Over the years lots of cities have wanted to replicate the success of Silicon Valley. I have some ideas how to do that, and why the way they've been going about it is wrong. 

I migrated to Silicon Valley in 1979, from Wisconsin, where I was previously a graduate student in Computer Science. I had an idea for a product, and a working prototype, and couldn't find people to partner with in Madison at the time. There were no computer stores, just a few consultants, and very few people with personal computers. I had one of course. I couldn't wait until they were commercialized. Too good an idea. 

I also got BYTE Magazine, and all the ads were for places in California, places like Sunnyvale, Mountain View, Palo Alto, Cupertino, Los Gatos. I looked at a map and saw they were all in about the same place. So I figured that was where I should go. So I loaded up my hippie van and drove there, and found lots of people who, like myself, were excited about the future of personal computers. 

If I were a country musician I'd go to Nashville. But I made software so I headed to Palo Alto. That's how I explained it to my friends at the time.

I was drawn to the original Silicon Valley, very early for PCs, so I have some idea what the appeal is and how I decided to go there. The most important thing was seeing that other people like me were already there. That was what drew me there.

Today there is no BYTE Magazine, but there are other ways developers find out where they should go. If I were going to do it today, I'd try to build an association between my locale and a popular open source project, or a new project started by someone who has credibility as the leader of a project. It's the technology and personal leadership that matters. Company names matter too, but only if you're hosting the company headquarters.

Usually localities go for large capital investments, tax breaks for big corporations, or infrastructure like high-speed networks. Your small state or city might find the money to do that, but that's not your advantage, there almost always are other places with more money to throw at BigCo's and infrastructure. and the money is better spent on education and open source, where it will have a more lasting effect. 

Those two things, education and projects, will do more to fuel the development of your tech culture than the presence of an outpost for a company headquartered in the valley. They will always see it as an satellite, the important jobs will still be in Menlo Park, Cupertino or Sunnyvale. 

Expanding on this idea -- a university-sponsored open source project would be even better. It can easily and inexpensively host conferences. People will come to your school to get educated on the project, and graduate having already made contributions. Mid-career project participants could return to university to teach, or to gain new skills. And the students in turn would put down roots in your community so they'd be more inclined to stay there after graduating. 

In my own case, had Madison had enough other people to work with, I might have stayed in 1979, because it has a lot of things going for it as a tech center. At least it did at the time. 

So what would step one be? Recruit the people who start open projects, and give them incentives to relocate. Treat them the way you'd treat talent for your local sports teams. Money is a good thing, but you should also have liberal laws, great things to do. Even if nerds never go out they like to think they could, and their families like it. 

Consider Boulder as an example, they have the Rockies and a fine university. Good places to eat and hang out. Bike paths. Denver, a major city, is nearby. A big airport. All the basic ingredients for booting up a tech culture, and that's exactly what's happening in Boulder. 

Fast Internet would be good too. And a liberal governor and attorney general who's ready to push back against oppressive policies from Washington.