The Last Refuge

“Patriotism,” said Samuel Johnson, “is the last refuge of a scoundrel.” In software, it seems, despair more often takes the form of lawsuits, and, in particular, intellectual property claims of the most dubious variety. If your market share keeps slipping, and you’ve lost the ability to innovate your way back to victory, Plan B is to sue your competitors. When you can’t win through innovation, in other words, try litigation.

So far, thank goodness, this strategy has never worked. In fact, a review of some of the most famous cases in the rather dismal corpus of software-related lawsuits suggests just the opposite: that bringing such lawsuits may further distract companies from the business of serving their customers, hastening their fall from grace.

Three cases from the early 1990s amply demonstrate the futility of using the courts to regain a lost lead:

  • After Borland introduced a more sophisticated spreadsheet program, Lotus sued, claiming that Borland’s “1-2-3 compatibility mode” illegally cloned the structure of Lotus commands and menu structures.

  • When gamemaker Accolade reverse-engineered the protocols needed to make games for Sega’s Genesis console without first obtaining a license, Sega charged that Accolade had violated their right to keep those requirements a secret.

  • Once Apple found that the “inferior” Windows operating system was nonetheless winning customers, it sued Microsoft, arguing that despite a licensing agreement between the two companies, Windows had gone too far in borrowing the windows/desktop user interface embodied in early Macintosh computers.

    In each case, the company doing the suing had once enjoyed a dominant position in its market—dominance that had been eroded by new products from new competitors that users apparently liked better. In each case, litigation was a last-ditch effort to hold onto what was left of the plaintiff’s market share, or even somehow to turn back the clock. In each case, the plaintiff’s strategy was to stretch intellectual property law to protect elements of its software—a menu structure, a set of compatibility protocols, a design metaphor—that hadn’t previously been thought protectable.

    Lotus, Sega and Apple all lost their cases, and soon after they entered the darkest days of their corporate histories.

    Ten years later, there is plenty of evidence to suggest the software industry has learned nothing from its own history. In the “live by the sword, die by the sword” category of ironic litigation, consider the fate of struggling hardware and software maker Sun Microsystems. In a case related to the various worldwide antitrust actions against Microsoft, Sun managed a settlement with Redmond, in 2001, over alleged misuse of a license to redistribute Java, Sun’s semi-open Web standard for transmitting small packets of application code.

    Just a few months ago, however, Sun itself lost a jury trial in which Kodak, a company with problems of its own, claimed Java violated patents Kodak had acquired from Wang Laboratories. Before the jury could decide how much Sun owed, the parties settled for close to $100 million (Kodak had been seeking $1 billion).

    Or consider an even more delicious irony: a lawsuit filed against the Recording Industry Association of America by P2P software vendor Altnet. The RIAA, as you know, is busily suing the teenage customers of its member companies, the major recording labels, for copyright violations, a last-ditch effort to save the hopelessly obsolete “system” of reproducing music using expensive and now-unnecessary media such as CDs (not to mention the factories, supply chains and stores required to distribute them). The suit, filed in October, charges that the technology the RIAA is using to search your kids’ hard drives for infringing MP3s is itself an infringement of Altnet’s patents! So RIAA, now who’s the “pirate”?

    Last but not least is the growth industry of litigation involving the SCO Group and the Linux operating system. SCO, in a previous incarnation as Caldera, acquired certain rights to UNIX from Novell, which had acquired them from AT&T. But UNIX has been rapidly losing ground to Linux, the popular open-source operating system that mimics many behaviors of UNIX. So in 2003, SCO sued IBM, claiming various abuses of its intellectual property rights in UNIX. Like Kodak, SCO sought $1 billion in damages—apparently the going rate to save a failing business.

    Interestingly, SCO doesn’t claim that Linux copied its code from UNIX, but rather that companies embracing Linux had previously sworn allegiance to UNIX, and in making the switch to Linux violated promises made to SCO or its predecessors. The current complaint (SCO has started over twice already and is arguing for a third try) rests on the proposition that IBM’s shift to Linux violated various licensing agreements it had with AT&T and others.

    Since filing its initial suit, SCO has added additional defendants (including Linux users) and moved aggressively to pressure hundreds of companies to purchase “licenses” from them—licenses that would cover whatever interest SCO may or may not have in Linux—in short, a kind of immunity from liability in the event that SCO proves its principal case. SCO now seems determined to transform itself from a software company into a software insurance firm—insurance being a nice way of describing what this really looks like: protection money.

    Regardless of the merits of whatever claims SCO ultimately makes, the company’s pursuit of licensing revenue from non-customers is—whether by design or coincidence—likely to result not in more revenue for UNIX, but for Windows, SCO’s theoretical competitor. (Microsoft has been a licensee of SCO since 2003.) How? By generating anxiety among potential Linux customers, SCO’s actions will not win it new UNIX customers but instead encourage Windows customers at the margins to stick with Microsoft.

    Entertaining as these stories may be, these cases underscore an important reality about the law of software, which is that the system of handing out and enforcing intellectual property rights developed for books and machines has proven itself a poor fit for the unique features of the software industry. (More on that in future columns.)

    For now, here’s a simple but important lesson CIOs can learn from the sorry history of software litigation: If your most important vendors start to sue rather than compete, then the likelihood is high that they’ve lost their business edge and are unlikely to get it back. That’s your cue to start diversifying your vendor portfolio. The more strategic the product in question is to your business, in fact, the more urgently you need to look for alternative providers, products and partnerships.

    It’s also a further endorsement, if one were needed, to move as quickly as possible to open standards and nonproprietary architectures. Freeware, shareware, and the General Public License that accompanies Linux (Richard Stallman, who wrote the first open source license, refers to it as a “copyleft”) have all been around for decades, but have only been gathering momentum in the past few years. Think of their growing popularity as the beginnings of a popular revolt against the excesses of intellectual property law—a form of civil disobedience to protest the software wars.

    As the SCO case suggests, open-source developers and users are hardly immune from litigation, but it does seem to be the better hedge against collateral damage caused by a desperate combatant. The collateral damage I’m referring to is you.

    Larry Downes teaches law and business strategy at the University of California-Berkeley School of Information Management and Systems. His most recent book is The Strategy Machine: Building Your Business One Idea at a Time (HarperBusiness, 2003). His next column will appear in April. Please send comments on this column to

  • Get the Free Newsletter!

    Subscribe to Daily Tech Insider for top news, trends, and analysis.

    Latest Articles