Several standard-setting methods have developed. Only in a few instances have they depended on a monopoly to set them. The standard that's eventually embraced by mixed market forces might not be the best standard, but it's at least as likely to be a good one as a standard imposed on a market by a monopoly.
You're right, there are several methods that have evolved over the years for setting standards in the computer industry. I didn't mean to imply one method would dominate. However, in markets (or market subsegments, like the OS) where there are huge dollars at stake, the 'standards' have traditionaly been set by the market leader.
Commodities goods such as protocol stacks and wire protocols (e.g., TCP/IP and HTTP) are best left to standards bodies. Yet variants of each survived for a time, ultimately dying off. Standards here have fueled the growth in the computer industry by providing the bedrock of interconnectivity.
One could argue that the OS is also becoming a commodity. Over time, operating system components like domain security, network file systems, video and audio became very well understood. This gave rise to Linux. Well understood problems coded by enthusiastic programmers wanting to take on the evil empire by cooking their own OS. There are few real innovations in LINUX (except perhaps the process by which it grew, but that's another issue altogether).
I guess the cynical would argue that any company or entity that succeeded in making a Unix variant easy to install and easy to use is innovating. But they'd be wrong. OS's like MAC OS, OS2 and Win NT have paved the way for a linux to appear. It seems that the prospective LINUX user has already scaled his/her learning curve using these other operating systems.
The real irony is that had their not been a Microsoft Windows (or another truly dominant OS) there would probably not have been a LINUX. It was the push-me, pull-you characteristics of a platform standard that created the huge install base of PCs upon which LINUX can now reside.
Once the standard Win32 APIs were in place, the cost of sofware dropped, the cost of home ownership dropped, the risk of buying a PC dropped and the install base grew. I don't mean to discount Intel here, but clearly without a standard OS, it's all just silicon, microcode and a lot of one-offs. I can't wait for the EE crowd to flame me on that point.
But as I argued earlier the consumer really won. The OS market is nearing true commodity status. What does it cost to put an OS on a machine?
The huge profit potential in the OS market drove Microsoft to innovate and a survival of the fittest mentality prevailed (not necessarily based on just the technology). And MS won that fight. The open-source phenomenon would not have succeeded in the first 15 years of the PC revolution.
Of course Microsoft was aided by market forces that weren't completely understood. Increasing returns is the key. First mover advantage is real.
In the later days of the industrial revolution markets moved slowly. Profit motive increased the level of competition until finally the market was too crowded and merger, consolidation and bankruptcies would thin the market to a few strong companies. The US automotive industry is one example of this. It also took about 60 years for it to happen.
Today the technology markets evolve much more rapidly and progress from the nacent stages of capital attraction and competitiveness to market leader and standard bearer very rapidly.
The question is does our government really understand this phenomenon? I doubt it. |
