Because while the field of mathematics has probably not changed, the field of didactics has changed drastically, and the student expectations in 1980s and 2021 will be also drastically different.

Has it actually improved though? Every study I've seen says students are worse at maths today than 30 years ago, so to me it doesn't look like all that pedagogy research improved things. If things actually got better we should have strong evidence supporting that. Images are prettier, sure, but do students who study using modern books actually understand the material better after the course is done?

I think attention spans have shortened (which is arguably devastating for maths), and the materials considered "state of the art in modern pedagogy" are the ones that take that into account in ways that a book from 1987 does not.

As someone who had a hard time focusing in school, modern books with lots of text actually made it harder to focus than books with less text and more information per word. I can read 20 words and then think about those, I can't read the same information if it is spread out and hidden within a 2000 words text.

So maybe kids has problems focusing partially due to modern pedagogy? Every word you write down has a cost to the reader, and the less attention span the reader has the more that cost matters. Drown them in too many words and they will just zone out since their attention span didn't last long enough for them to reach the important parts of the text.

Drill exercises require very little attention time but are the keybto learning a topic and understanding it. This is nothing new but it is usually despised because “boring”…

Like food and architecture, math pedagogy has been affected by the progressive diseases of modern civilization, so new often does not mean good.

Most modern textbooks I used in university are complete garbage. A lot of concept vomiting without intuition building or real world insight.

Not so much with respect to Maths: either you integrate volumes or you do not, and the theory is essentially “dumb”.

As long as you know what a derivative is, all integral and vector calculus is just a comination of “look at the problem with an infinite loupe and then add all those values “.

DiffEq is more or less similar.

Classical maths was very well taught: examples and applications galore.

I doubt math professors who write textbooks that are not meant for the mass public (i.e. not part of a state mandated curriculum) follow the field of didactics.

I've read math textbooks from the 60s onwards. I do not see a trend vs time.