Old analog televisions had controls called horizontal and vertical. One kept the picture from falling over sideways, while the other kept the picture from sliding off the top or the bottom of the screen.

I think we need similar controls in our heads when we look at the product categories we call markets.

In the world of free software and open source, we mostly care about the horizontal: creating code, standards and other means for expanding markets sideways, toward the horizons. We work to keep both developers and users free from capture in the vertical markets we call silos, so they can do more stuff in more ways. We don't often ask “Where would we be without silos?” because we know. We would be free to build and use anything we want, out in the market's wide-open spaces.

Still, innovation happens in silos too. Many of the technical graces we take for granted would not have happened outside the walls of silos built by Apple, Canon, IBM, Intel, Sony and other large companies with ample and hardened intellectual property portfolios. Linux itself was developed originally (and still primarily) for Intel's x86 CPUs. Much of what we take for granted in chips from Intel and other makers is thick with intellectual property protections we would hate to see applied to our own software work. Every large maker of original electronic products (including all the companies listed above) produces between dozens and thousands of new patents every year, adding to portfolios that muscle licensing income and deals of many other kinds. Occasionally, companies do battle in court, but most of the time the dealing is quiet. If revealed, it is only through pro forma small-print disclosures in documentation.

What matters is that these portfolios give large makers the confidence and security they feel is required to produce original and appealing goods for which there is little or no competition. This is an ideal to which Apple, for example, constantly aspires. Back in 1997, not long after Steve Jobs returned to lead Apple after a long interregnum, he killed off cloners of the company's computers. Here's what I wrote about it at the time in an e-mail to Dave Winer (which Dave later published):

The iPod, iPhone and iPad each not only fulfilled all six of those requirements, but also redefined their market categories in the vertical dimension. That is, they grew the range of things that could be done with a given device, and the size of its marketplace. The iPhone in particular redefined the smartphone market and enlarged it far beyond the narrow range of possibilities allowed by combinations of mobile phone makers and mobile phone companies. New kinds of applications by the thousands burst out of the ground like a geyser. But all are contained inside Apple's silo, where they move only through the company's sphinctered approval and sales processes.

In fact, far more can be done on Linux (notably Android) and Symbian OSes than on iPhone's, just given the open nature of the former and the closed nature of the latter. But, thanks to Apple, there is much more to imagine doing outside that company's closed and private silo.

This is the point at which some suggest that open-source goods are derivative, rather than original. But that's not the case. We still happen to live in a time when investment in closed, and original stuff is greater than investment in the open and original kind. As long as that's still the case, we'll have the Apples of the world aiming for the heights while the rest of us build out the widths and the depths that characterize wide-open marketplaces.