AST week, the Supreme Court heard oral arguments in Metro-Goldwyn-Mayer Studios v. Grokster, a case that has important implications for the future of online innovation.

Grokster makes software that enables Internet users to share computer files on peer-to-peer networks. The technology has been used to distribute many kinds of content, including copyrighted digital music.

MGM and other entertainment companies want to hold Grokster liable for the copyright infringement that occurs when users download copyrighted music without paying for it. Grokster argues that there are many legitimate uses for its technology and that it is not responsible for those who use it to violate copyrights.

This is just the latest installment of a longstanding battle between technology companies and copyright holders. It is useful to look at the history of some of these past innovations in trying to understand what policies may be appropriate today.

In the early 1900's, the disruptive technology was player pianos. Manufacturers of player piano rolls purchased a single copy of the sheet music of a song, hired someone to record the music and then sold these mechanical reproductions to consumers. The songwriters held that this was copyright infringement, while the piano roll manufacturers pointed out that they had paid the appropriate copyright fees when they purchased the sheet music.

In 1908, the Supreme Court found in favor of the piano roll manufacturers, but practically invited Congress to consider new legislation on the issue. Congress responded with the Copyright Act of 1909, which created a new form of intellectual property, mechanical reproduction rights.

The new law required piano roll manufacturers to pay songwriters a fee for each song. Subsequently, mechanical reproduction fees have been extended to new technologies like phonographs, audio tapes, CD's and online streaming digital music.

In the 1908 case, songwriters did not try to ban player piano technology. They clearly recognized that the additional distribution of their songs was potentially advantageous. Their goal was simply to get a fair share of the proceeds from the piano roll sales.

Another directly relevant Supreme Court decision is Sony v. Universal City Studios, a 1984 case involving the use of video recorders in the home. The film studios argued that Sony should be liable for copyright infringement since its video recorders could be used to copy movies and television programs illegally.

The Supreme Court held that Sony was not liable since the VCR technology had "substantial noninfringing uses."

This phrase has since become the legal test of whether liability can be imposed. Under this doctrine, a company that sells a technology whose only use is to violate copyright could potentially be liable for infringement, but as long as there are substantial noninfringing uses there would be no liability.

The studios lost the Sony case, but it forced them to take the home video market seriously.

Their first instinct was to set a $50 to $60 price for videocassettes. But by choosing a high price, they stimulated the development of the video rental market, giving users inexpensive access to movies.

On the other hand, the availability of rentals stimulated the demand for VCR's. As VCR prices declined, more people bought them and the video rental industry flourished, creating a new, rapidly growing outlet for studio productions.

In the late 1980's Disney began to experiment with lower prices for videos, hoping to bypass the rental stores and sell directly to home users. Disney's 1987 video release of "Lady and the Tramp" was priced at $29.95 and sold over 3.2 million copies, making it the best-selling video as of that date. Its record was soon eclipsed by "E.T.," which sold 14 million copies at $19.95 apiece.

These examples convinced Hollywood that if it priced its product low enough it could successfully compete with the rental market. When DVD technology came along in 1996, Hollywood understood that pricing under $20 was critical. DVD technology has been hugely successful because the prices of the players and discs have continued to decline, making it highly affordable and widely used.

The critical lesson from the history of the VCR is this: If consumers have ways to share content, either via rental markets or via the Internet, you will have to set low prices to induce them to buy. But low prices may well stimulate enough volume to make up for the lost revenue.

Apple's iTunes, with its 99-cent price for songs, has driven this lesson home, but there are those who argue that prices should be even lower.

In 2004, RealNetworks experimented with charging 49 cents for digital songs and sold more than three million downloads in a three-week period. The chief executive of RealNetworks, Rob Glaser, says that "the pricing that will result in the biggest overall market for music will involve some kind of tiered pricing," with "new mainstream songs for 99 cents retail, and up-and-coming artists and back catalog artists at a lower price."

It is worth observing that this is similar to the pricing strategy used in the video industry in the 1980's: a high price for the videos that were likely to be viewed only once, making them natural candidates for the rental market, and a low price for videos that warranted repeat viewing, making them candidates for purchase.

So what should the policy be for new technologies like Grokster? I advocate the Pizza Principle: If you want everybody to get as big a slice as possible, you first have to figure out how to bake as big a pie as possible. Once you have a nice big pie, you can let people fight over how they slice it up.

With respect to technology, the Sony decision got it right: encourage technologies that create more total value. Then, let companies fight to find business models that deliver that value to consumers. They can be awfully creative when they are forced to be.

Hal R. Varian is a professor of business, economics and information management at the University of California, Berkeley.