It's actually worse than nothing, because it encourages operational reasoning rather than deduction. Maybe okay for physicists, but not math / C.S. majors.

(Apologies to physicists, you guys are the best at manipulating series )

OTOH, ignoring crucial, simplifying steps and just plowing ahead with a dumb approach is an excellent introduction to computer programming as widely practiced, a career path that will likely be pursued by your students with 1000-fold greater frequency than that of mathematician.

Being a terrble programmer who can meet deadlines can earn you cash.

I know some of my code could be optimized but eh, **** it. I'll do it later (before final product shipment) and charge them. I can even probably sell it as some hyper cool new coding practice when really I'm just mopping up the slop I got everywhere.

Save your neurons for the theorems. To invert G.H. Hardy: there is no permanent place in the world for beautiful code.

Well documented code stands a shot, though! You said something about optimization: the correct target of this activity is for humans, who read it.

Of course, maybe nobody will read it anyway, but I know at least one person who might want to.

Ahem.

(I've never studied QFT, but my understanding is that it's a bunch of truncating series solutions arbitrarily and waving your hands pretending you're being rigorous)

A theoretical high energy physicist once conveyed to me how he felt about mathematical rigor, by holding up a textbook entitled 'Functional Analysis', with his hand covering up the 'lysis'.

As opposed to Nonfunctional Ana?

whoops...

The 'l' found its way through to form anal

I like how this screenshot from a new work-in-progress JK level looks like something from my TODOA DM.

(Using one of the textures from DF Mod for JA also helps)