Where this guy goes wrong is that he assumes (without realizing it) that there's some limit on the number of startups there can be. But that's probably not true.
No one proposes there's any limit on the number of people who can be employed by companies with thousands of people. Why should there be any limit on how many fragments that effort can be broken into?
(if I'm correct in my admittedly superficial understanding of his work)
Although that's probably not what the article's author had in mind when writing it, and I think it's a bit tangential to the feeling that I get too that there is a bit of froth in the market.
I'd agree with you that there's plenty of room out there, but offhand, I'd say the question to ask is whether the "width" of what startups are working on is growing proportionally to the number of startups, rather than simply the "depth" (people working on similar things).
False Dilemma. The author frames it as if its a choice between starting a startup that has a high probability of failing or doing nothing. However, most people who don't start startups have to work work for big companies, which have all the same problems but worse.
There are opportunity costs involved with starting a company instead of working at BigCo, and if there is, as the article says, an "entrepreneur bubble", then some people are doubtless making the wrong choice.
the writer of the articles compares two unequal things. The 80s with this era... The situation is a lot different. People back then didn't have any other channel to reach potential users, than marketting, which they had to pay.
The web is a new place, and the market is broader and global. You can be next to your grandma in Greece and sell to a client as far as Brazil through word of mouth......
No one proposes there's any limit on the number of people who can be employed by companies with thousands of people. Why should there be any limit on how many fragments that effort can be broken into?