You learned your native language by immersion, not by swallowing and regurgitating spoonfuls of grammar and vocabulary.
I am usually of the school that finds more value in deeply understanding a single thing than having good knowledge of a lot of things.
When it comes to start-ups and hackers, do you think it's more important to have a deep knowledge of a few topics or a fair understanding of a lot of topics? I think it's more beneficial to have the kind of knowledge that the article described, than a true bare bones understanding of programming.
You don't need to know compiler languages to code the next big thing. It just gives people an indication to how smart the founders are, which means a lot.
I have been thinking a lot about this. My conclusion is that the optimal thing is somehow both. When I say both, it sounds bullshit, but it is not. I think the key is to be ABLE to study specific topics deeply, know what is deep understanding, know what really tight focus is, learn and practice one or two specific topics very deeply AND with this kind of developed mind, go into lots and lots of quite diverse topics following your interests and needs. And be brave to sometimes dig a bit deep into something which no one expected you to dig into.
So optimally your knowledge should be 'incalculable' by other people. A kind of interesting, a little bit strange knowledge-portfolio which is not very common.
TL;DR: Go very deep into something (extremely deep if possible), learn a little a bit of everything (extremely only the basics about extremely lots of things) to avoid knowledge holes, and go a little bit deep into unexpected things to make your knowledge-base incalculabe and interesting.
But it also depends to a large degree what you want to do with your life. A good junior-to-mid grade Infantry officer for instance does not really need a deep knowledge of any one particular thing, but he needs to have a broad knowledge of many things and be able to move seemlessly between them and relate them to each other.
On the other hand, for a mathematician, even saying he wants a deep knowlede of math is far too broad, it would be better to say he wants to focus in on something like Chaos Theory. Even that might be too broad and many fine mathematicians spend their careers working on a handful of closely related problems in a very narrow sub-field of a field of mathematics.