What It Will Take to Create the Next Great Silicon Valleys, Plural

The thing that's missing from almost all discussions of making a "Silicon Valley" somewhere else is what I think is the key to the original Silicon Valley's success: intergenerational investment. That is, a bunch of guys hit it big on semiconductors in the mid 20th century, and instead of just buying boats and joining the country club, these people set aside a portion of their fortunes to invest in the tech-related businesses of the generation that followed. And, crucially, without extracting onerous terms from those investments. That established a culture of relatively permissive angel investment that encouraged innovation.

Trying to get a startup off the ground in most other places it too hard because raising seed money is too hard, and even if you raise seed money, raising a series A is too hard, and so on. To make a Silicon Valley, you have to seed the ecosystem with an initial generation of successful entrepreneurs who are wiling to adopt the ethos of the SV angel investor, then provide an investment infrastructure to make sure that promising startups don't die on the vine.

And, as many others have pointed out, an open spigot of government money was necessary for the original SV to develop, so that's probably necessary for any copycat.

>> Imagine a Bitcoin Valley, for instance, where some country fully legalizes cryptocurrencies for all financial functions. Or a Drone Valley, where a particular region removes all legal barriers to flying unmanned aerial vehicles locally. A Driverless Car Valley in a city that allows experimentation with different autonomous car designs, redesigned roadways and safety laws. A Stem Cell Valley. And so on.

On the surface this sounds kind of neat; on a practical level, this sounds like a libertarian nightmare.

> In fact, this kind of competition is probably the only way to create successful innovation clusters that can compete with the huge advantage Silicon Valley already has. . . . That’s why turning Detroit into a commercial Drone Valley could draw the innovative people who in turn want to be near other innovative people around that domain.

I find this akin to describing how other cities should create their own Wall Street. I don't see the point. Wall Street and Silicon Valley are both places that benefit so much from the concentration of capital (intellectual and monetary), that there is little practical reason to have more than one of each in the U.S. And outside the U.S., places with capital and technical talent will find a way to put them together.

What I'm interested in is the harder question: what will it take to create the next Detroit, circa 1950? Not just a place where new technology is developed, where wealth is created, but a place where technology creates broadly-shared prosperity and good middle-class jobs.

>> That’s why turning Detroit into a commercial Drone Valley could draw the innovative people who in turn want to be near other innovative people around that domain.

Detroit was built the automobile. It's a bad comparison to drones. Every househould in America isn't going to own 2.2 drones. Drones aren't so incredible complicated where countless other companies will pop up, making tires and seats and radios and spark plugs and bearings.

Let's assume a successful drone-building company sets up shop and ends up providing 2,000 jobs. That's a great gain for Detroit, but you're still looking to fill a huge void in a town that used to have 10,000 workers in a single plant.

What will it take to create the next silicon valley? Money. Lots and lots of government money. Defense-sized pots of government money. Each year. For a decade or more. That, and the ability to live there and speak only English.

There's a phrase for things like electricity and smartphones: general purpose technology https://en.wikipedia.org/wiki/General_purpose_technology

As for more Silicon Valleys, it's telling that to my knowledge outside California have made non-competes unenforceable, although Massachusetts is getting a clue more than 2 decades after Route 128 died a hard and painful death.

Marc Andreesen believes that someone should not be commenting on the theory of Disruptive Innovation if they have a Ph.D. in American Studies from Yale.

And now he wants to tell me what?

A ubiquitous, world changing, radically different innovation (like computers) in which hundreds or thousands of companies can simultaneously participate.

Is there such a thing on the horizon?

I don't think drones are it. Sorry Marc. You were in the right time at the right place and had the right skills. Don't expect this combination of timing, luck and effort to be easily duplicated elsewhere.

Trying to attract investors to your city is how you get a gold rush.

Drone Valley? Yeah, the FAA would like a word with you about flying test planes over populated areas...

This blog post seems like a good opportunity to bring up another essay I thought was very good by Andy Grove.

Andy Grove: How America Can Create Jobs

http://www.businessweek.com/magazine/content/10_28/b41860483...

The part that really resonated with me here was the issue of scaling in manufacturing. Much of his essay takes on the notion that the US will be fine if innovation happens in silicon valley and manufacturing happens elsewhere (this is an oversimplification, I'd recommend you read the essay).

But what really resonated with me here is that scaling is innovation, and that by splitting the two and neglecting the second, the US has put itself at a disadvantage, not just in the jobs generated downstream from high end innovative work, but in the ability to do that innovative work itself...

"I believe the answer has to do with a general undervaluing of manufacturing—the idea that as long as "knowledge work" stays in the U.S., it doesn't matter what happens to factory jobs... I disagree. Not only did we lose an untold number of jobs, we broke the chain of experience that is so important in technological evolution. As happened with batteries, abandoning today's "commodity" manufacturing can lock you out of tomorrow's emerging industry."

One reason this resonates so heavily with me is that I actually got an MS in Industrial Engineering, and I, like almost all of my cohort, work in software now. The closes I came was writing software for manufacturing systems, and many of our clients were overseas. But I never did learn how to scale, and an entire generation of potential leaders in scaling up manufacturing the US pretty much had to find something else to do (perhaps using those stochastic processes to figure out how to get people to click on ads rather than reducing defect rates in the mass scale manufacting of solar panels).

When it's time to manufacture mass numbers of drones (per the discussion above, I could easily see the average US family owning quite a few drones). We may not make them in the US, but the reason might actually not be cost, it may be skill - we may simply not have the experience with scaling on physical systems - partly because we skipped a couple of generations and we just don't have the experience anymore.

Now (my own hobby horse), add in the fact that the US seems to be perfectly OK with watching the enrollment of US citizens in graduate engineering programs plummet (perhaps a rational decision at the individual level), and if high end work shifts along with production, we may find we can no longer wave the magic wand and get talented people to come here. Fix immigration all you like, the top talent may go elsewhere and no longer bother with the US.

I'm putting forth a bleak scenario, and I'll admit it contains a few slippery slopes, so this is meant as a call for concern, not a prediction. But I think that the US needs to pay a lot of attention to our future workforce in serious, large scale manufacturing engineering, and we're neglecting it, badly.

A dupe of https://news.ycombinator.com/item?id=7902282, but we won't bury it because (a) the current article is a bit longer and (b) the title of the earlier submission was Detroit-specific, which caused the discussion to be.