In addition to what others say said, computational complexity, is a big reason. Gaussian Process and Kernelized SVM have fit complexities of O(n^2) to O(n^3) (where n is the # of samples, also using optimal solutions and not approximations). While Neural Nets and Tree Ensembles are O(n).
I think datasets with lots of samples tend to be very common (such as training on huge text datasets like LLMs do). In my travels most datasets for projects tend to be on the larger side (10k+ samples).
Maybe it's improved since I last used it (seems to still be an issue per a 1 minute google search), but OpenSCAD doesn't really have easy support for dimensioned drawings.
While it was very handy for my programmer brain to create a few 3D printed things, when I wanted to create a drawing for something I'd make myself, adding dimensions seemed very unwieldly. I used a different CAD program for those projects (maybe Autodesk Inventor?).
DuckDB is awesome! I find it the easiest way to ingest data from various sources then query it into a form I can do analytics on.
The datasets I work on are a bit too big for pandas, but spark is way overkill for them. DuckDB lets me efficiently work on them using only a single computer.
I used them ~2006 when I was college. Someone I know who just graduated this year in mechanical engineering also found and used them independently from me.
In my social circles, we've always talked about Gilbert Strang as the best math professor who never actually worked at the university we went to.
Surefire makes good comfortable reusable earplugs that aren't too visible (if you buys the transparent ones). They're not as effective as foam earplugs, but often that 10-12dB is good enough! Some models have holes you can plug/unplug to increase or decrease the amount of sound blocked.
They're often promoted for firearms use. However I don't own a gun and I enjoy mine for loud clubs and the like.
Indeed, also Augustus Caesar owned (or at least claimed to own) Egypt as his own private property. That would be difficult to quantify in modern terms.
I think datasets with lots of samples tend to be very common (such as training on huge text datasets like LLMs do). In my travels most datasets for projects tend to be on the larger side (10k+ samples).