Long Content vs The PG Way

Ed [2015-06-10 Wed]: Obsolete essay - I’ve since changed my mind. The arguments here are correct, but the premise is wrong (about the value of writing essays in high volume).

Entry question: What sort of approach should I take towards essay-writing?

Tradeoffs:

Well, let’s look at the problem in more detail. What are the requirements?

Timeless Essays

I want to write Long Content - content that will stand the test of time. I have never been a fan of the blogging format. Almost all of my favourite online authors are on a static site - Paul Graham and Steve Yegge. I think Eliezer only did daily posts cos he writes faster that way.

Create a Vacuum in my Head

I want to write quick and dirty mini-essays on pretty much every single idea that seems important to me. That will clear my mind and allow me to think greater thoughts1. I will be able to process my existing ideas and mine them for gems systematically, and free my brain to explore new territory. Crucially, I will be forced to actually think up new thoughts. I won’t be able to rest on my “laurels” (the supposed “good ideas” in my To-Do List).

I want to explore the fresh and exciting essays that I normally miss out on. I shouldn’t have to wait till I work through a whole backlog of ideas before I can start having new ones2. Working on new ideas all the time means working at the edge of my abilities.

Refine my ideas

I want to keep updating my ideas. As I get more evidence, I should evolve my ideas. Slowly, my essays must cut closer and closer to the truth.

This is basically Rapid Prototyping. I improve my essay over several iterations. And return after some time to give myself more objective feedback. After a bunch of iterations, my essay will become highly accurate and insightful.

Note that this whole process usually takes me around a week. I rarely update an essay after that.

Idea Bank

I aim to record (crudely maybe) every single worthwhile idea in my head into an essay here. That will let me clarify each one by explaining to a neutral, smart, imaginary reader. I can act on these beliefs with high confidence because my logic is explicit and well-tested.

This will be my Idea Bank.

Feeling of Closure

Finishing an essay means taking a problem through to the solution while making my reasoning explicit and readable. What happens if I have an unfinished essay lying around?

An unfinished essay lingers in my mind. It behaves like a To-Do List. It drains my Will Power just by existing. My impulse is to leave it alone, but I feel I have to complete it - I mean I thought it was important enough to start, right… so I fight the impulse inside my head. My Will Power goes down, and all the while nothing gets done.

Worse, the longer I put it off, the less motivated I am to come back and finish it off. The reward has taken too long to come - f*ck you, says my mind. My mind won’t give me any motivation for that task (high Delay).

This implies that having unfinished essays around will make you more likely to procrastinate (which creates more unfinished essays). The last few times I went for three or more days without publishing anything, it was because of drafts that I was “still working on”. Each time, I had broken my rule of publishing something each night, and thus ended up with unpublishable essays.

Beautiful Gems

Frankly, I want to feel proud of my work. I need to look back at what I have done and get a feeling of accomplishment. I want to feel the awesomeness of having a good finished product in my hands. Anytime I feel doubtful about my own abilities, I should get proof that I am capable of producing gems of ideas.

Force me to publish everyday

I want to make meaningful progress every single day. And the only honest measure for that is to publish. I shouldn’t say “Oh… I’m still ‘working on it’” and procrastinate. Not publishing is a sure sign that I’m doing something wrong.3 Each time I did that in the past, I was refusing to do the hard work necessary to complete an essay.

Publishing something everyday is, imho, the game-changer.

High Standards

The hardest part of writing an essay is in getting it from “ok” to “awesome”, not from “blank” to “ok”. The last 10% is where you make the largest gains. You have been Tightening the Loop till now. You have crossed off all the dead-end ideas and hit upon the core of your idea. Every move you make here will give huge dividends. Quitting with an “ok” essay means giving up on your biggest insights. Instead, step on the gas.

However, you have to think hard here to actually solve the problem. Till now, it was a cakewalk. You just wrote down whatever went through your mind without worrying about order or logic. It is in the last 10% that you actually transform your draft into a meaningful essay. You rearrange and rewrite everything to point in the direction of your core idea. You trim off the side-branches that go nowhere. And you discover some more corollaries that hadn’t hit you so far.

Perhaps the most important reason to release early, though, is that it makes you work harder. When you’re working on something that isn’t released, problems are intriguing. In something that’s out there, problems are alarming. There is a lot more urgency once you release. And I think that’s precisely why people put it off. They know they’ll have to work a lot harder once they do.

– Paul Graham, The Hardest Lessons for Startups to Learn

I suspect that when I write “in-progress” essays, I tend to skip all of this. I just “do my best” in a short period of time and then bail. My mind says “it’s ok, it’s ok… it’s just an in-progress essay. No need to worry about completeness. Move on, move on”. Well, I’ve got some news for you, mind. You have to worry about completeness. You have to hold yourself to a high standard. That is where you will build most of your skills and get your great ideas. That is part of the Hacker Spirit.

If it is possible to make yourself into a great hacker, the way to do it may be to make the following deal with yourself: you never have to work on boring projects (unless your family will starve otherwise), and in return, you’ll never allow yourself to do a half-assed job [emphasis mine].

– Paul Graham, Great Hackers

It’s not drudgery. Great work is its own reward. When you match your skills with work that is barely within reach, you will go into flow.

Self-contained Ideas

Compact, self-contained packages of knowledge are much easier to manipulate in your head than big, open, unfinished, unrefined chunks of information. This is why decoupling is such a big deal in programming - small (ideally side effect-free) functions rock. You can’t put a handle on 2000 words that don’t quite go in the same direction. They are all over the place. You can’t use them to come up with further ideas. You can’t use them to solve problems efficiently.

The way to do this is to capture the core of your essay in a simple sentence - something you can say in less than two seconds4. (Make something users want, Shut Up and Multiply, Be Relentlessly Resourceful.) Now, when you want to use the idea from the essay, you don’t have to recall the mass of words you wrote, just the simplified slogan - “How do we create wealth? Make something users want.”

So, I need to get every essay to a compact state where I have a good handle on it.

New and Exciting

I want the reader (aka me) to get to the latest and greatest essays with ease.

I can easily demarcate the New and Exciting essays, when I do it the “write one essay and write it well” model. Under the “continually update” model, I don’t get to see the newest essays. There’s lots of things being changed, but most of them are old essays. The best I can do is put up the most recently updated essays - not much of a dopamine hit there.

High Volume

I have a lot of things to figure out and I’m willing to put in the time to do it.

(Counter: Do you really need to figure out a lot of things?

“Most things make no difference.” – Tim Ferriss

Wouldn’t you rather spend a lot of time on the few important essays?)

Having Insights vs Verifying Ideas

There are two ways of writing essays. One is the normal model, where you work on one essay at a time and complete it before moving on to the next. The other one is where you keep building up an essay over time. This is Gwern’s idea of Long Content - keep accumulating evidence and stuff.

I think Long Content works best when you’re trying to verify an idea.

You want to see how true an idea is. So, you keep accreting statistics and data and quotes and putting them on one page for easy future access. Gwern is very good at this, I think. He writes in a careful, scientific-paper kind of way. This is great for monolithic, comprehensive documents about one topic - like his page on Spaced Repetition, DNB, etc.

However, like Paul Graham, I write to figure things out. I want to come up with Insights. This is the major selling point of Essays. You start with chaos, and a couple thousand words and few hours later, you end up with order. As of now, I value generating Insights over verifying them rigorously.

(Counter: What now, after the Technique Trap insight?)

My essays are firstly for coming up with ideas. Testing and verifying an idea is easier than coming up with it.

Also, Long Content is best for when you will update the same page over weeks and months and years. Empirically speaking, I haven’t felt like updating any of the ideas I have written one month ago.

(Counter: Maybe that’s because they’re mostly crap. No offense.)

The maximum time for which some idea stays alive in my head is around a week, it seems. I got some more ideas about the Bayes Theorem essay while I was doing other stuff and so I went and updated it. But, after a week, it’s pretty much done. I have moved on. A week gives me plenty of time to see the essay from a distance and edit it objectively. Cool. I hear Paul Graham takes about a week to refine and update and polish an essay.

(Counter: Rather, the maximum time for which I have high motivation for an idea is around a week. Doesn’t mean I have reached the core of the matter. That may take months as my understanding evolves.)

Do one thing and do it well

I do want to update my ideas over the long term. Just not in the same essay. I will write a fresh essay that talks about the new evidence and my updated idea.

Plus, I want to make each of my essays explore one idea and explore it well. As opposed to covering a whole bunch of ideas in one go. Why? So that I can link to that essay wherever I need to use that idea. It will be easy to look up an idea (just read its essay) and easy to look up where an idea is used (see which essays links to this essay).

Plus, if the idea is wrong, that whole essay and everything depending on it gets disqualified easily. Whereas if I mix up many ideas in one essay, it will be tough to update my beliefs.

Most importantly, it will be easier to hold in my head. One essay, one idea. I can use the title of the essay to keep it in my head.

You can use the Long Content model for each specific idea, instead of for a whole topic. Once you have generated and identified an idea, you can test and verify it over time. If you come up with some evidence, just add it to that idea’s essay page. And if you come up with something major against it, you can write a new essay talking about your updated idea.

How to organize my essays then?

What strategy will satisfy the above requirements well?

Ok. We need a strategy that satisfies the above requirements.

First of all, I think you really need a static site for writing timeless essays. In a blog format, old essays are all too easily forgotten. Plus, there is little incentive to update old posts. They are meant to be read on that particular day or that particular week. Blog posts are the exact opposites of timeless essays.5

What about the writing strategy?

Long Content?

The Long Content idea of building up an essay over days and weeks works well for writing quick-and-dirty essays about anything and everything in my head. It frees up my mind for better things.

However, I do not get a feeling of Closure very well in this model. The essay lingers in my mind. And it’s easy to excuse myself from publishing each day, by saying that I’m still working on some essay - it’s Long Content, after all. Yes, ideally, I should publish whatever I have at the end of the night, but it doesn’t work in practice.

(Counter: So, maybe clean up whatever you’ve written so far and release it. Iterate later.)

The worst part is that I fail to hold myself to a high standard in this model. I can get away with writing “ok” essays. I don’t do the hard work necessary to finish the last 10% of the work, where you get the maximum benefits. I lose out on the skill gains and the joy of doing challenging work. It is very easy to end up with a lot of half-done essays this way.

(Counter: I think the problem occurs when you mark an essay as “draft” and thus excuse all kinds of sloppy behaviour - slapdash sentences, runoff paragraphs, etc. Instead, hold even your in-progress essays to the high standard of having a complete logical structure. Make sure it is a complete albeit imperfect “product” at every point.)

If an essay isn’t complete, I don’t feel like refactoring it - aka cutting out useless bits, adding more supporting ideas, making the words sharper, and re-arranging things. Most importantly, I fail to discover all the implications of the core idea. You need to put things a certain way for further ideas to strike you.

So, Long Content for writing idea-generating essays as a whole doesn’t seem to be working out.

The PG Way

Write mainly one essay at a time and refine it several times over a week.

You get timeless essays here, no problem.

The biggest advantage is that you get tremendous Closure in this model. You get a feeling of accomplishment, you feel motivated to do more, you are forced to publish complete essays everyday, and you create compact, self-contained essays representing an idea.

This is a proven method for having epic insights and you can do it in high volume. You can easily list out the new essays on your Archives page.

And you can hold yourself to a high standard. You can challenge yourself to take an idea through to the end and discover all its implications. It’s hard to get away with a half-assed job here. You know when you’re cheating.

The only question, though, is whether you can do lots of quick-and-dirty essays. That is a big strength of the Long Content model. Using the PG / normal essay model, I find that I end up with quite a few drafts - some of them essays I couldn’t finish, some of them parts of essays that I broke up during refactoring. I generate more ideas than essays, so naturally my backlog keeps growing. Not cool.

I want to clear my head and think about new ideas instead of trying to keep up with my To-Do List.

But how do you do that when you have to spend quite some time on each essay and complete it before moving on to the next?

One option is to give up on writing down all the ideas in your mind. Maybe just write on the topics that are most important. Choose your battles wisely and fight with everything you have.

I have to think more about this.

Notes

  1. When you implement ideas, you get even better ones.

    From Paul Graham’s You weren’t meant to have a Boss:

    An obstacle downstream propagates upstream. If you’re not allowed to implement new ideas, you stop having them. And vice versa: when you can do whatever you want, you have more ideas about what to do. So working for yourself makes your brain more powerful in the same way a low-restriction exhaust system makes an engine more powerful.

    ↩

  2. Having to wait to implement ideas probably inhibits new ideas.

    From Paul Graham’s The Other Road Ahead:

    If I’d had to wait a year for the next release, I would have shelved most of these ideas, for a while at least. The thing about ideas, though, is that they lead to more ideas. Have you ever noticed that when you sit down to write something, half the ideas that end up in it are ones you thought of while writing it? The same thing happens with software. Working to implement one idea gives you more ideas. So shelving an idea costs you not only that delay in implementing it, but also all the ideas that implementing it would have led to. In fact, shelving an idea probably even inhibits new ideas: as you start to think of some new feature, you catch sight of the shelf and think “but I already have a lot of new things I want to do for the next release.”

    … Plans are just another word for ideas on the shelf. When we thought of good ideas, we implemented them.

    ↩

  3. Not releasing early is about putting off the hard work.

    From PG’s The Hardest Lessons for Startups to Learn

    Perhaps the most important reason to release early, though, is that it makes you work harder. When you’re working on something that isn’t released, problems are intriguing. In something that’s out there, problems are alarming. There is a lot more urgency once you release. And I think that’s precisely why people put it off. They know they’ll have to work a lot harder once they do.

    ↩

  4. Simplify information to the point where you can say it in two seconds.

    From “Your Brain at Work” by David Rock (TODO: exact chapter).↩

  5. Recommendations for running a timeless site: Long Site↩

Created: November 14, 2014
Last modified: September 28, 2019
Status: finished
Tags: writing

comments powered by Disqus