In Picking Winners, Mark Goldberg makes the case that governments shouldn't intervene in picking standards for technologies. He cites beta vs VHS, Windows vs Linux, IM services and social networks, as some of his examples. In large part I agree with him, but let's dissect each of these in a little more detail, because they are each different in important ways.
In the case of beta vs VHS, market conditions decided the winner. Sony tried to monopolize the market for home recordings by creating a higher quality recording solution, and consumers flocked to the variety of video devices, choice of manufacturer, longer recording times, and lower prices of Matsushita's competing VHS standard. Matsushita was smart and played the horizontal licensing game to Sony's vertically integrated proprietary stack, niching Sony's beta as a professional recording solution only. Consumer choice was well served here.
In Windows vs. Linux, competition is made possible because the underlying technology stack — the PC — is a commodity built from a public specification. Moreover, the application specification is also publicly available, allowing anyone to build applications for these platforms. The historic anti-trust decision which required manufacturers to not build the Windows royalty into the PC if the customer didn't want Windows, paved the way for competition to emerge on an equal footing. Consumer choice is also well served here.
The world of IM services is sadly not one that we should hold up as an example. At the moment we have five mostly closed and mostly not interoperable platforms. Granted, there has been some movement as Yahoo and Microsoft have agreed to interoperate, but it still boils down to consumers choosing to either install all the IM clients, or deciding which of their friends and associates to not communicate with using IM. Consumer choice is poorly served.
Social networks are similar phenomena to IM networks. Despite the emergence of standards like FOAF, it is not possible at this time to easily move from one social network to another. With the notable exception of Facebook, it's also not possible to easily build applications for the network.
So what does this have to do with wireless networks? We can generally say that in technology products built from components where some degree of horizontal commoditization has occurred, whether through market forces or by fiat, there exist opportunities for businesses to prosper serving segments and niches of the market. In technology products that don't have that horizontal commoditization, these opportunities don't exist. The Canadian wireless network is, today, three vertically integrated networks built on two incompatible technologies. Moreover, a unified standard is unlikely to emerge because the three players in that space like the lock-in provided by the incompatible technologies. It makes it incredibly difficult for applications providers to do business here, and limits the choices the consumer has rather dramatically.
Let's take another example: cameras. I own Nikon gear, and have done so for nearly 25 years. It's the product of a decision my wife made in 1984 (buying a Nikon film camera), and an accumulation of expensive accessories (mostly lenses) that are incompatible with any other camera manufacturers products. When we bought the D50 last summer, no other manufacturer was really a consideration. We're locked in. No choice here.
How about a more universal example: power. At the turn of the 19th century, Edison lost the fight with Tesla over direct current versus alternating, but then the governments of the world implemented competing national standards everywhere. What would the world be like if there was a single unified power standard (let's say 220V AC), and one plug standard? Electronics products wouldn't all have converting power supplies, which convert the power from one standard to another at the edge of the grid, and the business of providing those endless cases of adaptors for travellers would cease to exist. We would be able to move anywhere in the world without having to buy new appliances at the destination. And consumers would be served with lower cost products that didn't require converting power supplies and and the confusion of multiple plug ends.
Power is a commodity, and it's a mostly open commodity due to the ubiquity of conversion technologies.
The wireless infrastructure is also a commodity. The Europeans got it right when they recognized this and mandated GSM over CDMA. One or the other didn't really matter, by the way. There are technical differences between GSM and CDMA, but at the end of the day nobody except industry people and operators care. What was important was standardization because it allows the ecosystem around the rest of the technology stack to flourish. By choosing not to choose, however, regulators have allowed an artificial lock-in to occur built around networking technologies. The absurdity of the latest round of telephones with CDMA built in for North America, and GSM for the rest of the world illustrates this problem very neatly. Two standards, two sets of royalties, and lord knows how many radios results in a limited choice of handsets, and more expense for the consumer. It's as if we allowed right-hand and left-hand drive cars in North America, built separate highways for each and invited the road operators and auto manufacturers to compete. Then, because some people's houses and businesses are on one road/auto standard and others on another, the industry decided to build cars with two steering wheels that can drive on either road standard.
Consumers are best served when commodities are delivered in standard ways. And because monopolies tend to act in the best interests of shareholders rather than consumers I would argue, in disagreement with my friend Mark, that when the market reaches a point where competition is not being served, standards should be dictated.