writing

In honor of my dropping by Inkhaven at Lighthaven in Berkeley this week, I figured it was time for another writing roundup. You can find #1 here, from March 2025.

I’ll be there from the 17th (the day I am publishing this) until the morning of Saturday the 22nd. I am happy to meet people, including for things not directly about writing.

1. Table of Contents.

2. How I Use AI For Writing These Days.

3. Influencing Influence.

4. Size Matters.

5. Time To Write A Shorter One.

6. A Useful Tool.

7. A Maligned Tool.

8. Neglected Topics.

9. The Humanities Don’t Seem Relevant To Writing About Future Humanity?

10. Writing Every Day.

11. Writing As Deep Work.

12. Most Of Your Audience Is Secondhand.

13. That’s Funny.

14. Fiction Writing Advice.

15. Just Say The Thing.

16. Cracking the Paywall.

How have I been using AI in my writing?

Directly? With the writing itself? Remarkably little. Almost none.

I am aware that this is not optimal. But at current capability levels, with the prompts and tools I know about, in the context of my writing, AI has consistently proven to have terrible taste and to make awful suggestions, and also to be rather confident about them. This has proven sufficiently annoying that I haven’t found it worth checking with the AIs.

I also worry about AI influence pushing me towards generic slop, pushing me to sounding more like the AIs, and rounding off the edges of things, since every AI I’ve tried this with keeps trying to do all that.

I am sure it does not help that my writing style is very unusual, and basically not in the training data aside from things written by actual me, as far as I can tell.

Sometimes I will quote LLM responses in my writing, always clearly labeled, when it seems useful to point to this kind of ‘social proof’ or sanity check.

The other exception is that if you ask the AI to look for outright errors, especially things like spelling and grammar, it won’t catch everything, but when it does catch something it is usually right. When you ask it to spot errors of fact, it’s not as reliable, but it’s good enough to check the list. I should be making a point of always doing that.

I did the ‘check for errors and other considerations’ thing on this piece in particular with both Sonnet and 5.1-Thinking. This did improve the post but it’s not obvious it improved it enough to be worth the time.

I will also sometimes ask it about a particular line or argument I’m considering, to see if it buys it, but only when what I care about is a typical reaction.

If I was devoting more time to refining and editing, and cared more about marginal improvements there, that would open up more use cases, but I don’t think that’s the right use of time for me on current margins versus training on more data or doing more chain of thought.

Indirectly? I use it a lot more there, and again I could be doing more.

There are some specific things:

1. I have a vibe coded Chrome extension that saves me a bunch of trouble, that could be improved a lot with more work. It does things like generate the Table of Contents, crosspost to WordPress, auto-populate many links and quickly edit quotes to fix people’s indifference to things like capitalization.

2. I have a GPT called Zvi Archivist that I use to search through my past writing, to check if and when I’ve already covered something and what I’ve said about it.

3. I have a transcriber for converting images to text because all the websites I know about that offer to do this for you are basically broken due to gating. This works.

Then there’s things that are the same as what everyone does all the time. I do a lot of fact checking, sanity checking, Fermi estimation, tracking down information or sources, asking for explanations, questioning papers for the things I care about. Using the AI assistant in its classic sense. All of that is a big help and I notice my activation requirement to do this is higher than it should be.

I want this to be true so I’m worried I can’t be objective, but it seems true to me?

Janus: i think that it’s almost always a bad idea to attempt to grow as an influencer on purpose.

you can believe that it would be good if you were to grow, and still you shouldn’t optimize for it.

the only way it goes well is if it happens while you optimize for other things.

More precisely than “you shouldn’t on purpose” what I’m saying is you shouldn’t be spending significant units of optimization on this goal and performing actions you wouldn’t otherwise for this purpose

I am confident that if you optimize primarily for influence, that’s full audience capture, slopification and so on, and you’ve de facto sold your soul. You can in theory turn around and then use that influence to accomplish something worthwhile, but statistically speaking you won’t do that.

Janus: Name a single account that explicitly optimizes for being a bigger influencer / “tries to grow” (instead of just happening as a side effect) and that does more good than harm to the ecosystem and generally has good vibes and interesting content

You probably can’t!

actually, https://x.com/AISafetyMemes is a contender

but i know they’re VERY controversial and I do think they’re playing with fire

i do consider them net positive but this is mostly bc they sometimes have very good taste and maybe cancel out the collateral damage

but WOULD NOT RECOMMEND almost anyone trying this, lol

AISafetyMemes is definitely an example of flying dangerously close to the sun on this, but keeping enough focus and having enough taste to maybe be getting away with it. It’s unclear that the net sign of impact there is positive, there are some very good posts but also some reasons to worry.

No one reads the blog posts, they’re too long, so might as well make them longer?

Visakan Veerasamy: An idea I’ve been toying with and discussed with a couple of friends is the idea that blog posts could and probably should get much longer now that fewer people are reading them.

One of the difficult things about writing a good essay is figuring out what to leave out so it is more manageable for readers.

But on a blog where there is no expectation that anybody reads it, you do not have to leave anything out.

My guess is this is going to end up being a barbell situation like so many other things. If you cut it down, you want to cut it down as much as possible. If you’re going long, then on the margin you’re better off throwing everything in.

I highlight this exactly because it seems backwards to me. I notice that my experience is very much the opposite – when I want to write a good short piece it is MUCH more work per token, and often more total work.

Timothy Lee: I think a big reason that writing a book is such a miserable experience is that the time to write a good piece is more-than-linear in the number of words. A good 2,000-word piece is a lot more than 4x the work of a good 500-word piece.

I assume this continues for longer pieces and a good 100,000 book is a lot more than 50x the work of a good 2,000-word article. Most authors deal with this by cutting corners and turning in books that aren’t very good. And then there’s Robert Caro.

Josh You: I think by “good 2000 word piece” Tim means “a 2000 word piece that has been edited down from a much longer first draft”

Even then. Yes, a tight longer piece requires more structure and planning, but the times I write those 500-800 word pieces it takes forever, because you really do struggle over every word as you try to pack everything into the tiniest possible space.

Writing a 100,000 word book at the precision level of an 800 word thinkpiece would take forever, but also I presume it almost never happens. If it does, that better be your masterpiece or I don’t see why you’d do it.

Dwarkesh Patel is using the Smart Composer Plugin for Obsidian, which he says is basically Cursor for writing, and loves it. Sounds great conditional on using Obsidian, but it is not being actively maintained.

Eric Raymond joins ‘the em-dash debate’ on the side of the em-dash.

Eric Raymond (yes that one): My wacky theory about the em-dash debate:

Pro writers use em-dashes a lot because many of them, possibly without consciously realizing it, have become elocutionary punctuationists.

That is, they’ve fallen into the habit of using punctuation not as grammatical phrase structure markers but as indicators of pauses of varying length in the flow of speech.

The most visible difference you see in people who write in this style that their usage of commas becomes somewhat more fluid — that’s the marker for the shortest pause. But they also reach for less commonly used punctuation marks as indicators of longer pauses of varying length.

Em dash is about the second or third longest pause, only an ellipsis or end-of-sentence period being clearly longer.

Historical note: punctuation marks originally evolved as pause or breathing markers in manuscripts to aid recitation. In the 19th century, after silent reading had become normal, they were reinterpreted by grammarians as phrase structure markers and usage rules became much more rigid.

Really capable writers have been quietly rediscovering elocutionary punctuation ever since.

RETVRN!

I too have been increasingly using punctuation, especially commas, to indicate pauses. I still don’t use em dashes, partly because I almost never want that exact length and style of a pause for whatever reason, and also because my instinct is that you’re trying to do both ‘be technically correct’ and also ‘evoke what you want’ and my brain thinks of the em-dash as technically incorrect.

That’s all true and I never used em-dashes before but who are we kidding, the best reason not to use em-dashes is that people will think you’re using AI. I don’t love that dynamic either, but do you actually want to die on that hill?

Tyler Cowen lists some reasons why he does not cover various topics much. The list resonates with me quite a bit.

1. I feel that writing about the topic will make me stupider.

2. I believe that you reading more about the topic will make you stupider.

3. I believe that performative outrage usually brings low or negative returns. Matt Yglesias has had some good writing on this lately.

4. I don’t have anything to add on the topic. Abortion and the Middle East would be two examples here.

5. Sometimes I have good inside information on a topic, but I cannot reveal it, not even without attribution. And I don’t want to write something stupider than my best understanding of the topic.

6. I just don’t feel like it.

7. On a few topics I feel it is Alex’s province.

I don’t have an Alex, instead I have decided on some forms of triage that are simply ‘I do not have the time to look into this and I will let it be someone else’s department.’

Otherwise yes, all of these are highly relevant.

Insider information is tough, and I am very careful about not revealing things I am not supposed to reveal, but this rarely outright stops me. If nothing else, you can usually get net smarter via negativa, where you silently avoid saying false things, including by using careful qualifiers on statements.

One big thing perhaps missing from Tyler’s list is that I avoid certain topics where my statements would potentially interfere with my ability to productively discuss other topics. If you are going to make enemies, or give people reasons to attack you or dismiss you, only do that on purpose. One could also file this under making you and others stupider. Similarly, there are things that I need to not think about – I try to avoid thinking about trading for this reason.

A minor thing is that I’d love to be able to talk more about gaming, and other topics dear to my heart, but that consistently drive people away permanently when I do that. So it’s just not worth it. If the extra posts simply had no impact, I’d totally do it, but as is I’d be better off writing the post and then not hitting publish. Sad. Whereas Tyler has made it very clear he’s going to post things most readers don’t care about, when he feels like doing so, and that’s part of the price of admission.

If you want to write or think about the future, maybe don’t study the humanities?

Startup Archive: Palmer Luckey explains why science fiction is a great place to look for ideas

“One of the things that I’ve realized in my career is that nothing I ever come up with will be new. I’ve literally never come up with an idea that a science fiction author has not come up with before.”

Dr. Julie Gurner: Funny how valuable those English majors and writers truly are, given how much liberal arts has been put down. Why philosophy, creativity and hard tech skills make such fantastic bedfellows. Span of vision wins.

Orthonormalist: Heinlein was an aeronautical engineer.

Asimov was a biochemistry professor.

Arthur Clarke was a radio operator who got a physics degree.

Ray Bradbury never went to college (but did go straight to being a writer)

I quote this because ‘study the humanities’ is a natural thing to say to someone looking to write or think about the future, and yet I agree that when I look at the list of people whose thinking about the future has influenced me, I notice essentially none of them have studied the humanities.

Alan Jacobs has a very different writing pattern. Rather than write every day, he waits until the words are ready, so he’ll work every day but often that means outlines or index card reordering or just sitting in his chair and thinking, even for weeks at a time. This is alien to me. If I need to figure out what to write, I start writing, see what it looks like, maybe delete it and try again, maybe procrastinate by working on a different thing.

Neal Stephenson explains that for him writing is Deep Work, requiring long blocks of reliably uninterrupted time bunched together, writing novels is the best thing he does, and that’s why he doesn’t go to conferences or answer your email. Fair enough.

I’ve found ways to not be like that. I deal with context shifts and interruptions all the time and it is fine, indeed when dealing with difficult tasks I almost require them. That’s a lot of how I can be so productive. But the one time I wrote something plausibly like a book, the Immoral Mazes sequence, I did spend a week alone in my apartment doing nothing else. And I haven’t figured out how to write a novel, or almost any fiction at all.

Also, it’s rather sad if it is true that Neal Stephenson only gets a middle class life out of writing so many fantastic and popular books, and can’t afford anyone to answer his email. That makes writing seem like an even rougher business than I expected. Although soon AI can perhaps do it for him?

Patrick McKenzie highlights an insight from Alex Danco, which is that most of the effective audience of any successful post is not people who read the post, but people who are told about the post by someone who did read it. Patrick notes this likely also applies to formal writing, I’d note it seems to definitely apply to most books.

Relatedly, I have in the past postulated a virtual four-level model of flow of ideas, where each level can understand the level above it, and then rephrase and present it to the level below.

So if you are Level 1, either in general or in an area, you can formulate fully new ideas. If you are Level 2, you can understand what the Level 1s say, look for consensus or combine what they say, riff on it and then communicate that to those who are up to Level 3, who can then fully communicate to the public who end up typically around at Level 4.

Then the public will communicate a simplified and garbled version to each other.

You can be Level 1 and then try to ‘put on your Level 2 or 3 hat’ to write a dumber, simpler version to a broader audience, but it is very hard to simultaneously do that and also communicate the actual concepts to other Level 1s.

These all then interact, but if you go viral with anything longer than a Tweet, you inevitably are going to primarily end up with a message primarily communicated via (in context) Level 3 and Level 4 people communicating to other Level 3-4 people.

At that point, and any time you go truly viral or your communication is ‘successful,’ you run into the You Get About Five Words problem.

My response to this means that at this point I essentially never go all that directly viral. I have a very narrow range of views, where even the top posts never do 100% better than typical posts, and the least popular posts – which are when I talk about AI alignment or policy on their own – will do at worst 30% less than typical.

The way the ideas go viral is someone quotes, runs with or repackages them. A lot of the impact comes from the right statement reaching the right person.

I presume that would work differently if I was working with mediums that work on virality, such as YouTube or TikTok, but my content seems like a poor fit for them, and when I do somewhat ‘go viral’ in such places it is rarely content I care about spreading. Perhaps I am making a mistake by not branching out. But on Twitter I still almost never go viral, as it seems my speciality is small TAM (total available market) Tweets.

Never have a character try to be funny, the character themselves should have no idea.

I think this is directionally correct but goes too far, for the same reasons that you, in your real life, will often try to be funny, and sometimes it will work. The trick is they have to be trying to be funny in a way that makes sense for the character, in context, for those around them, not trying to be funny to the viewer.

I notice that in general I almost never ‘try to be funny,’ not exactly. I simply say things because they would be funny, and to say things in the funniest way possible, because why not. A lot of my favorite people seem to act similarly.

Lydia Davis offers her top ten recommendations for good (fiction?) writing: Keep notes, including sentences out of context, work from your own interest, be mostly self-taught, read and revise the notes constantly, grow stories or develop poems out of those notes, learn techniques from great works and read the best writers across time.

Orson Scott Card explains that you don’t exhaust the reader by having too much tension in your book, you exhaust them by having long stretches without tension. The tension keeps us reading.

Dwarkesh Patel: Unreasonably effective writing advice:

“What are you trying to say here?

Okay, just write that.”

I’ve (separately) started doing this more.

I try to make sure that it’s very easy to find the central point, the thing I’m most trying to say, and hard to miss it.

Patrick McKenzie: Cosigned, and surprisingly effective with good writers in addition to ones who more obviously need the prompting.

Writing an artifact attaches you to the structure of it while simultaneously subsuming you in the topic. The second is really good for good work; the first, less so.

One thing that I tried, with very limited success, to get people to do is to be less attached to words on a page. Writing an essay? Write two very different takes on it; different diction, different voice, maybe even different argument. Then pick the one which speaks to you.

Edit *thatrather than trying to line edit the loser towards greatness.

There is something which people learn, partially from school and partially from work experience, which causes them to write as if they were charged for every word which goes down on the page.

Words are free! They belong in a vast mindscape! You can claw more from the aether!

I think people *mightoperationalize better habits after LLMs train them that throwing away a paragraph is basically costless.

Jason Cohen: Yeah this works all the time.

Also when getting someone to explain their product, company, customer, why to work for them, etc..

So funny how it jogs them out of their own way!

BasedBigTech: An excellent Group PM reviewed my doc with me. He said “what does this mean?” and I told him.

“Then why didn’t you write that?”

Kevin Kelly: At Whole Earth Review people would send us book reviews with a cover letter explaining why we should run their book review. We’d usually toss the review and print their much shorter cover letter as the review which was much clearer and succinct.

Daniel Eth: It’s crazy how well just straight up asking people that gets them to say the thing they should write down

Why does it work?

The answer is that writing is doing multiple tasks.

Only one of them is ‘tell you what all this means.’

You have to do some combination of things such as justify that, explain it, motivate it, provide details, teach your methods and reasoning, perform reporting, be entertaining and so on.

Also, how did you know what you meant to say until you wrote the damn thing?

You still usually should find a way to loudly say what it all means, somewhere in there.

But this creates the opportunity for the hack.

If I hand you a ten-page paper, and you ask ‘what are you trying to say?’ then I have entered into evidence that I have Done the Work and Written the Report.

Now I can skip the justifications, details and context, and Say The Thing.

The point of a reference post is sometimes to give people the opportunity to learn.

The point of a reference post can also be to exist and then not be clicked on. It varies.

This is closely related to the phenomenon where often a movie or show will have a scene that logically and structurally has to exist, but which you wish you didn’t have to actually watch. In theory you could hold up a card that said ‘Scene in which Alice goes to the bank, acts nervous and get the money’ or whatever.

Probably they should do a graceful version of something like that more often, or even interactive versions where you can easily expand or condense various scenes. There’s something there.

Similarly, with blog posts (or books) there are passages that are written or quoted knowing many or most people will skip them, but that have to be there.

Aella teaches us how to make readers pay up to get behind a Paywall. Explain why you are the One Who Knows some valuable thing, whereas others including your dear reader are bad at this and need your help. Then actually provide value both outside and inside of the paywall, ideally because the early free steps are useful even without the payoff you’re selling.

I am thankful that I can write without worrying about maximizing such things, while I also recognize that I’m giving up a lot of audience share not optimizing for doing similar things on the non-paywall side.

This isn’t primarily about how I write. It’s about how other people write, and what advice they give on how to write, and how I react to and relate to that advice.

I’ve been collecting those notes for a while. I figured I would share.

At some point in the future, I’ll talk more about my own process – my guess is that what I do very much wouldn’t work for most people, but would be excellent for some.

1. How Marc Andreessen Writes.

2. How Sarah Constantin Writes.

3. How Paul Graham Writes.

4. How Patrick McKenzie Writes.

5. How Tim Urban Writes.

6. How Visakan Veerasamy Writes.

7. How Matt Yglesias Writes.

8. How JRR Tolkien Wrote.

9. How Roon Wants Us to Write.

10. When To Write the Headline.

11. Do Not Write Self-Deprecating Descriptions of Your Posts.

12. Do Not Write a Book.

13. Write Like No One Else is Reading.

14. Letting the AI Write For You.

15. Being Matt Levine.

16. The Case for Italics.

17. Getting Paid.

18. Having Impact.

Marc Andreessen starts with an outline, written as quickly as possible, often using bullet points.

David Perell: When Marc Andreessen is ready to write something, he makes an outline as fast as possible.

Bullet points are fine. His goal is to splatter the page with ideas while his mind is buzzing. Only later does he think about organizing what he’s written.

He says: “I’m trying to get all the points out and I don’t want to slow down the process by turning them all into prose. It’s not a detailed outline like something a novelist would have. It’s basically bullet points.”

Marc is saying that first you write out your points and conclusion, then you fill in the details. He wants to figure it all out while his mind is buzzing, then justify it later.

Whereas I learn what I think as I write out my ideas in detail. I would say that if you are only jotting down bullet points, you do not yet know what you think.

Where we both agree is that of course you should write notes to remember key new ideas, and also that the organizing what goes where can be done later.

I do not think it is a coincidence that this is the opposite of my procedure. Yes, I have some idea of what I’m setting out to write, but it takes form as I write it, and as I write I understand.

If you’re starting with a conclusion, then writing an outline, and writing them quickly, that says you are looking to communicate what you already know, rather than seeking to yourself learn via the process.

A classic rationalist warning is to not write the bottom line first.

Sarah Constantin offers an FAQ on how she writes. Some overlap, also a lot of big differences. I especially endorse doing lots of micro-edits and moving things around and seeing how they develop as they go. I dismiss the whole ‘make an outline’ thing they teach you in school as training wheels at best and Obvious Nonsense at worst.

I also strongly agree with her arguments that you need to get the vibe right. I would extend this principle to needing to be aware of all four simulacra levels at once at all times. Say true and helpful things, keeping in mind what people might do with that information, what your statements say about which ‘teams’ you are on in various ways, and notice the vibes and associations being laid down and how you are sculpting and walking the paths through cognitive space for yourself and others to navigate. Mostly you want to play defensively on level two (make sure you don’t give people the wrong idea), and especially on level three (don’t accidentally make people associate you with the wrong faction, or ideally any faction), and have a ‘don’t be evil’ style rule for level four (vibe well on all levels, and avoid unforced errors, but vibe justly and don’t take cheap shots), with the core focus always at level one.

I think this is directionally right, I definitely won’t leave a wrong idea in writing:

Paul Graham: Sometimes when writing an essay I’ll leave a clumsy sentence to fix later. But I never leave an idea I notice is wrong. Partly because it could damage the essay, and partly because you don’t need to: noticing an idea is wrong starts you toward fixing it.

However I also won’t leave a clumsy sentence that I wasn’t comfortable being in the final version. I will often go back and edit what I’ve written, hopefully improving it, but if I wasn’t willing to hit post with what I have now then I wouldn’t leave it there.

In the cases where this is not true, I’m going explicitly leave a note, in [brackets] and usually including a [tktk], saying very clearly that there is a showstopper bug here.

Here’s another interesting contrast in our styles.

Paul Graham: One surprising thing about office hours with startups is that they scramble your brain. It’s the context switching. You dive deep into one problem, then deep into another completely different one, then another. At the end you can’t remember what the first startup was even doing.

This is why I write in the mornings and do office hours in the afternoon. Writing essays is harder. I can’t do it with a scrambled brain.

It’s fun up to about 5 or 6 startups. 8 is possible. 12 you’d be a zombie.

I feel this way at conferences. You’re constantly context switching, a lot of it isn’t retained, but you go with the flow, try to retain the stuff that matters most and take a few notes, and hope others get a lot out of it.

The worst of that was at EA Global: Boston, where you are by default taking a constant stream of 25 minute 1-on-1s. By the end of the day it was mostly a blur.

When I write, however, mostly it’s the opposite experience to Graham’s writing – it’s constant context switching from one problem to the next. Even while doing that, I’m doing extra context switching for breaks.

A lot of that is presumably different types of writing. Graham is trying to write essays that are tighter, more abstract, more structured, trying to make a point. I’m trying to learn and explore and process and find out.

Which is why I basically can indeed do it with a scrambled brain, and indeed have optimized for that ability – to be able to process writing subtasks without having to load in lots of state.

Patrick McKenzie on writing fast and slow, formal and informal, and the invocation of deep magick. On one topic he brings up: My experience on ‘sounding natural’ in writing is that you can either sound natural by writing in quick natural form, or you can put in crazy amounts of work to make it happen, and anything in between won’t work. Also I try to be careful to intentionally not invoke the deep magick in most situations. One only wants to be a Dangerous Professional when the situation requires it, and you need to take on a faceless enemy in Easy Mode.

Patrick McKenzie also notes that skilled writers have a ton of control over exactly how controversial their statements will effectively be. I can confirm this. Also I can confirm that mistakes are often made, which is a Skill Issue.

Tim Urban says writing remains very hard.

Tim Urban: No matter how much I write, writing remains hard. Those magical moments when I’m in a real flow, it seems easy, but most of the time, I spend half a day writing and rewriting the same three paragraphs trying to figure out the puzzle of making them not suck.

Being in a writing flow is like when I’m golfing and hit three good shots in a row and think “k finally figured this out and I’m good now.” Unfortunately the writing muse and the golf fairy both like to vanish without a trace and leave me helpless in a pit of my own incompetence.

Dustin Burnham: Periods of brilliance would escape Douglas Adams for so long that he had to be locked in a hotel room by his editors to finish The Hitchhiker’s Guide to the Galaxy.

I definitely have a lot of moments when I don’t feel productive, usually because my brain isn’t focused or on. I try to have a stack of other productive things I can do despite being unproductive while that passes.

But over time, yes, I’ve found the writing itself does get easy for me? Often figuring out what I think, or what I want to write about, is hard, but the writing itself comes relatively easily.

Yes, you can then go over it ten times and edit to an inch of its life if you want, but the whole ‘rewriting the same three paragraphs’ thing is very rare. I think the only times I did it this year I was pitching to The New York Times.

What’s the best target when writing?

Visakan Veerasamy: Charts out for ribbonfarm.

I do endorse the core thing this is trying to suggest: To explore more and worry about presentation and details less, on most margins. And to know that in a real sense, if you have truly compelling fuckery, you have wiggled your big toe. Hard part is over.

I do not think the core claim itself is correct. Or perhaps we mean different things by resonant and coherent? By coherent, in my lexicon, he means more like ‘well-executed’ or ‘polished’ or something. By resonant, in my lexicon, he means more like ‘on to something central, true and important.’ Whereas to me resonant is a vibe, fully compatible with bullshit, ornate or otherwise.

Matt Yglesias reflects on four years of Slow Boring. He notes that it pays to be weird, to focus where you have comparative advantage rather than following the news of the week and fighting for a small piece of the biggest pies. He also notes the danger of repeating yourself, which I worry about as well.

Thread from 2020 on Tolkien’s path to writing Lord of the Rings. I’ve never done anything remotely like this, which might be some of why I haven’t done fiction.

Roon calls for the end of all this boring plain language, and I am here for it.

Roon: I love the guy, but I want the post-Goldwater era of utilitarian philosophical writing to be over. Bring back big words and epic prose, and sentences that make sense only at an angle.

Eliezer Yudkowsky: I expect Claude to do a good job of faking your favorite continental styles if you ask, since it requires little logical thinking, only vibes. You can produce and consume it privately in peace, avoiding its negative externalities, and leave the rest of us to our utility.

Roon: Eliezer, you are a good writer who often speaks in parables and communicates through fiction and isn’t afraid of interesting uses of language. You’ve certainly never shied away from verbosity, and that’s exactly what I’m talking about.

Perhaps some day I will learn how to write fiction. My experiences with AI reinforce to me that I really, really don’t know how to do that.

I usually write the headline last. Others disagree.

Luke Kawa: Joe Weisenthal always used to say don’t write a post until you know the headline first. More and more on short posts I find myself thinking “don’t write the post until you know the meme you can post with it first.”

As I said on Twitter, everyone as LessOnline instead gave the advice to not let things be barriers to writing. If you want to be a writer, write more, then edit or toss it out later, but you have to write.

Also, as others pointed out, if you start with the headline every time you are building habits of going for engagement if not clickbait, rather than following curiosity.

Other times, yes, you know exactly what the headline is before you start, because if you know you know.

I confirm this is sometimes true (but not always):

Patrick McKenzie: Memo to self and CCing other writers on an FYI basis:

If when announcing a piece you make a self-deprecating comment about it, many people who cite you will give a qualified recommendation of the piece, trying to excuse the flaw that you were joking about.

You think I would understand that ~20 years of writing publicly, but sometimes I cannot help myself from making the self-deprecating comment, and now half of the citations of my best work this year feel they need to disclaim that it is 24k words.

You can safely make the self-deprecating comments within the post itself. That’s fine.

Don’t write a book. If you do, chances are you’d sell dozens of copies, and earn at most similar quantities of dollars. The odds are very much doubleplusungood. Do you want to go on podcasts this much?

If you must write one anyway, how to sell it? The advice here is that books mostly get sold through recommendations. To get those, Eric Jorgenson’s model is you need three things:

1. Finishable. If they don’t finish it, they won’t recommend it. So tighten it up.

2. Unique OR Excellent. Be the best like no one ever was, or be like no one ever was.

3. Memorable. Have hooks, for when people ask for a book about or for some X.

If you are thinking about writing a book, remember that no one would buy it.

Michael Dempsey and Ava endorse the principle of writing things down on the internet even if you do not expect anyone to read them.

Michael Dempsey: I loved this thread from Ava.

My entire career is owed to my willingness to write on the Internet.

And that willingness pushed me to write more in my personal life to loved ones.

As long as you recognize that most people will not care, your posts probably will not go viral, but at some point, one person might read something you write and reach out (or will value you including a blue link to your thoughts from many days, weeks, months, or years ago), it’s close to zero downside and all upside.

Ava: I’m starting to believe that “write on the Internet, even if no one reads it” is underrated life advice. It does not benefit other people necessarily; it benefits you because the people who do find or like your writing and then reach out are so much more likely to be compatible with you.

It’s also a muscle. I used to have so much anxiety posting anything online, and now I’m just like “lol, if you don’t like it, just click away.” People underestimate the sheer amount of content on the Internet; the chance of someone being angry at you for something is infinitely lower than no one caring.

I think it’s because everyone always sees outrage going viral, and you think “oh, that could be me,” and forget that most people causing outrage are trying very hard to be outrageous. By default, no one cares, or maybe five people care, and maybe some nice strangers like your stuff, and that’s a win.

Also, this really teaches you how to look for content you actually like on the Internet instead of passively receiving what is funneled to you. Some of my favorite Internet experiences have been reading a personal blog linked to someone’s website, never marketed, probably only their friends and family know about it, and it’s just the coolest peek into their mind.

I think the thing I’m trying to say here is “most people could benefit from writing online, whether you should market your writing aggressively is a completely different issue.” I wrote on Tumblr and Facebook and email for many years before Substack, and 20 people read it, and that was great.

I would not broadly recommend “trying to make a living off your writing online,” but that’s very different from “share some writing online.”

What is the number of readers that justifies writing something down? Often the correct answer is zero, even a definite zero. Even if it’s only about those who read it, twenty readers is actually a lot of value to put out there, and a lot of potential connection.

Paul Graham predicts that AI will cause the world to divide even more into writes and write-nots. Writing well and learning to write well are both hard, especially because it requires you to think well (and is how you think well), so once AI can do it for us without the need to hire someone or plagiarize, most people won’t learn (and one might add, thanks to AI doing the homework they won’t have to), and increasingly rely on AI to do it for them. Which in turn means those people won’t be thinking well, either, since you need to write to think well.

I think Graham is overstating the extent AI will free people from the pressure to write. Getting AI to write well in turn, and write what you actually want it to write, requires good writing and thinking, and involving AI in your writing OODA loop is often not cheap to do. Yes, more people will choose not to invest in the skill, but I don’t think this takes the pressure off as much as he expects, at least until AI gets a lot better.

There’s also the question of how much we should force people to write anyway, in order to make them think, or be able to think.

As Graham notes, getting rid of the middle ground could be quite bad:

Robin Hanson: But most jobs need real thinking. So either the LLMs will actually do that thinking for them, or workers will continue to write, in order to continue to think. I’d bet on the latter, for decades at least.

Perry Metzger: Why do we still teach kids mathematics, even though at this point, most of the grunt work is done better by computers, even for symbolic manipulation? Because if they’re going to be able to think, they need to practice thinking.

Most jobs don’t require real thinking. Proof: Most people can’t write.

One could argue that many jobs require ‘mid-level’ real thinking, the kind that might be lost, but I think mostly this is not the case. Most tasks and jobs don’t require real thinking at all, as we are talking about it here. Being able to do it? Still highly useful.

On the rare occasions the person can indeed do real thinking, it’s often highly valuable, but the jobs are designed knowing most people can’t and won’t do that.

Gwern asks, why are there so few Matt Levines? His conclusion is that Being Matt Levine requires both that a subject be amenable to a Matt Levine, which most aren’t, and also that there be a Matt Levine covering them, and Matt Levines are both born rather than made and highly rare.

In particular, a Matt Levine has to shout things into the void, over and over, repeating simple explanations time and again, and the subject has to involve many rapidly-resolved example problems to work through, with clear resolutions.

The place where I most epicly fail to be a Matt Levine in this model is my failure to properly address the beginner mindset and keep things simple. My choice to cater to a narrower, more advanced crowd, one that embraces more complexity, means I can’t go wide the way he can. That does seem right.

I could try to change this, but I mostly choose not to. I write too many words as it is.

The case for italics. I used to use italics a lot.

Char: “never italicise words to show emphasis! if you’re writing well your reader will know. you don’t need them!” me: oh 𝘳𝘦𝘢𝘭𝘭𝘺? listen up buddy, you will have to pry my emotional support italics from my 𝘤𝘰𝘭𝘥, 𝘥𝘦𝘢𝘥, 𝘧𝘪𝘯𝘨𝘦𝘳𝘴, they are going 𝘯𝘰𝘸𝘩𝘦𝘳𝘦.

Richard White: I’m coming to the conclusion that about 99.9% of all writing “rules” can safely be ignored. As long as your consistent with your application of whatever you’re doing, it’ll be fine.

Kira: Italics are important for subtlety and I will fight anyone who says otherwise

It’s a great tool to have in your box. What I ultimately found was it is also a crutch that comes with a price. You almost never need italics, and the correct version without italics is easier on the reader.

When I look back on my old writing and see all the italics, I often cringe. Why did I feel the need to do that? Mostly I blame Eliezer Yudkowsky giving me felt permission to do it. About 75% of the time I notice that I can take out the italics and nothing would go wrong. It would be a little less obvious what I’m trying to emphasize, in some senses, but it’s fine. The other 25% of the time, I realize that the italics is load bearing, and if I remove it I will have to reword, so mostly I reword.

Scott Alexander does his third annual Subscribe Drive. His revenue has leveled off. He had 5,993 paid subscribers in 2023, 5,523 in 2024, and has 5,329 now in 2025. However his unpaid numbers keep going up, from to 78k to 99k to 126k.

I’ve been growing over time, but the ratios do get worse. I doubled my unpaid subscriber count in 2023, and then doubled it again in 2024. But my subscription revenue was only up about 50% in 2023, and only up another 25% in 2024. I of course very much appreciate paid subscriptions, but I am 100% fine, and it is not shocking that my offer of absolutely nothing extra doesn’t get that many takers.

Paywalls are terrible, but how else do you get paid?

Email sent to Rob Henderson: The hypocrisy of the new upper class he proclaims as he sends a paid only email chain…

Cartoons Hate Her: Sort of a different scenario but most people say they think it should be possible to make a living as a writer or artist and still shout “LAME!! PAYWALL!” whenever I attempt to *checks notesmake a living as a writer.

Rob Henderson: Agree with this. Regrettably I’ll be adding more paywalls going forward. But will continue to regularly offer steep discounts and free premium subscriptions.

I am continuously grateful that I can afford to not have a paywall, but others are not so fortunate. You have to pay the bills, even though it is sad that this greatly reduces reach and ability to discuss the resulting posts.

It’s great to be able to write purely to get the message out and not care about clicks. Unfortunately, you do still have to care a little about how people see the message, because it determines how often they and others see future messages. But I am very grateful that, while I face more pressure than Jeff, I face vastly less than most, and don’t have to care at all about traffic for traffic’s sake.

Ideally, we would have more writers who are supported by a patron system, in exchange for having at most a minimal paywall (e.g. I think many would still want a paywall on ability to comment to ensure higher quality or civility, or do what Scott Alexander does and paywall a weird 5% of posts, or do subscriber-questions-only AMAs or what not).

Scott Sumner cites claims that blogging is effective. I sure hope so!

Patrick McKenzie suggests responding to future AIs reading your writing by, among other things, ‘creating more spells’ and techniques that can thereby be associated with you, and then invoked by reference to your name. And to think about how your writing being used as training data causes things to be connected inside LLMs. He also suggests that having your writing outside paywalls can help.

In my case, I’m thinking about structure – the moves between different topics are designed to in various ways ‘make sense to humans’ but I worry about how they might confuse AIs and this could cause confusion in how they understand me and my concepts in particular, including as part of training runs. I already know this is an issue within context windows, AIs are typically very bad at handling these posts as context. One thing this is motivating is more clear breaks and shorter sections than I would otherwise use, and also shorter more thematically tied together posts.

Ben Hoffman does not see a place or method in today’s world for sharing what he sees as high-quality literate discourse, at least given his current methods, although he identifies a few people he could try to usefully engage with more. I consistently find his posts some of the most densely interesting things on the internet and often think a lot about them, even though I very often strongly disagree with what he is saying and also often struggle to even grok his positions, so I’m sad he doesn’t offer us more.

My solution to the problem of ‘no place to do discourse’ is that you can simply do it on Substack on your own, respond to whoever you want to respond to, speak to who you want to speak to and ignore who you want to ignore. I do also crosspost to LessWrong, but I don’t feel any obligation to engage if someone comments in a way that misconstrues what I said.