The article highlights a handful of wild-success examples and thus suffers from selection bias. The difficultly of tracking failed start-ups has been discussed, but has anyone studied that failure rate? Seems there's a paucity of coverage in popular press, even if any actual sources exist. Playing up the genius, or the under-dog, seems common-- maybe there's a rule similar to "man bites dog".

The subtext of the article seems to be that US education doesn't give kids all of the tools needed ("Skills like sales, networking, creativity and comfort with failure."). But that doesn't imply we need to discard education whole-sale. We probably just need to iterate (more often)-- cull obsolete topics and start testing new subjects.

Perhaps a passionate, motivated person can find and digest useful materials more effectively on her own than if she is distracted by formal schooling most of the day. But we won't make progress in the information age without a good degree of careful thinking. "[B]etting on the engines of future job creation" may make for a good read, but historically-informed updates to standard curricula would probably do the job.