To Slightly Reduce How Much the Internet Sucks, Use Positive Reinforcement

catching flies, honey, vinegar, etc

Two months ago I made a decision about how I wrote here, due in part to the desires of readers and people I admire. The results have been, from a readership perspective and likely a financial one, clearly bad. I got a lot of negative reaction for “All This Shit is High School” (which objectively was a significant market success) and I kept hearing about it for a couple weeks afterward, so I deliberately avoided writing inflammatory media critique the past two months. The result? Views are dramatically down in that time, from an average of around 45K to an average of around 20K, and revenue went from consistently climbing to essentially flat. No doubt this is a multivariate phenomenon, but deliberately avoiding my biggest attention-getter must play a role. I’m a big boy and I made that choice as an adult, aware of the potential outcome, and it’s cool. I’m more interested in things it says about the broader internet. (Bad things.) But let’s back up first.

This post is inspired by this piece from Charlie Warzel, if only in the sense that I’m thinking about how to improve the internet. I can begin by repeating a common refrain here: there is no such thing as independent media; there are only different kinds of dependence. If your financial security is derived from the approval of others, you are not independent. You can be dependent on different people and that difference does matter. I have been remarkably successful here in a crowdfunding context but I probably would never have been able to get a staff writer job at any traditional publication. (Such a job would probably pay a third of what I’m making, but that’s for another time.) But my generous readers are themselves stakeholders whose interests I will inevitably weigh and value. A consequence of this dynamic is that “independent” media is subject to external pressures too, in ways both good and bad. If you don’t like something about what is typically branded as the independent media, you can yell about it, which increases engagement and helps who you want to hurt; you can hope that it will go away, which it almost certainly won’t; or you can try to use the power of incentives, that very universal dependence.

More and more it seems that those who are within the establishment media lament about those of us on the outside and our sins. I won’t get into the weeds of the bigger picture parts of that conflict, though you can certainly guess where my sympathies lie. But there is clearly a market for news and analysis of the type that has traditionally been produced by legacy media publications that comes from people who are distinctly not a part of those publications or their professional culture, and there is clearly great angst from those within that culture about this fact. The remarkable and immediate success of Krystal Ball and Saagar Enjeti’s new podcast is just one more data point, to go along with Patreon, Substack, Joe Rogan, and more. Whatever you want to call this new form of media, it’s not going away.

This isn’t just about competition. Somewhere along the way American liberalism shed whatever vestigial attachment it had to counterculture and antiestablishment sentiment and fell fully in love with the powers that be. I suppose that will be viewed as inflammatory, but I find it hard to deny. Anyone who lived through the Covid-19 pandemic has to concede that American liberalism has adopted a rabid defense of ideas and narratives that come from official or establishment sources, anyone with a degree or a title or a pedigree. The whole framework of “misinformation” fit into this attitude; there was the CDC and Dr. Fauci and the public health community, and then there was a bunch of quacks and conspiracy theorists - though that same public health community insisted that masks were unnecessary and that Covid was not airborne. Fights over fields like critical race theory and gender studies reflects strike me as of very limited real-world valence, but draw massive controversy because the increasing power of elite institutions within the left-of-center means that what’s taught and researched in universities takes on outsized prominence. What’s more, decades of conservative complaints that the media and academia are biased against them has only strengthened liberal attachment to those cultures.

What this means is that attacks on highly-regarded liberal publications like The New York Times and the Washington Post1, or on the media writ large, are perceived as inherently coded in our culture war. People who specialize in those attacks are thus targets for derision online, particularly from those in the media professions themselves. And since such consistent critics are frozen out of establishment media, for obvious reasons, they’re being funded by readers and listeners while their targets continue to operate under fairly traditional media salary arrangements, two distinct cultures operating under different funding models. In a sane world they would engage through their various publications at a certain level of distance so that substance could take priority. Unfortunately they’re all on Twitter together2 and snipe at each other all day long and it’s wearying and terrible - and self-perpetuating.

I hate to invoke Glenn Greenwald, as I will inevitably be accused of trying to cape for him, but he really is the perfect example: he has a large audience, has been successful to the point that he has an unusual amount of freedom, and he is very, very disliked by media liberals. There are a lot of reasons for that, part of which is that the antipathy is very much mutual and he participates as lustily in that war as his critics do. But the fighting’s all very ugly and it goes on constantly and it’s hard not to wonder why it continues. Well, the answer is pretty simple: it continues because, whether they’re conscious of it or not, the endless Glenn vs Every Liberal You’ve Ever Heard of War serves the interests of both. For the liberals, it demonstrates their commitment to the professional and social cultures in which they jockey for rank; perversely, criticizing Greenwald over his accusations of insiderism is itself a way to be an insider. For Glenn, it’s the oldest media nostrum there is: there’s no such thing as bad press. Greenwald might be one of the most hated people among high-follower Twitter users, but he also has 1.6 million followers himself, and not coincidentally tens of thousands of paying subscribers for his newsletter. Controversy is instant marketing. If you want to draw a crowd, start a fight. The hatred of the people who despise Greenwald’s work simple ensures that there will be more of it. Nants ingonyama bagithi baba.

So here’s my proposal: instead of constantly giving negative attention to those you don’t like, which helps their careers and thus incentivizes them to do more of what you dislike, why don’t you use positive reinforcement to get more of what you do like, or at least hate less?

I’m 100% serious here. Let’s start from the assumption that most of the people that media liberals hate are not just going to go away. I think browsing the most popular creators at crowdfunded platforms like Patreon and Substack will reveal that, whether you like it or not, a lot of people who establishment media folks view as problematic have durable audiences. As much as you might want them to, Chapo Trap House and Matt Taibbi and Red Scare and Jesse Singal (to pick just a few obvious examples/targets) have large and dedicated groups of supporters who will fund their work. While genuinely dissenting voices are being systematically pushed out of traditional media, a lot of people the liberal media intelligentsia doesn’t like have audiences that will support their projects financially, sometimes lavishly so. And the number of platforms out there has multiplied, meaning that even if any individual writer gets run off of one, they will likely catch on somewhere else, especially given that Ghost exists to make centralized content moderation impossible3. Despite the endless canceling discourse it’s almost impossible for someone to get canceled to the point of being legitimately silenced, provided they have a loyal audience. (Look at… me.) Canceling is powerless. Traditional media types care enough about these people to bother getting mad precisely because they are successful. So let’s just accept that people you don’t like will continue to say things you don’t like. The question is, can you perhaps influence them to do more of what you like and less of what you don’t?

I think maybe you can, if you bear in mind what I said at the top: that none of us are independent, that you’ve gotta serve somebody.

Things are going great with this project and my life, which was most certainly not true a year ago. I do have a consistent issue though. What a lot of thoughtful people who care about me think I should write, and what gets me the necessary subscription revenue to continue to do this as my job after my Substack Pro contract expires next February4, are very different things. Almost the opposite, in fact. Many of my readers, friends, well-wishers, and general observers want me to stop doing the posts that bring in money and only do the ones that don’t. And I’d like to do that, but I would also like to ensure that I don’t have to go on the job market anytime soon. You see my dilemma.

Isn’t that exactly what platforms like Substack and Patreon are supposed to ameliorate, the need for every particular expression of someone’s work to do big numbers, the click anxiety that has ruined so many legacy publications5? Yes, that’s the idea. Have they succeeded? It’s complicated, at least in my idiosyncratic experience. In the narrow sense, for me, sure. My views have fallen off of a cliff in the past several months, and yet subscription revenues are steady, even inching up - I’m projected to earn more by the end of the year now than I was at the beginning of June, at any rate, though not by much. And from the long-term perspective it’s all going great. I’m making much more than I thought I ever would at any job, so I have no complaints there. My readers are, by my lights, unusually amenable to my peculiar set of interests, the breadth and pace of my writing, and my commitment to pursuing whatever I want to think and write about at any given time. I am in a position where I can comfortably support myself and another adult financially, and if we were not in New York due to preference and circumstance I’d obviously be even more secure. I’m lucky, privileged, and grateful.

But in the broader sense, neither Substack nor any other kind of crowd funding scheme can make any of us truly indifferent to audience. We are still in the selling writing business. You sell writing in a market, and if people like what you’re selling, they’ll buy. That’s the rejoinder to critics of this wave of crowdfunded media - we’re succeeding because we provide something people want to consume, something they’re not getting from legacy media. But that also means that the market will always assert itself into your decisions as a writer. The reality is that you have to keep convincing people to recommit to your project, and if you aren’t drawing new eyeballs eventually projected revenues will go down - enough people will just naturally fail to resubscribe and won’t be replaced. Attrition. I do get occasional “I’M CANCELLING MY SUBSCRIPTION”-style cancellations, and would be worried if I didn’t. But I suspect most people just lose track or lose interest and don’t bother to renew. Apathy will erode your salary much more reliably than anger. What everyone being supported through Patreon et al has to fight is the gradual loss of interest over time. Just to name a telling subset, people’s credit cards expire and then they don’t remember or bother to reup when they get their new card. Out of sight, out of mind.

This has been the big surprise of my Substack experience, the frequency with which I both gain and lose subscribers. In fact, my expectations were completely subverted. My assumption was that projected annual income would grow slowly but that those who bothered to get out their credit cards would really commit for the long term. In fact the opposite has been true. At the beginning I set what I thought was a somewhat ambitious goal for subscription revenue at the end of the first year. Instead I hit that goal in the first six weeks. But I also found that people will unsubscribe quite readily, again probably less because of specific unpopular posts and more because of a string of posts that just don’t spark their interest. And this leads to my secondary surprise: I care about that. After 15 years of proud indifference to clicks and shares and all manner of external validation for my writing, I find that I want to have more readers rather than fewer and to make more money rather than less. I try to police myself and to remain as idiosyncratic and driven by whim as ever, but I cannot help but notice my own concern for paying down my student loans6. Perhaps the people with really high crowdfunded revenues are free from this sort of feeling because of that cash buffer. But I’m not.

There is no such thing as independent media; there’s just different kinds of dependence. If you’re taking people’s money you’re under their influence, period.

All of this is complicated by the fact that there is a very obvious and deep divide between what certain committed and passionate readers wants from my writing and that which is actually successful in material terms. This represents itself most clearly, and most ironically, on Facebook. I will share a piece that (alongside whatever substantive point I’m making) baits the media, and half of the comments will scold me that I should be spending my time on deeper, more thoughtful engagement. But those will be among like 100+ comments, and there will be 150 likes, and 15 shares. Then I will write and share exactly the style of piece those people say they want, and it will get perhaps a half-dozen comments and 20 likes and 0 shares. (I will then write a meta status update sighing about this dynamic, and I will get more engagement on that then I do on the posts which I am saying should get more attention, which serves me right.)

This is an consistent reality about writing for the internet, at least in my experience: the vast majority of the people who would prefer you to write something other than that which gives the most engagement will not reward you for doing that something else with their engagement. They have a vision of what your work could be that they will happily share with you, but they won’t actually read any of it when you try to put those principles into practice. Look, I’m not trying to be overly deterministic here; I still write mostly based on personal whim, I have zero long term tracking or plans for what I write in a given week or month, and I don’t sit around saying “how can I please those Facebook critics?” But it’s simply not realistic to be truly indifferent to quantitative rewards, and I would have to fly completely blind in this project to remain ignorant of the fact that the more I produce the content many people say they want, the worse this newsletter performs. Allow me to illustrate.

These categories are chosen purposefully. Getting salty about media is my most consistent click-generator and money-maker and is also what most people say they would prefer I do less often or not at all. Education research/policy posts are probably the single most-requested type of post in emails and comments, and because they could plausibly get a lot of engagement, it makes the comparison more fair. (As you see, the top-performing education post did pretty well.) I do other types of writing that don’t get many views at all, but I don’t expect them to. My favorite type of piece I do here is things like my review of Jenny Offill’s Weather, but I go in to those knowing that they won’t do numbers and I consider the opportunity to produce them part of the gift of artistic freedom I’m lucky to have. The issue is that, if I want this to continue being financially viable as my job long-term, it’s hard to look at that table and not realize what I’m losing by giving up the upper rows and emphasizing the bottom.

I have added a little annual projected revenue since the beginning of June. This represents real financial good fortune and I’m not in any sense complaining about where I am in terms of total dollars. But that top-ranked media post there - which, for the record, I think is true and good and well-written - generated about 12% of my current total annual projected revenue in one day. Like everybody else, I’m influenced by those kinds of incentives. What many of my longer-term fans would say is that this table represents exactly the wrong kind of incentives, and I don’t really disagree. But I sometimes wish readers were more aware of the fact that if I only write the stuff they like I won’t be able to really write professionally.

To be clear, I don’t think media posts are bad and education posts are good. I value my media crit stuff. (I’m particularly proud of this one, and my long-term/passionate readers loved it too.) I think there’s a profound paucity of genuinely adversarial media criticism out there, that institutions like Poynter or The Columbia Journalism Review are subject to profound regulatory capture (the people who do that “criticism” want to be perceived as part of media), and that everything I’ve written about media as a social cartel is both true and necessary. Indeed, those posts do numbers because they make media people mad, and they make media people mad because they say true things. A hit dog will holler. But it certainly is also the case that they invite more negative emotion than positive even as they pay the bills, that the people who are most intimately familiar with my work and supportive of me personally like it the least, and that I would prefer to do it when I think it’s just but not feel pushed to do it for market considerations. But writing that stuff is a profoundly reliable way to goose subscriptions, and I can’t unsee that. I can tell where the spikes are in the projected revenue graph, after all. It hasn’t been hard forgoing that type of writing the past couple months, but if subscriptions suddenly took a nosedive? The temptation would be powerful.

It’s similar with left-on-left critique; it’s popular, I think it’s important, and I think I’m right on the merits, but the incentive structure gives me pause. The third-ranked post on this blog in terms of views is my post attempting to deconstruct social justice politics in a way that’s not sensationalistic. I’m proud of it and I’m grateful for the attention. But if I’m not careful, the urge to play to conservative audiences who just want to see me skewer the woke could take my work in directions I don’t want it to go. I’ll always criticize liberals as I’m a lifelong leftist7 and deeply committed to improving the left’s project. But there too the mercenary incentives are clear; it’s trivially easy to get a certain type of conservative to share “left-on-left” critiques which then generate revenue. I have a couple posts sitting in my drafts that would no doubt hit 70,000+ views and get some subs, but I don’t want to unleash them because I make these critiques out of genuine commitment to a better left, not to manipulate conservatives into doing my marketing for me. The question is to what degree my project is financially viable if I forego the more inflammatory stuff and focus instead on the stuff that people say they like but which they don’t read or share. I genuinely don’t know the answer to that question. This might just be a “me” problem. But I don’t think so.

I loved writing this post on the thinking behind changes to math pedagogy. And as is so often the case here, the comments were so sharp and engaging and substantive. That kind of mature, sensible, diverse, and informative conversation is truly rare on today’s internet8. And some people corrected me on a few things, constructively, so now we both know more. It’s exactly what I want my blog to be. But the post only got 13,000 views, in large part because it got no pickup on Twitter. (Twitter is by far the largest external driver of traffic to this Substack, despite the fact that I don’t have a Twitter9.) It got 20 likes, 12 comments, and 1 share on Facebook. For context, a Facebook post making fun of Rotten Tomatoes I dashed off a few hours later that day got 74 likes and 34 comments. The math post did not generate any growth at all to the mailing list or subscriber revenues, at least as defined by Substack’s internal metrics. It was a market failure, in other words.

I wouldn’t mind subsidizing the thoughtful stuff with the occasional inflammatory piece, the old Gawker strategy of Neetzan Zimmerman publishing viral horseshit so that we could get essays by Tom Scocca. But in practice even publishing argle bargle once a month gets me a lot of heat I don’t want - many email lectures from well-intentioned readers who are, perfectly legitimately, urging me to align my writing with their interests as consumers of my writing; groans from friends who want me to live up to what they see as my potential; people threatening to cancel their subscriptions in ways that don’t change the revenue math but do affect me…. I never cared about any of this shit when I wasn’t accepting money. Whether that indifference was better or worse I couldn’t say. But this is now my profession, I’m very grateful to say, and things change when you take money. They just do. I get it: this is the basic bargain with running a newsletter rather than writing for a high-circulation publication, it’s a remarkably privileged problem to have, and there’s nothing novel about discovering a conflict between art and commerce. Every writer knows the pain of writing something they truly love and seeing it disappear without notice. But I think these dynamics have real consequences for the future of media.

I took a swipe at Adam Serwer in a post, which was admittedly jerky, and it was like the Bat Signal went off for other liberal journalists - they immediately engaged on Twitter. Engaged negatively, but that distinction makes no difference for the purposes of attracting attention and subscriptions. Recently I wrote a critical but substantive and not inflammatory post about modern liberalism and referenced some of those very writers who had gotten mad. None of them seem to have even noticed I wrote it. I get that there are obvious reasons that people don’t want to engage with me, but to only engage with me when you’re mad makes it profitable for me to try and get you mad. More broadly, I wonder if they’re aware of the incentives they create when they only react to people outside their social sphere if they feel their friends have been insulted. Writers love to complain about how feckless readers only want clickbait, but given how Twitter drives traffic and generates attention I think the real problem is their own behavior towards other writers.

All of this leaves me unsure about what to write about, even when I feel something is important to engage with. I have a (harsh) post about a certain cable news personality sitting in drafts. It’s good and it uses this person as a lens to understand broader dynamics, important ones. But inevitably people in media would call it “personal,” which along with “harassment” is one of the ways they forbid legitimate criticism. So do I publish it or not? Some people would no doubt prefer that I don’t. But the media does not do real self-criticism, and the media matters. So it feels like a dilemma.

Back to the beginning. Perhaps you can make the internet a little better (I stress, a little) by doing good by your enemies. I ask you to recognize that hating your foes helps them, and while giving them certain kinds of attention won’t hurt them, it could help change the bad incentives that have us stuck in this hellhole.

If you get annoyed at Matt Taibbi for being mean to liberal journalists, you could yell about him on Twitter, which gets him views and clicks and subscriptions. You could ignore him and hope he’ll go away, which seems unlikely given that he has a high profile, several books, a popular podcast, half a million Twitter followers, and does TV hits all the time. Or you could try sharing his interview with Dennis Kucinich or his review of Kucinich’s book, both of which say interesting and useful things about the left and where we’re all headed. You’re annoyed that Glenn Greenwald goes on Tucker Carlson? You’ve tried yelling, calling him a Russian asset, saying “HE LIVES IN BRAZIL!” Has any of it worked? Then why not try reading and reacting to his fascinating piece on the anthrax scare of 2001? That’s a bizarrely underdiscussed moment in American history, and I don’t see why even committed liberal culture warriors couldn’t find value there. If you think Jesse Singal’s work on trans issues is problematic, you should pause and consider that all the hate he’s gotten over it has enabled a successful newsletter and a big-money Patreon podcast. Yelling on Twitter about him hasn’t worked10! It serves his interests! So why not consider reviewing his book, which is not about trans issues but dubious psychology research that goes viral?

An essential point: positive reinforcement here does not necessarily mean praise, it means engagement, including critical engagement. Actual critical engagement, I mean - not photoshopping Michael Tracey into a clown costume so your colleagues at Vice can give you 212 half-hearted likes, but actual “let me take this seriously enough to praise what deserves praise and then to unsparingly criticize what deserves to be criticized.” Meaningful criticism hurts much more. Trust me. And it will help break the terrible viral cycle where the pieces that inspire the most rage are the ones that receive the most attention in an attention economy, incentivizing more.

I don’t think the internet bad boys are going to consciously want to do more of what you reward with meaningful engagement. But I can guarantee you that they will do more of what you freak out about because oppositional psychology is a thing, and so are clicks. I’d love to see a mass psychology experiment along these lines. I have no illusions that people are going to actually do what I’m advocating for here, but I genuinely think they should. If nothing else, you would think members of the legacy media would eventually realize that their hate can be the best thing for someone’s career, in part because many ordinary people hate the legacy media. (Don’t shoot the messenger.) Chris Rufo will be able to coast to a comfortable career as a right-wing celebrity for life. Conservatives didn’t do him that favor. Liberal journalists did.

I wrote this piece about the strengths of standardized tests awhile back and I really thought it might break out and get a wider audience. It’s about education, which everybody wants me to write about. It interfaces directly with hot-button political issues, notably liberal assaults on the SATs. I focused on educational assessment in my PhD and wrote a dissertation about standardized tests of college learning, so it would seem to fit with my strengths. But the piece never really connected with the crowd the way I hoped it might, as you can see in the table. That’s disappointing - yes, for engagement and its attendant professional benefits, but also because I’m very passionate about the issues discussed and wish I was more a part of the conversation about them. It’s not a tragedy; my readers responded very positively and I’m grateful. I’m lucky in that I’m not really personally motivated to be widely read as such. I could certainly spend the rest of my life writing for the audience I have now and be happy. But as someone who lost everything at a young age, I’ll never be able to turn off the part of my brain that sees financial ruin right around the corner, and it can be hard not to dial up the old outrage machine. You can say that it won’t work next time, but people have said that before, and it’s been working for me for almost 15 years. It’d be better for me personally if the financial incentives better aligned with what people see as my strengths as a writer - and better for everyone if that was more broadly the case.

So what I’m saying to the media types with their large networks and influence is, you can once again tweet about how I’m a loser who shouldn’t have a platform when I call some bluecheck a dickhead or whatever, and in doing so draw readers and signups to this newsletter. Or! You could share that post on standardized tests, even if it’s just to say that you think it’s wrong. I suspect the latter strategy will get you more of what you want, in the long run, than just doing what you always do, as counterintuitive as it seems to bless your enemies with attention. It’s worth a shot. Could it be worse than the status quo? At this point I think people would be willing to try just about anything to fix the state of online conversation. It can’t stay this bad forever… right? If you will it, it is no dream.


There is much that I could say in defense of naming those two ostensibly non-partisan papers as liberal publications, but perhaps the most parsimonious is simply to point out that, were you to poll the employees of each, the dominant majority would self-identify on the lefthand side of the political spectrum. As is true in higher education, you can’t have that kind of numeric dominance without tangible pressures emerging. You just can’t.


Part of Joe Rogan’s tremendous success as a figure of dogged admiration, despite the antipathy of so many influential people, stems from his inaccessibility, I am convinced. He looks like a Buddha compared to his many critics because he isn’t on social media complaining about them, which only reinforces his authority among his legion of fans.


(Which makes treating Ghost as the platform that’s safe from bigoted views so, so weird. )


Again, I am on pace to make Substack more money this year than they will pay me under the terms of our contract. I mention it only because there is still a persistent narrative - you can guess where it comes from - that they gave me an advance as some sort of gift, which would be a profoundly weird thing for a for-profit company to do, especially given that I had never met or interacted with anyone involved in the company prior to starting to write here.


Go through Newsweek’s tweets and count how many are about snakes.


Or, like, buying cool shit. I’m susceptible to that too.


Gating comments only for subscribers does wonders. It really does. I get that people find it inegalitarian and feel that it violates my commitment to open exchange/angry yelling. But this is the only way I can have the kind of conversation I want to have without heavy-handed moderating, which is both ideologically untoward and a lot of work.


And never will again, don’t worry.


You should, however, yell at him for not having a logo for his newsletter. Amateur hour.