I think you're spot on. The challenge is to find away to evaluate both on a level playing field. A CS grad from different caliber universities probably (on average) have different levels of competency. Even graduates from the same university won't have the same level of skill. For traditional colleges, employers can cut through these asymmetries by using heuristics like institutional prestige and metrics like GPA.
The key is to find a method of credentialing general enough that it can apply to both traditional, college-educated job applicants and non-traditional ones alike.
Whether MOOCs will be enough for those non-traditional applicants to be successful in this modern credential system is a separate issue, and, as you point out, far from certain.
A new form accreditation is really what's holding back a huge change in higher education. MOOCs right now, at most, can replace only traditional professional development classes. They're still not a drop-in replacement for college degrees.
That's because of accreditation. Bachelors degrees are valuable, in large part, because they're a common currency: employers know when they see a potential hire has bachelor's degree, he/she at least spent four years learning at a high level and fulfilled some level of competency in his/her chosen major.
Right now, there's no way for employers to make similar judgments about people who have obtained their education entirely online, so it's really hard to get a serious job in, say, software development, with just a few Coursera courses on your resumé. In that case, you'd need to build up a portfolio of OSS projects, etc. Whereas a newly-minted bachelor's in CS will get your foot in the door somewhere, even without that extra work to back it up.
One approach to accreditation/evaluation are domain-specific exams, like the Boards in medicine or the Bar in law. But just passing an exam doesn't necessarily communicate the same thing as a degree, and thus doesn't really solve the problem. There are also, no doubt, disciplines not well suited to this form of accreditation. This approach (Alyxandria) seems more focused on accrediting courses (which solves the same problem) and does it by peer-reviewing those courses, which I think is a very interesting, credible, and scalable approach.
Right now, LinkedIn might be the closest competitor out there to this. They offer a form of peer-review for one's skills with their endorsement feature. Another company that was trying to tackle this problem is Accredible[1], but it seems they've now pivoted to include many more features than peer accreditation — perhaps at its expense.
It'll be really cool to see how this problem is solved in the long run. I think "accrediting" individuals, rather than institutions or classes that individuals can then take, will be the winning strategy, as education becomes increasingly unbundled. That is, if articles, YouTube videos, etc. are to be considered legitimate tools of learning in the future, as college classes are today, then accreditation of individuals will be the only sensible approach.
Agreed. Seems like the author is looking for something more along the lines of "socially optimal." Also seems like he mixed up which outcome is Nash and which is socially optimal. (Deny, Deny) = (Defect, Defect) = Nash Eq. (Confess, Confess) = (Cooperate, Cooperate) = socially optimal.
I disagree. The problem is frankly that we've come to expect too much from modern medicine. Doing a triple bypass surgery on a 90-year-old is expensive, plain and simple. Yes, we can save his life, but we have to be willing to pay the cost—and we aren't.
Medicare is a monopsony. They represent such a large share of patients for many practices that they set their own prices. And when people expect medicare both to pay for their triple bypass and not to go broke at the same time, what we get bankrupt hospitals and over-worked doctors.
Being more responsible about the way we apportion healthcare is the only reasonable option.
Yes it is. Nearly every reason on there is partisan. Say what you want about "trustworthiness" and the fact that having an "untrustworthy" person on a cloud service provider's board is worrisome. Frankly, it's not. Even if Rice were untrustworthy, I don't think my data would be in any danger.
If you're going to make complaints on partisan grounds, at least don't veil them as some kind of assessment of character.
Looks awesome. A great checkout experience is always a delight to use. Even Amazon's feels kind of clunky if you have to enter credit card information.
I totally agree. This is the way most of economics is taught; concrete numbers provide evidence for ideas/models, but when learning their concepts, they simply slow you down.
Yeah that caught my attention too. Personally, I really don't want my HN to be different from anyone else's. I come to HN not only to see stories that will interest me, but also to see what's captivating the community as a whole right now.
The key is to find a method of credentialing general enough that it can apply to both traditional, college-educated job applicants and non-traditional ones alike.
Whether MOOCs will be enough for those non-traditional applicants to be successful in this modern credential system is a separate issue, and, as you point out, far from certain.