What I think you could address here more thoroughly is how you propose in your model to deal with the natural tension/divergence between structure (especially legal) and innovation. The logic from The Black Swan captures this best, which is that the more you add structure to the system, the more that the “space of innovation” shifts to be outside of the realm you have circumscribed by that structure. In other words the more you “support” innovation with standardization, templates, roadmaps and tools, the more that true innovation will come from breaking that structure in clever and useful ways.

You are arguing here that there is not enough structure in the startup ecology to support maximal innovation, I would argue that’s not true. Existing technology and culture — broadly defined — as advanced to the point of crossing that tipping point. We are on the downhill slope where the momentum is toward empowering individuals to be innovative by themselves and in ultra-fluid processes (not structures) along with other individuals. So when you ask the question about “how we should be organizing startups”, this presupposes the wrong unit (collection vs individual) and and wrong focus (organization vs process). I suggest that the type of structure you are proposing will actually stifle innovation.

So what I would say is that you are on the right track but don’t go quite far enough in terms of your radicalness. Rather than support formation of structures which naturally self-reinforce and ossify, why not simply empower individuals to connect and communicate with one another in as many diverse ways as possible and let the individuals figure out everything from there? That is, focus creating/supporting the right culture — one that involves sharing, paying it forward, disregard for property, an ethos of cooperation, trial and error, etc — and don’t even touch things like roadmaps, guides and legal structures (except ones that loosen existing restrictions). In this way, you are creating an organic system, one worthy of the term “ecology”.

As fun as it is to “go meta”, it carries the distinct risk of building castles in the air, with no connection to the real landscape. If better collaboration between startups would be so beneficial, why hasn’t it been done yet? Certainly incubators have been tried before, and they’ve all flopped. Startups are connected by funders, often forming keiretsu-like groupings. Is something stronger than that really needed, or even helpful? Why?