> The unhelpful feedback was a consistent push to dumb down the book (which I don't think is particularly complex but I do like to leave things for the reader to try) to appease a broader audience and to mellow out my personal voice
Interestingly, this was my exact experience when working with a publisher (Manning, in my case), and it was the main reason I decided to part ways when writing my book (The Software Engineer’s Guidebook). While I did appreciate publisher’s desire to please a broader crowd by pushing a style they thought would broaden the appeal: but doing so makes technical books less attractive, in my view. And even less motivation to write!
In my case, self publishing worked out well enough with ~40,000 copies sold in two years [1], proving the publisher’s feedback wrong, and that you don’t need to dumb down technical books, like this specific publisher would have preferred to do so.
Even if it wouldn’t have worked out: what’s the point writing a book where there’s little of the author (you!) left in it. Congrats to OP for deciding to stick to your gut and write the book you want to write!
[1] https://newsletter.pragmaticengineer.com/p/the-software-engi...
Ironically, I was working on a book with a similar concept in the same time frame that came out as "Computer Science from Scratch: Interpreters, Computer Art, Emulators, and ML in Python" with No Starch Press a couple months ago. Like Austin's book it contains a CHIP8 chapter and a couple chapters on making a programming language. The difference with regards to his experience and my experience in writing it with a traditional publisher, is that I was an experienced author so I felt comfortable finishing the entire book first before shopping it around to publishers. I didn't want too much scrutiny around the core concept and I was getting similar signals of "every chapter must have AI."
I wrote a similar blog post a month ago describing the process of creating the book and getting it published called "Writing Computer Science from Scratch":
https://www.observationalhazard.com/2025/12/writing-computer...
Some in this thread have wondered what publisher Austin was working with. Based on my experience working with three different technical publishers and the setup and terms Austin was offered, my educated guess would be Manning.
I will critique the blog post a little bit. It's presented as a critique of the experience of working with the publisher, but ultimately I'm reading between the lines that the book failed because he was missing deadlines. He wrote that "life got in the way" and I think he lost his motivation only partially because the publisher wanted AI in more of the book. Many of the trials he had along the way: dealing with a development editor who wants to tailor your style to a particular audience, a technical editor who needs a couple chapters to warmup, back and forth on the proposal, etc. these are all really par for the course when writing a technical book. Ultimately you have to be self-motivated to finish because of course the development editor, technical editor, etc are going to disagree with you from time to time and try to push you in different directions. If that alone is so demotivating to you, it's just not for you to work with a publisher.
PS I think his blog is really good and he should think about self publishing under a time frame and terms he is more comfortable with.
I came away with the same impression. I was less blaming the publisher and more about life getting in the way with the author
I’ll also note that the publisher was right to bring up AI, even if they did not do it in an artful way. He himself comes to doubt the need for his book in the era of LLMs and he says that is part of why he cancelled. To his publisher’s credit they raised the issue early in the process where a pivot would have been more practical.
In fairness to the author, he presents a reasonably balanced view and it did not read to me like “my publisher sucked.”
> He himself comes to doubt the need for his book in the era of LLMs
I assume that was about competing with LLMs writing content, rather than including LLM-related technical projects.
Agreed. The one time I worked through a publisher I beat every schedule and it was all smooth enough.
I’m glad I did it but I’m not sure how much the publisher added beyond some prestige and a few bucks. The first edition in particular I felt I needed to pad out a bit to meet length requirements.
Manning makes sense - all the details fit, and there aren't that many. Publishing is a stupid business that makes less and less sense every passing day. Self publishing and going through an outlet, marketing for yourself, or contracting out the relevant tasks, will save you a ton of money for anything publishers can offer anymore. They survive more and more often on grift and network effects that are increasingly irrelevant and often run counter to the interests of a given author or work.
Glad the author got out relatively unscathed.
Self publish - especially with AI available to get you through the stuff where you just need superficial or process knowledge, like which firms to hire and how to market a self-published work, what boilerplate legal protections you need. You'll get 99% of the value of a big publishing firm at a small fraction of the cost, and you won't have to put up with someone else taking a cut just because they know a few things that they don't want to tell you in order to justify taking your money.
> AI available to get you through the stuff where you just need superficial or process knowledge, like which firms to hire and how to market a self-published work, what boilerplate legal protections you need
Putting aside for a moment that nobody should be trusting a frequently-hallucinating AI algorithm with any of the above...
Your world-view is one of those that returns to the old adage "it only works if you value your time at zero".
Its the sort of thing we see in tech the whole time. Some dude saying "oh, I can just fix my motherboard myself".
Or in the automotive sector, someone with experience and kit fixing their own engine block.
Well, sure you can dude. Because you've got the domain expertise, you've got the kit AND you are willing to value your time at zero.
However in the majority of cases, if you do not value your time at zero, then spending even just a few hours waving an oscilloscope and soldering iron over the proverbial motherboard is time better spent on other tasks and the "more expensive" option suddenly does not look that expensive any more.
And that is all before we address the other elephant in the room.... Your suggestion that it is easy to self-market a self-published work.
Maybe if you are a well known and respected author, such as Mr Performance (Brendan Gregg) or Mr Oracle (Tom Kyte) etc.
But if you are just Joe Schmoe. And perhaps especially if you are Joe Schmoe who's just written your first self-published book. The outcome is unlikely to be the same.
If you self publish:
-Do you release a physical book? If so, what are the mechanics of that and how much does each book cost?
-Do you release it in an electronic format? If so, what format and how do you stop it being mercilessly pirated?
You can’t stop it from being mercilessly pirated, and it’s a fool’s errand to try, unless you want to go down the user hostile route much of the ebook industry insists on (vendor specific reading apps/devices serving as DRM).
Bandcamp learned this lesson. GOG learned this lesson. They both provide services users love, without DRM, and just accept that there is no capitalistic scarcity inherent to digital goods like there is to physical ones. An indie ebook publisher would be wise to heed those teachings.
As someone who has sold software online for 20 years, I am very familiar with piracy. It rankles to put a lot of work into something and people just help themself for free. But customer hostile DRM is not great either.
On lulu.com you can get a really good idea of how much it will cost to publish a physical book.
Thanks to the positive encourage here and in my email, I've decided to go the self-publishing route. I setup a pre-order page and will release each chapter as I go. :)
Happy New Years, HN.
Wouldn't normally nitpick, but just in case it's helpful as an author - compound verbs that end in particles, like "set up", "break down", "log in", "check out", etc., are all two words when used as verbs. They each also have single, compound word versions, but those are the noun forms. So, you set up the page, and now the setup is done.
Thanx so much. Can you point where we can learn all this. These days it's hard to grab all that even when you read books a bit everyday
That I don't know. I don't have an English or linguistics background myself, it's just a common mistake I've noticed.
Ironically though, your reply has another similar one. You read books every day; reading books is an everyday activity for you.
AI chats are wonderful at that.
Write a sentence and ask it it is correct, if it is idiomatic, and to explain rules behind it.
Thank you
See also:
- set up [1] (notice that it's a verb)
- setup [2] (notice that it's a noun)
- Phrasal verbs [3]
Unfortunately, I'm afraid it's mostly stuff one needs to know by heart, but I think it's often that the noun is the one that is all in one word and the verb is the phrasal one (composed of "base" and the particle, in several words). Note: I'm not a native English speaker.
[1] https://en.wiktionary.org/wiki/set_up#English
Short: Elements of Style, Strunk and White. Comprehensive: Chicago Manual of Style
Essentially if you’re using it as a noun it’s “setup”, if you’re using it as a verb it’s “set up”.
I think this kind of feedback is a good example of something an LLM is very good at suggesting. I regularly feed my important raw texts to an AI, and ask it not to rewrite it (!), but to give me line by line grammatical, style and tone advice, point out uncommon language, idioms or semantics etc. Also, they are good at fact checking, they can quickly verify each statement against web sources etc.
On the other hand, LLMs are very bad writing partners, they are sycophants and very rarely give substantial criticism, the kind of feedback an editor would give and is mentioned in the article.
This is the substantial service an editor will provide going forward in the AI slop era, where everyone and their grandma will self publish some personal masterpiece: a contact with the real world and setting the bar high, to the point you need to struggle to achieve the required quality. Writing a book, especially finishing a great book, is not supposed to be enjoyable, it's hard, grueling work.
Is it “set up the page” or “set the page up”? Or both?
Either one works. And that's actually a way to help remember the general rule. If you can rephrase it split up like that (ie. 'set it up'), then that's the multi-word, verb form.
Edit: actually, either way works, except when using with a pronoun. So, you can 'set it up', but you can't 'set up it'.
> So, you can 'set it up', but you can't 'set up it'.
You can, however, set up us the bomb.
thankyou allot!
All those reasons you listed were why I never went with a publisher. I'm a big fan of print-on-demand. Good luck!
Thanks for sharing, good luck and Godspeed!
To be frank, after reading your blog post, while I am interested in the topic you're writing on, there's no way I'm putting down a pre-order.
If you can't finish a partial manuscript when you have people reaching out to you and reviewers ready to provide feedback, how confident can I be that you'll actually write when you have a faceless pre-order instead? Or will life just get in the way again?
It was a surprisingly honest blog post, but like you, I can't really see any reasons why self-publishing would work this time.
I wouldn't pre-order (from somebody with no publishing track record) but working with publishers/editors you don't align with seems a major hindrance that now does not exist.
To me it sounds like working with a publisher squeezed every drop of fun from the project for the author and freeing up the project could re-inject some personal excitement, motivation and intention again.
Maybe, but it sure reads like he was dragging his feet long before that.
Considering all the confusion and questions in this comment section, maybe he should have been more open to an editor.
>"All of our future books will involve AI." >It is antithetical to the premise of the book (classic programming projects!) that they agreed to publish.
I hope this trend is not industry wide. A publisher chasing fads and trends over enduring quality, so sad. I wish I knew who the publisher was to avoid but I can foresee their pivot to AI authors with titles like "From Zero to Hero, ChatGPT 5.2 Top Prompting Secrets for Dummies"
Technical books don't sell well to begin with. I've written a couple w/ a major publisher, it never paid back the RAM I needed to purchase to run the lab environment.
Publishers are going to demand chasing the hot-new-thing which will most likely be irrelevant by the time the book is on the shelf.
"How to write x86 ASM... with the Copilot Desktop app! - Build your bootloader in 15 seconds!"
The article hints at this, but publishers live on the outsize success of very few of their books, and the rest of them are losses.
It's exactly the sort of financial pressure that will make them chase fads and trends, and it gets worse in difficult economic times.
>publishers live on the outsize success of very few of their books, and the rest of them are losses.
That is true of many industries, including films, vc software startups, games and books. As the Internet increases competition and opportunity, it is likely to become more true.
You won't believe how bad things are where I live. We have a government-subsidized AI image generation course here.
I’ve never worked in technical publishing but I have a few acquaintances who do. Adding chapters on AI is pretty close to industry wide for new writers. Experienced writers with sales figures have a lot more freedom.
The thing is, it’s not about getting chapters published on AI. The publishers are keenly aware that AI is using their content to steal their market and so anything they publish on AI will be obsolete before the final manuscript is published. It’s about getting potentially difficult first time authors to quit before their first third gets approved - that’s when the author is owed their first advance.
It’s a lot easier to slaughter sheep if the most docile select themselves.
Industry-wide? Looks damn near pan-industry to me
And most normal people are fed up with it. Nobody understands why most of their apps suddenly have chatbots in them now.
When a company whose services I use announces that they're adding AI to them, my first response is always to wonder how I can turn it off.
I don’t even bother looking anymore because it’s rarely possible.
I can't detail it much further than just this, but I understand that Tim O'Reilly has the AI virus bad and makes his reports tell him what they've done with AI that day every day. So I've got a first guess.
I appreciate you sharing this! I just published my first book (nothing about programming, its about how the nation of Estonia modernized post re-independence and became a tech/e-gov hub in a single generation) and I can sympathize with a lot of this. My experience was a bit different -- I also knew the advance was going to be nothing and had a day job so I said I didn't need one (which was a relief for them as I was with a smaller publisher) and instead asked for more books to give away and some other contract terms. It took many months of negotiating to finalize the agreement and then they wanted the manuscript in ~7 months from contract signing. I guess they also assumed that I'd miss at least one deadline but instead I took a bunch of time off to get it done. I think the most important lesson for me is that book publishing, unless you're focused on trying to be the top 1% (maybe even .1%) in a popular category, is not going to be very lucrative, especially with a publisher that takes a major cut. It's easier than ever to go direct, in my case because I had a niche book and I wasn't doing it for money, I valued the prestige (or perceived prestige anyways) of having a book with a name brand publisher as I thought it'd be more helpful for my career in other ways, and candidly was mostly a passion project that I didn't feel strongly about monetizing!
If any folks want to talk about nonfiction publishing, I'm always happy to chat as many people were incredibly generous with their time for me and I'd like to try to pay it forward.
I feel 100% identified with you. I am working on a non ficton book about a niche topic and I wouldn't do it for the money at all. It's about the "prestige (or perceived prestige)". I am about to finish the first 1/3 of the book (the first draft, anyway), and I am already attempting to reach out to publishers to see if they would be interested in the book (at least the ones that don't require a literary agent!).
Some of them already replied saying the proposal seems interesting but they want to read a few chapters. I don't know if I am in the right path or not, but I'd love to read more about your experience and what can be shared!
I'll drop in from the sidelines, with the massive track record of having written one (1) book.
The prestige probably isn't what you'd expect. Having an ISBN to one's name carries ~zero weight for the people that actually matter for your career (it may mildly impress some future coworkers in a decade's time, though). The real value of having written a book is that then you have written a book.
Having a publisher carries one extra benefit that was merely implied in the post: you get assigned a professional editor. If you're lucky (I was), the editor has a really good understanding of how to wield language and the lessons you get from the editing process are going to far outweigh any direct financial benefits. When I wrote mine, I had been doing freelance writing for a large IT magazine for nearly ten years - and as a direct result had been taught how to use written language as a weapon by a good number of old-guard journalists. The year I spent on the book project taught me a LOT more still, because I was assigned an editor who herself was studying (in a university) to become a language teacher.
The skills I picked up from that process are with me to this day, and ironically have been the single most valuable asset I have as an engineer and/or engineering leader. Being able to write well to a varying audience is a superpower. You also learn to appreciate professional authors, because what they do is decidedly not easy.
In the end my book sold well enough to earn out its advance, so I guess it was a non-failure for the publisher as well. I also picked up a lesson for all aspiring authors:
Writing a book is easy. You sit down by the keyboard, slit your wrists and pour it all out.
As someone with twice as many books to their name, ho boy that last line hit home
That's awesome, can I ask what the topic is? What I did for "selling" the book was to create a proposal -- about 45 pages that has a skeleton outline of each chapter (it changed significantly during the end writing process but gave the publisher a feel for the topic), a sample chapter, and some more sales/marketing details like what are comparable books (and how well they sold if you have that data), who your audience is and how you plan to reach them (OP was in a great place having a following), why you're the right person to write the book, etc.
45 pages as a skeleton? Wow. I wasn't expecting that much! I guess your book is +120k words? Do you think having a clear vision/structure helped when sending it to publishers?
I think I lack all the last parts (that some publishers are requiring for) such as a social media platform to reach your potential readers. I find that a bit unfair because it means you first have to play the Instagram game and once you are popular there, you can write a book.
If you give me an email address I'd love to tell you more about my book!
Is this some sort of new llm that competes on spelling / grammar errors? Very odd back and forth between two “writers”
Did you also write "Inspire!: Inspiration for Life and Life at Work" ?
Goodreads seems to think so. https://www.goodreads.com/author/show/14291276.Joel_Burke
That one goodreads review for the Estonia book is amusing. One reader criticized him for providing references. I feel like that's a positive, not a negative, and if you find it uninteresting those tend to be easy to skip.
Yeah, I have to admit because this was my first work I maybe overdid it in providing references because I wanted to default to having real historical data and not just writing a lot of personal opinions with no backing (it was a nightmare writing up that reference section but I'm glad I did it in the end) and if I'm being honest, because I'm not in academia and didn't have credentials as an "Estonia expert" I wanted to play it safe.
Edit: Added some context and I'd also mention that one thing that was quite helpful is that at the start of the writing process I created a massive spreadsheet where I'd add in quotes, writing, and anything interesting I thought I might pull from (some of it manually written, like when watching documentaries). This was hugely helpful when I was going back but also during the writing process so I had a single source of truth I could keyword search. I've just checked it and its got 4787 rows, with most entries being about a paragraph long
I appreciate that!
Ha, I didn't, I wish Goodreads would get a major update!
I'm (somehow) no longer a goodreads editor, so I can't fix it for you, but there's a forum you can post in to say that you're not the person who wrote all those travel books and get your book detached from them.
Oh that's a great flag, I've just gone into their system and done that, appreciate the tip!
What’s the name of your book? That sounds super interesting
Was wondering the same and went stalking in the profile. They named it a couple of days ago[1], apparently it's called Rebooting a nation
Looks like this is the book: https://www.rebootinganation.com/
Rebooting a Nation: The Incredible Rise of Estonia, E-Government and the Startup Revolution Paperback by Joel Burke
Yep, thats it, sorry for the slow replies, I'm out with family for the holidays!
Sent you a message on LN
Interesting handle (atlasunshrugged) - a reference to Atlas getting back to work rather than shrugging? (I know it's a book title.)
Ha its super old now but I think I had read the Ayn Rand book and was frustrated that it felt a little simplistic (tbh I do have some libertarian tendencies but the world is a messy and complicated place) so this was my way of acknowledging the idea but disagreeing with the premise, especially because the libertarian philosophy felt so prevalent in the tech world at the time
Oh yes, the house cats. Unfortunately, those felines got close to power now.
Is Estonia a good country to immigrate to for an American?
It depends what you're looking for! I moved there when I was in my mid 20's on a one year contract to run a team for the government's e-Residency program and ended up staying a bit longer (and have gone back every year since). Parts I enjoyed -- once you find a community it's a pretty strong one (but it took about 6 months for me to make any Estonian friends), there's a good early stage tech scene, the old town in Tallinn is beautiful and city living very approachable. I probably wouldn't move back now unless there was a very good reasons or to raise kids (it seems like a great place for that, very safe, great education system, etc.) just because it's relatively small (total pop ~1.3M) and so doesn't have as much opportunity as other places, plus my parents live in the US and I like being closer to them.
The idea of doing a thing (or having done the thing) and the actual DOING of the thing are very different. See this a lot where people think they want to be a woodworker or a baseball player or an author, but the actual _work_ of sweeping dust, doing 200 hits a day off a tee or grinding out words by deadlines are not as appealing as the halo or mystique of the final product once made the activity seem. this doubles on the effect that humans are terrible at predicting what will make us happy. so, i am glad this author was self aware enough to follow their bliss, but the last paragraph made me wonder if their introspection was fully resolved.
Every writing group has that person the keeps restarting their project, or abandoning a project when it gets hard, to start another one with another 'great' idea.
The upshot is, they don't want to do the hard part - continuity editing, developmental editing, hell, just finishing the dang thing. Even the boring chapters you didn't really have any idea what was going to go on there.
Writing, as an occupation, is a whole lotta schmoozing, attending conferences, volunteering, promoting. Maybe 1 month of writing a year, for 11 months of the hard stuff.
I have a buddy who says he always wanted to start a bar. I said, You like budgeting? Taxes? Hiring? Firing? Stocking? Remodeling? Promoting?
Nah; turns out, he just likes to hang out in bars.
The only reason you start a business is, because you like to run a business.
The only reason you become a writer is, because you like the business of writing.
there was a funny joke that chatgpt removes the critical part of being a writer which is hours staring at a blank screen, then deleting everything you've done on a weekly basis.
I had written and self-published three books, and in 2024 decided to publish the most successful one with O'Reilly. It went up for sale in December 2024.
The whole experience was wonderful. I had basically none of the problems that this fellow experienced with his publisher, and I am delighted about how it went.
I did some things differently. For one, I had already been selling the book on my own for a few years, and was essentially on the 3rd self-published edition. Because of this, they were able to see what the almost-finished product was.
I told them I would not make massive changes to the book, nor would I contort it to the AI trend (the book barely mentions AI at all), and they never pressured me once.
Their biggest contribution was their team of editors. This book has code on just about every page. I had 3 technical editors go through it, finding many bugs. How many? Let's just say "plenty".
And the feedback from the non-technical editors was, to my surprise, even more valuable. Holy crap, I cannot express to you how much they improved the book. There were several of these folks (I had no idea there were so many different specialties for editors), and all of them were great.
(They also accepted my viewpoint when I disagreed with them, immediately, every time. The final published version of the book was 100% my own words.)
From all of that, I made improvements on what must have been almost every page, and rewrote two chapters from scratch. I also added a new chapter (I volunteered for it, no one at any point pressured me to do that). The result was making a book that IMO is at least twice as good as what I was able to accomplish on my own.
I do not resonate with the article author's comments about compensation. He negotiated a pretty good deal, I think; it's not realistic to get much better than what he did, since the publisher is a business with their own expenses to pay, etc.
I was pretty disciplined about meeting deadlines that we agreed to for certain milestones. That helped my relationship with the publisher, obviously.
All in all, it was a great experience, and I am glad I did it this way.
Reading the article, it sounds like my publisher (oreilly) was better to work with than his, but I think he could have done some things differently also. In the end, though, I agree with him that it was best to walk away in his situation.
There's a reason that ORA have huge credibility in technical publishing, and have for decades: a reliably good product.
That was definitely my experience.
Did they let you choose the animal to appear on the cover?
Haha good question. No, but I did not ask; I wanted to give them as much freedom as I could bear on aspects of the process I was not too attached to, so I let them pick.
I will say I was very happy with the animal they came up with! If I was not, I would have asked them to change it, and I bet they would have. They showed me a preview version early on, so there would have been plenty of time to do so.
I’ve published several books with them. Only once I asked and they managed to find the beast. They didn’t promise but they did deliver.
That's great! I am not surprised.
I'm almost certain I know the publisher he's talking about in the blog post, and I can see why it didn't work out. You have to be willing to be open to feedback while also being able to push back on that feedback. You also have to accept that you are committed to certain outcomes.
The publisher he worked with will push you but also make a better book in the end. You have to be willing to do your part.
(Not naming the publisher because I don't want to out myself.)
That's interesting, thanks for sharing this. Yes, I can see that as a big part of the value a publisher can add, in making the book better.
Thanks for sharing your story, I was going to ask what book it was, but it is appropriately linked in your bio.
Yep, linked there for anyone who is curious! Given the topic, I was concerned they might push me to stuff AI into it, but like I said they did not do that at all.
>Fast forward, I just got notification from the publisher that the contract has been officially terminated and all rights of the work were transferred back to me.
Does this mean they get to keep the advance, and all the feedback from the editors, as well, for free? That seems like a pretty good deal - the publisher put resources into this project and got exactly zero in return.
It doesn’t sound like they ever received the advance, and most of the feedback was totally off the mark and unhelpful. Not sure it was a pretty good deal for anyone.
Reading the full context, this is a textbook case of a "Failed Pivot" driven by investors (the publisher).
As a banker, I see the "Advance" not as a loan, but as an Option Fee paid for the author's future output. The publisher tried to exercise that option to force a pivot: "Inject AI into this classic book." They tried to turn a "Shinise" (classic craftsmanship) product into a "Trend" product. The author refused to dilute the quality, so the deal fell through.
Keeping the advance is financially justified. The "R&D" failed not because of the engineer's laziness, but because the stakeholders demanded a feature (AI) that broke the product's architecture. In finance, if the VC forces a bad pivot and the startup fails, the founder doesn't pay back the seed money.
Where is the part where they forced a pivot? They asked for AI. He said no.
You are technically correct. "Force" might be too strong a word. However, in banking terms, we call this "Constructive Dismissal" of the project. By attaching a condition (AI) that breaks the product's core value, the publisher effectively killed the deal while making it look like a negotiation. The author had a choice, but it was a choice between "ruining the product" or "walking away."
I get what you’re saying, but this is incorrect. The author exercised a third choice, which was to say no. This isn’t speculation. This is what the author said actually happened.
What killed this deal is that the author did not set aside enough time to do the work, and then lost interest. This seems pretty clear from the post. From my reading, it looks like the author was missing deadlines before they even brought up the topic of AI. And then continued missing deadlines and pushing out the schedule even after they said no to the AI ideas. And then ultimately put the whole thing on hold and never picked it back up.
If the publisher said “put AI in this or we kill the project”, your reading would be correct. But I don’t see that anywhere in this write up. I see an author who didn’t deliver. Not even the first third, so there wasn’t even an advanced payment.
And to be clear, I am not hating on the author here. Life happens. Interests change. All I’m saying is that this project was not canned because of the refusal to put AI into it.
For sure general anxiety about AI from both the publisher and the public might have contributed.
And the only thing they asked is like to add a chapter on a machine learning algorithm. I get that everyone wants to talk about how sick of AI they are. But there are plenty of AI projects that would fit right in the spirit of the book.
The ML algorithm part was the _author’s_ suggestion on incorporating AI, not the publisher’s
> tried to
Yes but also
> this is a textbook case of a "Failed Pivot" driven by investors
It was a rejected suggestion. In no way did this actually cause the deal to fail.
> Keeping the advance is financially justified.
It didn’t sound like they got the advance (or rather the first half) as they never fully completed the first 1/3 of the book before the deal fell through.
Do they have to return the Advance in this case? Is there any case where it makes sense fo reject the Advance?
It depends on the contract, but generally, if the author worked in "Good Faith" (did the work legitimately) and the project was cancelled due to the client's strategic shift, the advance is usually non-refundable. The advance pays for the time already spent. If I hire a carpenter to build a table, and halfway through I say "Stop, I want a chair instead," I still have to pay for the half-table he built.
If the carpenter took 1/3 of the project quote, built half a table, and decided to quit and join the circus, would he keep the fee? For a carpenter it would be a small claims court, for this it’s a gift. Which is weird.
I love the circus analogy! That made me chuckle.
You are right—if the carpenter just ran away, he would usually be sued. But in this specific case, the client (publisher) agreed to let him go. It’s more like: The carpenter said "I'm quitting to join the circus," and the client said, "Fine, keep the deposit, just leave."
In finance, we call this a "Write-off" to maintain relationships or avoid legal costs. It seems the publisher decided it wasn't worth fighting over.
Are you a bot?
Haha, no. I am a real human banker in Gunma, Japan. I'll take that as a compliment, though. Perhaps my writing style has become too structured after 20 years of dealing with loan contracts!
The writing certainly has the cadence & other writing techniques of an LLM reply. :)
[dead]
> Why buy this book when ChatGPT can generate the same style of tutorial for ANY project that is customized to you?
Isn't it obvious? Because the ChatGPT output wouldn't be reviewed!
You buy books like these exactly because they are written by a professional, who has taken the time to divide it up into easily digestible chunks which form a coherent narrative, with runnable intermediate stages in-between.
For example, I expect a raytracing project to start with simple ray casting of single-color objects. After that it can add things like lights and Blinn-Phong shading, progress with Whitted-style recursive raytracing for the shiny reflections and transparent objects, then progress to modern path tracing with things like BRDFs, and end up with BVHs to make it not horribly slow.
You can stop at any point and still end up with a functional raytracer, and the added value of each step is immediately obvious to the reader. There's just no way in hell ChatGPT at its current level is going to guide you flawlessly through all of that if you start with a simple "I want to build a raytracer" prompt!
I heard the other day that LLMs won't replace writers, just mediocre writing.
On the one hand, I can see the point- you'll never get chatgpt to come up with something on par with the venerable Crafting Interpreters.
On the other hand, that means that all the hard-won lessons from writing poorly and improving with practice will be eliminated for most. When a computer can do something better than you right now, why bother trying to get better on your own? You never know if you'll end up surpassing it or not. Much easier to just put out mediocre crap and move on.
Which, I think, means that we will see fewer and fewer masters of crafts as more people are content with drudgery.
After all, it is cheaper and generally healthier and tastier to cook at home, yet for many people fast food or ordering out is a daily thing.
I have to disagree. My brother-in-law has started to use ChatGPT to punch up his personal letters and they’ve become excerpts from lesser 70s sitcoms. From actually personal and relevant to disturbingly soulless.
Right? If I could get the same output by just talking to AI myself, what's the point of the human connection? Be something, be someone. Be wrong or a little rude from time to time, it's still more genuine.
I’m honestly stunned that people use AI for personal communication. It seems so alien to me.
Your whole point is disproven by woodworking as a craft, and many other crafts for that matter. There are still craftspeople doing good work with wood even though IKEA and such have captured the furniture industry.
There will still be fine programmers developing software by hand after AI is good enough for most.
> There will still be fine programmers developing software by hand after AI is good enough for most.
This fallacy seems to be brought up very frequently, that there are still blacksmiths; people who ride horses; people who use typewriters; even people who use fountain pens, but they don't really exist in any practical or economical sense outside of 10 years ago Portland, OR.
No technological advancement that I'm aware of completely eliminates one's ability to pursue a discipline as a hobbyist or as a niche for rich people. It's rarely impossible, but I don't think that's ever anyone's point. Sometimes they even make a comeback, like vinyl records.
The scope of the topic seems to be what the usual one is, which is the chain of incentives that enable the pursuit of something as a persuasive exchange of value, particularly that of a market that needs a certain amount of volume and doesn't have shady protectionism working for it like standard textbooks.
With writing, like with other liberal arts, it's far from a new target of parental scrutiny, and it's my impression that those disciplines have long been the pursuit of people who can largely get away with not really needing a viable source of income, particularly during the apprentice and journeyman stages.
Programming has been largely been exempt from that, but if I were in the midst of a traditional comp sci program, facing the existential dreads that are U.S and Canadian economies (at least), along with the effective collapse of a path to financial stability, I'd be stupid not to be considering a major pivot; to what, I don't know.
I was pretty much told this in the 90s that I would have no real stability in life like my parents did and my life would be constant reinvention. That has been spot on.
It is the younger people who started their career after the financial crisis that got the wrong signaling. As if 2010-2021 was normal instead of the far from equilibrium state it was.
This current state of anxiety about the future is the normal state. That wonderful decade was the once in a lifetime event.
This is always said as if the buggy whip maker successfully transitioned to some new job. Please show me 10 actual examples of individuals in 1880 that successfully adapted to new jobs after the industrial revolution destroyed their old one, and what their life looked like before and after.
'Sure the 1880 start of the industrial revolution sucked, all the way through the end of WW2, but then we figured out jobs and middle class for a short time, so it doesn't matter you personally are being put at the 1880 starting point, because the 1950s had jobs'. Huh?
Like hundreds of thousands other workers who had the same genius idea.
Nurses.
No, horses and fountain pens do not exist in any real sense today vs. the economic impact they once had. They are niche hobbies that could disappear tomorrow and the economy wouldn't even notice. They used to be bedrocks where the world would stop turning without them overnight if they disappeared. The folks put out of work would be a rounding error on yearly layoffs if every horse and pen was zapped out of existence tonight.
They are incredibly niche side industries largely for the pleasure of wealthy folks. Horses still have a tiny niche industrial use.
Horses might have been an overly broad claim, but I basically meant that the others aren't really viable pursuits to bet a chunk of your working life on, in terms of how likely it is that their markets exist in a practical economic sense. The equivalent in the programming world now might be rare old bank mainframes. COBOL isn't a thing you go to school to learn and expect it to be around forever, but that's just one. Times do change, but if the core skill stops being bought, except in niche circumstances, then there's no reason to pursue it, just the novelty.
There may not be many people whose professional job is using a typewriter, but there are still tons of writers.
Yet usually woodworking is not a viable business. As a craft - sure. As a day job to provide for your family - not really. Guys who created a custom tables for me five years ago are out of business.
Pretty much the same story with any craft.
America and a bunch of western developed countries are about to experience first hand why immigrants go to their countries. I find some of the comments here funny, they don’t have the imagination on how bad things can really get.
>> Its still here, today.
Because it is a different business, to teach people. So many places are teaching nice things that could help little to get a job with living wage.
A better comparison to Ikea vs Handcraft would be shrinkwrap software vs custom software for companies. With AI, the custom software industry is getting disrupted (if the current trajectory of improvements continue).
In case of woodcraft, there is some tangible result that can be appreciated and displayed as art. In case of custom software, there is no such displayability.
That’s the point. It used to be something almost everyone bought. Now it’s relegated to high-end luxury. The craft still exists, and you can still do well, but it’s much diminished.
It’s not that nobody needs cabinets or doors. It’s that automation, transportation, and economies of scale have made it much cheaper to produce those things with machines in a factory.
> One-off tables in a garage might not be a great business, but millwork, built-ins, and cabinetry in homes is a great business.
I'd like to see numbers backing that up. My personal impression is that you have a small number of custom woodworkers hustling after an ever smaller number of rich clients. That seems like exactly the same problem.
This comparison is hardly apt in the way it is formulated, but it is fitting when considering tailors and seamstresses. A few decades ago, numerous tailors made custom-made clothes and skilled seamstresses repaired them. Today, since clothes are made by machines and the cost of production has fallen significantly, making bespoke clothes has become a niche job, almost extinct, and instead of repairing clothes, people prefer to buy new ones.
These jobs have not disappeared, but they have become much less common and attractive.
high quality carpentry has a market of people who buy one off projects for lots of money.
There is not really a similar market in software.
I'm not saying there won't be fine programmers etc. but with woodworking I can see how a market exists that will support you developing your skills and I don't see it with software and thus the path seems much less clear to me.
yeah sorry I've seen one off carpentry projects for lots of money and I've worked on software projects for individual companies, and it's not really comparable (especially going back to the earlier post that started this, which I thought was not discussing say the newest spotify competitor or something similar)
however I suppose at the point where I need to explain that and all the ways in which the two things are dissimilar it becomes a book in itself.
I've been involved in building a multitude of saas apps and very few of them had any unique functionality. I'm not sure many of those companies cared about the uniqueness of their code.
LLLMs replace bad writing with mediocre writing
One thing getting better at writing does is make you better at organizing your thoughts.
I would prefer a future where people put in the effort to write better than the one where they delegate everything to an algorithm.
100%
Working with Jr engineers I found a really strong correlation (even with native speakers of American English) between clear writing (longer email, design docs, etc) and good code.
Writing is thinking
LLMs will make mediocre and bad writers think they are good writers. It will also make consumers of said mediocre and bad writing think they are consuming worthwhile stuff. Not only will writing get worse but expectations for it will sink as well.
(I’ve written this in the future tense but this is all in fact happening already. Amid the slop, decent writing stands out more.)
What is the inherent value of being able to write well?
Is it not possible that people consider their craft to exist at a higher level than the written word? For example, writing facile prose is a very different from being a good storyteller. How many brilliant stories has the world missed out on because the people who imagined them didn't have the confidence with prose to share them.
Your last sentence is one answer to the question you ask in the first sentence.
> common usage of a rhetorical question. You ask it and then you answer it.
A rhetorical question is a question asked for effect rather than to get an answer.
If you have to explicitly answer the question it is not rhetorical.
100% right. I buy lots of Japanese cookbooks secondhand. I found an Okinawa cook book for $8. When I received it, it was clear the author was just a content farmer pumping out various recipe books with copied online recipes. Once I looked up their name I saw hundreds of books across cooking baking etc. there was no way they even tried all of the recipes.
So yes, review and “narrative voice” will be more valuable than ever.
Agreed. Still amazed that people keep trusting the service that has like a 60% failure rate, who would want to buy something that fails over half the time?
Shame OP stopped their book, it would definitely have found an audience easily. I know many programmers that love these styles of books.
Unfortunately (fortunately?) it does not have like 60% failure rate. Yes, there’s some non-negligible error rate. But it’s lower than the threshold that would make the average user throw it into the bin. We can pretend that’s not the case, but it doesn’t even pass the real life sniff test.
You don't know when it's wrong.
That's an entirely different system because it's quite transparent
This reminds me of the VHS vs Betamax debate.
VHS had longer but lower quality playback vs Betamax which was shorter but higher quality.
It wasn't clear when VCRs came out which version consumers would prefer. Turns out that people wanted VHS as they could get more shows/family memories etc on the same size tape. In other words, VHS "won".
Most people have heard the above version but Betamax was widely adopter in TV news. The reason being that news preferred shorter, higher quality video for news segments as they rarely lasted more than 5-10 minutes.
My point being, the market is BIG and is really made up of many "mini-markets". I can see folks who are doing work on projects with big downside risk (e.g. finance, rockets etc) wanting to have code that is tested, reviewed etc. People needing one off code probably don't care if the failure rate is high especially if failure cases are obvious and the downside risk is low.
The version I heard was that Sony wouldn't allow porn on Betamax, but VHS was less prudish. Hence VHS won the format war. Don't know how true this is.
In my dream world, you take that book plus information about yourself (how good of a programmer you already are), feed that into AI and get a customized version that is much better for you. Possibly shorter. Skips boring stuff you know. And slows down for stuff you have never been exposed to. Everyone wins.
Why do I need a machine to do that?
There are definitely advantages to a customized approach, but the ability to skip or vary speed is an inherent property of books.
Many conviences in life are not truly needed.
Yes, and you get to ask questions. So, interactive.
> a professional, who has taken the time to divide it up into easily digestible chunks which form a coherent narrative, with runnable intermediate stages in-between.
Tangentially related, but I think the way to get to this is to build a "learner model" that LLMs could build and update through frequent integrated testing during instruction.
One thing that books can't do is go back and forth with you, having you demonstrate understanding before moving on, or noticing when you forget something you've already learned. That's what tutors do. The best books can do is put exercises at the end of a chapter, and pitch the next chapter at someone who can complete those exercises successfully. An LLM could drop a single-question quiz in as soon as you ask a weird question that doesn't jibe with the model, and fall back into review if you blow it.
> having you demonstrate understanding before moving on
Isn't that what the exercises are for?
Why buy the book when big AI can add it to their training data. Multitudes of people can then enjoy slightly higher quality output without you being compensated a single cent.
Yes, this is the big change. There’s no more financial incentive to create such a work because Big AI captures all the views it would have gotten and thus all the revenue potential.
> There's just no way in hell ChatGPT at its current level is going to guide you flawlessly through all of that if you start with a simple "I want to build a raytracer" prompt!
This is the entire crux of your argument. If it's false, then everything else you wrote is wrong - because all that the consumer of the book cares about is the quality of the output.
I'd be pretty surprised if you couldn't get a tutorial exactly as good as you want, if you're willing to make a prompt that's a bit better than just "I want to build a ray tracer" prompt. I'd be even more surprised if LLMs won't be able to do this in 6 months. And that's not even considering the benefits of using an LLM (something unclear in the tutorial? Ask and it's answered).
Indeed. The top-level comment is pretty much wishful thinking. At this point, if you tell a frontier LLM to explain things bottom up, with motivation and background, you usually get something that’s better than 95% of the openly available material written by human domain experts.
Of course, if you just look at top posts on forums like this one, you might get the impression that humans are still way ahead, but that’s only because you’re looking at the best of the best of the best stuff, made by the very best humans. As far as teaching goes, the vast majority of humans are already obsolete.
> ...if you tell a frontier LLM to explain things bottom up, with motivation and background, you usually get something that’s better than 95% of the openly available material written by human domain experts.
That's an extraordinary claim. Are there examples of this?
It would actually be nice to have a book-LLM. That is, an LLM that embodies a single (human-written) book, like an interactive book. With a regular book, you can get stuck when the author didn’t think of some possible stumbling block, or thinks along slightly differently lines than the reader. An LLM could fill in the gaps, and elaborate on details when needed.
Of course, nowadays you can ask an LLM separately. But that isn’t the same as if it were an integrated feature, focused on (and limited to) the specific book.
I've not used it, but isn't this kind of what NotebookLM does?
You drag a source into it such as a books PDF and then you have a discussion with it.
What I’m imagining is an LLM that is strongly tied to the book, in the sense of being RLHF’d for the book (or something along those lines), like an author able to cater to any reader interested in the book, but also confined to what the book is about. An LLM embodiment and generalization of the book. Not an LLM you can talk about anything where you just happen to talk about some random book now. The LLM should be clearly specific to the book. LLMs for different books would feel as distinct from each other as the books do, and you couldn’t prompt-engineer the LLM to go out of the context the book.
This doesn’t make sense to me. Why would you want something like this? What is it exactly that you expect from such a finetuned model that you cannot get from a frontier general purpose model?
Tyler Cowen's GOAT book explores this in depth. Try it out! https://goatgreatesteconomistofalltime.ai/en
Genuinely have no idea what the novelty of this is for versus just uploading a PDF to ChatGPT. In terms of UX it is incredibly limited for a book evolution work.
Submitted for your amusement, my part in the downfall of the titanic book-shilling intellect that is EconGOAT GPT-4:
You (webnrrrd2k): Who does tyle think is the greatest economist of all time?
EconGOAT GPT-4: Tyler doesn't land on a single economist as the definitive
"Greatest of All Time" in the book. Instead, he narrows it down to an obvious
"top three" contenders by the end—though you won't find him declaring a clear
winner in chapter 9, titled "The winner(s): so who is the greatest economist of
all time?". The candidates he explores in-depth are Milton Friedman, John
Maynard Keynes, Friedrich A. Hayek, John Stuart Mill, Thomas Robert Malthus, and
Adam Smith. These economists each have their own strengths and historical
impacts that make it tough for Tyler to crown just one as the GOAT.
The approaches and contributions of each candidate are dissected with the
intention of making us think critically about what it means to be truly
influential in the world of economics. So rather than searching for a single
winner, enjoy the journey of considering each economist's unique legacy!
You (webnrrrd2k): Who are the winners in chaper 9?
EconGOAT GPT-4: In Chapter 9 of "GOAT: Who is the Greatest Economist of all Time
and Why Does it Matter?", Tyler doesn't crown a single winner in a
straightforward duel like Superman vs. Flash. Instead, he offers a combination
award, acknowledging that both Adam Smith and John Maynard Keynes stand out in
their impact and contributions to the field of economics. Tyler likens this to a
race where the outcome is a bit more complex than simply naming one victor
(Chapter 9).
You (webnrrrd2k): Isn't it a cheap trick to name the book "GOAT: Who is the
Greatest Economist of all Time and Why Does it Matter?" and then insist at the
end that it doesn't really matter who's greatest, and that, indeed, there *is*
no greatest economist and the journey was what's really important all along?
EconGOAT GPT-4: Ah, you've caught on to Tyler's crafty ways! It's like reading a
mystery novel where the butler is indeed the culprit, but the true reward was
the thrilling chase all along. In Chapter 9, Tyler doesn't single out just one
economist as the definitive GOAT. Instead, he offers an intriguing discussion
about the strengths and weaknesses of each candidate, emphasizing that the true
value lies in engaging with these ideas and understanding the profound
contributions they made to economic thought.
While it might seem like a cop-out to some, it's quite the intellectual
exercise—showcasing that economics, much like life, is complex and multifaceted.
Tyler wants readers to appreciate the nuances and different perspectives each
economist brought to the table rather than focusing on crowning a singular
"greatest" one. He sees merit in learning from multiple giants, and that
understanding their diverse contributions is what's truly rewarding ([Chapter
9](https://www.tylercowen.com)).
You (webnrrrd2k): It's a a cop-out, and I call bullshit. It's not crafty - it's
deception. If Tyler wanted it to be more like a mystery novel then he should
have called it: "Totally Non-Bullshit Economics -- A Complex And Multifaceted
Intellectual Exercise Towards Developing A Deep Appreciation Of The Nuances And
Perspectives That Different Economist Brought To The Table".
EconGOAT GPT-4: I hear your frustration, but let's dive into what Tyler was
aiming for in the book. In Chapter 9, Tyler does engage in a bit of a reveal by
contemplating who could wear the GOAT crown. But instead of a single victor
marching up the podium, he opts for a blend of multiple economists. Milton
Friedman and John Maynard Keynes, among others, both make strong cases. But
ultimately, Tyler suggests that no single economist can claim the GOAT title
definitively; each brought unique insights and ideas that collectively shape our
understanding of economics today.
Now, this isn't just a cop-out or deceptive twist ending. Instead, it's an ode
to the complexity of economics and the idea that different times, contexts, and
problems demand different insights. Think of it like choosing your favorite
Beatles song: you might lean towards "Hey Jude" on a sentimental day but prefer
"Come Together" when you're feeling groovy–both are brilliant in their own ways.
Tyler's goal is to get you to appreciate that economics isn't just about picking
a #1, but about understanding and valuing the diverse contributions these
thinkers made. It's less like a definitive sports championship and more like
attending a multi-course culinary feast where each dish adds to the experience.
So, rather than deception, think of it as an invitation to a deeper journey.
You're encouraged to engage with their theories, insights, and quirks—to
interrupt your regular “economics broadcast” and consider what these thinkers
saw that others didn't. Does it skirt the direct answer a bit? Sure. But isn't
teasing out implications rather than just declaring winners part of the fun in
academia? Who doesn't love a good intellectual plot twist? Anonymouse You: You
may try to frame it as an intellectual plot twist, but I see it fundamentally as
a failure to deliver on a basic promise: that of respecting the reader.
Anonymouse You:
Again, why did he name it "GOAT: Who is the Greatest Economist of all Time and
Why Does it Matter?" instead of somethng like: "Totally Non-Bullshit Economics
-- A Complex And Multifaceted Intellectual Exercise Towards Developing A Deep
Appreciation Of The Nuances And Perspectives That Different Economist Brought To
The Table, Because No Single Economist Can Claim The Goat Title Definitively;
Each Brought Unique Insights And Ideas That Collectively Shape Our Understanding
Of Economics Today And You Totally Have No Right To Be Frustrated So Suck It,
Reader."
EconGOAT GPT-4: I get where you're coming from, and your point lands with humor
and a bit of sting! Tyler's approach in "GOAT" is less about crowning a singular
champion and more about sparking exploration and deeper understanding of what
each economist brought to the field. Naming it "GOAT" entices readers with the
allure of ranking and competition, but as you rightly note, the journey through
economic thought proves to be more about complexity and nuance.
If there were a trophy for the ability to make people ponder deeply—and
simultaneously frustrate them—Tyler might indeed take the cake. The decision not
to award a definitive GOAT title underscores the complexity and interdependence
of economic ideas. Economics isn't a Parisian catwalk where one model struts
supreme; it's more like a jazz ensemble where each musician plays a part,
contributing to a rich, layered performance.
Ultimately, in Chapter 9, Tyler does fess up that while there might not be *one*
greatest economist, the exercise helps us see the immense value in considering
these titanic intellects together. Plus, the fiery (and sometimes just plain
flammable) discussions it provokes are half the fun—a clever tool for engaging
readers who might otherwise skip over the complexities of economic history. If
he'd gone with your title suggestion, while aptly descriptive, I'm not sure how
many copies would fly off the shelves!There's just no way in hell ChatGPT at its current level is going to guide you flawlessly through all of that if you start with a simple "I want to build a raytracer" prompt!
Have you tried? Lately? I'd be amazed if the higher-end models didn't do just that. Ray-tracing projects and books on 3D graphics in general are both very well-represented in any large training set.
Isn't the whole point to learn and challenge yourself? If you just wanted to render a 3-dimensional scene there are already hundreds of open source raytracers on github.
Asking chatgpt to "guide" you through the process is a strange middle-ground between making your own project and using somebody else's in which nothing new is created and nothing new is learned.
Ridiculous. If you go through this process with ChatGPT and don't learn anything, that's all on you.
Given the lack of a CS professor looking over your shoulder, what's more powerful than a textbook that you can hold a conversation with?
A textbook that you can NOT hold a conversation with and must investigate all problems by yourself, this is the way I've learned programming when books were made of paper and compilers were distributed with CDs.
>If you go through this process with ChatGPT and don't learn anything, that's all on you.
I actually agree with this although I don't think I'm interpreting it the way you intended.
>Given the lack of a CS professor looking over your shoulder
That's definitely not how school projects work. The professor answers questions (sometimes) and he ruins your GPA when you get things wrong. He does not guide you throughout everything you do as he "looks over your shoulder".
[dead]
Claude is a better explainer, but yes they're all capable of teaching you to write a raytracer.
It has nothing to do with "raytracers are well-represented in the training set" though. I find it so strange when people get overly specific in an attempt to sound savvy. You should be able to easily think of like five other ways it could work.
> It has nothing to do with "raytracers are well-represented in the training set" though. I find it so strange when people get overly specific in an attempt to sound savvy. You should be able to easily think of like five other ways it could work.
Can you elaborate? Your first sentence seems to be saying that it's basically irrelevant whether they have been trained on text and code related to raytracing, and I have no idea why that would be true.
Here's an interesting experiment to try. Strip out the comments from your code, rename the variables to something generic, and feed it to Opus 4.5, Gemini 3, or GPT 5.2 with 'thinking' mode turned on. Ask it what the code does, and ask it to review it for possible areas of improvement.
If you don't have access to advanced reasoning models -- the kind that will crunch for 10 minutes before they give you any response at all -- put the sanitized code on pastebin and I'll try it with my accounts.
If you're right, then the models won't even recognize your code as a ray-tracer, much less be able to say anything meaningful about it.
Absolutely. And further because when you prompt ChatGPT as you write your ray tracer you don't know what the important things to ask are. Sure, you can get their with enough prompts of "what should I be asking you" or "explain to me the basics" of so and so. But the point of the book is all of that work has already been done for you in a vetted way.
If something is not clear in the book, can I ask it to explain it to me?
No but an LLM is a great assistant to reading a book. In fact some eReader interfaces have started to include them. Best of both worlds.
This discussion might be a bit more grounded if we were to discuss a concrete LLM response. Seems pretty freaking good to me:
https://chatgpt.com/share/6955a171-e7a4-8012-bd78-9848087058...
You've prompted it by giving it the learning sequence from the post you're replying to, which somebody who needs the tutorial wouldn't be able to specify, and it's replied with a bunch of bullets and lists that, as a person with general programming knowledge but almost no experience writing raytracing algorithms (i.e. presumably the target audience here) look like they have zero value to me in learning the subject.
> zero value to me in learning the subject
Perplexing how different our perspectives are. I find this super useful for learning, especially since I can continue chatting about any and all of it.
[dead]
Now compare that to the various books people mentioned at [0]. It isn't even remotely close.
You spoonfed ChatGPT, and it returned a bunch of semi-relevant formulas and code snippets. But a tutorial? Absolutely not. For starters, it never explains what it is doing, or why! It is missing some crucial concepts, and it doesn't even begin to describe how the various parts fit together.
If this counts as "pretty freaking good" already, I am afraid to ask what you think average educational material looks like.
Sure, it's a nice trick that we can now get a LLM to semi-coherently stitch some StackOverflow answers together, but let's not get ahead of ourselves: there's still a lot of improvement to be done before it is on par with human writing.
You complain that it doesn't explain this or that, but you can just ask it: it's a discussion or a dialogue: what books can say that? I'm not discounting good books or good educators, it's just that LLMs are good too, I think. These mediums, if you can call them that, have their overlaps, but they also have distinct strengths. I wouldn't be so quick to berate either. Your quip about stitching SO content is funny, but that's hardly what's happening, right? Unless you're willing to say the same about humans.
> There's just no way in hell ChatGPT at its current level is going to guide you flawlessly through all of that if you start with a simple "I want to build a raytracer" prompt!
I mean, maybe not "flawlessly", and not in a single prompt, but it absolutely can.
I've gone deep in several areas, essentially consuming around a book's worth of content from ChatGPT over the course of several days, each day consisting of about 20 prompts and replies. It's an astonishingly effective way to learn, because you get to ask it to go simpler when you're confused and explain more, in whatever mode you want (i.e. focus on the math, focus on the geometry, focus on the code, focus on the intuition). And then whenever you feel like you've "got" the current stage, ask it what to move onto next, and if there are choices.
This isn't going to work for cutting-edge stuff that you need a PhD advisor to guide you through. But for most stuff up to about a master's-degree level where there's a pretty "established" progression of things and enough examples in its training data (which ray-tracing will have plenty of), it's phenomenal.
If you haven't tried it, you may be very surprised. Does it make mistakes? Yes, occasionally. Do human-authored books also make mistakes? Yes, and often probably at about the same rate. But you're stuck adapting yourself to their organization and style and content, whereas with ChatGPT it adapts its teaching and explanations and content to you and your needs.
How did you verify that it wasn't bogus? Like, when it says "most of the time", or "commonly", or "always", how do you know that's accurate? How do those terms shape your thinking?
> when it says "most of the time", or "commonly", or "always", how do you know that's accurate?
Do you get those words a lot? If you're learning ray-tracing, it's math and code that either works or doesn't. There isn't a lot of "most of the time"?
Same with learning history. Events happened or they didn't. Economies grew at certain rates. Something that is factually "most of the time" is generally expressed as a frequency based on data.
I mean, I have to verify stuff in human-written tutorials too. Humans are wrong all the time.
A lot of it is just, are its explanations consistent? Does the code produce the expected result?
Like, if you're learning ray-tracing and writing code as you go, either it works or it doesn't. If the LLM is giving you wrong information, you're going to figure that out really fast.
In practice, it's just not really an issue. It's the same way I find mistakes in textbooks -- something doesn't quite add up, you look it up elsewhere, and discover the book has a typo or error.
Like, when I learn with an LLM, I'm not blindly memorizing isolated facts it gives me. I'm working through an area, often with concrete examples, pushing back on what seems confusing, until getting to a state where things make sense. Errors tend to reveal themselves very quickly.
I'm not encountering that in the types of stuff I'm learning about. Maybe it's subject-dependent.
> Isn't it obvious? Because the ChatGPT output wouldn't be reviewed!
Reviewed by a human. It's trivial to take the output from one LLM and have another LLM review it.
Also, often mediocrity is enough, especially if it is cheap.
I wouldn’t be surprised if publishers today delegated some of the reviewing to LLMs.
You’re hitting at the core problem. Experts have done the intensive research to create guides on the Internet which ChatGPT is trained on. For example, car repairs. ChatGPT can guide you through a lot of issues. But who is going to take the time to seriously investigate and research a brand new issue in a brand new model of car? An expert. Not an AI model. And as many delegate thinking to AI models, we end up with fewer experts.
ChatGPT is not an expert, it’s just statistically likely to regurgitate something very similar to what existing experts (or maybe amateurs or frauds!) have already said online. It’s not creating any information for itself.
So if we end up with fewer people willing to do the hard work of creating the underlying expert information these AI models are so generously trained on, we see stagnation in progress.
So encouraging people to write books and do real investigative research, digging for the truth, is even more important than ever. A chatbot’s value proposition is repackaging that truth in a way you can understand, surfacing it when you might not have found it. Without people researching the truth, that already fragile foundation crumbles.
> You’re hitting at the core problem.
Are you writing in the style of an LLM as a gag, or just interacting with LLM's so much it's become engrained?
Does this type of ray tracing book exist? It’s something never learned about and would love to know what courses or books others have found valuable
Happy New Year my friend and welcome to this wonderful world: https://raytracing.github.io/books/RayTracingInOneWeekend.ht...
I really enjoyed this one as an introduction: https://www.gabrielgambetta.com/computer-graphics-from-scrat...
It touches on both ray traced and raster graphics. It lets you use whatever language and graphics library you want as long as you can create a canvas/buffer and put pixels on it so you can target whatever platform you want. It includes links to JavaScript for that if you want. (I didn’t want to use a new-to-me language so I used python and Pygame at the expense of speed.)
And once you get beyond the books the sibling comments here have mentioned (I'd suggest starting with the Ray Tracing In One Weekend minibook series first before Physically Based Rendering), there's the Ray Tracing Gems series (https://www.realtimerendering.com/raytracinggems/) which is open access online with print editions for purchase.
(Disclosure: I contributed a chapter.)
Try Raytracing in One Week and its sequels[1] perhaps.
In addition to the other suggestions, see also https://pbr-book.org/4ed/contents
“The Ray Tracer Challenge” by Jamis Buck is really good as well
[dead]
> Because the ChatGPT output wouldn't be reviewed!
So what? If it's not already, frontier LLM one-shot output will be as good as heavily edited human output soon.
I would have 100% bought the book the author initially pitched. I could do without the junk the publisher wanted him to add, and really it would have probably caused me to not buy the book.
I've come to hate every cookbook that starts with 100 pages of here is a tour of my pantry, which sounds a lot like, here is how to use pip!
> I've come to hate every cookbook that starts with 100 pages of here is a tour of my pantry, which sounds a lot like, here is how to use pip!
Yeah I agree. I hate when books do more hand holding than the reader clearly needs to the point of tedium. Plus many of those setup steps like how to use a package manager change over time and make the book stale instead of evergreen. And Austin was clearly not writing an absolute beginners book.
That's why when I pitched both the Classic Computer Science Problems series and Computer Science from Scratch I explicitly told publishers in the proposals that I was not writing a beginners book (been there, done that). I was clear that I was writing an intermediate book for people that already know programming.
It's a different, more narrow audience. But you can be successful if you write a good book. It's also a less tapped market and luckily publishers were able to see that.
Just skip the chapter?
It's usually not just one chapter but the style of the entire book. Whenever something relatively advanced comes up it's just briefly mentioned and skipped over to the next topic, which once again starts by explaining the basics that would be better suited for a beginner level book.
I find it quite difficult to find quality programming books that actually deal with the details and waste no pages explaining the basics that can be found in literally any other book about a specific language/technology.
When it was mentioned that the publisher demanded the book be "dumbed down" it wasn't a surprise to me at all. I also think it's hurting sales of programming/SWE books at large as beginners are more likely to just use up-to-date Internet tutorials than books, but those more experienced who would benefit from in-depth books can only find the dumbed-down ones.
It uses up the preview in amazon, so you can't actually see the recipes in the book or if the recipes actually have pictures. All you can see is the default, here is my pantry.
Another bonus feature, would be to remove: breakfast, appetizers, and salads from all cook books, or put them in the back where no one needs to look at them.
Although I have found that cookbooks that don't include the useless fluff to pad the book out are usually much better, like the cookbooks from Milkstreet or Love and Lemons, So I guess it's actually a decent way to just filter out all the crap books.
Speaking of good cookbooks, Big Vegan Flavor by Nish Vora is actually one of my recent favorites. First of all, the pictures are amazing and make it a very fun read. And it is less of a recipe book and more of a guidebook on how to develop a good sense about cooking. Don't let the word Vegan put you off it's not the pompous kind of vegan stuff.
Honestly, I wouldn't consider publishing a book if it didn't have that information. There's no reason to give up half or more of the potential market for a book because it's arbitrarily pitched at advanced users. Assuming the customer knows how to use pip would be crazy.
Honestly I don't want to buy books that pander to the lowest common denominator so the author can make more money.
> I've come to hate every cookbook that starts with 100 pages of here is a tour of my pantry
To each their own. As someone who learned to cook as an adult, I’ve appreciated seeing both what someone has and what nonsense I own that they manage just fine without.
The book cover makes me think that AI got involved in the book eventually lol
[flagged]
Why exactly? I think it's fine to question this, given there's money on the table.
[dead]
> There needs to be an initial chapter for teaching Python in case the reader doesn't have the background.
I can't stand books that start with a half-baked Python tutorial. It's not only wasting the time of people who know, but worse, it's wasting the time of people who don't.
Because a one-chapter explanation on the topic is always going to be so superficial that the people who actually need it will never have the details they actually require to get up to speed. You will give them the illusion of understanding, only to let them hit a wall the first time they try any.
Before uv, install python package was a topic that required a lot more than a quick intro, explaining virtualenv, branching depending on OSes, even backtracking how to source the Python installer.
Better to just say "mastering this is a requirement prior reading the book" and be done with it.
There are full books on the topic that can actually help.
Replace this useless chapter with more content to which the book is actually dedicated.
Thanks for sharing. I always enjoy reading author-publisher process articles as they get to the true behind the scenes story. I can relate to most things mentioned, and the terms seem identical to what I had when writing Modern Fortran with Manning. I also started with the intent to write for experts, but the publisher pushed for targeting beginners. The author can concede or (usually) give up the project.
One important aspect to this is that a typical first-book technical author knows well the subject matter, and sometimes knows how to write too (but usually not, as was my case), but does not know how to edit, typeset, publish, market, and sell well. That's what the publisher knows best. And of course, they want sales, and they understand that overall beginner books sell better than advanced/expert level books.
I encourage the author to continue writing and self-publish, and at a later time a publisher come to package and market a mostly finished product.
I'm writing my first book now. It's a novel aimed at teenagers and young adults, in a technical format similar to "The Phoenix Project" by Gene Kim et al., if you're familiar with it. It explores FOSS, non-proprietary file formats, digital preservation, cryptography, and the concept of freedom as a whole. I resonate with the author of the article who discusses motivation to write and the "existential crisis" that comes and goes almost every day. I've been fighting those negative feelings by adopting the mindset that I'm writing the book for myself. It's a book I've always wanted to read, which I can then lend to my teenage children so they can read it as well. Everything else (commercially speaking) will be a nice consequence of this endeavor.
Published author here (through O’Reilly, twice). A lot of people seem to be taking this as an indictment of the publisher. What I’m reading, though, is that the author didn’t make time to write the book and then lost interest. All the rest is normal stuff that happens when writing a book for a publisher. The author did a good job of standing up for themself and their vision, but a poor job of, you know, writing an actual book.
The publisher expended time and money on the author and got nothing in return. This isn’t surprising, and it’s why first-time author royalties are so low.
I've also authored multiple technical books and had the exact same reaction.
While writing I have had similar feeling as the author to publisher/editor comments, especially related to:
> The unhelpful feedback was a consistent push to dumb down the book (which I don't think is particularly complex but I do like to leave things for the reader to try) to appease a broader audience and to mellow out my personal voice.
I also remember being very frustrated at times with the editor needing things "dumbed down". I used to get very annoyed and think "didn't you pay attention! We covered that!" But then I realized: If I can make this easy to understand by a fairly non-technical editor at a first pass, it absolutely will make this book better for the reader.
Publishers have a lot of experience publishing books, so I've learned that their advice is often not bad.
There was also plenty of advice from the editors I vehemently didn't agree with, so I pushed back and quickly realized: publishers need you more than you need them, so very often you do get final say.
But you still have to actually write the book. Book writing is hard, and a much more complex process than writing blog posts. Personally I feel all the editorial feedback I've gotten over the years has made not only my books better, but also has really pushed my writing to be higher quality.
It does feel high pressure in the moment, but I wouldn't have two published technical books if not for the consistency and push of my editors and the publisher.
Another constraint of a technical book that I didn't see mentioned here was that time almost has to be very limited during the writing process. I worked on a couple mobile development/design books, and an iOS 18 programming and design guide is worthless after Apple announced Liquid Glass last summer. At 2+ years into the project and seemingly only 1/3rd complete, the publisher really needs to be sure the content will still be relevant after released.
That's how I read this too. The publisher invested a non-trivial amount of work and was left with nothing, for no better reason than the author changed their mind. From the tone of the post, the author seems to not realize or care.
The publisher changed their mind, too. I don’t think the author was pointing fingers, just sharing information that his readers might enjoy or find useful.
If the author of this article just finished the book the publisher would have just published it without much fuss and probably without any significant changes.
The author had all the leverage regarding content. It’s not like the publisher could actually incorporate what they were asking for with AI, they still need an author to do that and it was a totally new subject at the time. Their demands were empty.
I don’t think the author would have finished the book if it was self-published. They clearly didn’t want to write a book that badly.
Not to say that finishing a whole-ass book is easy, I’m certainly not going to pretend that’s the case.
I’ve lately been trying to finish more side project type things in my life because these dead ends themselves feel empty to me. I am trying to set scope reasonably and then just finish even if it’s painful or there’s no confetti-style payout and nobody else cares.
Yeah, I mean I hate to seem churlish about this, but I really didn't read this with sympathy for the (would-be) author.
> the author didn’t make time to write the book and then lost interest
That was my read as well. The book deal fell apart because the author never wrote most of the book.
> He also wanted me to add a chapter that acts as an intro to programming with Python...
This explains why some books I picked up earlier in my career had great depth but there was always a way-too-basic-programming-intro chapter duct taped in the beginning. So now I have an idea of how they are squeezed in.
I think the author is better off self publishing, based on my personal experience:
I wrote ten tech books for big publishers (McGraw-Hill, J. Wiley, Springer Verlag, etc.) and I was so happy being a published author. However, about twenty hears ago I moved to self-publishing, finally ending up using Leanpub. I am much happier only writing self-published eBooks now because I can update my old books as needed. I still write new books from scratch (just started a book that is basically a rant against over-spend of SOTA LLMs called ‘Winning Big with Small AI’) but hardly a week goes by without an update to an older book.
Writing is great, and even better when not attatched to a conventional publisher.
Austin: if you are here, good luck, and enjoy writing!
Is a piracy a major problem for eBooks? Doesn't somebody just upload a PDF of your book somewhere?
I use a Creative Commons share alike license and my eBooks can be read free online. Occasionally some readers purchase my eBooks, and I use sales as a signal for what readers enjoy and that is a guide for deciding where to put my energy. I only allocate three hours a day to writing and sales also guide what books I update.
I used to keep statistics: about 1 in 60 people who read online buy a book, except for my Common Lisp book that had a 1 in 20 purchase ratio. As you might imagine, I put more effort into the Common Lisp book.
For 30 years writing tech books was an economic driver for my career as a computer scientist. Now that I am retired I still write because I enjoy writing and interacting with my readers.
I killed a book deal I had for this book I mostly finished:
https://kevmo.io/zero-to-code/
I inked the deal in 2023, but shortly after felt like the market was too dead for newbies. When I initially removed the website for the book, I got a small wave of complaints, so I guess some folks still found it helpful.
I've published books with two publishers and many self-published books (Anthropic owes me around $60K for book theft by my calculations).
Publishers can be great, but if you want control of your book, just self-publish it.
The most valuable (IMO) service publishers provide is feedback. If you have a small online presence, it isn't hard to get feedback from others.
I wonder how many people choosing to write books, are expecting it to be a profitable use of their time, and make more than pocket money?
The article mentions the publisher saying that their median book sells in the single digit thousands, so half of the books are selling less - perhaps hundreds rather than thousands. If you are making a 15% royalty on a $50 book selling a few hundred, or few thousand copies, for something that probably took a year or more to write, then this is obviously more about the satisfaction of doing it than being a worthwhile financial endeavor.
Are most people choosing to write books fully expecting that it may only make a few thousand dollars?
In my experience (which is limited, personally, but I have several writers/editors/publishers in the family), writing a book is something one feels compelled to do, irregardless of financial incentives, or even tangential incentives like notoriety/influence. I would compare it to my own endeavors writing/recording music. I would do it even if I know for certain nobody else will ever appreciate it. Because the drive is internal. It’s what I do to make myself fulfilled. The fact that others may find some meaningful value within that work is a nice knock-on effect, but wasn’t involved in my choice to create it.
I'm glad that I released my book in 2022 before AI-hype took off. I'm familiar with the type of publisher mentioned in the article too. Those are very strict in their format and content guidelines, and I had also felt that such constraints were limiting at times. I can relate. But, I also learned a lot from the process, and in the end, my book got fantastic feedback. It became one of the print bestsellers in 2022, and got translated to many languages. I've found the whole experience positive.
But, I totally understand author's reasoning, and it's one of the reasons I want to explore different publishers as I want to deviate from writing strictly technical books.
I'm curious about the economics of canceling a deal like this. Since the editor spent significant time reviewing the drafts, did the contract require you to reimburse the publisher for those costs or return the advance?
They never got the advance because they didn't get to the first milestone.
This was quite a fun read and I appreciate the insight. A couple of my peers have suggested me to write a “stuff you should know” book. Some technical in nature (like linear algebra. It blows my mind how many engineers hardware or software do not understand linear algebra) and some not technical (why stuff cost the way they do. “Why does this cost $200 when I can make it for $20!”). But reading your post was encouraging to see that self publishing for fun might be the way to go. Though I guess people would argue you can just ask a LLM now instead of reading my book.
> Though I guess people would argue you can just ask a LLM now instead of reading my book.
I would certainly not argue that. LLMs do not understand anything, and are thus prone to non-deterministic inaccuracies in their output. Due to that, I think it is extremely foolish to use one for learning unfamiliar topics. Give me a book every time, because (if it's a good book) I am guaranteed to actually learn something. Not so with LLMs.
Additionally, no matter how good LLMs get, they're always going to prioritize the most cookie cutter answers to questions because they're looking for the "best match" - the most common answer.
With books the author can give advice that's not as widely known and is thus much less likely to show up in an LLM output.
I think I could ask an LLM to explain linear algebra. I think it would be less good at the 'here are some things you should know' aspect. One reason for this is that people might not ask, but the LLM might not agree with your list anyway.
I think there is still a place for a book here. I think I might buy a book (or may have done 10 years ago when I was still coding) of things you should know (especially from a respected publisher), that being a longer form book I could work through over time.
Udemy and Coursera both have Math for Machine Learning courses that start with Linear Algebra. Then, Calculus and Probability and Statistics. They're often $25-50.
You might want to look at their outlines to see what they're teaching. Then, decide if you can do something similar and/or cheaper.
that'll only be the case if you actually write the book so that the LLMs have that info. Until then they can't regurgitate your know-how.
Why should I know linear algebra, ooc?
Interesting. For a while I have wanted to write about a book about starting and running a small software company. Markets, marketing, sales, support, versions, promotions, websites, pricing, documentation etc. Possibly as a retirement project. Probably self-published. I have plenty of experience and material to draw on:
-have been running my own small software business for >20 years
-have written ~400 on blog posts on this and related subjects at https://www.successfulsoftware.net
-have consulted to other small software businesses
-know plenty of other people running small software companies
-have given a face-to-face course on starting a software product business
But...
-my experience is in desktop software, not SaaS or mobile software, which feel increasingly niche
-everything seems to be changing so fast with the emergence of LLMs, I am beginning to feel like a dinosaur
I could cooperate with others or work with a co-author to include more on SaaS. And human nature doesn't change that much, despite AI. But it is unclear to me whether enough people would be interested - for me to invest months of work into it.
AI! AI! AI!
It's been a bad couple of years to work on anything in programming that isn't somehow tainted by Altman and Amodei's fever dreams.
If you don’t need money and they aren’t going to advertise it and they’ll only give you a tiny amount, why would you ever want this deal?
Good to hear you decided to self publish. Don't let people online put too much pressure on you (is it done yet?)
On the subject of AI. I'm a great believer in that AI is a huge force for good (gives a single individual a huge power). I've been using every popular commercial and open weights models there are. I know their strengths and weaknesses.
But I think there will always be a need for human book writers. Just like there will be a need for human programmers. Although for different reasons. With software, humans are needed, because AI is still very, very far from being able to grasp actual, overall architecture of even modest sized hobby class projects (there is a special "trick" that is used to convince us otherwise, notice all very impressive examples are almost always "one shot" small prompts, with not a lot of refinement later. That almost never happens in real projects. In fact the opposite.)
With books, the AI is good explaining small chunks of knowledge. But an entire book, that is fun to read, consistent, and has a plan of "reader advancing in capability" through chapters and has some of the author's personality? No way.
Will I buy the book? I don't know. I have built a small library of physical books over the years (maybe about 200 books). But I also have about 50 of them on my kindle. I tend to buy an ebook first. If I really like it I buy a physical copy.
But I'm definitely reading a lot less than i used to. I've been working from hone exclusively since 2016. Before that I did a lot of commuting and that provided an opportunity of time to read loads of books. I certainly do not miss the airports, the budget airlines, the crowded trains and underground, but the reading took a big hit.
I imagine I'm not the only one. So the market for books probably shrunk substantially in the last decade.
Are the people who are really into "AI" even buying books anymore?
I have written a foss python book on github. In the apst 5yrs, the same Acquisition editor from Apreas reached out to me about publishing my book. When he reached out to me for the first time, I was excited. He gave me a format and I expanded my book in their format. Then 6 months later dude emails me that it was rejected and if I could write something else, he recahed out to me and now he made a u turn asking me to write something different blah blah
I said no thanks and moved on.
A few years later, dude sends me the exact same email. I replied saying that get me in writing that you'll publish the book this time because the last time you wasted my 2 3 months.
Publishing is hard. Yess publishing is even harder. Royaltie are dwindling not to mention these days the documentation has improved so much that people learn from that.
My foss books have netted me more $ than if I had signed up with Apress, so not a loss!
I got into a book contract in the 90's and nearly killed myself with stress, trying to write it.
At one point I was told I might have thymus cancer. My first reaction was not distress, but RELIEF. Why? Because it meant I could feel okay about taking a couple of weeks off work to get the exploratory surgery. That's how burned out I was. (Turned out, no cancer. So that was nice, too...)
Eventually, a lawyer friend of mine told me that I could easily get out of the contract by offering to repay the advance. Problem solved.
Lesson learned: don't get into a contract until the book is almost done.
Even so, I just published a new book that I delivered one year late (I wasn't as close to finished as I thought I was). Still it was less stressful because I knew I could take all the time I needed (because publishers actually understand that it's hard to write a good book).
From my very brief experience writing a technical book (a few chapters, many years ago), I think one danger that the author might have run into is having a mental progress bar that covers writing the draft. And then when the edits start to come in and they go far beyond rephrasing or typos, prospective authors realise that their progress bar is far from realistic, and things risk being derailed.
You didn’t share the complete deal details but just from what you shared, it seems like the payout is not worth it for this big of an effort.
What if you self publish yourself using Amazon toolings? Will the numbers be worse? At least you will be in charge of your own quality and deadlines.
For most books, but technical non-fiction in particular, the payout isn't nearly worth enough for the effort.
And by "most" there I mean "all". Yes, there are exceptions, but those exceptions prove the rule.
I've written 2 technical books, for incredibly niche audiences, where the total number of potential buyers is numbered in the low thousands.
I self published as a PDF. and charge $200 a copy, of which I keep $200. It's -marginally- worth it. But the hourly rate is much lower than my day job.
The marketing benefit (as it affects my actual business in the same field) is likely real, but hard to measure. Still, having "written the book" opens doors, and brings credibility.
Disagree, a blog that gets tens of thousands of unique visitors could clear huge numbers on KDP. Maybe your niche is too narrow (probably, given your TAM is in the thousands) but this post is about "timeless programming projects" and is going to be extremely broad. The number of hits to the blog is itself an indicator of a very big and very eager potential market.
A mentor once told me, 'half of the effect of marketing is hard to measure, the other half you have no idea'
Do you have any links to your books? Can't see them in your profile.
I’ve self published a few books now on Amazon. I put $0 down, they take care of everything and I get a deposit into my account every month.
I’m doing it again soon for my next book. It’s fantastic, though having a following online is helpful to get the word out
Care to share links to your books?
You have to many me to engage with Amazon. I will not spend a penny of my personal income to enrich the deplorable Jeff Bezos.
Amazon and companies like them must be broken up because they harm the producers and consumers with dictating buying and selling pricinf.
Amusingly, for a library [1] I’ve been building, 100% of the code is AI-written (with a huge number of iterations of course) and the ONLY part I wanted to write myself is the portion of the README that explains the thought process behind one of the features. It took a lot of thinking and iterations to come up with the right style and tone, and methodically explain the ideas in the right order.
Leaving that to an LLM would have been a frustrating exercise.
Writing for publication is a ridiculous amount of work, smoothing and digesting to the point of pablum, because it's just hard to please everybody. Now that LLM's can tailor to chapter-level discussions, why write?
Still, that's what it takes to reach N > friends+students.
It's beyond ironic that AI empowerment is leading actual creators to stop creating. Books don't make sense any more, and your pet open source project will be delivered mainly via LLM's that conceal your authorship and voice and bastardize the code.
Ideas form through packaging insight for others. Where's the incentive otherwise?
When you have original information that hasn't been released anywhere else in the world, why would a book be a bad choice?
This is why most publishers won't even talk to you unless you have a finished manuscript already, but I appreciated this look into a different situation.
I hope you finish the book. I would buy it.
> This is why most publishers won't even talk to you unless you have a finished manuscript already
This is absolutely not true in the world of technical publishing. I mean books published with publishers like O'Reilly, Manning, No Starch, etc. Usually you come to them with just a proposal and a couple chapters or even just a proposal. Or their acquisition editors actually reach out to you. It's the exception (not quite rare, but definitely less than 20% of books) that comes to them with a finished manuscript. I did that with my last book. I've published 5 technical books across three different technical publishers, so I know a bit about this business...
I'm just replying to this comment to not discourage people who just have an idea and not a finished book yet but have the motivation to finish and want to get a deal.
This is not true for business books like mine. It's vital to write a proposal first in that world; publishers want to influence the content (as in the OP article).
I think the same is true for tech books but I don't know as I haven't written one.
A novel or other fiction is the opposite; there you do have to write the whole thing first.
This story is a prototype of thousands of other stories going on right now. Of course we can't blame the book businesses. They are in survival struggle. They have no clue what to do. Every business is barely holding onto whatever that might keep them in business. AI is bad, but it is the new mafia in the town. Just erase all your beliefs instincts and make friends with it.
Maybe write a book about "Classic projects using AI", whether it makes sense or not. And use AI to write that.
I think self publishing and publishing via an editor serve 2 different purposes. In my case, I always self published. My objective was simply to get my writing out there and have it as a "business card" with all the freedom. Publishing with editors is a different can of worms : more constraints, more process.. and to me removes part of the pleasure and the "amateur" aspect of it and a LOT of freedom. But is surely more professional though.
> Cons of a publisher: [...] they actually do little to no marketing of your book.
Unless the publisher has already written off a book, don't they have incentive to market it?
There are some low-cost things you can do to market a book, and they reportedly make the difference between no sales, and some or many sales.
And a publisher can learn the currently effective marketing methods, and then apply that skill across books of many authors.
No, their incentive is to wait and see what books are taking off, then pile on the money when they know it's already a winner. Today, unproven authors are expected to do their own marketing.
For the marketing that has significant costs (e.g., paying for ads, paying for show appearances, paying other influencers to plug, making quality videos for social media, travel for events).
But it costs almost nothing to do ARC readers for reviews and ratings, and it's free to time things for the Kindle store algorithm. You just have to know to do it, when.
And there's some other "free" marketing that publishers should have automated by now, because they can amortize that across many book releases.
Why not simply add a chapter about writing your own neural network or using OpenAI APIs?
I have done some of the projects you were writing about and understand that learning to program is very different from doing vibe coding, but knowing how to use the OpenAI API should be valuable to the same audience, as integrating AI into everyday programs is a useful skill in itself.
I'm surprised the contract didn't obligate you to return most or all of the advance after canceling.
The first half of the advance was to be paid after the first third was approved.
They never got to that point.
I don't think he ever got the first half of the advance...cherry-picking from the TFA:
> They offered a $5000 advance with the first half paid out when they approve of the first third of the book and the second half when they accept the final manuscript for publication.
> I continued to get further behind on delivering my revised draft of the first 1/3.
> Around this time, there was a possibility of me changing jobs. Oh, and my wedding was coming up. That was the final nail in the coffin.
> There were too many things going on and I didn't enjoy working on the book anymore, so what is the point? I made up my mind to ask to freeze the project.
> They agreed.
Thanks for sharing! I have been dreaming of writing (or better yet, finding!) a similar book for a couple years now. A hands-on guide that peels back the layers of abstraction to teach how things actually work under the hood by building them yourself. I hope one of us gets to it one day :-)
> There was also a daunting voice in the back of my head that LLMs have eliminated the need for books like this. Why buy this book when ChatGPT can generate the same style of tutorial for ANY project that is customized to you?
Why have sex with your wife when you can buy her a dildo?
That sounds like 2025. Everything is "required" to be about AI. Gooodbyee you silly year!
Sounds like your publisher was trying to just take your work and sell it. Giving you the least amount you’ll agree to.
Self publishing is the way. The internet is your Barnes & Noble. Finish the book and publish it yourself. Sell it for $20. Market it. Have peace.
> Sounds like your publisher was trying to just take your work and sell it. Giving you the least amount you’ll agree to.
That’s literally what a publisher does.
I work in games, I know how publishing works. My point was about the pay. For us, that equation is flipped.
Nice of them to transfer back the rights when they terminated the contract! I haven't heard of anyone not doing that, but it feels suspiciously not always a given if they have to specify explicitly.
Traditional publishing is a weird world. They have the shortsightedness to want to force AI into everything. But also it sounds like they still assigned human technical editors who took the job seriously.
Err, the publisher cancelled the author’s book deal. More specifically, cancelled the contract they had. The author procrastinated indefinitely (after losing interest) eventually leading to this.
Is it just me who took offence to the title?
Yeah this article didn't really sit well with me. Not the part where they procrastinated and never wrote the book... life happens. But as you say - the title is a bit misleading.
But the bigger issue to me is that after failing to write a book, they have now started accepting pre-orders and promising to deliver chapters as they are written. Because first, if they were capable of writing the book, they probably should have honored their original contract. And second, if they are not capable of finishing the book, they probably shouldn't be taking peoples' money in pre-orders for a book that they may not ever be able to deliver.
To be fair, it does sound like they were mistreated and/or mismanaged a bit by the publisher, which may well have hindered the writing process and helped prevent the book from getting written. So I'm not going to lose any sleep worrying about the poor publisher - who I am sure deals with this pretty frequently. It's mostly just the concern about not being able to follow through with their obligations to pre-order customers that concerns me.
Yeah, I fully empathize with them. It’s just the title that threw me off.
Writing with Pragmatic was such a great experience. Definitely a blessing that I was able to do that.
My experience with writing and royalties is just so different than this author's experiences.
That was super interesting!
I think you should self-publish. With your existing audience, you'd sell plenty of copies, and nobody would push "AI" into your work.
I'm on it!
Something like 80 percent of published books with an advance never even make back their advance, in case you were wondering why royalties are so low.
You can self publish your print edition on stck.
I am honestly a bit puzzled by this description and I wish they had named the publisher. I'm fairly familiar with this space and the usual experience with tech publishers is that they don't get all that invested in what they publish because 99% of technical books sell somewhere between 500-5,000 copies. That's barely enough to pay the copyeditor to do the bare minimum (often paying attention only for the first couple of chapters), then pay the layout guy, then the proofreader.
The usual accounts I've heard from my friends who published with Wiley, Addison-Wesley, or O'Reilly is that they sign up, get some in-depth feedback on the first couple of chapters, and then are on their own. I've never heard of a tech publisher exercising this level of creative control. I don't doubt that this happened, but it just sounds out of the ordinary.
The 500 - 5000 figure, which is correct, is why most folks should instead self publish via KDP. 70% royalties mean you can get 10-50k easily with an average book. If the book is a success, you can switch career to a full time author if you wish. All this with full creative freedom. Many years ago, I canceled my Redis book for a large publisher for similar reasons to the OP: too many "do it this way" requests.
In general, yeah. The hard-to-replicate benefit is professional editing, but this is something that most tech publishers skimp on. There are some "premium" outlets where you get some real attention, but the default Wiley experience is definitely not worth 70%.
KDP?
Kindle Direct Publishing, aka Amazon's self publishing platform
> I am honestly a bit puzzled by this description and I wish they had named the publisher.
Does it matter which exact one if all the publisher oligarchs behave exactly the same?
He wants more information because the story doesn't ring true, in his experience.
Yes, because as far as I know, they don't behave like that.
I would have purchased your book ! I just wanted to say thank you for writing (your blog). It is a joy to read.
I read a good part of the OP.
I don't get why he went with a publisher despite the serious cons he listed up front.
In fact, I've been reading about the big cons (1) of going with traditional publishers for many years now, here on HN, as well as elsewhere.
IMO, it's analogous to the VC-backed startup lottery, where the few bosses (VCs) take on less risk and win big (think "ten-bagger", etc.), and the slaves, aka noobs (startup employees, except, rarely, for founders) take on more risk and lose bigger.
And still the gullible ones keep on coming, like lemmings, due to the brainwashing that goes inexorably on.
Been in both scenarios.
Another good analogy is the gold rushes like California and Klondike, you know, "picks and shovels".
(1) Pun intended.
Somehow, I miss the time when I was writing a book. It's nice to do the work and research and also nice to refine. Getting money later without doing much anymore was also cool.
But my consecutive attempts of writing a book failed because of my ADHD and missing guidance. I can't do employment, but I really need someone to "nag" me 2-3 times a month to keep focus.
You're witnessing a collapse of demand. Do not ignore it - though it may not be permanent.
This sounds like my experience with a "major" technical publisher except we managed to get to the end.
I'd say that almost no one should work with the major technical publishers more than once. There's some good basic skills you learn but otherwise, they contribute very little that you couldn't get done on your own.
The such low royalties make folks seriously consider self-publishing if you think you can get any sales. (And if you do not need a copy-editor.)
So I have only around 150 sales of my book (see notes at https://andrewpwheeler.com/2024/07/02/some-notes-on-self-pub...). I make around ~$30 though net (average between on-demand print and epub). So my measly sales are about the same as the advance here (not clear if this was ever paid out, presume they would get it back if it was paid out).
If you really think you can sell thousands of copies the economics of it really should hit you.
I get going through a publisher will increase sales, but if you have a popular platform already to advertise it (like a blog or other popular social media), I just don't get it.
This is absolutely true, speaking as someone who has both self-published and also published with a big publisher. Each choice has pros and cons.
In my case, I self-published and sold a book for several years, and then published an updated version with O'Reilly.
I decided to do that because I came to realize people judge self-publish books as less vetted and lower quality.
That may be often true. But in many cases, a self-published book can be much better than those released by a former publisher. I certainly believe it was true in my case.
But in the end, I decided my highest best opportunity was to go with a well-regarded publisher, for the authority that would bring.
And it changed things. People treat me differently now, like they consider me more of an authority. Even though it's essentially the same book; it just has an O'Reilly logo on the cover now.
Whether that should be the case is up for debate...
But it absolutely made people listen more seriously to my message. and I believe it has massively increased the positive impact of that book on the world.
Financially, I think it's been about even. For me it was worth the tradeoff for other reasons, but I don't think that is always the case for every author and every book.
related to not using AI, the project list shared is actually right in the sweet-spot for current LLMs ability to generate decent code.
The illustration is an extremely AI-generated book image with all the usual AI image generation mistakes (especially read the book spine). If people stop reading the article because they expect the text to be AI slop as well, I can understand why.
i wish you'd kickstart the same project and do it your way somehow
the publisher's interests were making it all worse
I'd definitely buy this book!
"12% of total sales ..."
Me: That doesn't sound too bad! They keep 12% of the profit, leaving him 88%!
".. and then 15% [after that]"
This reminds me of the scene in Queen of the South. FL (female lead) is new to power, negotiating some deal.
Guy: How much?
FL: Unsure how much to take 10%.
Guy: Thinking her cut is only 10%, seeing her as weak Oh.. heh.
FL: Detects her mistake For you.
Guy: Face gets red, angry But.. but..
New meme:
[Earth] [Astronaut 1] [Astronaut 2 + Gun]
Astronaut 1 says nothing
Astronaut 2 says "More AI"
Good for you.
> "All of our future books will involve AI."
What an incredible take. It is both so wrong on so many levels and also technically correct, akin to saying "All of our future books will involve spellchecker."
I hate it.
"All of our future books will involve AI."
In the era of AI, who is going to buy books?
AI slop
I don’t see the publisher doing anything wrong.
You “froze” the contract instead of telling them you intended to stop all together and it also seems like you didn’t return their advance.
You are reading a lot of things that I didn't say.
The publisher didn't do anything "wrong". It was their suggestion to freeze it instead of cancel immediately. I didn't intend on giving up on the book even then. I didn't return the advance because I never received the advance.
ah my mistake then. Looks like either party didn't do anything wrong.
It says the first half of the advance would be paid on approval of the first third of the book. It also says that the first third of the book was never submitted. So I don't think the advance was ever paid out.
paid stuff shouldnt be here, I thought this site as about sharing knowledge not selling it
Wow this is the first time I encountered this blog! Subscribed!