KD^3C^3 - 20250105 His duende was winning
Feeling a lot of imposter syndrome this week. Well not imposter syndrome exactly, but I don’t think we have a catchy name for “You have been doing a thing consistently for nearly seven years and you haven’t gotten any better at it” which is what I’m feeling with this newsletter.
Small nitpick: It hasn’t actually been seven years yet. It’ll be seven years in March, but I’m bad at remembering anniversaries of things, if I don’t have a calendar alert set to remind me. So the start of a new year is as good as anything.
There’s a lot of reasons Malcolm Gladwell’s Ten Thousand Hours concept is bunk (check out the If Books Could Kill Episode on his book Outliers) but an important reason is that doing something over and over again doesn’t automatically make you better at it. I’ve been breathing for roughly three hundred thousand hours and I’m not the world’s best breather. I’m probably not even in the top ten. To get better at something takes time and reflection. I’ve been putting the time in, when it comes to writing this newsletter, but I don’t know if I’ve gotten any better, and that’s probably because I’ve done very little reflection.
If this thing is just a personal public diary, then that’s probably ok, it doesn’t have to be polished every week. Or any week, for that matter. And my habit is really one of just pushing this out into the universe. But it hasn’t exactly made me better at it.
Do I want to get better at it? There’s a joy and freedom in not feeling the need to be the best at everything, even in being bad at it. At the end of the day there’s still a hat where there never was a hat.
Is the creation of something that didn’t exist before sufficient, though? I used to say yes, but now the web is being flooded with slop generated by Large Language Models and Image Generators and they’re all creating things that never existed before, right? That seems to counter my normal argument that creation is an inherent good, but maybe I need to think more about what I mean when I say creation.
There’s an idea originally proposed by David Bye called the Workmanship of Risk. I’m going to dramatically oversimplify it, by boiling it down to the idea that when creating something there is a risk that the creator, or the artisan if we’re comfortable calling this art, can ruin the work. This is traditionally applied to physical objects made by hand, such as painting or sculpture or even the manufacturing of clothing by hand. A factory that stamps out forks can fail to make forks, but because of a failure in the machinery or even in the design stage, but the system is designed to make a fork the same way every time.
With the written word, it almost feels like there isn’t a risk of this nature. If I make a typo I can (and often do!) go back and fix it before publishing. So is the written word something hand made? I think it is if we expand the risk of failure to also consider the risk of lost time. If I spend hours writing a scene for a play (a thing I have done in the past) and at the end of that time, the scene has to be deleted from the final work that time has been lost. If I were a better playwright, I could have skipped writing the scene entirely.
So is the slop generated by computer programs regurgitating stolen art (written and visual) hand made? No. At best it’s manufactured, but it’s a black box system. You input a request and the machine spits something out on the other end that has a high mathematical probability of looking authoritatively like something you requested. But there’s no risk of the “prompt engineer” (ew, i felt gross even typing that) making a bad decision that can ruin the work. Because the prompt isn’t the output. It’s not a direct one to one process. You can enter the same prompt dozens or hundreds of times and get different results each time. Nothing the hypothetical artisan does has a direct impact on the result. Rather they are pushing buttons that whip random monkeys in the infinite room full of typewriters and sometimes something useful comes out.
This has gotten way from me a bit. Do I want to be a better writer? Yes. I don’t need to be the best, but I want to look back at my work seven years from now and see some improvement. The only problem is I don’t know what that improvement looks like.
Small nitpick: It hasn’t actually been seven years yet. It’ll be seven years in March, but I’m bad at remembering anniversaries of things, if I don’t have a calendar alert set to remind me. So the start of a new year is as good as anything.
There’s a lot of reasons Malcolm Gladwell’s Ten Thousand Hours concept is bunk (check out the If Books Could Kill Episode on his book Outliers) but an important reason is that doing something over and over again doesn’t automatically make you better at it. I’ve been breathing for roughly three hundred thousand hours and I’m not the world’s best breather. I’m probably not even in the top ten. To get better at something takes time and reflection. I’ve been putting the time in, when it comes to writing this newsletter, but I don’t know if I’ve gotten any better, and that’s probably because I’ve done very little reflection.
If this thing is just a personal public diary, then that’s probably ok, it doesn’t have to be polished every week. Or any week, for that matter. And my habit is really one of just pushing this out into the universe. But it hasn’t exactly made me better at it.
Do I want to get better at it? There’s a joy and freedom in not feeling the need to be the best at everything, even in being bad at it. At the end of the day there’s still a hat where there never was a hat.
Is the creation of something that didn’t exist before sufficient, though? I used to say yes, but now the web is being flooded with slop generated by Large Language Models and Image Generators and they’re all creating things that never existed before, right? That seems to counter my normal argument that creation is an inherent good, but maybe I need to think more about what I mean when I say creation.
There’s an idea originally proposed by David Bye called the Workmanship of Risk. I’m going to dramatically oversimplify it, by boiling it down to the idea that when creating something there is a risk that the creator, or the artisan if we’re comfortable calling this art, can ruin the work. This is traditionally applied to physical objects made by hand, such as painting or sculpture or even the manufacturing of clothing by hand. A factory that stamps out forks can fail to make forks, but because of a failure in the machinery or even in the design stage, but the system is designed to make a fork the same way every time.
With the written word, it almost feels like there isn’t a risk of this nature. If I make a typo I can (and often do!) go back and fix it before publishing. So is the written word something hand made? I think it is if we expand the risk of failure to also consider the risk of lost time. If I spend hours writing a scene for a play (a thing I have done in the past) and at the end of that time, the scene has to be deleted from the final work that time has been lost. If I were a better playwright, I could have skipped writing the scene entirely.
So is the slop generated by computer programs regurgitating stolen art (written and visual) hand made? No. At best it’s manufactured, but it’s a black box system. You input a request and the machine spits something out on the other end that has a high mathematical probability of looking authoritatively like something you requested. But there’s no risk of the “prompt engineer” (ew, i felt gross even typing that) making a bad decision that can ruin the work. Because the prompt isn’t the output. It’s not a direct one to one process. You can enter the same prompt dozens or hundreds of times and get different results each time. Nothing the hypothetical artisan does has a direct impact on the result. Rather they are pushing buttons that whip random monkeys in the infinite room full of typewriters and sometimes something useful comes out.
This has gotten way from me a bit. Do I want to be a better writer? Yes. I don’t need to be the best, but I want to look back at my work seven years from now and see some improvement. The only problem is I don’t know what that improvement looks like.