Remodeling the Units and Flows of Intellectual Production (How Academia Got Pwned 8)

This is the eighth post in a series about the glorious completion of my academic career, the internet, and the future of intellectual life. This will probably become a book. If you'd like to hear about that when it happens, be sure to subscribe.

[This post dives deeper into some back-of-the-napkin financial projections for a different kind of intellectual production model, fit for a new kind of high-brow but radical and truly independent, internet intellectual. It speaks to the question of what I plan on doing next, but it might be applicable to others also. If you're not interested in this aspect of How Academia Got Pwned, you can feel free to skip this one, as you'll probably find it quite boring.]

Let’s say we start with the idea for one book. What is the best, most important book one could start writing today, if one knew in advance one would never need permission at any time from conception to publication? Develop the idea simply to the point where one can draft a full outline. To fix ideas and generate some forecasts later, let’s assume we’re talking about a high-brow non-fiction book with a word-count in the conventional range, but on the lower end (erring on the side of brevity for harebrained schemes is probably best). Say, 50k words total. To keep numbers tidy, say the book has 10 chapters of 5k words each, and each chapter has 5 sections of 1k words each. To say that you “have an idea for a book,” means you can sit down and simply outline the one main point of each section, in each chapter. You don’t need to know how exactly, and obviously it can change as you go, maybe drastically, but let’s say this is what it means to officially have the plan for a book. Anyone with experience writing long stuff — if they have an idea for a book — could probably do this part in a day or two.

And once you can do this, you pretty much also have a course syllabus. For each chapter, just rewrite the chapter headline into something that sounds more like a lecture title. Of course, the hard part is having and executing the content. But a book outline, like a syllabus, is just a skeleton for planned content. We’ll come back to this later.

At this point, you have to throw up a website for the project, just because you need some place for people to go if they come across your stuff and want to learn more. Don’t overpromise though, because you don’t know where it’s really going to end up. Ideally, a website would simply allow people to receive updates about the project.

On this point, here is a long aside (skip the whole paragraph if you’re not interested by the next sentence). In a world where people can get kicked off any social media platform any day, radical intellectuals should privilege email. It’s funny how one of the most direct, decentralized, and non-corporatized channels of digital communication is dominated by businesses and scams. If you work a demanding professional job, what I’m about to argue will be inapplicable to you, because your inbox is already overloaded (that’s not an argument against email, it just reflects the degree to which demanding professional jobs crowd out free communication). Email gets a bad rap because it is the medium of choice for spammers, but that’s just because smart, evil people are usually the first to exploit what works the best. I’m signed up to tons of email newsletters by weird creative people and bloggers and I love them, much better than checking websites, or my RSS reader, which is higher-volume. I never miss anything new, unless I just don’t feel like reading that day, in which case I can skip and delete in only a second. I would really like to see email newsletters become more popular among radical intellectuals. If we’re trying to create independent work, we want to make it as easy as possible for interested readers/watchers to sign up for email updates, without becoming marketers. No manipulation, no overly aggressive popups, minimization of distasteful instrumentalism without foregoing strategic (instrumental) intelligence, which would be tantamount to exiting the meme pool. If one sticks to these rules of thumb, we shouldn’t cross the line into distasteful or exploitative or corny, like marketers and self-help gurus often do. I don’t mean to be drawing hard rules here, and one of the wonders of life on the internet is you can experiment with dozens of different methods at any time, but these are the rules I’ve followed as I’ve built up my systems so far. I’ve put a lot of effort into making it easy for people to sign up for email updates, for instance, but I’ve put equal care into not doing most of the ugly stuff shameless marketers do. If I ever deviate from this I expect one of you to tell me with immediate and brutal honesty! True intellectuals cannot allow themselves to become marketers because it is directly contradictory to their mission (indeed this is a major failure mode of our project, because the temptation will always be great). On the other hand, we cannot afford less than optimal communicative efficacy, and for obvious reasons, the current tools of the marketer represent the current perfection of communicative efficacy.

Now let’s say you commit to writing 1k words per day, 5 days per week. For the average person this might be a bit of a slog, but for an academic suddenly released from all the other nonsense, it’s not hard at all. On this schedule, it will only take you 10 weeks to write a first draft of your book. In large part because you wasted no time asking anyone about anything.

Now, let’s assume you really went off the rails and this monstrosity you’ve drafted is hopelessly undesirable to any currently existing publisher. You could sell it directly to readers who are as off-the-rails as you, but you have no readers yet. No problem, just go get some. This is where the massive efficiency gains and positive externalities come in. In these 10 weeks, you’re not just writing a book, you’re also making lecture content at the same time. I’ll explain…

If you think it's impossibly difficult to gain readers, you don’t understand yet. By virtue of the very fact that you are sabotaging your own prestige by slumming it in the digital ghetto, you’re going to get at least a few eyeballs giving you a peek, if only to understand what’s wrong with you. Over time, the ones that are temperamentally/ideologically adjacent to you will stay and the others will go, and you’re disproportionately likely to win continued attention by virtue of the fact that if you’re doing this you probably have something real to say. Everything I’m laying out is for people who have real work to do; that’s why this isn’t a self-help commodity I’m writing here, because I’m not at all pretending that anyone can do this. If you’re not smart and disciplined with real intellectual work to do, nothing I’m theorizing here can make you “succeed” and I won’t take likes or dollars to feed anyone that impression.

Also, you can’t be pessimistic from looking at current academics who occasionally blog. For most academics who blog, their blog doesn’t quite develop into much (in terms of content quality or following), but that’s in large part because it’s often a safe, sterile, side-project given only the scraps of energy left-over from an already too exhausting career. So you can’t compare what I’m sketching to such examples you might have in mind.

Getting followers isn’t complicated. Having things to say is the hard part, but if you do, then all it takes is consistency and time. It basically boils down to posting original and interesting content consistently (note, it only has to be original and valuable to a tiny fraction of the internet population for you to eventually have a respectable audience of loyal and highly interested readers).

How convenient for blogging that your book outline is really just a long list of things you want to write. And how convenient it’s already broken into chunks of one thousand words, about the average length of a thoughtful blog post. So there you have it, you will blog 1k words per day, 5 days per week, for 10 weeks. Each 1k bit should be written for blog readers in mind, but all the blog-specific parts can just be cut out at the end, when you turn your collection of blog posts into the book. Make sure your blog gives people an option to sign up for updates, only if they want to, of course. You could mention the book and course project in your blog posts but in my experience, ambiguity here is your friend; it keeps the blogging fun and unencumbered when don’t feel externally committed to some long-term result (you should be long-term committed to your plan, in your own mind, but when it feels like something you owe the world, it can have negative effects I think.) Unless you are very confident it will happen, in which case it’s probably good to let people know what’s coming (like with these posts).

It would be amazing if there was some way to convert this 1k chunk each week into another piece of content, also weekly, that had all of the following properties. It can’t require too much heavy cognitive lifting, because 1k words every day will take its toll. This second derivative should mostly require simple labor. It should allow for some improvement/revision/addition on the 1k words. It should be accessible to a different type of person, on a different social network, and therefore provide new value relative to the 1k words, even if it’s admittedly somewhat derivative. Finally, and crucially, it should also represent a piece of content that could enter (either directly or indirectly, after editing) into the course corresponding to the book. It turns out that the genre known as the Youtube video has the virtue of meeting all these criteria. Every week, post one video presenting a summary of that week’s 5k words, bonus points for lecture slides or visuals edited into the video, but just do the best you can. At first you won’t receive many views, though you’ll get some, and the real value comes later. Just like the blog, make sure your videos give people an option to subscribe via email, if they want to. At the end of the 10 weeks, you’ll also have 10 lecture videos. You already have a syllabus. This means you pretty much already have a course, and you didn’t even need to put your pants on, let alone shower every day of the week.

Getting an ebook and paperback on Amazon is nearly instantaneous, once you’re happy with the final product. Price the ebook for a few bucks, and the paperback for about $15, which is pretty conventional. You will make about $4 of that $15 for each copy sold. Add some extra content or editing to the course videos, or cut off the last few minutes of each Youtube video, slap a modest little paywall on the course. Email your subscribers to let them know.

Depending on your current audience, this first production cycle might make you only peanuts. Personally, I like to concoct my own schemes always assuming the worst case scenario, but allowing oneself the motivational gains from pondering the best case scenario, as well. There is obviously a lot of uncertainty in this kind of exercise, but we can account for that uncertainty explicitly with this great new tool called Guesstimate.

Guesstimate is just a spreadsheet that lets you input distributions rather than individual values. Have you ever tried multiplying distributions? It’s prohibitively time consuming to do in Excel or by programming it yourself, so Guesstimate does it all for you in the background (Monte Carlo simulations). It’s really great for thought experiments such as this, where I have no idea how many books/courses I would sell each month, but I could much more confidently guess an interval within which the number probably lies.

For units sold, we'll set the minimum very conservatively low (0), and the maximum to the low end of what I would consider good but realistic. For books, you can get a rough estimate of base-rates by estimating sales from Amazon rankings, using tools such as this. I’ve seen pretty random non-fiction authors achieve 100 copies/month, so let’s go with that as the best-case scenario. Some notable examples by Bronze Age Pervert, Elizabeth Sandifer, or Vox Day have exceeded this significantly, and I don’t see why such results would be out of reach, but I'm trying to be harshly realistic so I'm not even going to consider these known success cases. For courses, the number of units likely to sell is harder to estimate because most successful online courses today have some direct money-making value. I have a lecturer friend who made a pretty random political science course on Udemy and he reports that it does better than you would think. Let’s say my course would sell somewhere between 0 and 20 units per month, on average. We’ll tell Guesstimate to use a log-normal distribution, which means lower values are much more likely. We’re thus being doubly conservative and biasing the estimate downward, to discount for the well-documented over-confidence of entrepreneurs.

Now we need to input expected earnings per item. Most paperbacks are priced about $15; Amazon offers a royalty rate of 60%, minus shipping. Since they do all the printing and all the shipping, this is about $4, which means there is about $4 leftover for you, per copy. Many online courses have succeeded at very high price points, but that’s usually for learning content that will make the buyer money somehow. I’m interested in the market for disinterested truth-seeking content, so presumably people are willing to pay less for such content. So let’s imagine, conservatively, that I could produce a course worth only $20 (one-time purchase).

Run some Monte Carlo simulations, and we can produce a confidence interval for our annual income derived from our one book+course combo. From left to right, the image below shows how Guesstimate converts our inputs to an estimate.

The featured number in the center of each box is the mean value, and the range represents the 95% confidence interval. You can see the model (and I think copy and edit it) here.

According to this, in the worst case scenario, this 2.5-month exercise makes us only $520/year but at the higher end of a not-inconceivable range of possibilities, it could be making $7400/year. I feel comfortable with the plausibility of this model because the projected results are really not great, so it feels realistic. The best guess, the mean, would be about $2300/year. If I could do this amount of work for a whole year, then I could do this four separate times. That would mean that I could reasonably expect to be making $9,200/year after the very first year, and it’s not inconceivable I could be making up to $29,000/year after the first year. As I said, I like to assume the worst, so let’s take it for granted that I will actually be making $2080/year after the first year.

There are two realizations that started to make this kind of vision irresistible to me: Even in this latter case, which seems virtually guaranteed as the worst that could possibly happen, I would consider such a year of my life subjectively worthwhile. Sure, we’d make next to nothing for a whole year, but the intrinsic value of producing totally autonomous work for a whole year is very high to me. I already have about 3 books outlined with parts drafted to varying degrees of completion, and I suffer a lot from not having the time or permission to get them out there. Aside from money, I would be tremendously satisfied to get them done and in print. Even if it financially fails, it seems worth it to me. If I have to get a real job afterward, that’s fine because I got three damn books done and I can do my future books at a slower, and more contented rate on the side of a real job. The past five years I’ve been so restless and frustrated at constantly doing career-maintenance work and waiting on others. There's no way I could write more than one book every 3-5 years doing everything the old-fashioned way as a career academic is supposed to.

The second point is simply to recall that, after the work is produced, the future income it produces is passive. So even just $2k/year starting this year will make $100k for my family over the next 50 years, even if I have to get a real job one year from now. So even if it fails to become financially sustainable this year, it’s not like this one year of writing would be one big selfish waste of time. This point also raises the possibility that, if I’m only partially successful in the first year, I could end up doing quite well by carrying on this work with some additional part-time paid work on the side. Given the cumulative and passive nature of this income, a few years of hustling on this as well as part-time paid work could get me to writing full-time, say, after 5 years of building my back catalogue, maybe.

The cumulation factor is worth pausing on. In the self-publishing world, most people report that your back catalogue of books gains in sales when future books come out. You are learning how to do things every cycle, so presumably your work increases in quality and value each cycle. And each time, your audience is somewhat larger. If there are multiplicative interactions among any of these increases, then even a very modest initial start could sooner than later produce an inflection point in your income growth rate (e.g., one of your videos takes off in the Youtube algorithm, drives many new email subs, which makes your next release more successful, which makes your back-catalogue more successful, and as a result your one random spike on the Youtube algorithm is giving you some non-linear takeoff in your bottom line.). It’s not about seeking superficial virality (a surefire way to never say or make anything that matters), but quite the opposite: it’s about having in place a sturdy machine that allows you to produce high-quality work over time while also absorbing the randomness of virality or platform-anomaly into the productive apparatus, rather dissipating in Twitter beefs that go nowhere.

In my own personal case, I wager that my financial projections are a bit rosier than this, because I have other sources of income support, such as 52 patrons and 10 participants in my private seminars. If these increase even modestly each month, then that’s another pathway to pwning academia no doubt, but I have tried to explain my thinking with reference to a more general production model that could potentially be taken up by any educated and disciplined person with knowledge to share, regardless of whether they have patrons or other income streams. I also have the advantage of already having a lot of writing and teaching materials sitting on my hard drive, which I can use out of the gate. While this makes my own experiment not totally applicable to any person interested in such a path, it does speak to my point about how academia, specifically, gets pwned. For academics are uniquely capable of defecting effectively and sustainably. Well, if my calculations are correct, that is...