Last summer I took CS184 Startup Engineering. It was a free Stanford Computer Science Course on Coursera, taught by Balaji Srinivasan.
It was about 10 weeks long. There were ~100,000 students enrolled. It was free. And it was, far and away, the most valuable thing I've ever done in my life. It was an order of magnitude more valuable than my 4 year undergraduate education at the University of Connecticut.
Since taking that course, I've pushed hundreds of code commits into the Airbnb codebase (I work on the online marketing team here). They're small pull requests , and I'm rarely ever writing anything from scratch, but the number of engineering hours I've saved by being able to write my own PRs is extremely valuable.
Out of curiosity, I asked a few of the recruiters at Airbnb what putting CS184 on my LinkedIn means. I explained to them what it was, how much value I got out of it, and how much value the company got out of it. I showed them the course, and the certificate you get when you finish. Everyone had the same answer: "It doesn't mean much".
My bias is, online education still has that "University of Phoenix" stigma. How valuable the course actually is still doesn't seem to mean anything yet. Maybe, in general, that's correct. Maybe most online courses still suck. But I can very much verify that life-altering, immensely valuable online courses exist.
This idea of online learning and more specifically, credentialing, looks more like it's a social engineering problem rather than "knowledge delivery" problem.
I've been geeking out on institution-agnostic credentialing and accreditation for a little while now. If you're interested in this space please get in touch with me - email is in my profile.