Ask HN: Are we still writing and reading "tech books" in 2025?
In the late 90s I bought my first "tech book" at 12. I have fond memories of _Sams Tech yourself Perl in 24 hours_. I had read the first few hours already. My mom would drive my brother and I 30 minutes to Borders what felt like every month. I always gravitated to the tech books.
As I walk into my office today I see two walls full of book shelves crammed with forgotten and well annotated books. Most inherited from my former boss who retired at the beginning of COVID.I have books on Unicode, XML, Java, Struts, Information Architecture, Sed, and a heck of a lot more. Some older books too. Like a 1990 edition of "Full Text Databases".
But it got me thinking, he last purchased a book about a decade before he retired. I haven't been to a book store in ages. Heck, I barely use my companies subscription to Proquest/Safari books online anymore. I was once enamored with all of the digital copies of Tech Books I had at my fingertips.
I've been at conferences with the writers of some of these old books I have in my office. Many aren't writing books anymore. A few will share horror stories about their own book writing experience.
I imagine my experience is not unique around HN. So to my question, are we still producing and consuming "Tech Books" in 2025? I love textbooks, it feels like it should be the ideal for passing on a lot of crystalized knowledge to the next generation. But many tech books are just dealing with flavor of the month tools to help you be reasonably proficient, and they quickly become outdated in months / years. I think the financial incentives are also fairly limited; nobody is getting rich from writing a technical textbook. Sometimes there are books which have become the standard for some domain, but they were written from an older perspective which is no longer as true for the modern era. I've heard it claimed that the dragon compiler book is a good example of this, focusing too much on limiting memory usage at the cost of speed. Another problem is the limited collective curation efforts. There's tons of these "awesome list of things", but they give you a flood of options instead of picking out the best ones. This is a problem because people probably aren't enthusiastic to read multiple books on a topic, and there's an ever growing mountain of content to sift through. Books are still super relevant If you read something written by a field specialist you don't just learn something, you get the core of the issue. All the mechanics that lead to it, tradeoffs, and the possible paths for the future. That makes the book relevant for many years Though it is hard to make ourselves read long form content nowadays where short content is pushed to us every day Also seeing so much short content make us think the books are outdated. When in reality it keeps churning the same implementation details and getting us burned out for not progressing. And bleeding keeps being reinvented on set concepts And might be a good hiring tip, read a book a year, put it on your CV You get new career prespectives/vision, opportunities, libraries ideas, companies ideas, research ideas I used to read tech books left and right. I did C++ instead of Perl in 24 hours. Back when the dinosaurs roamed the earth... Writing, yes. Many people are still writing tech books. https://www.oreilly.com/library/view/beyond-vibe-coding/9798... Very relevant book and likely is worth reading for some people. Reading? I havent read an actual book on tech in ages. Absolute utter waste of time. If i'm not learning the bleeding edge where there is no book, I'm doing something wrong. https://www.oreilly.com/search/?q=devstral&rows=100 1 result that's not relevant? Sure there's some claude code books available but you can watch like 1 youtube video or just use it and know more than the book author. IT is changing far too quickly to be waiting on the book publication process. The death of technical books (they're not entirely dead -- I still buy some -- but the field is a pale memory of what it used to be) is one of the larger losses to our field. Those books did things that simply can't be replicated in other forms of media. Being able to drill deep into complicated issues and take the time/space to talk about them with clarity is valuable. Nothing online really does this, and I think it's because the medium doesn't really allow for it. My consumption of R books was off the charts. 2008 and into 2018.
With Julia, I used LLMs and maybe a book
I really know R. I wrote some good Julia programs but I do not understand it to the level of R.