The need for new material in something like a math book is truly rare. In the past twenty years, we've developed new instructional technologies, but a textbook really shouldn't be tied to a specific brand of calculator or programming language. I'm currently teaching precalculus using a 1994 edition. It's got some exercises to be done using a computer, and has some sample code in BASIC, but they're easily adapted to modern tools. I'm also making use of a textbook published in 1967, because it does a better job at explaining certain topics. Sure, the word problems have prices in cents instead of dollars, but that doesn't make it any less effective.
I'm using those old books because they are the best books I have available. I have newer books available, but they eschew clarity for colorfulness and newer "teaching practices" that only get in the way of actually teaching.
Or did you think everyone learns best the same way?
Wouldn't that make sense? I feel like bad textbooks have given the idea of text books a bad name. And I don't think this is a "no true Scotsman" fallacy. It's more an issue of mostly horrible Scotsman written to align with ridiculous state standards.
Textbooks are more than just reference manuals. They're meant to teach. Teaching practices have most certainly advanced over the years.
I agree the price of new textbooks is ridiculous considering most editions change very little year-to-year, but please don't say "Nothing new needs to be written." It's not so much what is written— it's how it's written.