As the Microsoft Corporation anxiously awaits the verdict of Judge Thomas Penfield Jackson, whose decision may determine its fate (Will certain practices be prohibited? Will the company be broken apart?), the software giant could be forgiven for feeling that the greatest threat to its dominance of the computing industry is the U.S. Department of Justice.
And yet it can be argued that an even greater threat to Microsoft's dominance is a motley collection of free software tools and operating systems often called "open-source software" or "free software." Two key characteristics distinguish open-source software. First, unlike most commercial software, open-source software code is intended to be studied and improved upon by other programmers. Second, all such improvements must be revealed publicly and (under most licenses) distributed freely in a process that encourages continual innovation. Ranging from an operating system called Linux, named for a student from Finland who wrote a key part of its core programming code, to a Web server named Apache, put together by a band of volunteer programmers as literally "a patchy" set of updates for older government-funded software, open-source programs are emerging as not just inexpensive but also more robust and dynamic alternatives to commercial software.
But to say that open-source programs are just now emerging is not quite right. Largely funded by the federal government, open-source software was the creative force behind the explosion of the computer industry; it also drove development of the Internet and still comprises much of the Internet's inner workings.
So what happened to the original wave of government-backed free software? Basically, as the federal government let go of its commitment to integrated and open-software standards, it opened the door for private companies to take over the market with "proprietary" software programs--programs whose coding was kept secret and whose operating protocols were not public or standardized. Which is the main reason the average PC user is now stuck dealing with operating systems that are compatible with only some software programs, and with software programs that are incompatible with many other software programs. This, in turn, is how the way was cleared for a corporate giant like Microsoft to dominate the computing world by making its own proprietary software (namely, the Windows operating system) the de facto standard.
Should Public Policy Support Open-Source Software?
Find out more by reading the TAP Online Controversy.
Open-source software has re-emerged today largely as a reaction against the Microsoft proprietary standard. Companies that have seen their own proprietary alternatives crushed by the Microsoft steamroller have turned to alliances with open-source software to combat the Windows monopoly. For this resurgence to succeed, however--and if it is to provide an effective antidote to the Microsoft monopoly--the federal government must revive public policy that supports open-source computing. Fortunately, the outline of such a policy can be gleaned from recent history.
A Brief History of the Internet
In the 1950s, President Eisenhower founded the Advanced Research Projects Agency (ARPA), whose role it was to coordinate overall research-and-development spending by the military. ARPA gave funding to a range of university and government projects, but among the more significant grants it distributed was the $3 million per year given in the 1960s to Project MAC at MIT, to encourage the spread of time-sharing computing on the then-breakthrough minicomputer technology. Project MAC and related projects were largely responsible for cultivating the ethos of software and hardware innovation--"hacking" in its early, nonpejorative sense (before it became associated with electronic vandalism)--that fueled the computer revolution.
Another project that ARPA funded in the 1960s was the Augmentation Research Center (ARC) at the Stanford Research Institute. Decades before most people saw electronic mail or word processing software or a "mouse," ARC was developing and experimenting with these concepts. A quarter-century before downloadable pornography or the Drudge Report appeared, the researchers at ARC were using windowing environments and hypertext-linked documents in their daily work.
Together, Project MAC, ARC, and other ARPA-funded institutions formed the collaborative research network that would eventually become the Internet. In 1969 the first four "nodes" on the network were linked together; by October 1972, when the so-called ARPANET was first demonstrated publicly, there were 29 nodes in the network. ARPA, continuing to oversee military-related research and development, presided over the creation of an array of software tools--such as the TCP/IP protocols and the File Transfer Protocol (FTP)--needed to manage and extend the computer network.
Preserving the norms promulgated by ARPA and the community of researchers was essential to preventing individual profit-taking from undermining open standards. In 1973, when an ARPANET contractor refused to reveal the source code it was using, ARPA threatened to hold up the contractor's other federal contracts unless the company released the code publicly. This loudly established the tradition of open-source code in the development of Internet standards.
ARPA's network harnessed the resources of universities to provide a continuous stream of free software to improve its functionality. Innovations could ricochet almost instantly across the nation, even the world, without the friction of normal distribution or purchase costs. A classic example of such an innovation is SENDMAIL, a program hacked by a Berkeley student named Eric Allman to assist network managers in processing e-mail traffic; Allman's program is used today to direct over 75 percent of Internet e-mail traffic. Others at Berkeley created the Berkeley Internet Name Daemon (BIND) program, which directs Internet traffic by site name--like www.prospect.org--so that we don't have to type numbers like 188.8.131.52 into our Web browsers.
While free software continued to enhance the spread of computer networking, what ultimately made the Internet what it is today were the "Gopher" browsing software developed at the University of Minnesota and the Web browsing software developed in government labs in the early 1990s. With these innovations, the Internet became, in essence, one giant hard drive that could present its vast resources to the user through a few quick keystrokes or a click of a mouse.
A text-based version of the Web was initially designed at the Centre Européen pour la Recherche Nucléaire (CERN) in Geneva, and in 1993, computer science students at the National Center for Supercomputing Applications (NCSA), at the University of Illinois, created Mosaic, the first Web browser that added graphics to the traditional text display. Web sites exploded across the Internet; so did the browsers needed to view them. Both were based on the open standards defined by the Internet programmers working with the federal government.
Government funding helped support new software as a font of innovation on the Internet; government supervision helped maintain the standardization required for easy compatibility among the wide range of computers increasingly sharing resources on the Net. Despite odes to the "anarchy" of the Internet, its creation conformed to government specifications while marshaling the broad professional, volunteer, and eventually commercial resources of the emerging computer elite. In fact, the government's very deftness in pooling these resources helped render it effectively invisible. The broad consensus the government built over the Internet's shape gave the network's design so much legitimacy that it was seen less as a creation of "the government" and more as a creation of society as a whole. At each step of its development, ARPA and associated government agencies expanded participation to an ever-widening set of experts and technological leaders who, in turn, would encourage others in their academic, scientific, community, or business realm to support the effective development of the Internet.
The Story of UNIX
UNIX, a computer operating system, is the poster child for open standards. Its history demonstrates not only what happens when government policy supports open standards, but also what happens when the government withdraws that support.
Created initially at Bell Labs, UNIX was the first operating system developed to be independent of specific hardware (earlier systems were designed for specific machines). It was especially popular with ARPANET programmers working to create an integrated set of software tools for managing their emerging network, so UNIX's early life was due largely to government subsidy. In the late 1970s, ARPA funded researchers at the University of California, Berkeley, to make substantial improvements in UNIX; for example, it paid to integrate Internet protocols into the software, allowing different computers on the network to talk to each other. Then, for a minimal licensing fee, Berkeley distributed its version of UNIX, now including Internet protocols, throughout academia and, after a time, into industry.
But what made UNIX nearly universal in corporate and highend computing in the 1980s was decisive action by the federal government. In 1986 the government passed regulations forbidding any company that did not offer a standard UNIX from bidding on any government computer contract. In 1988, when the Air Force declared the Digital Equipment Corporation's proprietary version of UNIX ineligible for government contracts, the entire higher-end computing world took it as a signal that they'd better adhere to strong UNIX standards. The result was that by the early 1990s, millions of high-performance computers used in government, university, finance, and engineering offices were already Internet-compatible.
But by then, the government, perhaps cowed by the posturings of a new breed of technolibertarians, had begun backing away from active supervision of Internet standards. The immediate result? Commercial warfare over standards, as Sun Microsystems, Hewlett-Packard, and other companies lined up behind different variants of UNIX.
The Browser Wars
In the 1980s, Sun Microsystems had hitched a ride on public UNIX standards to explosive growth. In the 1990s, Netscape Communications, the Bay-area firm that became ground zero of the Silicon Valley Internet explosion, was more brazen: It began its life with a direct assault on the original governmentbased Web standards created by NCSA.
The initial Web "browser," Mosaic, was part of a larger federally funded project to create graphicsbased collaborative tools. Mosaic was first introduced on the UNIX platform in January 1993, and debuted on Macintoshes and PCs eight months later. Copyrighted by the University of Illinois, Mosaic could be downloaded for free by individuals and companies wishing to use the Internet for internal communications. Through a commercial partner, Spyglass, NCSA began widely licensing Mosaic to computer companies including IBM, DEC, AT&T, and NEC.
Meanwhile, near San Francisco, Jim Clark was watching and plotting. Clark, the former CEO of computer maker Silicon Graphics, was a veteran of the UNIX standards wars; he understood how much money could be made if a company could control the standards of this new Internet tool. Clark left his job, founded Netscape, and set out to destroy Mosaic. He hired Marc Andreesen, a member of the Mosaic team, along with a number of other NCSA programmers, to design what the company called, in-house, "Mozilla, the MosaicKiller." Within six months, Clark's team succeeded in creating a powerful browser with easytonavigate features and a speed faster than Mosaic.
On its face, this was a salutary development; the government was encouraging a wide variety of firms to make these kinds of improvements. But Netscape did something else--it went beyond the HTML standards embedded in the Mosaic browser, and included the ability to display text formatting that didn't exist in the NCSA version. This made Netscape more attractive to users than Mosaic, but it also meant that Web pages designed to work with Netscape would not be readable by other browsers and would create conflicting standards.
But it was Netscape's next move that proved most devastating to the culture of open standards: Clark and Andreesen began giving Netscape Navigator away for free over the Internet, without paying any of the required license fees to NCSA. Netscape in effect "dumped" its browser onto the Internet, thereby undercutting the rest of the commercial browser companies, which were following government rules requiring a per-copy license fee to NCSA. Netscape, as the sole enhanced commercial browser flooding the Internet, was able to destroy NCSA-led standards and take over Internet-standards creation itself, the idea being that it could then sell high-priced server software that operated according to Netscape's proprietary standards.
In the 1980s, when open UNIX standards were threatened by private standards, the federal government intervened. So what did the government do to support NCSA's standards in the face of Netscape's assault in the early 1990s? Nothing. It made no investigations into possible monopoly practices, brought no lawsuit alleging intellectual property infringement, established no requirement that the federal government would use only NCSA-approved codes in government Web sites, made no announcements that it would refuse to buy any Web servers based on such nonstandard formatting. In short, it gave no sign at all of opposition to Netscape's take-over of the standards.
Netscape realized that its defiance of open standards did not go over well in the programming community. Thus, with an eye toward an inevitable battle with Microsoft, it tried to burnish its reputation among programmers by working with the old Internet fellowship of engineers on new standards. And when Microsoft appeared on the scene with its Internet Explorer browser--which appeared automatically on every Windows desktop--the grumblings over Netscape's proprietary advantage diminished. The greater fear now was of Microsoft taking over the whole computing world. Indeed, when Microsoft began giving its browser away at the end of 1995, any alternative browsers gave way to the twocompany fight between Netscape and Microsoft.
The promise of open source
Why did the government retreat from its upholding of open standards? After all, the government's abdication has left Internet development increasingly in the hands of selfinterested companies seeking commercial advantage rather than maximum innovation and compatibility for consumers. The government's retreat was driven by a number of factors. First, Main Street Republicans expressed their usual ideological opposition to government regulation. Second, in the aftermath of the Cold War, the general public appetite for government involvement in high-tech endeavors waned. Finally, and probably most importantly, private industry--drooling at the profits to be made in commercial production of Internet software--lobbied the government to butt out.
While the Clinton administration did make some cursory gestures toward asserting a public interest in the development of what it called the National Information Infrastructure (NII), privatization proceeded apace. And what limited funds the Clinton administration did allocate for encouraging community and local government development of the Internet were vociferously opposed by conservatives in Congress. When the Republicans seized control of Congress in 1994, even those funds were initially cut off and ultimately sharply curtailed--even as the need for such funding exploded with the expansion of the World Wide Web.
When the government was the principal market for Internetrelated computer services, before the Internet was commercially viable, private industry happily benefited from government spending on network technology. But once a commercial market for Internet services appeared, private industry saw government involvement in a very different light--as a threat to corporate control of information markets. All of a sudden, companies that had started life as extensions of the government were seeking independence, realizing they could reap great profits if the government receded. Even some of ARPANET's founders, who initially cultivated the ethic of freely sharing information and software, have migrated to the private sector and are now fighting to reduce the federal government's role and to increase private ownership of intellectual property.
And yet even in the midst of this tidal shift toward privatization and proprietary systems, an undercurrent is beginning to flow in the opposite direction: Open-source software is resurgent. Technology firms have realized, after trying to create substitutes for government through private consortia and other standardsevaluating bodies, that none of these has the ability government does to transcend particularistic company concerns in favor of the public interest.
Many in Silicon Valley have become aware of the danger that proprietary technologies pose to the trust needed to sustain the collaborative model that has fueled the region's explosive growth. The greatest danger to such collaboration is, of course, Microsoft. Microsoft used a combination of its early alliance with IBM and predatory pricing tactics to build its operating-system monopoly. From there, Microsoft extended its proprietary standards into the market for largescale business computing, which had formerly been the province of mainframes or UNIXbased network servers. And while the Internet at first appeared to be a dagger at Microsoft's throat, threatening to disrupt its monopoly, Microsoft soon saw that manipulating Internet standards in a proprietary direction could extend the company's control throughout the whole world of corporate computing. By 1997 Microsoft NT computer servers were outselling UNIX servers. It was clear that in the absence of strong standards and government support for such standards, proprietary models had a decided advantage in yielding the market stability and monopoly rents that a company like Microsoft could reap.
Open source struck back. Many Silicon Valley firms saw an alliance with the global open-source software model as necessary for survival in the face of Microsoft's onslaught. These firms would forgo some profits in order to maintain the priority on innovation that gave them an advantage in the remaining commercial aspects of technology development. Sun developed its semi-open Java language, used to create programs that could run on any computer, no matter what its hardware or operating system. Netscape, meanwhile, took the unprecedented step in March 1998 of publicly revealing its browser source code--the usually topsecret guts of any commercial program. The idea was that Netscape could marshal the resources of the global programming community, still supported at many universities with federal government research funds, and then continually upgrade both its browser and server software.
While Netscape's actions did not stop the Microsoft onslaught (eventually Netscape was bought by America Online), it did highlight the promise of publicinterest-oriented software development. In fact, despite the combined market heft of Microsoft and Netscape, the most popular Web server on the Internet remains Apache, a version of the original free server software developed at the NCSA. Though much of NCSA's original work was privatized in the early 1990s, a geographically dispersed group of software programmers, some at universities and some in private business, began collaborating in 1995 to update the NCSA server to increase its power and manageability. The result was a Web server now used by organizations as diverse as McDonald's, Yahoo!, the FBI, and IBM, which passed over its own Lotus Domino server in favor of Apache when it put its "Deep BlueGary Kasparov" chess match on the Internet. Similarly, one of the favorite Web programming languages is a free language called Perl that has been modified and improved through a global network of collaborators.
Finally, Linux, a free and freely modifiable version of UNIX developed by a student at the University of Helsinki in 1991 (in collaboration with hundreds of Internet enthusiasts around the globe), has become the fastest expanding operating system in the world. Millions of copies have been installed on computers worldwide. Based on what are sometimes called "copyleft" principles, the Linux operating system can be downloaded for free over the Internet or packaged with documentation and sold for modest amounts backed by technical support by companies like Red Hat and Caldera. Extremely popular in developing nations like South Africa, Mexico, Cuba, India, and the Philippines, Linux also began to eclipse other forms of UNIX in the United States partly because of its price but also because many people considered it technically the best operating system in existence.
Over the past couple of years, new partners for Linux and other open-source software have begun to emerge. Corel--maker of WordPerfect--announced it would be releasing a full suite of office applications for Linux. IBM announced that its Web tools would fully support the Apache Web server and would join with other "Apache Group" developers to help improve the Apache code. HewlettPackard, Compaq, IBM, and Silicon Graphics have all agreed to install and support Linux for their hardware customers. Most dramatically, the country has seen companies like Red Hat and VA Linux--seller of Linux-installed hardware--conduct billion-dollar IPOs based on open-source software.
Subscribe to The American Prospect.
Where do you want to go tomorrow?
Does Judge Jackson's decision matter? Or is open-source software about to slay the Microsoft monopoly beast on its own?
While Microsoft executives testifying in court played up the marginal danger of Linux to its market share, the reality is that Microsoft server sales are still growing faster than the overall market. Linux is growing, but mostly at the expense of commercial versions of UNIX, whose growth essentially halted in 1998. Neither Linux nor a rejuvenated Apple is undermining Microsoft's complete domination of the desktop market. Worse, the Microsoft competitors who are for the moment collaborating in support of Linux might easily--especially with a little prodding from Microsoft--break into competing camps that would reduce Linux and other open-source software into a competing muddle of standards. The threat of a split among the Linux collaborators, especially with billions at stake, makes them continually vulnerable to the centrally controlled Windows standards.
Nevertheless, the limited success of Linux and other open-source software does have implications for the Microsoft antitrust trial. The availability of open-source software is not an excuse to find Microsoft innocent of the wholesale monopolistic abuses that the trial has exposed. But it may become one of the remedies that the court and other government agencies use to rein in Microsoft's monopoly power. Many, including my former organization NetAction, have proposed remedies to Microsoft's monopoly abuses, from breaking up the company into multiple competing units, to courtordered limits on its licensing agreements, to forcing Microsoft to reveal its source code to prevent inhouse coders from having advantages over competitors in nonopen-source markets.
While such government restrictions are likely necessary, none of them speaks to the issue of creating a strong viable alternative to Microsoft. The federal government already spends billions of dollars on software research, purchases, and implementation. If it consolidated those resources in support of open-source solutions, it would not only expand many of the clear advantages that open-source software delivers, but it would also simultaneously undermine the Microsoft monopoly. If the government revived its collaboration with top programming talent to define the best standards and used its purchasing power to require that those standards be met in government contracts, this would go a long way toward challenging the Microsoft monopoly and preventing fragmentation of standards throughout the open-source universe.
Many critics of the Microsoft suit raise reasonable concerns that a purely negative, restrictive approach to punishing Microsoft might inhibit innovation at the company without necessarily creating a viable competitor. Promoting open-source software is the positive policy option that the government should employ to encourage the sort of innovation and competition that created the Internet in the first place and is needed now to end the Microsoft monopoly. ¤
To read the full NetAction report on which this article is based, go to