Perils of open source

Some folks might say that "open source" has perils of its own, that it "doesn't work" or some such. However, does this sentiment match reality? Practically speaking, here are some examples of perils of open source.

Makerbot Industries

How does Bre Pettis go from this:

"... this is something we learn as kids, that sharing is good. When you share something you get more back from it. But we forget this as adults. So, with open source hardware, we're back to that. When you get a makerbot, you're not just getting a machine. You're getting the knowledge of how it works; you're getting information about everything that puts it together. If you want to modify it, or you just want to learn about it, if you want to hack it, you can do it." Makerbot TV

To this?:

"... how do we encourage people to grow the projects into business, and how to grow their business into successful large businesses, that can hire a lot people, and let people shift from doing open source from something they do on nights and weekends, into something that do for their day job." Makezine

"Until we merged with Stratasys, we spent a lot of our efforts … respecting their IP, and now we kind of have our hands in their cookie jar," GigaOm

It seems the narrative is that the right path for open source is to engage in large-scale entrepreneurship, business development, and corporate models. Perhaps the question should be not whether open source is viable for business, but rather, whether business-as-usual is viable for open source.