I used to practice anti-trust law, back in the old days where there was such a thing. In thinking about the DOJ case several things occur to me:
(1) Software is different.
A classic abstract example in anti-trust law deals with marginal cost. In many markets the cost of making an additional item declines (i.e. capital is required). If two competitors compete, one will eventually obtain all of the market because one will get ahead of the other on the price curve, his product will cost less to make, he can price it lower, etc, etc.
Real life is largely void of such examples as the cost of selling tends to increase.
Software is an exception. The product is intellectual property and the marginal cost is often near 0. What is the cost to Microsft of the hundred millionth copy of Win95? The cost of printing a certificate-- Microsft has offloaded the other costs (maintenance, repordcution, etc.) to its buyers. Moreover its sales model is incredibly lean-- because despite the misconception Micorsft does not market to the end consumer -- Microsft sells to computer OEM's -- a market in which 90% of the business is probably held by 20 firms.
Software is an exception in another way. At one level there is no competition by statute. Windows95 is a monopoly, no one else can make it, because Congress, in its undoubted campaign financed wisdom, has extended copyright and patent law to embrace software. You go to jail if you make Windows95 without Microsoft's consent.
This matters because customers become "committed" to Win95. My copy of WordPerfect8 will not run on OS/2, or beOS. or Unix.
Thus in several ways software is an exception more radical than the extremes that wre dreamed up by profs when I took anti-trust law.
Throw into this mix the weakening of antitrust law that has ocurred over the last three decades. We say, for instance, that there cannot be predatory pricing if a product is priced above marginal cost -- Turner & Areeda's concept that may have made sense when dealing with machines and factories, but which becomes nonsesnical when applied to software.
We have effectively eliminated Robinson-Patman, the statute which could protect ainst attempts to bu market share, not by repealing it, but rather by strained cavortings of reasoning and definition ("and thus we can see, that in a legal sese, no one is really harmed when the plaintiff was bankrupted') to render it a nullity.
What Microsoft is,is wrong. It is offensive in a sense of economic fairness. It stifles creativity. It harms consumers. Today you pay three to four times for your operating system what you paid 8 years ago. In a year or two you probably will pay 6 to 8 times as much as you are forced to NT, whether ou like it or not. Sure, modern operating sytems offer much more -- but in other software we have seen prices plunge even as features multiply.
We need, I think to rethink anti-trust law, and its applictaion to software, and we need to rethink the consequences of applying copyright and patent law to software.
Regards,
John
John H. Lederer
<johnl[_at_]ibm.net>
Received on Sun Oct 26 1997 - 13:40:38 GMT
This archive was generated by hypermail 2.2.0 : Mon Mar 26 2007 - 00:35:27 GMT