I hated writing until I learned there’s a science to it (2024)
science.org
science.org
- Ira Glass
I think most of us have experienced this. I consider myself an above-average writer and I absolutely hate everything I write.
But the problem, for me anyway, is that it's exceedingly difficult to know what to work on next in order to improve. In that regard writing is entirely unlike a lot of sports.
My throws are bad? Better throw 100 passes a day, every day, until my muscle memory is there. I'm getting beat deep? Better work on my fitness. Maybe I'll never get to where I want to be, but at least I know why.
But improving one's writing is seemingly impenetrable, to me. I read what I write and it sucks but I have zero intuition about how to un-suck it. I fucking wish I could write like Heller, or Didion, or Tolkien. Not even in terms of writing novels but just the quality of their prose.
listening to narrations of vast variety of poetry and narrating something yourself will help you develop your specific voice and read with more intent.
you may not even need the "science of writing" this article describes. let yourself just... be with text.
In the beginning it's great to practice your art by yourself with lots of safety, but sooner or later you're gonna want to to ask the public/community at large what they think of what you do, so you can get external feedback from people other's who love the same thing. I think this is probably the only way to actually get better, you need to connect with other people around it, and get their point of view. I've found this true for any creative endeavor I've tried to get better at.
Receiving criticism is probably as hard to get good at as giving criticism, so don't let the harsher stuff get into your skin as some people aren't so good at giving criticism, but you'll find lots of other useful advice that you'll agree with, and find directly actionable :)
Even if you don’t improve for 100 days straight, small successes accumulate. In a decade you will have transformed yourself.
There is just no way you won’t improve significantly if you keep trying new things and bring yourself to fail ever day.
What helped me was the saying “your first million words are gonna be shit”. I still distinctly remember, four or five years into writing every day, when things finally clicked, my voice came through, and my sentences became fun. It is delayed gratification to the max.
Thank you for this.
How do LLMs do it? They don't learn by writing a lot they learn by reading a lot.
Literally they learn by both, training process is filled with "exercises" for the weights to be shaped by, not just "seeing" tokens :) At least the typical LLM that outputs tokens have to actually "practice" doing so.
When I am teaching anyone any of these skills, the first thing I say is “are you ready to be bad at this for a long time?” Sometimes it catches people off guard. On the other hand, if someone says “yes” then I know that they are going to be a good learner.
I have done quite a bit of painting and caulking for a guy who's not in the profession. I despise both with a passion, though, especially caulking, and I have never once been satisfied with a single paint or caulk job I've done. I feel like I'm the embodiment of "be bad at this for a long time," although I'm objectively probably halfway decent at it.
That is to say I think Ira Glass' quote of "You've just gotta fight your way through" to get where you want to be seems especially meaningful in the context of something like painting, where most everyone _can_ do it (or writing / storytelling in Ira's case), but very few are actually good at it.
There's one UK guy on YouTube that convinced me of the evils of water/iso sprays and the beauty of the proper silicone caulking tool.
The little wedge shaped caulking tools btw are not enough, as you need some stick to it so you can get around certain angles/items.
I eventually got rather good, albeit slow, and now can easily finish a wall where you can’t find butt or tapered seams with a flashlight, with minimal sanding. It took many hundreds of hours over the years, and a clear idea of what the bar was, for me to get there. The results still bring me joy, but more also the intuition built up around working with mud translated to a quick ramp up for more ambitious projects with stucco and concrete.
At least you limited yourself to human scale hand-y work.
An engineer type can go down some dark yak holes trying to find solutions to achieve inhuman flatness
It's about honing ability and competence.
I only just learnt about https://en.wikipedia.org/wiki/Wabi-sabi
Perhaps "artisinal" encompasses the idea?
Ira Glass is imperfect perfectionism.
When my mom was in her late 60's having never done any work with drywall or mud, taping, floating, etc and attempting to make lemonade out of a situation where every room in the house was being sheetrocked because Dad was actively using some of his many skills to convert their house into a home, she became creative and produced a collection of decorative walls that anyone could admire.
There are rag-rolled walls with layers of colors over imperfect textures. There are walls painted a neutral background color and then combed with spectacular streaks. Some rooms are wavy and others are vertical lines. She layered colors and textures because the drywall, after all the work was done had textural flaws and places where use of a single color would make all the imperfections pop like Shiprock from the New Mexico plain. It was hard making 1/4" (6.3mm) drywall hide all the changes to the structure and the shiplap that had happened in the 80 years (80 years) since the house was built.
My favorite walls are in a hallway with nearly 10' (~3m) ceilings. She used a variable depth texture, thicker than anywhere else in the house and knocked relatively flat though with plenty of knife swirls randomly distributed. In the heavy texture she used a leaf print to impress hundreds of oak leaves from floor to ceiling in random orientations as if they are all falling. The colors are autumn colors with a light base and darker accents that create additional shadowing. It really is beautiful and is quite original.
Every time I walk that hallway I hear:
The leaves are falling all around, time I was on my way.
Thanks to you I'm much obliged for such a pleasant stay.
But now it's time for me to go. The autumn moon lights my way.
With the original oak hardwood floor painted a nice checkerboard pattern and the trim all Dad's handiwork it is really great.
And they had these minor-superhero things they could do.
Like he could hammer around a corner. You would think it wasn't a big deal, but he could put wire staples places where a beginner or a fancy staple gun couldn't reach.
Because of the way the strands are laid, wire has a direction and way it "wants" to go. I'd been an electrician twenty years by that point and knew how to work it. Not strength. Not that I said any of that.
Might not be as strong or fast as he used to be, but knows how to do it smartly.
Sometimes the problem truly requires strength and brute force, but not usually.
Even something like taking up concrete. It seems straight forward but you can waste a lot of energy hitting the wrong places.
Friend of mine worked in Prisons for a while, and has a way of telling the difference between people with a lot of muscle and people who know how leverage their musculature to high effect.
You get enough experience and you can turn a bolt or screw and tighten it to stretch a little.
Not too little (doesn't hold parts together and/or can come loose).
Not too much (uneven tightening or snap/strip something).
Just right (torque wrenches make this precise).
Oh, I love this so much. It communicates so many important things in one go.
My kids must be absolutely sick of me saying in a dozen different ways "you just have to do it a lot."
This also reminds me of how I've often responded when people ask me about learning to code. I ask them if they're ok with sitting in front of a computer for many hours.
It sounds obvious, but most people (in my experience) simply aren't ok with that, and hadn't considered it.
I wonder if this puts people off who shouldn't be put off.
I certainly don't think I'd want that for myself if asked. But coding doesn't feel like that. Sure, I sit in front of a computer for hours, but productive sessions generally involve losing track of time and realising you've been sat there for hours
That is, verifying a solution is much easier than finding it!
P != NP
Does anyone truly think we’ll have another generation of nerds hacking away for hours by hand to hone their trade? Or will it turn into a craft like woodworking?
I see my sons tinkering with software now in the same way I did. They want to play games and make games and build their own tools. I don't know whether they'll wind up building software as a profession, but I'm sure they'll have the base skills needed to.
One of the worst parts of software is that you develop this deep expertise in something that has less to show for it to people than a guy dicking around on a guitar or someone who made a stool after 40 hours of experience.
People aren’t decompiling their iPhone apps to admire your handiwork and evaluate how thoughtfully you modeled the domain.
Will artisanal human-driven taxi drivers exist in the future? No, Waymo is even more expensive yet people prefer it. The human aspect turns out to be a liability and less of a human expression people would go out of their way to experience.
Damn, on point. So hard to go over that hump but so worth it, every time, in retrospect.
> "The secret of the creative process is in 98% of sweat, 1% of talent and 1% of luck."
Did you have to be born to the right parents at the right time, or just avoid a car accident?
And the ability or desire to work hard has some very soft dependencies.
For a concrete example: in his book "On writing" Stephen King details his life up until the point he hit it big with "Carrie". You could say he was lucky for the book to sell as good as it did, but that would require ignoring that he had been writing (and getting rejected!) non-stop for roughly 20 years.
I’ve heard that to be a good writer one should also be a good reader. Maybe they’re a good reader and had one great story in their mind. Having produced it, they were satisfied?
I just read a bio of the author Jack London. He suffered terrible poverty and dismal employment prospects as a youth. Writing was an escape. He produced a lot of mediocre work in order to get paid.
Yes, but you need to learn to linger over the text. To read it slowly in order to analyze how all the different parts fit together, to critically examine the overall structure, to consider how you would have or could rewrite it, to drill down into even the choice of words and how synonyms could have changed meaning and interpretation, to re-read it over and over across the years to see how your own life experiences change how you read the exact same content.
IMO not many people are either built for that, or are willing to expend that effort. Even I stumble - a lot.
Why contrarian? This comment further elaborates on the very idea I was pointing towards. From what well does a singular piece of writing spring?
And you describe it here—quite possibly from someone who reads deeply, and writes infrequently.
That turned out to be the best life experiment we ever did together. They are teens now and dealing with far more writing every day than just a couple of paragraphs. The other day I found some cards they've written for Father's Day and other holidays over the years, and I can't even tell you how impressed and proud a parent I feel whenever I see their writing. That single skill manifested in improved overall literacy and discipline. My daughter received the Presidential Award of Academic Excellence. My son was accepted to an elite college with a scholarship. He's a competetive swimmer with dozens of medals. His team competed at the state level and even set state records. They are going to be fine. And the only thing I had to do is to teach them to face the thing they hated doing. One paragraph a day.
what country do you live in where this is a normal thing?
do you mean a family groupchat in general? like no one around you has a family groupchat on imessage/sms or insagram or whatsapp or something? i know "third party" messengers are weirdly unpopular or at least fragmented in the US but..?
The comic has been quoted online many times, but I had to buy an old copy of Michigan So Far on eBay to actually see the comic: https://25thandClement.com/~william/links/img/Writing-Guindo...
I (now) fondly remember getting my first English paper back as a hs freshman with an F. Shook me to my core. I had, after all, gotten into this good school as one of the select few. That was a good lesson and it's served me pretty well.
"Good" Fiction writing is an inaccurate science but has a similar trajectory to what the author went through. To become good at it you _need_ to read other people's works (the good AND the bad stuff) to figure out for yourself what makes that writing stick out to you, and you need to learn to love to edit, and to show people what you did.
The most time consuming portion of the writing process is the editing process in my opinion. It's also my most favourite part. You take a half-formed idea and you cut. And you tweak. And then you cut some more, until paragraphs start to take the shape of the story you actually wanted to tell, and sentences become so load bearing you can't remove any of them without altering everything around it. It's a puzzle with no real "solution" other than what I feel works.
Really, it's only after I kept at this for a while (and put things out there and didn't get bad comments at all!) that I started to get a little more confident in myself and begin to go to writing groups and such. It's hard work but it's worth it, just like any skill.
Often it forces a clarity that only comes from writing ideas down in a way that's necessary to explain your results to your peers.
The process itself sucks, but the outcomes are often quite satisfying and rewarding.
Nowadays the intro/motivation/problem statement / related work (citation tax) / formulation/<actual results> / simulations / conclusions / futurework format is just soul crushing.
Today there is much more research being done and published, and fully reading a paper is a special case. Papers are now more structured, and the primary use case is quickly skimming over the paper to determine if it merits more thorough reading. (Usually it doesn't.)
But I actually do think it's good to force people to think about the "why" question a lot.
And anyway, aren't reviewers themselves supposed to be able to connect dots?
When I was an active reviewer I tried not to ever say "Who cares" in so many words and just focus on the technical contributions. There's enough overinflation/fraud in papers that we dont need to be encouraging it by asking them to over-over inflate a possible reason for the work that is beyond the obvious ones that reviewers in the same field would appreciate.
All IMHO. There's no "right", just "preferred".
That would not make me hate writing less.
It was through the language of English that I got to love writing and reading (mostly Audible). It was Writing in a Foreign Language again, but this time, it fitted more to my style, also because it’s straight to the point and you have so many of the same words to explain a specific sentence so well. I wrote more at https://www.ssp.sh/brain/writing-in-a-foreign-language just in case of interest.
What's the null hypothesis? Is it possible to disprove your assertion?
Except there usually is a right answer, only it tends to be buried under several layers of “how do you want this to be interpreted?”
The one thing I have learned about writing is that the same sentence can have wildly different interpretations between people if it’s written sloppily enough. Your core meaning could still very well be there, but because the prose was sloppy, it opened the door to alternative interpretations that you cannot control.
Now sometimes there is no way to refine it further. But most of the time it can be.
And then, even after sharpening the prose, you need to take your reader into account. What works for an adult might not work for a teenager. When I was a kid, I remember having a doctor in the hospital ask me if I was nauseous, and I replied, “there is nothing wrong with my nose”. I had never come across that word before, and associated the sound that went into its pronunciation with the closest other word I knew - the word for nose. The doc had failed to take into account people with a more limited vocabulary, setting up a chance for medical misinterpretation - I had indeed been sick to my stomach, but because he didn’t ask about it in those simpler words, he almost didn’t learn about it.
Writing is even more bereft of context, and so you need to not only sharpen the prose to cut off undesirable paths of interpretation, but also write for what your audience knows. And sometimes this can be one hell of a rabbit hole in of itself.
I always thought that even very small children knew that. If someone wasn't aware of this for years even after many attempts, I wouldn't trust them to do any sort of research.
Somebody should have put a red line through every word of this.
disclosure: I used the science of editing this comment after writing it.
Storytelling is something that connects science and the science of writing for me. I've received advice to tell a story when writing scientific papers, but I haven't heard a concrete and sufficiently complete explanation of what that really means. For example, when describing a new model, what is the story you tell and how do you tell it? A book I find useful on the structure of storytelling is "Beginnings, Middles & Ends" by Nancy Kress. It helps me think about how to tell a story in a scientific paper, even though the book has nothing to do with that.
Some things that I have learned from of them:
Write for yourself first and get to the end. Rewrite to add in all those things you didn't put in the right place the first time.
Speak at least the dialogue out loud. Spread the description around.
Read some of the worst to remind you that even they got published. Copy the greats for practice on dialogue, or description or whatever you want to work on as deliberate practice.
Try different things like write your story as a game, or a puppet play, or stage play, or screenplay, or radio play. Draw a storyboard or animatic. Go to the park and write what you see. Have your characters in a room together and eat a pie.
1. Write stuff
2. Make it better
3. Continue with step 2
It even is an algorithm.
1. Write down the problem.
2. Think very hard.
3. Write down the solution.
- Use a word like "science" to lure in the geeks
- (you don't even need to know what science is, its ok)
- Some of the geeks will push your headline to top of HN just because it had the right word in it
- Put some filler about life being hard in the article, so those who actually read it have to waste ten minutes of their lives (proving your point).
- Profit and glory!
With science writing, in terms of published papers, this makes perfect sense. You want to avoid imparting your own biases and interpretations in favour of letting the data speak for itself. That means the author needs to quite severely “step aside” from a literary perspective.
However, I wonder if this workshop is still relevant in the age of LLMs...?
'Write shit. Edit it. Repeat.'
What a disappointment.
"Sucking at something is the first step towards being sorta good at something." - JAKE THE DOGMake a first draft that is bad, and improve it from there.