One thing that LLMs will give us, is the ability to make really interesting and exciting tutorials. We can do that by hand, now, but it's a lot of work. I know of which I speak. I've been doing tutorials forever, but they haven't been particularly popular. Part of that, is that I'm pretty bad at self-promotion, and having that eau d'expert scent about you, is a big selling point. I really don't waste much time, trying to convince people that I have something they want, so they don't want it.
That part will still be necessary. I suspect LLMs can help polish and optimize self-promotion, but that also means everyone else will be doing it, and the competition is likely to be stiff.
I spent a lot of time thinking about the state of the industry and my feelings about it. I realized the time I most enjoyed and was happiest was when I was early in my career, learned a lot and tried to actively be a better software engineer.
I decided to "go back" to that place. I've started reading engineering books, using physical media instead of digital media, etc.
In addition to learning, I've been writing programs that are 'useless' but things I want to make (even if no one ever sees them). I'm also building a product with unconventional technologies simply because I enjoy it and think it's a better way of doing things.
I find it interesting that course/book authors are blaming LLMs and likely not the more obvious example of how the US has been in a faltering k-shaped economy for years where the industry you want sell material to has dealt with constant layoffs + anti-worker tactics over the same time period. If people aren't secure in their jobs, they aren't going to waste money buying something they don't need.
Also people just age out of buying this stuff, what use do I have for programming books after working 15 years in the industry? The docs have always been sufficient to learn on your own IME. That plus realizing there are other fun things to do with your one single life than constantly "upskilling."
Use to be a massive subscriber to frontend masters, but the site has honestly sucked for years where they rotate the same dev influencers teaching the same courses year after year. Hard to justify subscribing to something that hasn't really innovated or done interesting projects in half a decade.
For reference 50% of spending comes from the top 10% of earners. To be in the top 10% you need to make 250k. So it is definitely true that we have a K shaped economy, but I don’t think that’s a great explanation for why tech workers are spending less on tutorials.
Additionally many tech workers’ companies will subsidize spending on learning materials. So we should expect demand to be more stable than for things like bicycles which are or TVs which aren’t usually directly subsidized.
Programming tutorials are still alive, coding tutorials are dead.
Coding is going to be limited to hobbyists and specialized professionals, time of mass coding is behind us. Like the time of mass knowledge of assembly, human computers, etc.
Learning programming now is like learning how to use a cotton gin or something, its an antiquated hobby at best now, not a profession so it shouldnt be a surprise it isnt lucrative anymore
You're correct that writing code by hand has been heavily displaced by AI programming, but understanding how things work isn't really an "antiquated hobby" and this mentality causes a lot of people to incorrectly assume that latent knowledge of how to program is no longer useful.
It's still pretty undisputed that if you know how to write code, you can generate better AI code than someone who doesn't, and do it with less tokens. I think it's more like in university where in other engineering disciplines you learn the math and underlying concepts and are forced to do them by hand. This gives you an understanding of what you're doing, but in the real world you, ya know, just plug that stuff into matlab.
One thing that LLMs will give us, is the ability to make really interesting and exciting tutorials. We can do that by hand, now, but it's a lot of work. I know of which I speak. I've been doing tutorials forever, but they haven't been particularly popular. Part of that, is that I'm pretty bad at self-promotion, and having that eau d'expert scent about you, is a big selling point. I really don't waste much time, trying to convince people that I have something they want, so they don't want it.
That part will still be necessary. I suspect LLMs can help polish and optimize self-promotion, but that also means everyone else will be doing it, and the competition is likely to be stiff.
I would not pay for something so obviously extremely AI generated. If your sales are slipping, I don’t think this helps.
Strange… I have purchased more technical books this year (and read them) than the last five years combined.
What was the motivation?
I spent a lot of time thinking about the state of the industry and my feelings about it. I realized the time I most enjoyed and was happiest was when I was early in my career, learned a lot and tried to actively be a better software engineer.
I decided to "go back" to that place. I've started reading engineering books, using physical media instead of digital media, etc.
In addition to learning, I've been writing programs that are 'useless' but things I want to make (even if no one ever sees them). I'm also building a product with unconventional technologies simply because I enjoy it and think it's a better way of doing things.
I've essentially gone back to me, circa 2004.
Any recommendations?
I have all but too many recommendations... perhaps I should write about them and share it.
I immediately got the feeling this was AI generated, and sure enough 100% according to Pangram. Good thing his tutorials are dead I guess.
I looked one up yesterday. For Pure Data.
LLMs hopeless in that domain I have found
Don't worry, LLMs will do a great job once they add the tutorial you found to their training data.
I find it interesting that course/book authors are blaming LLMs and likely not the more obvious example of how the US has been in a faltering k-shaped economy for years where the industry you want sell material to has dealt with constant layoffs + anti-worker tactics over the same time period. If people aren't secure in their jobs, they aren't going to waste money buying something they don't need.
Also people just age out of buying this stuff, what use do I have for programming books after working 15 years in the industry? The docs have always been sufficient to learn on your own IME. That plus realizing there are other fun things to do with your one single life than constantly "upskilling."
Use to be a massive subscriber to frontend masters, but the site has honestly sucked for years where they rotate the same dev influencers teaching the same courses year after year. Hard to justify subscribing to something that hasn't really innovated or done interesting projects in half a decade.
For reference 50% of spending comes from the top 10% of earners. To be in the top 10% you need to make 250k. So it is definitely true that we have a K shaped economy, but I don’t think that’s a great explanation for why tech workers are spending less on tutorials.
Additionally many tech workers’ companies will subsidize spending on learning materials. So we should expect demand to be more stable than for things like bicycles which are or TVs which aren’t usually directly subsidized.
Programming tutorials are still alive, coding tutorials are dead.
Coding is going to be limited to hobbyists and specialized professionals, time of mass coding is behind us. Like the time of mass knowledge of assembly, human computers, etc.
[dead]
Learning programming now is like learning how to use a cotton gin or something, its an antiquated hobby at best now, not a profession so it shouldnt be a surprise it isnt lucrative anymore
You're correct that writing code by hand has been heavily displaced by AI programming, but understanding how things work isn't really an "antiquated hobby" and this mentality causes a lot of people to incorrectly assume that latent knowledge of how to program is no longer useful.
It's still pretty undisputed that if you know how to write code, you can generate better AI code than someone who doesn't, and do it with less tokens. I think it's more like in university where in other engineering disciplines you learn the math and underlying concepts and are forced to do them by hand. This gives you an understanding of what you're doing, but in the real world you, ya know, just plug that stuff into matlab.
Sure maybe for another 3 months