If that's what he meant, then I have to agree with a few of the others -- it's a terribly written article even if the points are good. I appreciate the explanation!
I didn't think the article was terrible at all.
My understanding is that the 5-year failure rate of new businesses is about 40-50%. Given that an average 2x-year-old has a much lower chance of being successful enough in his first couple of jobs to (a) stay there for 5 years and (b) have it still be worth coming to work every day, this isn't a bad failure rate.
In any case, the article isn't making a case for "not doing a startup". It's making a case for being realistic about outcomes. This is a good thing. If you're going to fail, fail fast so you can move to the next thing.
Whole startup org charts (or use-of-proceeds slides) are based on the same stale shake-and-bake playbook, which is probably why so many of the resulting companies have the obligatory 4-person marcom team with the VP/Marketing, the Dir/Marketing whose job depends on managing booths at conferences, the PR/AR person who adds the obligatory budget line item for Gartner and Schwarz Communications, and "Sr. Marketing Manager" whose job it is to project manage the people that team outsources copywriting to.
