Considering that the maxim for the new millenium is "Safe is risky and risky is safe" I think he might be doing the best thing. All in all, it seems like today the best way to approach things is to be different and his hiring policies are definitely different. Sure, he might fail, but would he be able to succeed if he did the same thing as everything else?
I'm sure there might be problems with choosing the wrong criteria, but I can't see how "elitism" in itself can be a cause of problems in the hiring process.
Yes, I can see that as a possible problem. I've seen fewer cliqueness in this crew (prehaps because they all come from varied educational and professional backgrounds), than I have in groups pulled mostly from tech academics and companies.
What about the children of poorer families who don't have the luxury of going to college. The above would be fine if parents/students were paying the full cost of the facilities, but in most cases its government/charity subsidized.
A better way to teach how to learn to learn, would be to teach a class in exactly that. The motivational speaker industry does a better job in "meta" subjects, such as "getting things done", "how to be successful", than college does. Paul Graham's essays for startups are more effective than college courses in entrepreneurship.
I would go further to say that the current system is disadvantaging rich countries immensely. I typical person without a college degree is likely to think he shouldn't even attempt to seek work such as IT (it's for those college educated types), and instead heads into the local service economy (further depressing wages there), instead of earning money for the country in globally based economy.
The certification system he's talking about would massively accelerate global outsourcing of US jobs.
ISO 9000 certification in the 80s and 90s played a huge role in sending US manufacturing overseas. People were reluctant to believe that factories in various second and third world countries could be just as good. Then the certs proved pretty well that they were!
Same deal now with personnel. Employers are sometimes reluctant to believe Indians or Vietnamese are on par with a four year American college grad. Sufficiently standard tests would prove they are.
In other words, the US has coasted on reputation in a lot of ways and empirical approaches can damage that reputation.
But you're assuming outsourcing is bad for individuals in the US, whereas most indications are the opposite, it increases their wealth. The point about the nongrad IT worker is that he would be able to create more wealth than he would as a local service worker (in the same way as offshore workers do). Increasing production (and hence wealth) from all workers, whether offshore or in US is better for society than allowing inefficiencies and protectionism to prevail.
The rust belt was not sustainable. From about 1945 until the 70's, the US had an unfair competitive advantage over the rest of the world: our industrial capacity wasn't blown up. This led to an artificially large manufacturing sector in the US (we were exporting goods to the rest of the world). Once they fully recovered (roughly the 70's), there was no need for this.
If you want to bring back the rust belt, you'll probably need to start a large war in Europe and Asia...
The decreasing cost of manufactured goods, increase in quality of them, and the increase in GDP that has occurred over the past 30 years. (You can argue much of this can be attributed to technology, but a large share of it is due to outsourcing - most things in dollar stores and walmart are not due to technology but outsourcing).
The rustbelt is caused by their inability to compete with the better quality Japanese cars (and Japanese salaries are higher than in the rustbelt, you can't make the cheap labour for that).
Real median wages are flat or down in the US for the last 30 years.
You can talk about the theoretical upsides to global free trade all you want. I'm a believer. But 2 billion people jumped into the labor pool and the reality in the US is that they made a wave that drowned a lot of people. Real wages are depressed. That's not to say there's any real alternative to full open participation in the global market, but your free-market cheer and pom-poms don't capture the full reality from a US worker perspective.
I don't think you can blame globalization for that,
-technology has eliminated many jobs
-companies have become more efficient at "desklling" workers into more narrowly defined roles (and so paying lower wages)
if you want to attack free markets, why not go further and advocate barriers on inter-state commerce within the US.
In the UK, the higher education system is as you desribed in point 1, you get a degree in exactly your subject. yet they have no shortage of people who can debate on the issues you mention in point 2.
Requiring everyone to learn some nonpractical things is a waste of societal resources. If some people want to learn something just for the sake of it, they could join reading groups/attend lectures on their own time.
I'm trying to draw a parallel between how content on the web is being created and how an artificial mind might be created. If you sat down at tried to engineer a giant encyclopedia within one company/community, you would end up with britannica (within one company), and wikipedia (within a community of enthusiasts), However, the web at large is much larger than both of these because of the economic incentives for people to create articles. Many people try and fail (in effect working for free), but the progress is rapid, as in evolution.
In the same way, incentivizing people to produce classifiers would have the similar effect of rapid progress.
If you think of intelligence as being the ability to predict accurately, having a giant web of classifiers that predict accurately could be construed as a form of artificial general intelligence, or a human type artificial intelligence. Having an enough data about the world, because you have created many classifiers that can recognize the semantic events in video and speech, would allow you to make all the same types of recognitions and predictions that a human would make.
