2020-01-23

Writing Every Day

I don’t write every day. Perhaps I should. Writing often is a great way to build a visible and public body of work, improve as a writer, and work through a great many ideas with more technical completeness than would come from idle thought.

Let’s explore why we might want to write every day.

First of all, by “write” every day, I of course mean “publish”. I do in fact write every day, even if only messages and notes, but also a great many unfinished essays. They don’t reach the point of completion. The point is to consistently drive it through to completed product that can be shared with a community of friends, ie you guys.

A classic parable comes to mind: A pottery master broke her students into two groups. One group was to be judged on quality of the work produced alone, the other on quantity alone. After a period of instruction and production, the group assigned to produce quantity was producing higher quality than the group targeting quality. Repetitive practice produced the better results than up-front effort.

The application to writing is obvious: publish every day, whatever you have, regardless of quality, and your writing will be better than it would if you only published polished “high quality” work.

Is it true? Nobody wants to do bad work. You will always be trying to do good work. Your ability to actually do good work is a function of your skill at the outset of the work, not effort. Skill can be trained up with lots of practice. All that trying to target immediate quality does is slow down your practice.

The axiom doing the work there is that skill outperforms effort. Why might that be? Skill is knowing what to do at what time. Having those motor and attentional patterns drilled in to perfection. Knowing not to go down some bad path, which would introduce an uncorrectable imperfection and require starting over. Having direct experience with all the other ways you could do it, and knowing that this one is the best. Skill results from lots of experimental practice. Effort, being extra careful with your intention, cannot make up for lack of skill, detailed technical knowledge.

Another parable comes to mind in the same vein, from Hamming’s “You and Your Research”. He asks a mentor how John Tukey does such good work. The mentor’s answer is blunt: “You would be surprised Hamming, how much you would know if you worked as hard as he did that many years.”

Knowledge and work are like compound interest. (Not that that’s anything our generation has ever seen). Finishing one project gives you a base from which to do more. The more you push things through to completion, especially solving the general problems implied by the specific, another theme of Hamming’s talk, the bigger your library of knowledge is, and the better your next thoughts will be. Writing is where you flex your brain and work out your thoughts.

So let’s write more.

On the compound interest theme, published work connected to a body of work is worth more than the sum of its parts, and continues to earn interest as long as it is accessible. There are a lot of intensifications from scale and repetition. This insight I owe to David Perell.

ReviewBrah’s youtube channel didn’t start out with multimillion view counts, nor would any one of his videos earn that on its own. Nor is his reach limited to a couple videos with a million views; all of them get tons of views. This adds up to a huge impact, and a huge empire. The best writers I have read are like this.

Eliezer Yudkowsky, the big man of modern “rationality”, wrote every day for three years, and it created an impressive following of some of the smartest people on the Internet.

Besides intensifications from repetition, there’s what happens to the particular pieces of writing.

Deadlines, especially short deadlines like every day, force two things: they force you to actually do it, which is obviously good, and they force brevity.

With longer work, it’s very easy to get writer’s block. How many half-finished essays do you have that you started over a couple of hour-long sessions, and then got bogged down and never finished, and started another? I have hundreds.

Frequency forces smaller size, which makes writing easier to produce, by keeping it on the faster initial part of the sublinear effort vs results curve for writing. Empirically, my best work is done in two hours and weighs in at about a thousand words. By the time I’m in the multiple thousands of words, it can take weeks to slog through. Maybe it’s just me, but I suspect it isn’t.

Twitter is powerful because it forces brevity. It’s easy to casually read a tweet, or casually write a tweet. You don’t get bogged down in all the background arguments and contingencies. You just get the idea out there and see what your friends think.

On the reading side, brevity is great. You want to get through the core insight fast. If someone linked you an essay for some important idea, you don’t want to see some monster of twenty thousand words that you need to put on your reading list. You want to blast through it in less than five minutes. If it’s good, you start to want to explore that author’s other works, but you don’t want to have to read all that background up front.

Everybody links particular wikipedia pages. Nobody wants to link a serialized wikipedia as a whole. The enhanced linkability feeds back into the value of the public body of work. Frequency forces brevity, which enforces this property.

So let’s write more.

I think I’ve made the positive case well. What about the other side? Why shouldn’t you write every day?

The first and most obvious issue is time. It takes an hour or two to write and publish your daily essay. In some lines of work, you just can’t spare that. Or at least you can’t arrange your schedule and self-discipline such that it actually happens.

This is legitimate, but I think in many cases, especially for any line of work with an intellectual or public relations component, it’s simply worth the time to systematically collect and work out your thoughts, and tap into that compound-interest machine.

The second issue is the downside risk of sloppy work. If we live in a hostile world of attack mobs and bullying, saying the wrong thing online can get you in a lot of trouble. You’re more likely to say the wrong thing if you’re saying a lot, and saying it at a lower level of quality assurance.

Downside risk is much more of a concern if you’re a thought criminal or have a habit of posting cringe. Otherwise, I think this is greatly exaggerated as a serious concern. Most people will skim over your half-baked nonsense and read the good stuff, unless you’re really off your rocker.

Finally, you might have nothing important to say or think about. Maybe this is a problem that some people have. I certainly have more that I want to think through and say than I think I ever could. Each thought opens five more threads. Maybe I could put a dent in it by writing a lot more. Maybe you could too.

So let’s write more.