think is the history of Wikipedia or something like it? If you and the people you were working with had not come along when you did and done what you did, and the field were left open, would Wikipedia be more maximizing engagement to a for-profit company? It wouldn’t be called Wikipedia, but it would be some kind of online encyclopedia, something maybe more like a Facebook model.

Or do you think there’s something intrinsic about your enterprise that it was going to be a not-for-profit, no matter who did it?

WALES: It’s a very good question. In many ways, it’s contingent. I think it didn’t have to exist at all. The idea of a publicly written encyclopedia using a Wiki model is not necessarily something that was inevitable. Then a separate part of that is, was it inevitable that we would choose a nonprofit, nonadvertising business model? Definitely, that was not inevitable. That was a set of conscious choices that we made.

I used to joke it was either the best decision or the worst decision I ever made, but then I stopped even making the joke because even if I said it was a joke, I would get headlines saying, “Oh, Jimmy Wales regrets that he’s not a billionaire.” That’s not what I said, actually, and it’s only a joke anyway. It’s clearly a great decision, but ultimately, it didn’t have to be this way. It could have been different. In fact, you could imagine several different models.

Let’s say that I was still in charge of everything, and instead of putting it into the nonprofit structure, I’d kept it in a for-profit structure, but it pursued — like a Craig Newmark of Craigslist — a model where I say, “Look, I’m going to maintain complete control of the stock, and I’m not going to raise venture capital money. I’m going to run it as a very profitable hobby and get some donations. Or maybe I’ll stick a little ad here or there, but I’m not going to try to monetize like crazy.”

That’s possible, although hard to really imagine, just as Craigslist is hard to really imagine. It’s like, Craig’s made a ton of money, but he could have made 10 tons of money. He’s just the person for whom a ton is plenty. It’s an interesting question, really.

COWEN: What are the greatest threats to the future of Wikipedia?

WALES: Well, this is what I always have trouble with because I always say I’m a pathological optimist. That’s part of why Wikipedia exists as it is, because I just think we can find a way, and it’ll all work out, and it’ll be fine.

But the things that I would say that we focus on and think a lot about are, really, it’s about community health. How do we make sure that the community is happy, productive, not full of trolls, thoughtful, kind, all of those great Wikipedia values? Which are like all communities, all groups of people.

We have a certain set of values that we aspire to, and we live up to them to a lesser or greater extent on a day-to-day basis. How do we make sure that we are still mostly doing, already, things that we’re mostly getting right, knowing that it can’t be perfect?

Financially, I feel like we’re okay. The fundraisers do very well. We’re very conservative with money, so I think that, as an immediate threat or an obvious threat, not so much. Although technology shifts, and so, will the technology move people to voice assistance where they don’t actually come to Wikipedia, but they just get the information from us and end up not donating money? That’s a possibility.

Obviously, there are other potential threats, but I feel confident that we can defeat them. But as I say, I’m a pathological optimist. There’s a lot of noise these days about repealing Section 230 of the Communications Decency Act, without which Wikipedia couldn’t exist. It’s completely impossible. I can talk more about that. I can talk for an hour about that if we wanted to.

But because we’ve got some noise around that, not only from the right — Donald Trump has been tweeting all caps, “REPEAL SECTION 230.” I’m not sure he even knows what it is — but from the left as well because people are concerned about the power of the big tech companies, and they wonder if Section 230 has something to do with that, and so on and so forth. So, I think that is under threat, and I think that that is a serious threat to anything that’s open on the internet.

COWEN: There’s something called Conquest’s Second Law, which I believe is itself on Wikipedia [nope]. It suggests that all organizations that are not explicitly right-wing over time evolve to become more left-wing. We’ve seen this with the Ford Foundation, Pew Foundation, Rockefeller. Do you think Conquest’s Second Law is essentially correct?

WALES: I don’t know. I wouldn’t think so, although . . . I don’t know. [laughs] I have no idea. I think we definitely have organizations that I wouldn’t say you could easily categorize in contemporary terms as either left- or right-wing, I’m thinking in particular of the Catholic Church, which has maintained quite steady certain values and ideas for a very, very long period of time, with minimal shift over time.

If you try to characterize religion as either right- or left-wing, typically, in the US, you would tend to say right-wing, but if you look at the current pope and certain elements of the church’s positions, you would say they’re pretty left-wing in other ways. I don’t think they easily fit into that, but the broader point is, can an organization maintain an explicit set of values over a very long period of time that is neither left- nor right-wing in its nature? I think so. Certainly, it seems possible to me.

COWEN: Let’s say, in the US at least, that politics continues to polarize along educational lines. Now, obviously, most Wikipedia contributors, people on the board, are going to be well educated one way or another, if only self-educated. Is it then really still possible for Wikipedia to maintain what would be seen as a neutral stand by people on the other side?

WALES: That’s a really tough question. I think that we can slice it up a little bit and get a little more refined and say, “Can Wikipedia maintain a neutral stand?” Yes. That, I think, yes. Objectively speaking, yes. But that would be recognized as such by some people on the other side or that would be universally recognized as such — that could be quite hard. We’re in an era where there’s such a decline in respect for traditional journalistic institutions across the board, that you begin to think . . .

I think about quality publications that matter, and so I think about the Wall Street Journal tending to be a bit more on the right, and the New York Times tending to be a bit more on the left. Then you think, yes, but where does Breitbart fit into that? Where do some of the new media publications fit into that? And where do people who believe that all of those things are part of some sort of the elite conspiracy to hide the truth about pedophiles or whatever? It becomes hard.

I think it’s very hard when you have people who have given up on the concept of objectivity, the concept of neutrality. That isn’t necessarily just on the right. Certainly, when we think about postmodernism and some of the critiques of the ideas of neutrality and objectivity that come from that end of the world, it’s a deeper philosophical question, I would say.

COWEN: How well and how fairly has the media covered Wikipedia?

WALES: Pretty well. Pretty fairly. It’s an interesting thing because there was a time — and even at that time, I did feel that it was a temporary phase — there was a time when it was a popular thing to do to troll through Wikipedia, find something wrong, and then write an inflammatory story about “Something is wrong in Wikipedia.”

I said at the time to the community, to volunteers, that there was a period of time when it became fashionable to write very easy, cheap news stories about something horrible on eBay. You say, “Oh, no, someone’s selling their baby on eBay.” Or they’re selling their soul on eBay, or they’re selling their baby’s soul on eBay, or they’re selling a gun or whatever.

After a little while, people realized actually, it’s not legal to sell your baby on eBay. It’s against the rules of eBay. And if somebody posted as a joke, it just gets taken down. Over time, the media, more broadly, got bored of writing a story like, “Oh no, someone vandalized Wikipedia.”

I remember one case in particular when a famous celebrity — what was her name? Anna something. She married a really old guy for his money and so on. Anyway, she died of a drug overdose, and somebody vandalized Wikipedia and wrote something terrible about her. It was a huge story the day she died because she had a drug overdose, this very famous tabloid-y person.

One of the journalists said, “Does this show something deeply wrong with the Wikipedia model?” And I said, “It shows something deeply wrong with what’s going on in journalism, that this celebrity dies and I get eight phone calls about vandalism in Wikipedia. That’s absolutely not the point.”

There have been times when I felt, “Oh, that’s actually not fair.” But in general, we’ve been treated well. I think, certainly over time, a lot of the worst fears that people had about Wikipedia, that it was just going to degenerate into a troll fest or something like that, haven’t proven true. I would say in recent years, once we’ve seen how bad social media can be, we’ve been treated very, very nicely.

COWEN: What’s the biggest problem with how Wikipedia has covered Wikipedia? There’s a whole page, as you probably know, “Why Wikipedia isn’t so great.” Is it too negative on itself?

WALES: When we set ourselves the mission to do something like that, we are very thorough and good at it, as we are Wikipedians. I actually think that’s really healthy. I’ve said to people, “If you want to understand critiques of Wikipedia, there’s a really good page about that in Wikipedia.” I think that’s great. I think that’s fantastic.

I would say that the areas where we struggle the most — and I alluded to this earlier, talking about a thing that I just happened to be involved with this morning — we’ve become very bureaucratic in a certain kind of way.

We’ve got a lot of history. We’ve got a lot of rules. We’ve got a lot of procedures. They’re all there for a reason. They all arguably do more good than harm, and yet, it is hard for people, new people to get involved. It’s hard for people to understand certain things.

A good example was, we got a little bit of negative press because a famous author wrote quite a . . . it was actually, I thought, quite an amusing piece about how something about his ideas was wrong in Wikipedia, and he emailed us a correction to say, “No, that’s not what I think.” He got something back saying, “Do you have a source?” He was like, “What better source can I have for what I think?” It’s amusing and funny, and it’s like, “Oh, isn’t that terrible? Wikipedia won’t accept an author’s own words.”

Except that the issue was, he emailed us from a random-sounding Gmail address. There was no proof it was actually him. He had a blog. He could have easily written a post to his blog, explaining something. It was actually more subtle than that. It was like, “No, we actually do care what he thinks, and we’re not out to do this.” But because he didn’t understand Wikipedia — by the way, it’s not his job to understand Wikipedia; I’m not saying that at all — we had an outcome that was okay, not optimal.

COWEN: How has the rise of mobile shaped the content of Wikipedia? Take a not-too-controversial article, say the Napoleon Bonaparte one. More and more people are reading it on mobile. More are editing it on mobile. What’s the invisible hand mechanism that connects the rise of mobile to how the actual words on the page might end up being different?

WALES: That’s a great question, and I define a great question as one that no one’s ever asked me, so very good. I would say that the connection is probably less than it should be. So, Wikipedia editing is still overwhelmingly a desktop phenomenon. Part of that is that mobile editing experience isn’t great. In the past, I would say the mobile editing experience isn’t great because we need to invest more money in making it better. But today I would say, actually, it’s probably about as good as it can be. I’m sure we can still improve it, and there are people who are working on that, but the issue is it’s a small-phone factor.

Like, how do you actually make it easy to edit Wikipedia on a mobile device? It’s quite difficult, and we look at things for mobile participation. Like, is there a way for people to quickly do a spelling fix or something like that?

Then the content — it’s written largely on the desktop, and therefore there is a divide in a sense between reading and writing, but that probably doesn’t matter so much because I think the reading experience for Wikipedia on mobile is perfectly fine. In some ways it’s better because the scan distance for reading is a little shorter, but it probably could be a little more optimized for reading on mobile devices.

COWEN: Now, you don’t allow paid-for editing, but I’ve known people who are paid editors — on Wikipedia, they write things. They’re not, for the most part, slanting content. They’re trying to fill in gaps or add detail on, maybe, the person they work for. To what extent do you think Wikipedia is actually somewhat parasitic on nontransparent, paid-for editing, and you get rid of the worst manifestations of it, but it’s still a big part of the underlying foundation?

WALES: It’s a really good question. It’s something that we struggle with. We know that it does go on, and we know that we don’t approve of it. Around the margins, there are a lot of different angles on what that looks like. I think one of the first things to note is that we try to make a distinction between paid editing and paid advocacy editing.

The difference is — let’s say you’re a university professor, and your university says, “Actually, as a part of your evaluation, we are interested in our professors being involved in public dialogue, in public debate. And actually, if you can contribute knowledge to Wikipedia and show us that, we consider that a positive thing, and you’ll get a good review based on that.”

Then, in some sense, they’re being paid to edit Wikipedia, but they are not being paid in the same way that the PR department of the university would be paid to make the university look good. In fact, they probably wouldn’t have much interest in making sure that page about the university itself is good. That’s paid editing, but not paid advocacy editing.

Paid advocacy editing is really the problem. In that case, people are of different minds. I’m quite a hardliner on this. I just think it’s not something people should be doing at all. In fact, I think what we need to do to cope with it is have better mechanisms that are less combative for people who have a conflict of interest to actually contribute constructively by not editing pages, but by making suggestions in some organized fashion that will actually be paid attention to.

But that gives rise to a lot of other problems. One of the problems is, there’s the old saying, which I should know who said it, but it’s very hard to get someone to understand something when their paycheck depends on them not understanding it. If you’re being paid to spruce up the page about your boss, and your idea of what’s neutral filling in the gaps and our idea might be substantially different simply because you come to it with an advocacy bias, and that’s problematic.

I think what works for people is if they come in in good faith and talk to the community. That’s by far the most effective way to get things done. But we also know there are people who are doing paid editing, and it’s something that we basically disapprove of because we also see the abuses of it, where people go around trying to charge people a lot of money, pretending things that aren’t true about what they can get done, and so on and so forth.

COWEN: I’m the rare person who actually has no sock puppets. Why not allow sock puppets? What exactly is wrong with them? So what if a person has more than one identity out there, as long as you can monitor the identity that is operating on Wikipedia?

WALES: That’s a great question. In fact, we do try to make a distinction between a sock puppet and a legitimate alternate account. We actually have procedures whereby you can declare a legitimate alternate account to the arbitration committee so that you’re insulated from any bad harms if it’s found out. Some of the keys are that we rely on trust.

One of the things that is really important to us — we do a lot of what we call “not voting.” It’s voting, but it’s really a straw poll. The votes — typically, they’re not the final word, and if somebody comes into a discussion and pretends to be five different people, arguing that something should be deleted, and there are two actual different people arguing that it should be kept, that’s deceptive. It kind of skews the balance.

People who are reviewing that say, “Well, I think we should keep it, but I see there are five people here with a different opinion, so maybe I’m wrong.” That bulking up your impact by double-voting on something, by pretending to be different people, is super problematic.

The other problematic sock puppeting is a sock puppet to conceal your conflict of interest. I remember we had one notorious case of a PR firm that had engaged in quite a lot of problematic editing.

One of their accounts — it made a lot of edits and pretended to be a retired fellow who was a car collector. There were all these pictures of old cars and so on. They had a whole persona created that seemed like a lovely chap who just liked to edit Wikipedia, but in fact, it was just somebody at the PR firm who was giving a cover, and I think that kind of deception is problematic.

The good examples of multiple accounts would be someone who wants to edit in a controversial area. As an example, let’s say you’re a well-known person, and suppose you took an interest in our entries on pedophilia, not because of any prurient interests but simply because you think this is actually an important topic of social impact.

Well, you probably wouldn’t necessarily want to be known at your university as the guy who edits the pedophilia articles on Wikipedia. That’s just not easy for people to be open about, even if you’re doing all the right things. So you might say, “Yes, I actually want to edit in some areas of World War II history under this identity, but I’m going to do some work over here, and I really prefer it not to be tied back to my real-life identity.” And that’s kind of okay, as long as you’re not voting in elections with two accounts and things like that.

COWEN: Do you believe in what the Europeans sometimes called the right to be forgotten on the internet?

WALES: I have a real problem with the right to be forgotten as it is currently enshrined in European law. It’s a complicated subject, but here we go. First of all, I think privacy is very important, and I think that a lot of internet services need to be much more thoughtful about people’s privacy and really think harder about this. I also think a lot of governments need to think harder in the new era about privacy versus transparency about public information about their citizens.

As an example, I live in the UK, but I’m registered to vote in Florida. So I registered for a mail-in ballot. I got an email saying, “Here’s your ballot information, blah, blah, blah.” I can look up my ballot — not how I voted, but I can look up my home address, which I normally try to keep very, very secret.

I probably shouldn’t even mention this because trolls will now know where to go look for it. But my home address is there, and all you need is my name and my date of birth, and you can get my home address. And I’m like, “That’s not really necessary for public transparency.” Or is it? It’s a good question.

I think we need to struggle with some of those questions. Right to be forgotten — my biggest beef with right to be forgotten in the UK, currently, is that Google can be forced to take down links to content that is perfectly legal on a publisher’s website. In fact, the case that got this all underway is a link to a newspaper article, which is still online, but Google is not allowed to link to it, and there’s no judge involved.

When we think about this in an American First Amendment context, what you would say is, it wouldn’t do. It wouldn’t pass First Amendment scrutiny to say, “Oh, no, it’s perfectly legal to publish this magazine. It’s just not legal to offer it in stores.” Like you’re not allowed to. You have to hide it. We would say, “Hold on a second. That’s just not right.”

That is a violation of the freedom of expression if you can’t distribute your content because, clearly, if you’re not findable in Google, that’s a problem. Nobody’s going to read the piece. There’s no easy right of appeal. You could appeal to Google, and I do believe Google’s doing the best they can to balance people’s privacy interests with freedom-of-expression interests, but I don’t think it’s Google’s job.

I don’t think that we want to have a company, however well intentioned, to be tasked in society with deciding those kinds of issues where, really, if something is such an egregious privacy violation that it should be censored, boy, I really want to have a judge involved. I think that is something that we need to not have handled at the company level.

On things under- and overrated

COWEN: Now, in the middle of all these conversations, we have a section, overrated versus underrated. I’ll toss out a few ideas. You tell me if you think they’re overrated or underrated. Got it?

WALES: Great. Yeah, fun.

COWEN: Growing up in Huntsville, Alabama — overrated or underrated?

WALES: I would say underrated.


WALES: Well, Huntsville is a really interesting place. I grew up in Alabama, which is not known generally to be a super high-tech place. But Huntsville, Alabama, is where the NASA scientists are headquartered. They’re at the Space & Rocket Center, which means that, for a time in the early ’60s, Huntsville apparently had the highest number of PhDs per capita of any place in the US. Huntsville is quite a high-tech town.

Yet, at least for me, what I feel was important as well was not just the high-tech piece of it, but it is Southern in a good way. People are very polite. People are friendly. It’s an openness of culture, which I think has served me very well with Wikipedia because I’m basically just friendly to people, and that’s actually very helpful. So underrated.

COWEN: Transhumanism — overrated or underrated?

WALES: Bit of both. I’m not sure how it’s rated. [laughs] I think 99.9 percent of all people have never even heard of the idea. Probably, people who are massively into it overrate it, and everybody else underrates it. Who knows?

COWEN: The Marylebone section of London.

WALES: Oh, well, let’s just say overrated. It’s very nice there. I actually used to live there. I live in Notting Hill now. It’s really nice, but if you’re talking about living anywhere in London, I think you have to take a look at the property prices and you’ll say, “This whole place is overrated.” [laughs] It’s outrageous.

COWEN: Black-Scholes option pricing theory — overrated or underrated?


COWEN: I know you have an opinion here.

WALES: Again, I think there was a period of time when it was definitively overrated because it’s a genius sort of innovation, a very, very clever model. Actually revolutionized a lot about how people thought about finance and enabled a lot of interesting stuff in derivatives. Of course, as a model describing reality, it’s got serious weaknesses, which is why it’s been replaced, over time, with more sophisticated models.

Probably by today, people think, “Oh, Black-Scholes — that’s like a little toy model that doesn’t actually show anything.” That’s underrating it, but maybe at the beginning when people thought it was the Second Coming, that was probably overrated.

COWEN: Your piece criticizing Black-Scholes — is that the first thing you ever wrote?

WALES: I don’t think I ever criticized Black-Scholes. I wrote a piece that was an extension. I always like to think it was the last extension of the most obscure possible thing, which was the pricing of index options when all of the underlying assets follow a lognormal diffusion. It’s in the Black-Scholes tradition, but it’s the last little tick of something, and it was a huge, complicated mathematical mess. [laughs] But it wasn’t a critique of Black-Scholes. It was just trying to extend it to a different case.

COWEN: Ayn Rand’s objectivist epistemology— overrated or underrated?

WALES: Ooh, I would say underrated there. I think that’s an interesting one because, clearly, her politics and ethics and other aspects of her thinking are hugely controversial, and we could say lots of things about that.

But her epistemology is very interesting. I’m not an expert, but I used to be quite keen on this area, and I read a lot, and I thought a lot. I think there’re some real insights, interesting insights there about concept formation that have not been appreciated by people who just think, “Oh, she wrote these long, ranting novels about capitalism.” I think there was more to her thinking than most people realize, so underrated.

COWEN: What’s the weakest part of her philosophy?

WALES: I think that one of the biggest issues is that her writing exhibits a lack of kindness. One of her core values was chewing on ideas. Something she loved to do in her life, and it’s something that she advocated, is chewing on ideas. What that means is really thinking things through, considering alternatives, and so on.

That you wouldn’t necessarily know because once she chewed on the idea long enough, she has quite a flamboyant and dramatic way of presenting them, as if she’s just ranting to you about the world as she sees it. I think that’s problematic. I think it would be better if her public image were more about, “Let’s be thoughtful and think and reflect on complicated ideas because the simple answers are probably not the correct answers.”

COWEN: How have the last five years or so of your experience with the internet caused you to revise your self-described libertarianism, if at all? Is it working?

WALES: It’s a very good question. I’m always very cautious about the label libertarian just because I find the Libertarian Party to be quite questionable on many things. I guess where I come down is, I would say, on many, many issues, I like the concept that we have in US law around the First Amendment of strict scrutiny.

Look, if we’re going to have the government intervene in something, we need to be very thoughtful about that because we know very well from economics that many interventions don’t work the way most people think they work, and they actually can cause a lot more harm than good. Something very simple, like the classic example, would be, “Let’s raise the minimum wage to $40 an hour.” I think most people get that actually probably isn’t a great idea. That isn’t actually going to help poor people very much at all. Maybe a few of them, but mostly not.

I guess, through my life, as I’ve gotten older, I’ve become more a centrist or a gradualist. When I was 21, I might have been, “Crush the state and eliminate all this, and freedom is fantastic.” Now, I’m really saying, “Look, if you’re going to get anywhere, and you’re going to improve the state of the world, and you’re going to avoid authoritarian tendencies, and you’re going to avoid overreach by government, you’re going to have a prosperous society, you need to be very thoughtful about how you change rules, very thoughtful about how you get somewhere.”

My classic example would be zoning laws. If you overnight eliminate all zoning laws, I’m not sure you would get to the same place you would have gotten to had you had a more market-oriented solution from the beginning because zoning laws, in some ways, roughly replicate what a more market-oriented system might’ve come to. I don’t know, I guess I’m a gradualist in that sense.

COWEN: So you’ve become more Hayekian, you might say.

WALES: Yes. Maybe. [laughs] Possibly. Not sure.

COWEN: What would your management advice be for Facebook?

WALES: Ohh, for Facebook. Management advice, I don’t know. I always say I’m a terrible manager, but if I were giving advice to Facebook, I think they’ve got a real hard problem. To be a little bit sympathetic for a moment — although I’ve been quite critical of Facebook in many ways — one of the things that we find easier at Wikipedia is that Wikipedia has never been a place that’s a wide-open free-speech forum.

We don’t have to really deal with some very complicated edges around what advocacy is crossing lines that are just too offensive because, really, at Wikipedia, you’re not supposed to be advocating for any particular thing at all, except for how do we make Wikipedia better? How do we describe these ideas fairly? If you come into Wikipedia with some sort of an agenda, however mild, that’s probably not the right thing to do.

For us, these kinds of things — we don’t say, “Oh yes, people should be allowed to say whatever they want.” “Oh, but what if it’s so offensive?” They have to deal with that. It’s really hard because the point of Facebook and Twitter is to just post whatever you think and your ideas, and therefore, you’re going to find some people who have completely horrific ideas.

Given that they’ve got a very difficult problem, what I would say is, they need to really think harder about the model that they’re using for moderation, which I think doesn’t scale very well. Hiring people in sweatshop conditions to make content judgements or trying to get AI — which is not really ready to judge the nuances of human conversation — to do it, neither of those is scaling very well.

None of those works very well, as opposed to what I think is really more important: put more power into the hands of the community. Now, what does that look like for Wikipedia? We know that. What does that look like for Facebook? That’s a hard problem. I don’t have a simple solution for them.

The one thing about Facebook that I think is really, really interesting is, because of the structure of ownership of Facebook, Mark Zuckerberg has probably one of the strongest CEO positions of any company. It would be very difficult for anybody to oust him as CEO. Therefore, he can make decisions that would be death to a CEO who has to really panic about quarterly results.

He can say, “Look, I’m going to make some decisions that are going to reduce engagement on Facebook, but that are going to prevent a situation where the whole world thinks we’re destroying Western civilization. We’re going to maybe make half the money we were going to make, but in the long run, it’s good for shareholder value because we can demonstrate that if we carry on our current path, everybody’s starting to hate us, and that’s not good for business.”

He could do that. He has the power to do that, and nobody could really fire him for that. That means, I say, hold him responsible in a way. I hold all CEOs responsible for doing morally right things, but it’s a little easier to understand if you’re at a company that says, “Look, if I know what we’re doing isn’t great for the world, but it’s not illegal, and if I stop doing it, I’m going to get fired because the shareholders will revolt.” Then, it’s like, “Yes, you’ve got a tough problem.”

Mark — he doesn’t need any more money. He’s got plenty, and therefore, he is in a position where he could make decisions that are best for Facebook in the long run. I am not suggesting that he violate any fiduciary duties, but he could take a long view in a way that most CEOs, unfortunately, can’t.

COWEN: How well is the Facebook Supreme Court going to work?

WALES: That’s a very good question. I have to say, I don’t know enough about it. The kinds of things I would be interested in would be, what kind of power do they actually have? The Supreme Court idea Wikipedia had — not Wikipedia, Facebook — I don’t know how that’s going to work out. I’d like to know more about the structure of it and what actual power they have. If it’s merely an advisory body filled with worthy types that Facebook isn’t going to listen to, clearly that’s going to absolutely do nothing.

But if they are willing to take seriously the idea — and I think they do take this idea seriously — that it isn’t up to Facebook to be the judge of what’s true or not. I think that’s something that they don’t want that responsibility. Yet, because it’s a place where disinformation is being shared in a way that is damaging the world and is problematic, they also have to take responsibility. It’s like, where do they draw the line?

I always remind people, if we had a different set of historical facts, and five years ago — I think it’s longer than that now — ten years ago, Facebook can suddenly say, “We’ve actually decided to start shaping the public conversation by deciding what news is of quality or not, and we’re going to deemphasize things that we think are low quality.” We would all have freaked out and said, “Oh my God, it’s Orwellian. Facebook is now trying to decide for us what’s true and what’s not true.”

Now, suddenly we’re in this opposite universe, where everybody’s demanding, “Facebook, you must do something. There’s disinformation on your platform.” They’re in a tough spot, so I don’t know. The Supreme Court, good, yes, I don’t know. They better have some power. Otherwise I’m not sure.

COWEN: What is it you think that you might know about the education of children that other people do not?

WALES: That’s a good question. I don’t consider myself an expert on education, but I have three children, and I have my own experiences there.

COWEN: You are, then, an expert on education.

WALES: [laughs] I also can reflect on my own education. For me, I think if you put me in the camp that says, there is a big downside to traditional education in terms of, if you aren’t careful you can get into a very rigid process that is not good for creativity. I would say my ideas in that area are similar to — unfortunately, he recently passed away — Sir Ken Robinson.

For me, when I was young, I had a very unusual educational upbringing. Sometimes it’s reported it was Montessori. It wasn’t technically Montessori, but it was very open, and we had lots of free time. I basically spent a lot of time reading the encyclopedia. Incredibly valuable just to have time to explore, be encouraged in following your curiosity, as opposed to rigidly sitting in rows and meeting testing standards and all of that.

Having said that, I’m not opposed to having standards for schools, and tests are a good way of measuring that. But I think you need to really focus a lot on creativity because if we look forward, the world faces a lot of problems, and we’re going to need a lot of innovation to solve some of these problems.

COWEN: If you had $100 million to give away, putting aside your own enterprises, what would you do with it?

WALES: That’s a really good question. I liked seeing today that apparently, Michael Bloomberg is spending $100 million in Florida for ads to defeat Donald Trump, so I’d be very tempted to go that route.

Actually, I was disappointed to read something today, and I haven’t followed up, so I don’t know what the general public conclusion is. He had promised $100 million to help people who had been disenfranchised in the state of Florida because they had not paid some fines or some sort of fees to the government. Therefore, their right to vote had not been restored even though they had met all other conditions.

He was going to donate a lot of money. Then I saw an NGO who said, actually, the money never showed up. A lot of other people donated, but he didn’t, and they were unhappy with him. I don’t know about that.

Actually, if I stepped back, probably defeating Donald Trump isn’t how I would if I had one chunk of $100 million. For me, it’s about how do we think about restoring a sense of decency and thoughtful public engagement where we have an election, and I look at the policies and I say, “I’m going to vote for this person because I agree with that policy.” You say, “I’m going to vote for the other person because I disagree on that policy.”

Then we have an interesting conversation about the policy, and we may not agree, but we feel at the end of the day, we’ve chewed on some ideas. Whereas today, it’s the ranting and the scandal-mongering and all of that. It’s not healthy. It’s not really giving us leadership that we need.

COWEN: How happy are billionaires?

WALES: [laughs] I know quite a few of them. They seem pretty content.

COWEN: How happy are famous people?

WALES: That’s a good question. Obviously, that varies quite a bit. I suppose, actually, to be fair, it probably varies with the billionaires. That was a flippant answer. I don’t think money buys happiness, but the truth is, it buys freedom, so people then can pursue interests and do things like that.

The fame — I’ll give an example of someone who I know reasonably well, Sean Parker. He is both famous and a billionaire. I think if he were given the option to choose to give up one or the other, he would definitely give up being famous because the grief he has gotten in his life around his extravagant wedding and things like that are completely not aligned with the things he’s interested in.

He’s super interested in cancer research, and he donates money there. Whenever I talk to him, he’s really passionate about certain areas of medicine, not because he’s trying to make money — it’s just that’s what he’s interested in. That makes him happy, and being famous on the side is actually annoying. That’s me assessing him. I haven’t actually asked him that, but . . .

COWEN: How happy are people who have power?

WALES: That’s a good question. Even the word happy — what immediately comes to mind is President Trump. I don’t know. Does he seem happy? I don’t know. It’s an interesting question. I think different people have different personalities around power. Some people have it because they want to do constructive things in the world, and they like having power and so on, and they’re probably pretty happy.

Other people — I think they’re just lost souls, and they just want power and then attempt to be happy, and maybe it doesn’t make them happy. It’s a hard question. I remember a few years ago, I was on a street corner in New York City. I was just about to go into the UN. It was during General Assembly. The street was blocked off, and then an enormous motorcade went by: hundreds of motorcycles, New York police, then the Secret Service, Trump, and more hundreds of more motorcycles.

I thought, for a man who probably — and I don’t believe in diagnosing people with medical conditions without knowing — but probably qualifies for some sort of medical diagnosis of narcissism, “This can’t be good for him, but he must fucking love it.” [laughs]

COWEN: How happy are people who seem to have power but cannot, in fact, exercise it, due to constraints?

WALES: That’s an interesting question. Again, it really depends on who they are and how they got there. I would sort of put myself in that category. People think I have a lot of power with Wikipedia, but I’m very ceremonial. I give advice, and I try to be thoughtful. People do listen to me, so I have influence, but not power. I’m very happy about that. For me, I’m very content about it.

Other people, though, may have come to that situation in a way that they expected to have power and they don’t, so then they must be quite unhappy.

COWEN: Two last questions. First, can Wikipedia truly survive without you? I know you’re calling your role ceremonial, but you’re a kind of focal point, a final layer of adjudication. I once said to Vitalik Buterin in my chat with him, “Ethereum relies on you. Even when you do nothing, you’re the Russian sage in the background, like in an old Dostoevsky novel.”

Are you, in fact, the American — now partly British — sage in the background of Wikipedia who underpins the whole thing, even when or perhaps because you do nothing?

WALES: It appeals to my ego to imagine such that, obviously, without my sage wisdom and guidance and calm ceremonial neutrality, the whole thing would fall apart, but frankly, I really doubt that.

I do think that it is a complicated matter that I give a lot of thought to, and which I have given a lot — it’s not a new idea. How do we construct systems and models and procedures and processes in the community that strive for maintaining our values in the long run and improving the encyclopedia and building something that we value? I think we’ve done a pretty good job of that. We struggle at times, as I say, with excessive bureaucracy, but people are aware of that, and so on and so forth.

For me, I think the hard bits and the complicated bits are probably around the interface between the community and the Wikimedia Foundation. I’m not so sure that we have a long-term settlement of those issues that I feel confident can maintain without a lot of effort, not just on my part, but on the part of core people who are around to make sure that the foundation is both able to act as necessary, but doesn’t become . . .

Imagine that they’re going to pursue a model like a Twitter or Facebook moderation model, where they’re going to take on all the jobs of moderation, which, frankly, wouldn’t scale or wouldn’t work. That’s the area where I do feel I’ve got some work to do still.

COWEN: Last question: what will you do next?

WALES: Wow. I’m really focused a lot these days on WT.Social. WT.Social is rethinking social media, trying to create a new social media platform with a different business model. I would say it is a pilot project. We’ve got about 500,000 people on the platform, but many of them are not active. We’ve got a core community working on that. That takes up a huge amount of my time.

I could tell you — this is not announced yet, but as this won’t go out until November, it will be announced — I got this goofball idea that I’m really excited about, and I’m having fun with, which is, I’m putting together a system so that children can visit Santa Claus. This is why I’ve got the Santa background here. Children can visit Santa Claus virtually this year because people shouldn’t be going to shopping malls and stand in long lines to sit on an old man’s lap with coronavirus going around. So you can sign up and visit Santa Claus. That’s what I’ve been working on this week, and I’m having a lot of fun with it.

COWEN: Jimmy Wales, thank you very much.

WALES: [laughs] Great.