Reposting some of my best Inkhaven posts here. Will do so at a 1-2x/week cadence unless anybody objects. I'll start with the first post[1], about writing tips for myself and others:
Today is my first day at Inkhaven! It’s a program where I try to publish a blog post every day for all of November.
To set the stage for the rest of the program, I offer up a loosely organized list of writing advice I wished my younger self would’ve learned earlier. I intend to refer back to this advice often over the course of the program (and hopefully in future years). I’m vaguely optimistic that it can also be helpful to some other participants in my program, and maybe to other aspiring writers as well.
Source: inkhaven.blog
Core Principles: Write Fast, Write Weird, Get Feedback
Write like you’re running out of time
Why do you write like you’re running out of time? (Hey!)
Write day and night like you’re running out of time? (Hey!)
Ev’ry day you fight, like you’re running out of time
Keep on fighting. In the meantime—
The most important thing to do as a budding writer is to write a lot. Write a lot very, very, fast. Write a lot. If the goal of writing a lot is impeded by other advice on this list, you1 should heed the other advice too, but in general, prioritize writing a lot over other goals. For example, if you don’t have surprising things to say, write a lot anyway.
Who knows? Maybe you misjudged your audience and your “obvious points about field/question X” is actually novel to many readers. Probably not. But it’s possible!
Publish a lot, too. Many positive feedback loops only come into play once you hit “publish”, and some from the intention to publish and taking yourself seriously as a fast blogger.
Thanks for reading The Inchpin! Subscribe for free to receive one post a day over November and follow me along my blogging journey!
Try to say things people don’t expect
Many writers enjoy writing what readers expect them to say and re-emphasizing the exact same points over and over again in slightly different ways. For people who want to do this and can do this well (I enjoy Matt Yglesias and Bentham’s Bulldog as two examples of fulfilling this niche), more power to them!
But I at least a) know I’d be bored writing the same thing over and over, and b) realistically, know I’ll suck at it, compared to masters of the craft who live and breathe repetition in their field of interest.
So my comparative advantage is to write a wide breadth of articles, and be surprising in (almost) all of them.
This should be true recursively at all levels:
- Given the previous body of work, try to pick topics to write about that are surprising to your audience.
- Try to not get pigeonholed as the “Effective Altruism (EA) guy”, or the “econ reasoning guy”, or the “book reviews guy” or the “anthropics guy”
- Given the topic in question, your typical audience member should not be able to predict your conclusion based on the topic and a basic amount of knowledge about who you are and your past positions.
- Your arguments should be surprising
- Somebody who has not independently came to the same conclusions as you should not be able to predict the outline/structure of your arguments based on just your choice of topic and your conclusion
- Specific examples should be surprising
- This goes across both the field you’re talking about (if every other post on cognitive biases start with Lindy the feminist bank-teller, avoid Lindy the feminist bank-teller like a MAGA bank) and across your own past catalog of writing.
- Some people have a folksy charm of reusing the same 1-3 examples over and over again in every blog post, paper, or speech.
- I refuse.
- This goes across both the field you’re talking about (if every other post on cognitive biases start with Lindy the feminist bank-teller, avoid Lindy the feminist bank-teller like a MAGA bank) and across your own past catalog of writing.
- Specific jokes should be surprising
- This is probably obvious but bears repeating. Surprise is one of the most important elements of humor, and good jokes get ruined if they’re telegraphed too much.
- Your conclusion can be unsurprising, given the rest of the post
- A good argument naturally flows to a desired conclusion. So conclusions don’t need to be surprising
- On the flip side, do you even need a conclusion? Respect your reader’s time. If the conclusion is obvious, it might be better to leave it unsaid!
But in the service of saying surprising things, don’t say false things. It’s easy to be “surprising” or high-entropy simply by not having a restriction on “truth” capping what you can say, or via lacking a detailed world model. Good nonfiction writing should avoid lying, by either commission or omission.
Don’t say random impertinent things, either (#sorandom). Instead, aim to say things that in retrospect would come across to readers as surprising but inevitable.2
Figure out ways to get fast + reliable feedback
Try to figure out ways to get fast, regular, and reliable feedback. From yourself, AIs, early readers, and eventually mentors and intended readers. Like almost all other areas of human endeavor, one of the best ways to improve is via deliberate practice, and one of the cornerstones of deliberate practice is immediate and reliable feedback.
Unfortunately, when you first start out, all your feedback methods suck. Your intuitions suck (unless you are a writing god, which is unlikely). Your AIs will be pretty miscalibrated on what your intended outcome is (plus they’ve been primed by your early drafts, which suck). Your readers will suck. You probably don’t have mentors. You certainly aren’t getting the readers you want. Oops.
But actually, this is okay! Fast biased feedback can still help you improve!
For example, one of the earliest ways I remember improving as a writer is via writing lots of Quora answers (starting back ~2014 when Quora was halfway decent). The training signal I got on which answers were good was a combination of my own thoughts, Quora upvotes and downvotes, and the occasional high-quality comment. Was the feedback great? Of course not! But nonetheless, the early feedback got the ball rolling, and over the course of months and years, I learned to write more quickly, pay attention to what members of my audience like, structure my arguments better, have more engaging examples, etc, etc. All good things!
Many people have the (false) belief that GIGO (“garbage in, garbage out”) is a fundamental law of nature. I instead think of it as a weak heuristic that is frequently wrong. When it comes to writers and writing, compressed garbage can (sometimes) instead become diamonds.
So all in all, it’s very important to continuously write for an audience, and write in a manner that allows you to collect a ton of feedback from yourself and others. Don’t be afraid to (mostly) ignore the bad feedback3, but treasure all the feedback you receive regardless, and create gradients that make it easier for others to give you feedback, and for you to receive it.
A relevant and related model is the generator/discriminator model from machine learning (aka babble/prune). Writing a lot is your babble, good feedback (from yourself and others) is your pruning mechanism.
Relatedly, publish. Publicly. A lot. 80% of the ways you receive feedback are ~effectively closed to you if you only write drafts for yourself, or circulated among a small number of peer editors.
Find Surprising Things to Say
It’s great that you want to write a lot of surprising things in different ways that are conducive to receiving lots of feedback . But how do you write lots of new and surprising articles without lying, or just saying technically true but effectively impertinent facts?
“Surprising but inevitable” as a broad North Star
Aristotle said a good story’s ending should be “surprising but inevitable.” Likewise, I think it’s a very important goal of nonfiction. This helps maintain a high-novelty/-surprise factor of your writing, while still making each article internally cohesive and driven by a specific, coherent, logic.
For example, in my Rising Premium of Life post, which until then primarily drew on economics data and modeling on humans about changing preferences in valuation of life this century, I had a long, seemingly random, detour into facts about the lifespan of bats, mice, and other animals. But by the end of the article, the connection was apparent, and perhaps even inevitable – once you see the analogy between intrinsic and extrinsic animal lifespan differences and how safety might beget more safety, it’s hard to unsee.
Likewise, my Why Reality Has a Well-Known Math Bias post opens with the seemingly absurd image of a shrimp physicist trying to do advanced fluid mechanics calculations before giving up, quitting shrimp physics in favor of going back to shrimp grad school in the shrimpanities. Laughable, and yet the metaphor is directly connected to multiple arguments later on about how best to think about the unusual effectiveness of mathematics in the natural sciences.
Source: Gemini Pro 2.5/ my prior article
Among other bloggers, Richard Hanania is halfway decent at this.4 His arguments for leftist censorship are utterly surprising given his own history of being suppressed. Yet his arguments have a sort of internally cohesive logic that’s simultaneously intellectually challenging and funny. Scott Alexander is, of course, a master at crafting surprising but inevitable arguments across a wide range of domains.
Read a lot and do the research
To find novel or at least surprising (to your audience) ideas to write about, it helps to read a lot, and persistently.
My first piece that went semi-viral (~6k upvotes, ~500k views) was this Quora answer on “honest college majors.” It’s pretty silly, but I think what elevated this Quora answer was that instead of having a single stereotype that other people complain about (“humanities + social science majors are too leftist! X majors are unemployable! ABC majors is too hard/easy!”) I had already by that time had a reasonably decent introductory understanding about every academic major I joked about. None of my jokes were particularly revolutionary, but I think my breadth set me apart (survivorship bias joke in biomedical engineering, replication crisis joke in psychology, LSAT joke in Pre-Law, decision theory joke in history) and made it more interesting than the other answers that only had 1-3 novel angles to go for.
It’s also good to broadly use multiple separate ~independent quality filters for your readings, and be extremely wary of systematic biases in your access to information.
For specific articles, try to understand the State-of-the-Art of a field before opining on it. This is easier to do in some fields than others! For example, in my anthropics/mathematical effectiveness post, the question spanned many different academic fields (physics, philosophy of math, philosophy of science, etc) and I know there’s a real chance I missed important existing work. Nonetheless I gave it my best shot, skimming many papers I can find online, spinning up multiple different AIs to ask questions about fields I have less familiarity with, and asking three different philosophy academics I know to review my draft before publishing it. Even so, there’s a decent chance I missed something critical5. Still, you got to try!
If you’re opining on a known field, it behooves you to understand the existing academic consensus (or non-consensus!) before opining otherwise. Pay attention to track records on similar arguments, the intelligence/education level of various proponents and opponents, amount of total collective research effort undertaken before researching a conclusion, etc.
Sometimes while researching an article, you realize that the article you originally planned is bunk (either because your ideas are false or because they’ve already been said better elsewhere). This is okay!
Personally at that point I just stop that writing project and move on elsewhere, but people who are overly perfectionist/have trouble following the “write a lot” advice might benefit from documenting their mistakes and discoveries before moving on.
When writing nonfiction, unless you are writing something highly personal and specific to you, it’s often good to read multiple takes before writing your own. This is most obvious for research-heavy writings but it’s easier to do, and arguably more impactful on the margin, for posts others won’t consider “research.”
In my intellectual jokes post, by the time I started the post, I already had read thousands of different jokes online. But just to be safe, I Googled “intellectual jokes” and read/skimmed upwards of a couple hundred existing jokes in various online compilations, just to make sure the nine jokes I could stand behind are genuinely the best/funniest intellectual jokes by my lights.
Could I have written an intellectual jokes compilation if I only knew ~200 jokes and only ~10 intellectual ones? Of course! I expect most of the existing compilations look like this! But the marginal cost of skimming as many jokes as I did really isn’t that high6, especially for a post that eventually got ~30k views, and I think my posts benefit from me caring a bunch more about quality in areas other people don’t seem to.
Find what others miss
A corollary of reading a lot and writing things that are surprising is that you want to identify what others miss.
For example, for my Ted Chiang review, I reread several of his stories and read 10+ prior book reviews before starting to write my own (As of today I’ve probably read/skimmed >30 reviews). This helps ensure that my book review covers not just points that are interesting to me, and that Chiang does unusually well, but also specifically points that other book reviews overall missed.
Similarly, in my honey post, I biased towards covering subareas of bee welfare (e.g. positive vs negative valence), that other people did not. I didn’t try to answer thorny questions of normative ethics, bee consciousness, or intensity of valenced experiences, because my impression is that those were already (relatively) widely discussed elsewhere.
Don’t be a hater
To quote a wise woman “Haters gonna hate hate hate” This is not just definitional but causal, hating makes you have more of a character of a hater.
Many people start writing blogposts because someone’s wrong on the internet and they need to be set right. This is a perfectly fine sentiment to carry you through a post or two (and I’ve certainly fallen prey to it myself). But as an overall rule, it’s dangerous to see yourself as primarily a critic (or worse, a “hater”). It’s better to be motivated by love, curiosity, the Good, and other positive traits, rather than hate.
Know your audience
I’ve said before that you should write things that are surprising. But surprising to who? The simplest answer is “your intended audience”.
Who is your intended audience? That’s what you have to find out! Imagine who you want to read your work. Better yet, interview your real readers! (or people who are almost-readers).
Build a rough psychological profile of the people who you want to read your work, and try to say things that are true but surprising to them! Try to say things that slot in well (“inevitably”) with their own mental self-narratives, while at the same time being genuinely surprising.
Make Good Arguments
Own your words
Say what you believe and believe what you say. This is important.
There are different ways to achieve this. Classic style achieves it via purity of style. I like to just use first-person language (“I believe”, “I think”, “I aim”). The important thing is to avoid a weird academic-ese where you use weaselley, reflexive, language.
It often helps to write in your own voice, and write as you speak, though this is not strictly a requirement. Sometimes it helps to affect specific voices, or over-emphasize some aspects of your personality (see later sections on style).
Talk to your readers as if you are talking to an intelligent, educated, friend, or (as I do sometimes) to your younger self, who is intelligent but hasn’t yet stumbled upon the exact same insights as you.
Show your work with data
Include graphs, sources, calculations. Anticipate and diffuse obvious objections. Learn from people who are good at writing with data, and use it natively.
Draw on interdisciplinary knowledge
I’m not an expert on any specific field. And if I am, it’d be in relatively niche and inherently inter-disciplinary “fields” (like “EA grantmaking” or “pandemic forecasting”). So it’s pointless to hide that and try to “compete” with the experts in their native turf.
Instead, it’s better to rely on my own breadth and interests to make cohesive interdisciplinary arguments on topics of interest.
For example, my first anthropics post draws not just on anthropic reasoning, but also on philosophy of science, physics, history of science, evolution, math, philosophy of math, and even AI. The Ted Chiang review similarly draws not just from understanding Chiang’s own work or other science fiction, but relies on a confident understanding of social criticism, modern philosophy and economic growth models.
But at the same time, don’t “dazzle” people with knowledge for the sake of impressing them! Make focused arguments necessary to express a point (or occasionally to be interesting/funny), and never try to come across as too intelligent for your readers!7
Polish Your Craft
The following four points on craft are areas that I suspect most aspiring nonfiction writers can benefit from.
Make focused stylistic choices and stick with it (within a post)
As I wrote in my field guide to writing styles, a mature writing style is defined by making a principled choice on a small number of nontrivial central issues: truth, presentation, cast, scene, and the intersection of thought & language.
A good artifact of writing maintains not just surface similarities but has a quiet consistency that underlies the work.
Open Asteroid Impact, for example, never broke character. There were jokes I considered including but rejected because while funny in isolation, they would’ve ruined the mirage that I was running a Serious Company completely lacking in self-awareness, and thus make the website overall less funny.
While I believe a lot in experimenting in general across the course of your writing career, I think most posts should pick a side on each of the central issues and stick with that. Don’t have your book of prophecy be 1337spk.
Pay a lot of attention to titles
If Coates instead titled it “Some Arguments in Favor of Considering Ethnicity As One of Several Factors When Deciding Appropriately-Sized Transfer Payments” the article would become substantially less memorable
Spend significant effort on your title (>10 minutes per serious post, and sometimes closer to an hour). It’s one of the most important components of a successful blogpost.
I think this is very unintuitive to people, myself included. From the perspective of a writer, a title is just one line to write (and often one of the least interesting lines, as it’s unlikely that you as a writer discover something novel in the process of creating a title). But from the perspective of a (potential) reader, titles are quite important, as:
- It is probably one of the most critical pieces of information for readers to decide whether something is worth investing the time to read and engage with, particularly if linked elsewhere.
- It helps set the tone for how readers should engage with the rest of your essay.
- Positive example: Some unfun lessons I learned as a junior grantmaker
- Short, to the point, and contextually useful (someone reading this title knows both where I’m coming from, as well as have some bounds on the limitations)
- Negative example: Why short-range forecasting can be useful for longtermism (old title)
- The old title was bad because it led readers to believe (falsely) that the post would give an argument for the statement in the title, rather than just assume that the title is a placeholder.
- It also predictably led to people being confused about the intended purpose of the post, which is more “this thing would be cool, would love to find collaborators.”
- imo the newer title Early-warning Forecasting Center: What it is, and why it’d be cool is comparatively much better, though still not ideal.
- Positive example: Some unfun lessons I learned as a junior grantmaker
- For low-information readers, a title helps them decide whether/how much to be angry at you.
- This can either be desirable or undesirable, depending on your intended purpose.
- As I personally don’t like it when randos are angry at me, I try to screen all my titles for a “will this title seriously offend low-information people who just randomly see this title on Twitter/Facebook/Hacker News?” check.
- All else equal, your title is the substring of your text that is most likely to be taken out of context, so some prudence is warranted.
So titles are very important to readers! Since they are so important to readers, a writer trying to cooperate with (current and future) readers should also treat titles as important!
In addition to wanting to be cooperative with readers, in the current algorithmic information environment, you also have to somewhat “fight”/argue for your post’s value in seeking your readers’ attention.
Finally, good titles should capture the essence of your point after it’s been argued. So you and your readers can refer to it again in the future. So it behooves you to come up with good titles (or a title; subtitle combination) that both encapsulates your primary thesis and also is intriguing enough for readers to click through.
I often leave my titles blank before finishing a post (or have a purely descriptive internal title like “writing styles post” or “war post”), and use AIs to generate title ideas after I finish writing a post. Sometimes the AIs can help discover real gems, like elevating The Secret Third Thing from what I thought of as a throwaway Twitter reference to a poetic phrase that encapsulates a core argument of mine in the Ted Chiang review.
More often, I go back and forth with AIs and use AIs as inspiration but settle in on a cleverish multilayered title like Why Reality has a Well-Known Math Bias, where the final title and pun was entirely my own, but I probably wouldn’t have generated the title if AIs didn’t suggest the direction of other (worse) puns first.
Unfortunately sometimes I fail at a title, even for pieces I otherwise like. For example, I don’t like to refer to the Why Are We All Cowards? by its final title, since it’s both clickbaity and imprecise compared to the actual argument, but “the rising premium of life post” unfortunately sounds like uninteresting gibberish to people who don’t already understand the post’s central concept.
Vary your sentence lengths, but bias towards shorter
Self-explanatory.
Pay attention to reading level and word complexity
When you see online advice on complexity of language, many people go for advice like Orwell’s, where you “Never use a long word where a short one will do”, etc, unless you’ll otherwise say something “outright barbarous.”
I mostly disagree. I want to be more ecumenical with my advice. I think complexity of language has its place, and I think there are many times where you ought to prefer more complex language even if simpler words are “good enough,” even when the simpler language is not “outright barbarous.”8
For me, I think better advice is to never write with words you personally are uncomfortable with, and to write with words and methods of expression that are comfortable for your actual audience. This means understanding your actual audience (see above), and developing models of what words would and would not be too complex for them.
As a ballpark, I think for popular pieces, you ought to aim for writing ~4 years of specialization/grade levels below where you think your audience’s actual reading level is (since people are less comfortable with writing at the limits of their ability). But this is just a ballpark. I’m more confident that writers should pay attention to their reading levels/complexity of words and be intentional with their choices9 than that they should land at a specific place.
Polish Your Craft II
These are craft points that I suspect are more specific to the types of writing that I want to do.
Optimize your posts for both skimmability and non-skimming experiences
Some posts (e.g. Rethink Priorities reports, academic papers in the natural sciences) are primarily made to be skimmed by the vast majority of readers. For those works, the core success criterion of the writing is skimmability (ease of parsing on a quick skim) and the rest of the writing is practically more about proof-of-work and/or evidence you know what you’re talking about, than it’s actively meant to be read.
Some writings (e.g. Great American Novels) are written primarily for Deep Reading (™), and the authors would even be offended if you try to skim them! Some of those works are deliberately engineered to be harder to skim.
For my own blog posts, I try to anticipate both skimming and non-skimming use-cases. So I try to write articles that are both easy to skim and easy to not skim. This is a real tradeoff I’m making, and other bloggers might want to only pick one of the two sides to focus on.
Use headers strategically
I try to have clear, meaningful section breaks that preview content. This makes it easier for readers to skim my content, for returning readers to jump to sections they’re interested in, and for me (and hopefully others!) to link specific sections to use as lemmas/subarguments in discussions with others, etc.
Integrate visual and other non-writing elements
Most writing advice given before the last ten years, and often many that are not, ignore the online nature of the vast majority of written content we produce and consume these days. They assume modern writing still looks like Ye Olde Essayes hosted HTML-only on some blogging server.
Needless to say, modern blogs don’t look like this at all anymore, especially the more successful ones.
Charts are helpful not just for demonstrating content, but for breaking visual monotony. Images, graphs, tables, videos, subscription links, they are all great too.
To quote Scott:
The clickbaiters are our gurus – they intersperse images throughout their content. The images aren’t always very useful, they don’t always add much, but now it’s not just a wall of text. It’s a wall of text and images.
Of course, the best visual and other non-text elements actively advance your core arguments, not just enhance the vibe or make for a more pleasant reading experience.
See here, here, here, and here for non-writing integrations that I especially liked.
Lead with concrete examples/analogies
When illustrating Wigner’s puzzle, I didn’t introduce it with an academic discussion on the unreasonable effectiveness of mathematics in the natural sciences. Instead we got an image of a harried shrimp physicist at the bottom of a turbulent waterfall. When introducing questions of bee welfare, I didn’t start with complex moral deliberation or questions about arthropod phenomenology, I started with a relatable (to Berkeleyites, anyway) anecdote of holding up a line at the local bubble tea shop. The Puzzle of War was introduced with graphic descriptions of Qin dynasty China (and then mellowed out with specific examples of fictional wars between game-theoretic elves and dwarves).
As for why you want to do this, Scott Alexander puts it best.
Use microhumor liberally
Jokes are good. Mediocre jokes aren’t as good as good jokes, but they are better than no jokes. And microjokes…well the homeopathic Law of Minimum Dose will say that microjokes are actually the funniest jokes of all!
Deploy footnotes liberally (for asides)
In every article, a strong temptation I have in my heart-of-hearts is to include every detail, no matter how irrelevant. But in my head (of-heads?), I know that this is a terrible idea. Readers are busy people, and they have much better things to do than wade through 10,000 words on the rising premium of life10.
But how do I balance my emotional need for completionism and including every detail (plus irrelevant asides) with my intellectual desire to be a good citizen who presents completed and compact finished works?
Answer: Footnotes! After an initial draft where I barf everything on the page, in the editing phase I can tighten up my language and remove unnecessary cruft11. And when there are paragraphs I especially like but I know I ought to cut, footnotes come in handy!
That said, it’s also possible to go overboard with footnotes. I’m less aggressive with cutting footnotes than in the main text, but I still average ~10 footnotes instead of like 25.
Ship It Anyway
Experiment widely
Bruce Lee is an expert of kicking, but not necessarily of substack blogging ...
Bruce Lee once said “I fear not the man who has practiced 10,000 kicks once, but I fear the man who has practiced one kick 10,000 times.” Well, writing is rather unlike kicks. If you write the same story, blogpost, or paper 10,000 times, people aren’t going to fear you. They’ll just be confused, and/or bored.
As a starting writer, it’s hard to know which articles a) you especially enjoy writing, and b) the marketplace of ideas would reward you for. Rather than perfecting a specific genre or sub-style of essay, it’s probably much more productive to experiment very broadly and sample across a wide swathe of possible pieces you can write.
Experiment widely with a) different styles of writing, b) different topics, c) different venues to publish in, d) different levels of post-writing edits and perfectionism, e) different intended audiences, and f) different registers and degrees of linguistic complexity, and so forth.
Integrate feedback across posts
Getting feedback is great! But often, the most critical/useful feedback essentially makes your piece unsalvageable if you were to take it too literally.
Wait…unsalvageable? How is it still useful, never mind “most critical”? Simple: you take the feedback you receive and apply it to your next post.
Seeing a published piece of yours continue to have critical flaws isn’t pleasant, but it’s vastly preferable to constantly rewriting a piece to address what might well be an unfixable flaw. And it’s of course better than pretending to seek feedback but not updating!
Learn from better and worse writers trying to do what you’re trying to do
It’s good to learn from better writers than you. For me, I learn the most from Scott Alexander, because I respect him and he has a writing style that I find pleasant to emulate.
It’s also good to learn from worse writers than you. Why? Well, first of all when someone else makes the same mistakes as you, you might be able to understand why those are mistakes from a distance of some remove.
Secondarily, someone being worse than you at one dimension (or in aggregate) doesn’t mean they’re worse than you in all dimensions of writing. Just as averaged out faces can be very beautiful and ML student algorithms trained on human game-play can achieve superhuman performance, you too can learn to exceed your masters even if you only ever train on worse (overall!) writers than you.
Finally, it’s harder to see progress if you only compare with writers that are far better than you. By comparing with worse writers, you can learn and improve on more realistic axes.
For me, the writer I learn the most from is my younger self. This might be surprising, since in some sense my younger self is almost by definition a worse writer than me. Nonetheless, I find it valuable to reread my older pieces with clear(er) eyes. Seeing mistakes in my older pieces, as well as things past-me did well, has been very illuminating in improving my present writing.
Be willing to ship unfinished work
Finally, be willing to ship imperfect and unfinished work! This is somewhat redundant with past advice, but still bears saying.
Shipping a lot of imperfect work means you can:
- Write a lot of posts
- Write a lot of different posts
- Learn a lot about yourself and what you enjoy and hate
- Get a lot of public feedback that you can apply to future posts
- And many other virtues.
That’s why I’m in Inkhaven where I have to publish a blogpost every day for a month! 😱😱😱
You, too, might benefit from being extremely prolific. But that’s only possible if you give up on perfectionism and be open to publishing the imperfect.
So publish even when you aren’t fully ready, even when you haven’t edited everything through, and even when your conclusion isn–
- ^
The reception to this post has been odd. It didn't get that many likes on Substack, but it has been liked (and is the only post of mine to be liked) by some pretty big writers on Substack. I'm not sure if this is evidence my writing advice is good or not, however, consider the alt-text here: https://xkcd.com/125/