> 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.
His last letter was an update on a serious health issue. I care. It's not supposed to be a yuck fest, especially inorganic AI sludge.
I’m honestly stunned that people use AI for personal communication. It seems so alien to me.
You don’t think your comment would be improved by including an emoji of an alien? And maybe a telephone?
The “tells” for AI are - as of this writing - very obvious. I assume they will disappear over time.
The obvious tells are obvious. The less obvious ones fool you so you don't learn them. There's no feedback. You sound like a schoolteacher saying "My students don't cheat on exams. I'd know if they did!" For starters, you can just tell the AI to write in the style of a HN commenter, or whatever.
LLMs are for producing work that should not be done. So it can definitely comment in the style of a comment that's superfluous.
I keep testing with LLMs and it's super bad in writing convincingly in the style of a good writer. That's because good writing is driven by intent, and LLMs don't have intent.
I'm baffled by AI fans who seem sceptical that writing styles exist, and that discerning styles is just part of reading any text at all.
AI fans seem to be people who literally can't tell good from bad, and get upset when you maintain that you in fact can. They think you're having them on.
> tell the AI to write in the style of a HN commenter, or whatever.
This might turn out to be one of those tricks like Ctrl+C Ctrl+V that surprisingly few people discover.
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.
No job is special, even though many programmers like to think of themselves as so. Everyone must learn to adapt to a changing world, just as they did a hundred years ago at the turn of the century.
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.
Yep, could be right. It might have only ever been a few stalwart professions that were expected to be constants. But I think the cost of life during the pre-2010 era absorbed the reasons those anxieties existed, whereas the severity of the rise in that cost of necessities since is the problem. As in, having an expectation of a volatile income-earning life is one thing if a house costs $80k or rent is $400, but having a volatile life with rent for the smallest serviceable apartment being $2-3k, and the same house costing $2m; that lack of stability isn't priced in by the markets
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?
I don't have a dog in this disagreement, but putting the bar at "dig up the personal details of 10 different individual people and the changing dynamics of their lives over decades _starting from 1880_" is a pretty insane ask I'd imagine. How many resources for reliable and accurate longitudinal case studies from the 19th century are there really? I suppose we could read a couple dozen books written around then but that's just making a satisfactory reply so prohibitively time intensive as to be impossible.
Indeed, and when 10 were pulled up by zozbot234, they say that doesn't count. This sort of discussion is not really useful in my eyes, shifting goalposts around and not saying what one means.
I agree, but I do wonder if because those times were generally less specialized, urbanized, etc.. it would have been more possible to simply pivot to another non-specialized "job", because you were either uneducated and poor and needed to be able to do everything, or born rich and able to do the one special thing your whole life. Like when the buggy whip maker couldn't sell whips anymore, they just did 4 of the other jobs they had to do anyway.
The classic old person advice is to just walk in somewhere, give the owner a stern handshake, and you got a job, and if that job could pay your mortgage, then problem solved. Whereas now, to become a buggy whip maker (or whatever), we've developed yhe expectation that you go to school for 4 years and start out at the bottom of the income ladder. If the income we need to pay for the basics (which admittedly are different) requires a lifetime of experience, then it's impossible to pivot
1. Samuel Slater: Textile mill worker → Factory founder
Before: Born to a modest family in England, Slater worked as an apprentice in a textile mill, learning the mechanics of spinning frames.
After: In 1790 he emigrated to the United States, where he introduced British‑style water‑powered textile machinery, earning the nickname "Father of the American Industrial Revolution." He built the first successful cotton‑spinning mill in Rhode Island and became a wealthy industrialist.
2. Ellen Swallow Richards: Teacher → Pioneering chemist and sanitary engineer
Before: Taught school in Massachusetts while supporting her family after her father's death.
After: Enrolled at MIT (the first woman admitted), earned a chemistry degree, and applied scientific methods to public health, founding the first school of home economics and influencing water‑quality standards.
3. Frederick Winslow Taylor: Machinist → Scientific management consultant
Before: Trained as a mechanical engineer and worked on the shop floor of a steel plant, witnessing chaotic production practices.
After: Developed Taylorism, a systematic approach to labor efficiency, consulting for major firms and publishing The Principles of Scientific Management (1911), reshaping industrial labor organization.
4. John D. Rockefeller: Small‑scale merchant → Oil magnate
Before: Ran a modest produce‑selling business in Cleveland, Ohio, struggling after the Panic of 1873 reduced local demand.
After: Invested in the nascent petroleum industry, founded Standard Oil in 1870, and built a monopoly that made him the wealthiest person of his era.
5. Clara Barton: Teacher & clerk → Humanitarian nurse
Before: Worked as a schoolteacher and later as a clerk for the U.S. Patent Office, earning a modest living.
After: Volunteered as a nurse during the Civil War, later founding the American Red Cross in 1881, turning her wartime experience into a lifelong career in disaster relief.
6. Andrew Carnegie: Factory apprentice → Steel tycoon
Before: Began as a bobbin boy in a cotton factory in Scotland, later emigrating to the U.S. and working as a telegraph messenger.
After: Invested in railroads and iron, eventually creating Carnegie Steel Company (1901), becoming a leading philanthropist after retiring.
7. Lillian M. N. Stevens: Seamstress → Temperance leader
Before: Earned a living sewing garments in a New England workshop, a trade threatened by mechanized clothing factories.
After: Joined the Woman's Christian Temperance Union, rising to national president (1898‑1914) and influencing social reform legislation.
8. George Pullman: Cabinet‑maker → Railroad car innovator
Before: Trained as a carpenter, making furniture for a small New England firm that struggled as railroads expanded.
After: Designed and manufactured luxury sleeping cars, founding the Pullman Company (1867) and creating a model industrial town for his workers.
9. Mary Elizabeth Garrett: Schoolteacher → Medical education reformer
Before: Taught at a private academy in Baltimore, earning a modest salary.
After: Used her inheritance to fund the Johns Hopkins School of Medicine (1893), insisting on admission of women and establishing the first women’s medical school in the U.S.
10. Henry Ford: Farmhand → Automobile pioneer
Before: Worked on his family farm in Michigan and later as an apprentice machinist, facing limited prospects as agriculture mechanized.
After: Built the Ford Motor Company (1903) and introduced the moving‑assembly line (1913), making automobiles affordable for the masses.
Come on. These are edge cases. We are talking Joe average that went from having their own business, to working an average job. You know, the mass scaled transfer that will be what happens for most of society when people say 'they will find new jobs'. Not the small amount outliers.
The guys that ended up spending their lives living in boarding houses with other men, never starting a family. The guys that ended up living in tramp camps traveling the country looking for work. The families that ended up as migrant workers. The people that broke and lived in flop houses or those long term hotels in downtowns.
That is the real picture no one shows. What does 'find new work' after the industrial revolution look like for the average person, and the answer isn't 'become Henry f'ing Ford' now is it post WW2 style middle class employment.
OK, here's 10 concise, plausible examples of how average individuals in the late 19th century might have adapted to new jobs after economic change (e.g., industrialization, mechanization, decline of artisan trades, agrarian shifts). In order to focus specifically on the "average", we now choose to simply build illustrative, composite scenarios based on common historical patterns (as opposed to specifically documented people) so you can see typical before-and-after lives.
1. Blacksmith → Railroad Carriage Foreman
Before: Local village blacksmith, shoeing horses, forging tools; works long hours in a small shop; income modest but steady; reputation tied to local farmers.
After: Takes a job at a nearby railway workshop as a metalworker assembling and repairing iron carriages; works in shifts under foremen; more regular wages and slightly higher pay; moves to a rail town, his kids gain access to a school; he loses some independence in his new setting but has marked gains in job security and cash pay.
2. Handloom Weaver → Textile Factory Machine Minder
Before: Home-based handloom weaver producing cloth on commission; schedule flexible but income fluctuates with orders; household-centered labor.
After: Employed in a textile mill running power looms; fixed hours, supervised by overseers; steady wage, production quotas; exposure to somewhat harsher factory conditions, but more reliable monthly income and potential for overtime pay.
3. Sailmaker/Canvas Artisan → Industrial Tent or Sail Factory Worker
Before: Small workshop making ship sails and canvas goods for local ships; skills highly specialized and tied to maritime trade.
After: Joins a larger factory producing canvas goods and tents for military contracts or industrial customers; moves from bespoke work to standardized production; learns new machine operations; steadier demand, less artisanal pride but higher throughput.
4. Small-Scale Farmer → Agricultural Wage Laborer / Seasonal Harvester
Before: Owner-operator of a few acres, subsistence mixed with small-market sales; vulnerable to crop failures and market price swings.
After: Sells or loses land, takes wage labor on larger farms or in orchards; seasonal work, long hours during harvests; cash wages replaced self-provisioning, children sometimes work; potential to move seasonally to find work.
5. Canal Boatman → Dockyard or Stevedore Worker
Before: Independent canal boat operator transporting goods along waterways; income from tolls and freight; lifestyle itinerant but autonomous.
After: With canal traffic declining due to railways, he becomes a dock laborer unloading goods at a port; work is bunched into long, physical shifts; income less variable but less autonomy; often joins laborer networks or unions.
6. Carpenter (Craft) → Construction Gang Member on Urban Projects
Before: Skilled carpenter building houses and furniture for local clients; runs small crew, has flexible custom work.
After: Migrates to a city for large-scale building projects (tenements, bridges); works as part of an organized gang, specializing in one repetitive task (e.g., formwork); wages steadier including opportunities for overtime, though with less creative control.
7. Tanner/Leatherworker → Factory Leather Stitcher / Machine Operator
Before: Family tannery producing saddles and boots for local markets; knowledge-intensive, smelly but respected craft.
After: Enters a leather goods factory operating stitching or cutting machines for mass-produced footwear; learns machine maintenance; working environment more regulated, discipline stricter, pay more regular.
8. Rural Cooper (Barrelmaker) → Brewery or Canning Works Employee
Before: Independent cooper making barrels for local brewers and farms; demand falls as metal containers and standard packaging rose.
After: Hired by an urban brewery or canned-goods plant to maintain wooden vats or operate filling machinery; moves closer to urban amenities; job is less entrepreneurial but offers steady pay and sometimes benefits.
9. Watchmaker/Clock Repairer → Precision Machine Operator or Assembler in Electrical or Watch Factories
Before: Skilled artisan repairing and crafting clocks, serving local clientele; income from repairs and small commissions.
After: Joins an urban precision workshop or early watch factory assembling parts or operating lathes; narrower tasks but higher volume and a connection to emerging industrial technologies; more predictable pay and potential for apprenticeships for children.
10. Fisherman (Small-Boat) → Cannery Worker in Coastal Town
Before: Independent small-boat fisherman selling catch to local markets; income seasonal and weather-dependent; family-based operation.
After: Cannery expansion offers steady work in fish-packing plants, processing, salting, or canning; shift work with set hours and piece-rate opportunities; his income is quite a bit steadier, though the job is somewhat repetitive and sometimes hazardous.
Here's a sketch of the bigger picture we can see from these scenarios:
Greater wage stability: Moving from piecework, self-employment, or seasonal income into factory, railway, or dock wages usually meant steadier cash pay and more predictable household budgeting.
Larger scale means less autonomy: Many tradesmen lost control over hours, pace, and methods, trading independent decision-making for supervised, repetitive tasks.
Increased urbanization and mobility: Job shifts often prompted moves to towns or industrial centers, changing family networks and access to services (schools, markets, hospitals).
Routine and discipline: Factory and industrial work imposed fixed schedules, time discipline, and stricter workplace rules compared with artisan or farm life.
Skill reorientation: Some workers adapted existing skills to machines and assembly roles, while others lost craftsman status as specialized knowledge gave way to standardized production.
Household changes: Women and children were drawn into wage labor more frequently; domestic production declined as cash purchases rose.
Health and safety trade-offs: Improved incomes sometimes came with worse working conditions, crowding, and new occupational hazards.
Social mobility mixed: For some, steady wages enabled upward moves (education for children, home purchase); for others, loss of independence and precarious labor markets limited long-term mobility.
Community and identity shifts: Occupational identities tied to craft or land weakened; new worker solidarities and urban social institutions (mutual aid, unions, churches) grew in importance.
Consumption patterns: Predictable wages led to more regular purchase of manufactured goods, changing diets, clothing, and household items.
Don't move the goalposts, you said find 10 and they did. Keep in mind it's much harder to find evidence for average people than for "successful" people, by definition.
I didn't move the goalpost. The point was to become better informed what 'find a new job' looked like at the start of the industrial revolution, not to own me. When people say 'in the industrial revolution, people found new work' they aren't referring to Henry Ford and you know that. You know their answer wasn't relevant to the discussion.
What did losing your stables/small scale manufacturing/family farm/etc and adjusting to post industrial revolution 'finding a new job' look like? Saying 'Henry Ford existed' doesn't add anything to understanding that.
> You know their answer wasn't relevant to the discussion.
You're the one who asked the question...
Sorry but if you're going to mean something other than what you say, then you should specify that in what you say rather than have people guessing or answering questions you yourself asked then saying, but no that's not what I mean. Well, yeah, we don't know unless you say so, and that's exactly where the goalpost moving is, you literally did it in the above comment.
Got it. When tech bros/AI proponents tell a single mom with kids who is worried about losing her middle class job to ai that 'new jobs come along' what they are saying is 'just be Henry Ford'. If you want to stick with this response great. It's on point for the whole pro-AI set.
'new jobs come' is very much pushed by the AI set when people bring up concerns. I was simply trying to define what that looks like. Apparently it's 'just become Henry Ford or a Rockefeller'. Good luck with that talking point.
No one said become Henry Ford. You keep trying to put words in people's mouths because you come into this discussion with a preconceived notion that you're correct, meanwhile multiple people are providing evidence that you're likely not. If you have any actual points to share please do so but it doesn't seem that you do rather than saying the same two points over and over and moving those talking points once someone tests it. I don't know about you but I don't find that very enlightening discussion.
Do you think those people just starved to death? They had to find other jobs and they did. Now I'm sure I could find you 10 such examples if I trawl through historical records for a few hours but I'm not going to waste my time like that on New Year's Eve.
Why are you constructing a strawman in your second paragraph? No one said or even implied that, you just made up your own quote you're attacking for something reason?
People imply that the jobs shakeup will 'work itself out', but tend to imply that working out looks like post WW2, when in reality we are at 1880 level change that took decades to 'work itself out' and working itself out included in 2 world wars during that time.
No one is waiting 70 years to work itself out, you're operating on a false premise that I don't even know where it comes from, people were not starving jobless for 70 years straight. Like I said, people found other jobs, within a decade span at most of their occupation being automated.
People definitely were starving and looking for work at the start of the industrial revolution. Your response is why what I said isn't a strawman. Your response of 'trust me bro, it worked out' very much is. The economy/job/living situation in 1880-1940, the 'working through it' phase of the industrial revolution, was very shitty and nothing like post WW2.
Tramps/homeless/tramp camps was a part of life. Huge groups of men lived in 'boarding houses' their entire lives. Most people today don't even know what boarding houses were. There were huge populations that traveled as migrant field workers. The reality looked much different than post WW2 employment.
Sure you said people found other jobs. But really, lots of people didn't. Or didn't find jobs that allowed them to live outside boarding houses and have families. There is a whole big picture waved away with 'found jobs' that implies post WW2 jobs/lifestyles, when in fact that is not what happened post industrial revolution, and did not happen until much later, and was not guaranteed to happen ever.
Ratelimited so editing:
People handweaving away legitimate fears with 'people found new work' are implying it is on the level/quality of life. My 'strawman' is pointing out their 'found new work' was shittier work and a shittier life and that the handwaving/minimizing peoples fears with such a hollow statement is bullshit and should be called out as such. 'found new work' is nothing more than 'people didn't all die'.
> There is a whole big picture waved away with 'found jobs' that implies post WW2 jobs/lifestyles
No, you are the only one implying this, no one said anything about this which is why I said it is a strawman. I said nothing about the quality of work found, only that new work was found. That may or may not be the case for today's situation, no one can know and I am not in the business of prognosticating so.
Things that won't be automated anytime soon, like plumbers or electricians.
Or double down on applied ML?
Like hundreds of thousands other workers who had the same genius idea.
Nurses.
A lot of nurses leave the profession because of the abysmal financial and working conditions.
Unions and striking have been slowly changing that, thankfully.
> 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.
Did you respond with a fallacy of your own? I can only assume you’re not in or don’t have familiarity with those worlds and that has lead you to conclude they don’t exist in any practical or economical sense. It’s not difficult to look up those industries and their economic impact. Particularly horses and fountain pens. Or are you going by your own idea of practical or economical?
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.
The Mechanics Institute, where craftsman learned and offered their wares in my town, was founded in 1801.
Its still here, today.
I wouldn't dismiss an industry based on business failures. The restaurant industry still exists, despite it being almost a guarantee that you will fail.
There's also stores with hand-knitted clothes and bears, sculpters and painters.
Yes, all of these are niche - but they survive because they embrace a different business model.
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.
I wouldn't know. Been 25 years since I lived in South America. I live in one of those western nations - Australia.
>> 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.
It's not a university or TAFE.
The teaching they do is conferences - they host archeology, psychology, and engineering mini-conferences and talks, today. Whilst also being one of the biggest libraries in the town.
This isn't somewhere handing out a bunch of useless Certificate IIs. Its somewhere you need a degree to even get in the door.
Their primary income is actually from trade unions - almost all of them rent their office out of the building, because of its established connections to everyone who knows something. And all members get access to the library, equipment and most talks. (Though not the rat warrens that still run under the town from the gold rush.)
Right, nobody needs cabinets or doors because... AI. /s
I'm a professional woodworker. One-off tables in a garage might not be a great business, but millwork, built-ins, and cabinetry in homes is a great business. You're likely not exposed to cabinet or architectural woodwork shops that build high-end homes, or that just do renovation for that matter.
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.
There are still plenty of industries that won't trust AI generated anything unless it's gone over with a fine tooth comb, or maybe not even then. Devs will still have careers there. I'm talking about medical devices, safety critical systems, etc. In any case, I don't even believe AI gen code will get there anytime soon, but if I'm wrong that's okay too.
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.
Bro, there is a HUGE market for one off software projects for lots of money.
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.
We are all talking about exceptions to the exception at this point. It's typically uninteresting to discuss...
But at the risk of being uninteresting, I know of incredibly niche software projects that were done for a single rich benefactor as one-offs simply for their personal use-case and no one else. Years in the making and quite well paid. In one case effectively lifetime employment for the sole developer.
They are obviously incredibly rare, but they do exist.
However, they are totally irrelevant to discuss in this context since it's a rounding error of 1% of 1% of current developers who would get such roles either via luck or skill.
OK well I stand corrected then they do exist, because it is probably near 1% of woodworkers who do the quality projects I am thinking of - say table able to seat 20+ people in the shape of a dragon with extremely detailed scales and everything that takes 6 months or more to make - that kind of thing.
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.
SAP or Salesforce can do anything. Or you can write custom code for a fraction of the price for small/medium sized businesses.
Obviously, they care about the uniqueness of the code if they are one-offs.
My thought is that many of them thought their products were one-offs but were actually not in practice.
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.
That's a common usage of a rhetorical question. You ask it and then you answer it.
> 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.
People used to say the same things about Wikipedia. They weren't wrong. Still aren't. But it sort of misses the point.
That's an entirely different system because it's quite transparent
Sure, well cited LLM output is better than opaque. You still don't immediately know when it's wrong though. Same as Wikipedia.
Despite that, both are incredibly useful if used correctly.
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.
https://notebooklm.google
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:
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.
I didn't say "text and code related to raytracing" though. I (and the parent post) said "raytracers".
It's more important whether it knows basic concepts about computer graphics, linear algebra, etc. Reading the code of a raytracer is not that helpful because it's hard to extract general concepts from low level code like that.
Besides that, it has web search and research tools.
I just fed Claude Opus 4.5 the source of a raytracer I wrote actually, and it had reasonably good comments on it, but it knew less than I know and its updated version had a few more bugs and was missing non-obvious optimizations I'd added. (In particular it loves writing FP math as all doubles for no reason.)
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.
Would have to strip a lot more than that, like every printf and function name.
https://github.com/mrvacbob/atrace
I see what you mean, I was assuming you were referring to a smaller code base.
I'm sure Paul Heckbert's business card is in-distribution, but it'd be a good example of what I'm talking about ( see variation at https://fabiensanglard.net/rayTracing_back_of_business_card/ ).
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.
[0]: https://news.ycombinator.com/item?id=46448544
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.
So are you just verifying/factchecking everything it tells you? How is that a good learning experience? And if you don't, you are learning made up stuff, so not great either.
It's a good tool to learn stuff, I'm not trying to argue that, but one has to be fully aware of its shortcomings and put in extra work. With actual tutorials or books you have at least some level of trust.
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.
> Something that is factually "most of the time" is generally expressed as a frequency based on data.
that is exactly my point. This is purely anecdotal, but LLMs keep pretenting there is data like that, so they use those words
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.
[1] https://raytracing.github.io/
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.