How many people do you know who know things like statistics, analysis of algorithms, discrete math, calculus, to the same level as someone with a Master's degree in math or comp. sci. and just learnt that by watching YouTube videos from Stanford or MIT? I know some people (including myself) who got some exposure to some of this without getting a degree (I got one eventually) but finding someone as well rounded as a good curriculum is very rare.
Watching videos is more like thinking you're learning. In order to learn you need to practice and internalize.
Now you might argue all this stuff doesn't matter. And if you're just cranking out some mundane code maybe it doesn't. But sometimes it does. I'd imagine it'd be hard being a CTO of in a deeply technical business without having the in-depth theoretical knowledge but CTO can mean lots of different things in different companies.
All that said universities are far from perfect and their curriculum is a mixture of practical and theoretical trying to balance preparing future researchers and scientists and preparing the future workforce. Software development also tends to be very diverse. I do think software developers need some of this background and unfortunately it's much harder to get it without going to university...
The other thing is IMO universities can't make someone who has no aptitude to software a good software developer.
I've learned much more from watching statistics videos from top notch YouTube lectures (I like the Harvard ones) than from statistics at university, although there was a lot of it. That said I regularly need to look up statistics, probabilitiy distributions and hypothesis tests in books when I need them twice a year.
The main problem I see in startups I coach is they think they are "data driven" but marketing, sales and top management lacks the proper statistical understanding to make useful decisions. They forgot what they have learned at university and feel no need to relearn statistics for their job. They make all the rookie mistakes.
My counterpoint is that I know a good amount of people who have a ton of parroted knowledge, and know nothing of how to apply it. I know far, far, fewer people who know how to apply knowledge. There are even fewer people who can start applying freshly acquired knowledge within the same day.
A mindless search engine can parrot and rattle off 50 probability distributions. Only a human can tell you the implications of a specific distribution emerging from your data. Google and DuckDuckGo are free, I don't need to hire someone who can do the same as a search engine.
My opinion of what you refer to as learning, is that it is the absence of learning. It is one of the longest held misconceptions, spanning hundreds of years. It is why we need to retrain everyone fresh out of college. It is why junior vs. senior roles exist. Juniors have all of the rote knowledge and no practical experience. Seniors have immense practical experience and have probably long forgotten all the rote knowledge. The latter gets paid vastly more, the latter is far rarer.
With 50 I can't remember things I've learned with 20 and need to look up many things from my youth, so I guess I've never learned a thing. Thanks for that cherry blossom thought.
The point here is that it's easy to mistake understanding in the moment for durable learning. It's not ideal, and I won't say it's intentional, but I expect the repeated, spaced exposure process of reading, lecture, homework, exams and finals to yield a little more durable result than a focused video only curriculum with zero testing for retention. Students are notoriously bad at self-evaluation, conflating mastery in the moment with learning.
However, I think if someone tried to partner video lectures with a spaced repetition tool like Anki they _might_ be able to outperform the college treatment group. It's a tool I've been exploring, both personally and professionally. In my youth, I had learned the US state capitals (as many do) and Anki proved pretty quickly I had forgotten that; the same goes for say, bone names, constellations, reading music, foreign languages studied, etc.
> Watching videos is more like thinking you're learning. In order to learn you need to practice and internalize.
And this is key, I think. It's hard to really understand without practice. You feel you might understand but it's fleeting.
And, the reason many say their didn't get any use out of heir college is because... Well, they now know stuff. Since they know it it's not hard (for them, because they are already used to it) and they don't even realize how involved the learning process was.
I've seen many CS graduates who after in a few years after graduation don't remember even the most basic CS stuff. It looks like most students learn just enough to pass an exam and then forgot everything.
On other hand I don't have a degree but I wish I had. All I know I learned bit by bit over the years and somebody who graduated at the top of their class definitely knows listed topics better than me.
Watching videos is more like thinking you're learning. In order to learn you need to practice and internalize.
Now you might argue all this stuff doesn't matter. And if you're just cranking out some mundane code maybe it doesn't. But sometimes it does. I'd imagine it'd be hard being a CTO of in a deeply technical business without having the in-depth theoretical knowledge but CTO can mean lots of different things in different companies.
All that said universities are far from perfect and their curriculum is a mixture of practical and theoretical trying to balance preparing future researchers and scientists and preparing the future workforce. Software development also tends to be very diverse. I do think software developers need some of this background and unfortunately it's much harder to get it without going to university...
The other thing is IMO universities can't make someone who has no aptitude to software a good software developer.