There are several inexpensive ways in which government could help the open source effort:
Being the world's largest consumer of computer software, the U.S. Government has the ability to promote the widespread use and continued development of open source software through its purchasing policies. Not only would many government agencies benefit from the added reliability and security which OSS products provide, but the increased demand for these products would encourage more corporations and independent programmers to embrace OSS methods. This trend has already begun on a small scale: the U.S. Postal Service, for example, uses a highly modified version of Linux to read addresses on envelopes electronically. Many other agencies use Linux for network administration tasks, as it is considerably more affordable then the competing Windows NT software. A risk-free way to assess the benefits of OSS to particular government agencies would be for Congress to initiate a study by the General Accounting Office. The conclusions of such a study could serve as a road map for future software procurement. The study could address the following questions:
Another possible government action involves the vast pool of software created for internal tasks within the government and the military. Collecting nonclassified source code in a series of repositories for the purpose of allowing public access would benefit both government and the public. Companies and individuals will have access to the expertise of government and military software engineers, obviating the need to solve software problems which have already been solved. Additionally, if some individual or group takes an interest in improving some piece of software in use in a government agency, the agency will reap the benefit, at no cost to taxpayers.
Many agencies would no doubt object to the perceived security risk involved in disclosing government source code. As mentioned above, computer security experts consider this argument fallacious. A determined attacker can find security flaws in software with or without the source code, so concealing the source is actually more of a hindrance to those who could seek out and correct security flaws than to those who would exploit them. To put it simply, concealing source code leads to a false sense of security. Opening source code to the public, though it may create short-term apprehensions, will result in more secure software in the long run.
These actions can be addressed in terms of the ongoing effort to eliminate the Year 2000 Problem from government systems. Using nearly any recently developed OSS software assures Y2K readiness, and opening the source code of internally developed software allows for easier modification of that software to solve the problem. These issues should be brought to the attention of the Council on Year 2000 Conversion recently established by President Clinton.
Finally, as discussed above, open source and open standards go hand in hand. Simple, open communications protocols and standards of compatibility facilitate OSS development, as they form a fundamental building block of any OSS project. "OSS projects have been able to gain a foothold in many server applications because of the wide utility of highly commoditized, simple protocols," writes Microsoft's Valloppillil. Significantly, the strategy proposed in that document for competing against OSS is "extending these protocols and developing new protocols," which implies replacing open standards with proprietary ones. It is precisely this sort of predatory practice which the government should oppose, both on antitrust grounds and specifically to prevent Microsoft from using its control of protocols to interfere with OSS development. The government should more vigorously lend its support to the open standards developed by industry, such as the Internet Engineering Task Force's standard set.
Next: Conclusions and End Notes