I'm in a university dealing with this stuff and we get both ends at the same time, they (companies) want more immediate technical skills (e.g. programming) while also complaining that grads don't know their math fundamentals enough.

In an engineering department (as in not CS) I'd say the bias is overwhelmingly towards the math end of the spectrum, partly because faculty largely don't even keep up with the technology. You'd have to heavily revise your course constantly (imagine teaching how to use google products...). Plus it just feels less worthwhile teaching things that will become obsolete in a few years.

Oh and by the way we need to teach them to write better too. Kids can't communicate these days. And economics. And ethics. It ends up being a very tight squeeze we can cram in maybe two courses on programming if it's a top priority.

I've seen theories that it's more related to how, at least aerospace graduates, appear to have much less practical experience at least passed as theoretical knowledge, and churn out designs that might be generally good when assembled, be possible to assemble... but totally impossible to maintain. Things like too small tolerances on pretty much everything, having to disassemble huge portions of an aircraft because there was no maintenance hatch of appropriate size, etc.

In a way, F-35 (but also recent F-16) suffer from related issues, where the complexity of maintenance is staggering, and the tools you're given to help manage that suck too much.

> That's interesting. The popular sentiment in programming world seem to be Unis don't tech practical stuff enough,

I think the point might be that there might be a sweet spot somewhere between

- demanding Java code to be created by students using only Notepad and javac

and

- just teaching the latest js trends

I think it is fully possible - and a lot more motivating - to teach theoretical problems hand in hand with practical problems.