A writing commitment

I recently committed myself to publishing one piece of writing per week, in an attempt to improve my writing and produce a record of my thoughts. Here’s my reasoning.

On writing well, there seem to be a few key points that are widely agreed upon. First, it’s important to write often. You should accept half-finished work — it might come in handy later.

Second, you should get to the point. This is a large part of Strunk and White. It’s also the content of this Scott Adams blog post on writing, which is so short that it’s almost certainly worth reading.

A brute-force approach, where you set aside chunks of time and refuse to do anything but write within each, is useful (perhaps essential). The mechanism at work here is described by Richard Hamming in You and Your Research:

… the way to manage yourself is that when you have a real important problem you don’t let anything else get the center of your attention - you keep your thoughts on the problem. Keep your subconscious starved so it has to work on your problem, so you can sleep peacefully and get the answer in the morning, free.

Even if you start out with no good ideas, once your brain knows it won’t get any novelty unless it puts something on the page, it will eventually do so (Alexey has referred to this as “passing your brain’s shit-tests”).

Another trick is to put yourself into situations where you need to write. Financial constraints are one way — Winston Churchill wrote a lot because he needed to pay the bills. Public commitments are another. For example, I’m only writing this post because of the commitment mentioned above. I don’t want to be doing this right now — it doesn’t even seem valuable, just a collection of bits of advice that I’ve seen elsewhere. But I’m doing it anyway because I don’t want to go back on my word.

Finally, you should ask other people to read drafts of your work. You’ll try to convince yourself that it’s good enough, and they’ll point out all the ways that it’s actually terrible. This is great as long as they know what they’re talking about.

So, that’s the writing process solved. Great! The weekly commitment addresses points one and four directly: advertised in advance, it applies both internal and external pressure. It also addresses points two and three indirectly – brevity and focus feel more acheivable when you can see a deadline approaching.

Is production the right metric, though? The point of the Hamming quote above is that you should starve your subconscious into working on important problems. Is weekly writing important?

Not really. I like to write because I like producing things I can put my name on, and writing happens to be one of the production methods I enjoy the most. But that reward signal is basically independent of what I’m writing about, which means the topics I pick first will probably the easiest and most banal (like writing itself, perhaps the most over-written subject in the history of writing).

However, I still think it’s worth doing, because producing something is important, and managing production schedules is hard. Suppose you want to produce N words per week. You can commit to writing and publishing N words every week, or 4N words every month, or N/7 words per day. Which is best? My suspicion is that the longer timescales can lead to better work in principle, but that this takes some experience, finesse, and perhaps desperation, none of which I possess in the requisite quantities. In any case, the payoff from practice and increased fluency is worth it – you can always anneal the timescale upwards as you get better. Something is better than nothing.

2019-08-10