|
|
|
|
|
by leggomylibro
3167 days ago
|
|
Nah, MOOCs hold a lot of promise for other areas of study, but they definitely are not totally there yet. They seem to be pretty good in 'soft' fields that don't have much hands-on applications, but still pretty terrible for things like electrical/mechanical engineering or embedded development. Like, I'm interested in making things so that my code can interact with the real world, so I tried Coursera's embedded development class. Holy hell. It was awful. Way worse than a lot of free youtube tutorials I've watched, and they were charging money. The course files flat-out did not work. The course VM with the cross-compilation toolchain and everything was incapable of booting; good thing those things are easy to install. The assignments told you to do different things from the grading criteria; I wound up erring on the super generous side with grading, especially since the lectures were often largely unrelated to the tasks. And there was no embedded platform involved at all, full stop. In a months-long course. I learned much more in a week with Google and a $10 ST Nucleo board. That's where online education could still use a lot of work, IMO. The sort of thing that requires lab segments. And to be fair to Coursera, that's a tough thing to get right. It would be nice if they actually verified that they were selling courses that functioned at all, and that soured me personally to their platform, but at least you can get a refund. They also have a power electronics class I was interested in, but there's no chance in hell I'd risk it now. I don't mind throwing a few hundred dollars and hours after education, but that stuff is potentially really dangerous and I don't want to risk getting it wrong because a lecture on transformer winding wasn't vetted... |
|