I believe that it is not that style helps the content to be more right, not in the way PG believes (like in the example about writing shorter sentences), it is that a richer style (so, not shorter, but neither baroque: a style with more possibilities) can reflect a less obvious way of thinking, that carries more signal.<p>I'll make an example that makes this concept crystal crisp, and that you will likely remember for the rest of your life (no kidding). In Italy there was a great writer called Giuseppe Pontiggia. He had to write an article for one of the main newspapers in Italy about the Nobel Prize in Literature, that with the surprise of many, was never assigned, year after year, to Borges. He wrote (sorry, translating from memory, I'm not an English speaker and I'm not going to use an LLM for this comment):<p>"Two are the prizes that each year the Swedish academy assigns: one is assigned to the winner of the prize, the other is not assigned to Borges".<p>This uncovers much more than just: even this year the prize was not assigned to Borges. And, honestly, I never saw this kind of style heights in PG writings (I appreciate the content most of the times, but having translated a few of his writings in Italian, I find the style of PG fragile: brings the point at home but never escapes simple constructs). You don't reach that kind of Pontiggia style with the process in the article here, but via a very different process that only the best writers are able to perform and access.
This is wrong in so many different ways it's like an art piece. Every part of it that tries to defend the central thesis is actually disproving it. It's kind of funny actually. Here's a dude that's been writing for 30 years, and not only is his writing bad, his ideas are crap. It has the feel of somebody who's completely convinced of his own ideas, despite the fact that they're based solely on his personal experience.<p>I have a simple proof that the thesis is wrong. Take a moron, and have him work on a farm for 30 years. Then have him write a book about running a farm. Now, he's going to sound like a moron, and will write very poorly. But most everything he writes will be right. Despite his bad writing, he can still communicate his observations of how and why simple things work. So it's not hard to be right while sounding wrong. You just have to be a moron.
I feel the essence here is -- iterative writing improves both the prose and the core point.<p>When you write well, you iterate.
When you iterate, you improve both the prose and the core point -- because you crystalize ideas further.<p>This makes improvements in these seemingly perpendicular directions counterintuitively correlated.<p>Ironically I found this specific PG essay uncharacteristically obtuse. This could have been much shorter.
The point about end-notes being a mechanism to ease the strain of fitting tree-like ideas into a linear essay is lovely. It brings to mind David Foster Wallace's writing, which is obsessively end-noted and if you listen to his speeches, you can see that he basically tortures himself in sanding down his ideas, much like PG says.<p>PG's ideas in here, to the extent that I agree with them (which is not fully), does break down for ideas. Example being: brilliant engineers who are incredibly capable at having ideas and executing against them but incredibly incapable of communicating said ideas. Their ideas are very true, evidenced by their ability to produce real results, but also oftentimes ugly when communicated.<p>A final counterpoint is JFK's eulogy, which sounds amazing, but, after the initial emotional appeal wore off, I realized doesn't really have a strong unified thread running through it, and is thus forgettable in terms of the truths it ostensibly delivers. <a href="https://www.youtube.com/watch?v=DOiDUbaBL9E" rel="nofollow">https://www.youtube.com/watch?v=DOiDUbaBL9E</a>. Compare to "This Is Water" by DFW, which doesn't have the same epic prose, but is maybe the most true-seeming speech I've ever heard <a href="https://www.youtube.com/watch?v=DCbGM4mqEVw" rel="nofollow">https://www.youtube.com/watch?v=DCbGM4mqEVw</a>. It could be that PG's ideas were never intended for spoken speeches, but whatever this is still an adjacent truth != beauty example.
Paul Graham, as a writer who writes sentences that sound bad and who promulgates ideas that are stupid, is an expert in neither good sounding sentences nor sound ideas, hence is unqualified to hold forth on the topic.
Dear Paul,
I'm sure it has been said elsewhere in the comments, but ironically I struggle to agree with this essay...which happens to be nicely and succinctly written.<p>I'm arguing that it's your own bias generated from the synthesis of your own idea that selects for sentences that effectively express the idea, and nothing to do with the writing itself.<p>The anecdote about the puddle who suddenly gains consciousness and remarks that the world is so perfectly formed around it, that it's proof of divine creation, seems to apply here.<p>The author generates an idea and is trying to articulate it. A well written sentence or paragraph that flows, pleases the author. This is because the idea they are trying to express is done in a satisfying way.<p>Thus the more pleasing the writing to the author, the more efficiently it articulates the original idea.
It's the author's bias, based on their own idea, that defines the level of 'pleasingness'.<p>Lastly, Paul, do you think the LLMs are any less satisfied with their confident and irrational hallucinations, than they are with their more well supported claims? Further, if you weren't aware that the output was ridiculous, would you be able to tell a accurate statement from a false one?<p>Thanks for the essays. Love them.
>You can't simultaneously optimize two unrelated things;<p>What kind of illogical nonsense is this? I found the speed my car produces the best MPG and simultaneously found the best volume for the stereo.<p>I suppose his overall conclusion works in one way, the article is both poorly written, and devoid of anything useful. Nobody would read it if it wasn't from pg.
> It's hard to be right without sounding right.<p>This doesn't seem true in the age of LLMs, which are notorious for being confidently incorrect.<p>In fact, this whole article seems out of touch with the realities of where AI is going. In my opinion, good writing is dead. Or rather, good writing is commoditized. Good ideas are still very much alive, but if you have an idea and bad prose, iterating with an LLM will have a better end state than rereading your paragraph 50 times.<p>That said, if you're only writing to internalize your own ideas (journaling) then this makes more sense.
In french we have this very old expression : "ce qui se conçoit bien s'énonce clairement". "what is well understood will be clearly phrased".<p>It's been consistently used by parents and teachers since the 17th century, so i guess there must be some truth to it.
> I think writing that sounds good is more likely to be right.<p>Paul's point would make sense if his case was about greater verisimilitude, which might sound like splitting hairs, but is an important phenomenon in philosophy. Many dictators have sounded good but their core messages were abhorrent.<p>In the same vein, there are thousands of fiction books, some more brilliantly written than others, but nothing in that spectrum makes any of their stories any more real or true.<p>> I know it's true from writing.<p>Well, some things just appear to be true. I admire Paul's writings and I believe his honesty in trying to get to the truth, but in this specific essay, it seems like what he's alluding to is the appearance of truth. Good writing makes core ideas look more true, but it can't objectively have a relation to truth itself, only with our description of said idea.
I think a strong sense of confidence (perhaps overconfidence) and inflated self-worth are likely closely related to increased likelihood of success. “Those who dare, win” and all that.<p>But I think most of the opinions and advice rich, successful people like to share is just a side-effect, not a productive output, of these traits.
I'm solidly in the camp that believes that if Graham wasn't rich no one would read this stuff or claim to admire it. He also should have run this through a spell checker.
> The clue to the answer is something I noticed 30 years ago when I was doing the layout for my first book. Sometimes when you're laying out text you have bad luck. For example, you get a section that runs one line longer than the page. I don't know what ordinary typesetters do in this situation, but what I did was rewrite the section to make it a line shorter. You'd expect such an arbitrary constraint to make the writing worse. But I found, to my surprise, that it never did. I always ended up with something I liked better.<p>This is a well-known phenomenon, and yes, "ordinary" writers and typesetters do this too. These visual loose ends are called Widows, Orphans and Runts [1]. Writing that is less visually ugly on the page will seem to read better.<p>> Because the writer is the first reader<p>This seems like a derivative of a zen-like koan from jazz musician Winton Marsalis' "Music is always for the listener, but the first listener is the player" [2]. Interesting that he immediately starts talking about music there too.<p>> But while we can't safely conclude that beautiful writing is true, it's usually safe to conclude the converse: something that seems clumsily written will usually have gotten the ideas wrong too.<p>I think I would have enjoyed this read more if it was clear at the top that by the time he'd finished writing it, he disagreed with his initial assertion ("I think writing that sounds good is more likely to be right.") Without that, the article kind of feels like bait, and reading it plus writing this comment feels like me taking it.<p>[1] <a href="https://en.wikipedia.org/wiki/Widows_and_orphans" rel="nofollow">https://en.wikipedia.org/wiki/Widows_and_orphans</a><p>[2] <a href="https://www.facebook.com/watch/?v=162573063275994" rel="nofollow">https://www.facebook.com/watch/?v=162573063275994</a>
I think I agree with the central point here. I think the key phrase is “internal consistency”. This is also very true of programming. It’s difficult to build good software without having a handle on the subject matter (and/or a domain expert to get feedback from).<p>But often writing is also a process of discovery. Maybe you are trying to write something that hasn’t been written about before. This is like building software without a spec. You can still write well and be irrelevant, just as one can build great software upon bad assumptions and fail to sell it. This doesn’t make it bad writing in its own right, but it also may not be very useful to anyone. In my opinion both software and prose should be produced for a purpose.<p>Thus, if there’s no meaning to it, writing, like software, falls pretty flat.
Interested to know how much Immanuel Kant pg has read. Kant's whole project was about grappling with how the mind structures experience—and how language mediates, rather than transparently transmits, thought. You can never fully get to the thing in itself.<p>I can see how re-editing can make the ideas more coherent within pg's frame of representation, but I'm struggling with the idea that it makes them any more true.
There was a line in the (not great) Ferrari biopic where Enzo is explaining his philosophy and says:<p>> In all life, when a thing works better, usually it is more beautiful to the eye.
There are two sentences in this essay that I couldn't understand. Can someone help me?<p>1. "An essay is a cleaned up train of thought, in the same way dialogue is cleaned up conversation"<p>I thought dialogue and conversation were the same thing. What is the difference between them besides one being a cleaned up version of the other?<p>2. "If for some bizarre reason the number of jobs in a country were fixed, then immigrants really would be taking our jobs."<p>What does this even mean? Is it an exemple or an analogy? It sounds like at this point in the text there should be an analogy, but this sentence sounds like an example. So, which one is it?<p>Also, did anybody else got confused too?
I’m a sucker for when the form serves as an example for the author’s idea.<p>> If it were, it wouldn't be good, because the rhythm of good writing has to match the ideas in it, and ideas have all kinds of different shapes. Sometimes they're simple and you just state them. But other times they're more subtle, and you need longer, more complicated sentences to tease out all the implications.<p>From William Zinsser’s On Writing Well:<p>> The growing acceptance of the split in-finitive, or of the preposition at the end of a sentence, proves that formal syntax can't hold the fort forever against a speaker's more comfortable way of getting the same thing said—and it shouldn't. I think a sentence is a fine thing to put a preposition at the end of.<p>Another from the same book:<p>> CREEPING NOUNISM. This is a new American disease that strings two or three nouns together where one noun—or, better yet, one verb-will do. Nobody goes broke now; we have money problem areas. It no longer rains; we have precipitation activity or a thunderstorm probability situation. Please, let it rain.<p>> Today as many as four or five concept nouns will attach themselves to each other, like a molecule chain. Here's a brilliant specimen I recently found: "Communication facilitation skills development intervention." Not a person in sight, or a working verb. I think it's a program to help students write better.
So, this would seem to be fairly easy to test empirically. Get a reasonably object measurement of the quality of writing, and use it on something where you know if it's true:<p>1) court testimony which we know (from outside evidence) is either true or not true
2) scientific papers which we know to have been reproducible, or not
3) stock pundits predictions about the future of some company or other, which we know with hindsight to have been accurate or not<p>Much more convincing to me than any amount of good writing about writing, would be to have some empirical evidence.
> But while we can't safely conclude that beautiful writing is true, it's usually safe to conclude the converse: something that seems clumsily written will usually have gotten the ideas wrong too.<p>I wouldn't phrase it exactly this way, but this is an important point that I really struggle to get across. I regularly see proposals and such that are very challenging to reason about because of their writing. But when I ask for terms to be more rigorously defined, or for the document to be reordered into a more principled structure, some people seem to have a strong instinct that I'm just being difficult for the sake of it. I still remember one guy who insisted that I need to make a specific technical criticism or sign off, and absolutely refused to accept the answer that my structural feedback was intended to help us reason about the technical details.
...I find it a bit surprising that Paul Graham of all people, writing in May of 2025, managed to get through this entire essay without mentioning LLMs.<p>Because I think LLMs provide a clear counterexample to his thesis. They are quite good at the craft of writing--not perfect, but probably much better than the median human--and they are just as good when the content is true as when it's false. This quality ruins a lot of my heuristics for evaluating whether writing is trustworthy, because LLMs are so good at bullshitting.<p>So while I agree that for <i>humans</i>, writing that sounds good tends to also be logically correct, that clearly isn't inherent in all writing.
There is a famous line about legal writing: “There are two things wrong with almost all legal writing. One is its style. The other is its content.” [1]<p>PG going for the "they're connected" angle, not too convincingly as shown mainly in the paragraph starting "This is only true of writing that's used to develop ideas".<p>[1] <a href="https://yalealumnimagazine.org/articles/3774-legal-prose-and-verborrhea" rel="nofollow">https://yalealumnimagazine.org/articles/3774-legal-prose-and...</a>
> Bruce Lee:
Before I learned the art, a punch was just a punch.
After I learned the art, a punch was no longer a punch (jab, uppercut, etc) .
Now that I understand the art, a punch is just a punch.<p>I think a lot of that "shaking of the bin" to compress objects brings you closer to the final and concise level of talking about punching. That middle section is verbose, petty.<p>A great example of this is Nietzsche's "god is dead, and we have killed him." He just skips over the details, and nerd-bait debate about atheism that had been ongoing since Spinoza. There's no contribution he could have made to that debate. All had already been said. Nietzsche assumes the readers' familiarity, expresses his own take and opens up a possibility for a "what's next."<p>If Nietzsche had one more sentence, the entire impact would have been destroyed.<p>A more typical form of writing at this time would have been "<i>By rationally examining the philosophical basis for belief in god...</i>" This predictably yields a relitigation of the debate... the Richard Dawkins route, a very different book.
I do not agree with the premise of the article --- writing that sounds good is more likely to be right. I've seen enough beautiful lies, fictionalized versions of the truth, and cunning orchestrations of a string of well-woven sentences, none of which had any intention of revealing the truth, but of convincing the reader to believe it's true.<p>I propose the following -- writing the sounds good manipulates the reader into thinking that it is right. Feels better to believe it.
A bunch of people here are trashing @pg's writing and that's a bad use of time in dubious taste.<p>@idlewords already broke that game like 15 years ago: <a href="https://idlewords.com/2005/04/dabblers_and_blowhards.htm" rel="nofollow">https://idlewords.com/2005/04/dabblers_and_blowhards.htm</a>.
>When I'm working on an essay, I spend far more time reading than writing. I'll reread some parts <i>50 or 100 times</i>, replaying the thoughts in them and asking myself, like someone sanding a piece of wood, does anything catch? Does anything feel wrong? And the easier the essay is to read, the easier it is to notice if something catches.<p>This seems to me analogous to the process I've discovered with photography... the more you throw away, the better the remaining photographs.<p>Clearly, I'll need to adjust my habits. I usually re-read what I wrote a few times, then later a few times should said comments actually attract attention.<p>This seems to me analogous to the process I've discovered with photography... the more you throw away, the better the remaining photographs.
"Good writing" nearly always collides with something else, for example a writer paid by the word. Or a writer granted too little time to <i>compose</i> prose, as opposed to merely <i>creating</i> it.<p>A shorter exposition is nearly always (a) better, and (b) more work. I'm reminded of Mark Twain's remark, “I didn't have time to write a short letter, so I wrote a long one instead.”<p>A classic writing book, now nearly forgotten -- "Strunk & White"/"The Elements of Style" (<a href="https://en.wikipedia.org/wiki/The_Elements_of_Style" rel="nofollow">https://en.wikipedia.org/wiki/The_Elements_of_Style</a>) -- famously exhorts authors to "Make every word count."<p>An underlying cause is that people don't <i>read</i> enough, before presuming to <i>write</i>. This results in malaprops like "reign him in", an example I see almost daily now. (A monarch <i>reigns</i> over a kingdom, a cowboy <i>reins</i> in a horse.) Examples abound, this is a common one.<p>Even worse, I now see automatic grammar checkers making ungrammatical "corrections" (incorrections?) like replacing "its" with "it's," or the reverse, but in the wrong circumstances.<p>But my all-time greatest annoyance are constructions like "Similar effect to ...", which in nearly all cases ought to be "Effect similar to ..." with copious variations, all wrong. Online searches discover that, in many such cases, the wrong form prevails over the right one.<p>Someone may object that language is an art form without fixed rules, that seems right. But granted that truth, many popular word sequences sound like fingernails on a chalkboard.
> “So here we have the most exciting kind of idea: one that seems both preposterous and true.”<p>Am I missing something or is the “seems true” part taking too many liberties here?<p>If anything, as described in the previous few sentences, the premise seems false, not true.<p>Kind of ironic since the line sounds right but isn’t rigorously right, so it undercuts the main argument.
Didn’t the ancients define “sophistry” as (in part) exactly the kind of writing one does when influence is more important than truth?<p>It might be that it’s hard to create sophistry accidentally in one’s writing, but it’s certainly a possible - and common - trick.<p>The danger is when you convince yourself that what you’re writing isn’t sophistry… because - after all - it looks good.
> I know it's true from writing. You can't simultaneously optimize two unrelated things; when you push one far enough, you always end up sacrificing the other.<p>We know from experience that it’s possible. Many of the greats did both.<p>There’s a tweet where PG argues that Musk can’t be evil because smart people work for him. His reasoning is basically: “No intelligent person would work for someone evil, and I know many smart people who work for Musk. Therefore, he can’t be evil.”<p>But that logic doesn’t hold up. Our modern understanding of evil often involves some form of dehumanization, usually in the service of a so-called higher goal, which is used to justify the cruelty. The obvious historical example is Hitler. And to say that no smart people ever worked for him is absurd. Just look at Heisenberg or Heidegger. They were definitely “smart” for any definition of “smart”.<p>It seems like PG struggles to recognize what’s right in front of him. He tries to make abstract, high-level arguments that often contradict observable reality - and he rarely offers concrete and rational explanations to support them.
Paul clearly excludes non-native writers. I know excellent thinkers who struggle to express an idea in a given language that they're not fluent in.<p>Another related point: I've seen geeks who're solid in thinking but terrible at explaining what they think.
For all the writing there is about writing, Orwell’s six rules have yet to do me wrong:<p>1. Never use a metaphor, simile, or other figure of speech which you are used to seeing in print.<p>2. Never use a long word where a short one will do. (To add: short words are best, and old words when short are best of all)<p>3. If it is possible to cut a word out, always cut it out.<p>4. Never use the passive where you can use the active.<p>5. Never use a foreign phrase, a scientific word, or a jargon word if you can think of an everyday English equivalent.<p>6. Break any of these rules sooner than say anything outright barbarous.<p>Though there is one more that is deserving of seventh place:
7. Edit ruthlessly. “Murder your darlings.”
Paul Graham's notion that good writing and stylistic quality are mutually required represents a false dichotomy. He understands this, which is why the essay is so good.<p>Style is a distraction while clarity and sophistication are inherently aligned.<p>In reality, there is opposition between style and substance. Sophistication does not increase clarity, clarity is not always the byproduct of aesthetic refinement. True writing excellence lies in the deliberate orchestration of form and meaning.
Good writing has the benefit of helping others for many decades and centuries. That's a realization I came to recently. My goal now is to write a variety of essays, articles, and books on topics that I excel in.
If I am drunk, or have brain damage, my ideas will be bad and so is my writing.<p>It’s not about what makes writing good. It’s about what makes it bad. People don’t agree on good writing.<p>I think Paul is talking more about <i>not bad</i> writing. Otherwise, he’d be talking about poetry, which also asserts truth and style as inseparable. People disagree about poetry. Most people agree Paul is a good (not bad) writer, not a poet.
Graham says good writing sounds good and is more likely to be true. But his own writing is hard to read and confusing. His sentences are long and messy. If he’s right, then his own ideas must be wrong because his writing sounds bad.
To me this looks like a false dichotomy. Writing beautifully and having a good argument are not mutually exclusive. So I think the article sets out a false proposition and discusses it at length.
Reading Graham's essays on writing always puts me in mind of the videos where Mexican moms react to Rachael Ray trying to cook (<a href="https://www.youtube.com/watch?v=zFN2g1FBgVA" rel="nofollow">https://www.youtube.com/watch?v=zFN2g1FBgVA</a>), or a Malaysian guy has to watch a BBC cook make rice (<a href="https://youtu.be/53me-ICi_f8?si=0AaZ82dk_AYFqJAx&t=226" rel="nofollow">https://youtu.be/53me-ICi_f8?si=0AaZ82dk_AYFqJAx&t=226</a>).
If you paint - no, paint by numbers does not qualify - then this will seem obvious.<p>I believe there is another level beyond this. When you write, but not by numbers.
You'll have to excuse me from taking writing advice from the guy who claims discovering the word "delve" anywhere means it was written by ChatGPT.
What how does that concept work in poetry or lyrics, for example? Something could be completely fictional (and artificial) and still be exceptionally well written.
I'm sorry, but this is such a risky conclusion (that essays that "sound good" are also "right"). The assumption is that the author is not only with good intentions but also not ill informed. What if an intelligent yet either evil or simply confidently incorrect author falls in love with their ideas, and manages to edit it to perfection by "shaking the bin". Does that make the writing more true? We have seen in the past how very well-written and convincing texts have changed the course of history, some for the better, but sadly, some for the worse.
While writing, you edit your writing towards both sounding good and having solid ideas.<p>That's the entire essay's point and I largely agree with it.<p>Its why I think current LLMs are bad writers, not because their prose or ideas are off, but AI generated writing does not have the same quality of robustness from ideas that are thoroughly vetted through the author's editing of it.<p>At the end of the day, good writing takes a lot of time thinking through what you are really saying and standing behind it. LLMs cannot do that for you (yet).<p>LLMs can certainly be a helpful tool, mostly by unblocking authors via creating prose, honing in on accurate expressions, researching edge cases and suggesting arguments or counter arguments.<p>But the craft of sharpening an idea to a very fine, meaningful, well written point is something that is still far off.<p>Of course, until the next research paper completely proves me wrong.<p>Step 1. RLHF on real time edited documents
Step 2. Profit??
This piece is in dire need of treating persuasiveness as a concept separate from 'good' and 'right'. Persuasive writing considers its audience and identifies avenues to create a train of thought that feels natural and true <i>to them</i>. What PG is describing is essentially the process of writing something that you yourself find persuasive. That can be a useful technique if your audience is people like you, but it can also amplify your biases and isolate your work from other perspectives that could provide value (increase 'right'ness per PG's definition).<p>Writing to convince yourself can help to refine your ideas, but writing to convince your detractors can point out blind spots and encourage finding strong evidentiary support for your argument. Equating "good" with "I would find it convincing" decimates the value that PG rightly identifies in using writing to enhance your ideas rather than just convey them.
> If you have to rewrite an awkward passage, you'll never do it in a way that makes it less true. You couldn't bear it, any more than gravity could bear things floating upward. So any change in the ideas has to be a change for the better.<p>I don't get that point. I'd say that there are many ways a thought could get diluted, misinterpreted, turned into wrong conclusions, or made less clear on subsequent iterations.<p>Just like the metaphor of shaking a bin with different objects doesn't work if the objects are tomatoes, glass or cats.<p>PG's essays could be a case of this; other examples are often seen in politics, when complex topics are trivialized by demagogues. A lighter example could be "Uncleftish Beholding"[1], an attempt to write about the atomic theory in a way that (very arguably) reads better.<p>[1]: <a href="https://en.wikipedia.org/wiki/Uncleftish_Beholding" rel="nofollow">https://en.wikipedia.org/wiki/Uncleftish_Beholding</a>
And then you have people like Kant and Hegel who have both been criticized for their writing styles, but I would bet in 200 years time people would still be reading and studying them. And with Paul Graham they'd ask, "Who?"
The web site, on my browser, is typical of so many I encounter these days: patently unreadable. Chrome in dark mode renders it white-on-white text. I press contral-a, and it can be read, but it is still not easy.
This essay nails it, clear thinking really does lead to clear writing. It’s a good reminder that writing is less about sounding smart and more about being understood.
Oh Mr. Graham again. I’m not a fan. I tried to read along, I really did, and then I came upon this sentence:<p>“But not without method acting.”<p>This is one of the most terribly written sentences in the English language I’ve seen since getting out of jail in January. It violates every reasonable convention regarding communication. It is terrible and please take note that a person who put this sentence out into the wild, without intentional comedy, is a fucking terrible writer.
All the woke and progressive ideas "sound good" on paper.<p>But they are detached from reality, scientifically wrong, and lead to disastrous outcomes.<p>--<p>All of marketing "sounds good" and even "looks good" - that is the entire point - but it is actually all lies.
pg is an expert at cranking out vapid, insubstantial pieces of writing that the gullible eat up as the pronouncements of genius. If I actually wanted to be known as a <i>writer</i> the last person I'd consult for advice is Paul Graham.
> I think writing that sounds good is more likely to be right.<p>Yikes. I stopped reading right there… There's a lot of corrosive, racist writing on social media right now that "sounds good" to a lot of people.
Paul Graham clearly loves writing and spending many iterations on producing something perfect. That was also his approach with the Arc language.<p>I may be projecting my own preferences here, but such a person is likely to have an ambivalent relationship with LLMs, which just output bland mediocrity or falsehoods.<p>In his previous essay, he warned that one should not create things that make the world worse. He softened it up by saying, without proof, that creating awesome things is probably fine (are LLMs awesome, I don't think so?).<p>Now he talks about good writing. I get the impression that he is one of the last remaining humanists in Silicon Valley, who at least has doubts about the direction we are going in and would be happy if YC startups created something else.
What happened to this man? A few years ago he wrote a glorious Rosetta stone of second-order functions in several programmng languages. Now he's spewing tired tropes about journalistic-style writing.
The problem with successful tech figures is that, over time, they often become convinced they can succeed at anything. Musk is a glaring recent example - and now Paul. I do appreciate his essays on tech and related subjects, but not for their literary merit. If I’m seeking advice on writing, I’d turn to actual writers - people who’ve earned recognition and acclaim specifically for their work in that field.
Wow, is this ever out of touch. People are currently confronting smoothly delivered, glib sentences that are wrong at an unprecedented scale due to widespread adoption of language model AI.
In which Paul Graham (re)-discovers Aristotelian and medieval metaphysics and the unity of truth, beauty and goodness. Or maybe more pragmatic in the word of Andrei Tupolev, head of the Soviet design bureau of the same name, "An ugly plane doesn't fly"