_Source_by_an_Accidental_Revolutionary_ (Beijing and Sebastopol, CA: O'Reilly, 1999),
151.

[4-54] Wayner, 289.

[4-55] "IBM Extends Software Leadership on Linux," IBM Press Release, December 8,
2000.

[4-56] Economists have only begun to examine the incentives that might affect open code
projects. Josh Lerner and Jean Tirole have summarized some of the benefits, relative to
closed code projects, as follows: (1) lower costs due to (a) familiarity of the code from,
e.g., university training, and (b) customization/bug-fixing advantages; and (2) higher
benefits, especially due to signaling from (a) better performance measurement (easier to
demonstrate skill), (b) "full initiative" (because no supervisory involvement), and (c)
greater labor market fluidity. See Josh Lerner and Jean Tirole, "The Simple Economics
of Open Source" (NBER Working Paper No. 7600, December 2000).

James Bessen has offered a far more ambitious model of the benefits from open code
projects. In a paper not yet published, Bessen argues that open code provision is superior
to closed source provision when the public good (i.e., software) is a complex public
good. The intuition is that the costs of debugging complex software projects become pro-
hibitively high within a closed code environment. As Bessen writes, "[I]f a product had
100 independent features and each combination took only one second to test, then
the job could not be finished before the sun is predicted to swallow the earth even if
every human currently alive spent every second until then testing." James Bessen,
"Open Source Software: Free Provision of Complex Public Goods" (working version,
April 2001), http://researchoninnovation.org. Bessen concludes that "strong property
rights... may actually limit provision of complex [public goods]," and hence that open
code projects have a competitive advantage over closed.

[4-57] This argument is related to a point made by Carl Shapiro and Hal R. Varian in _In-_
_formation_Rules:_A_Strategic_Guide_to_the_Network_Economy_ (Boston, Mass.: Harvard
Business School Press, 1999), 256-259. As they argue there, in a market heavily depen-
dent upon standards, there is often an incentive for competitors to contribute to public
standards rather than to make the standard proprietary. The battle over the standard
HTML is an example of this. For a while, both Netscape and Microsoft tried to develop
extensions to HTML that were peculiar to their own servers and Web development soft-
ware. Their aim was to split the common standard base to induce the market to follow
their own design. Later, however, they both apparently decided that this effort to divide
the standard would not be profitable; better if there were more on a common platform
than fewer on a proprietary platform. Varian and Shapiro demonstrate the conditions
under which that behavior could be rational. See also Varian, "Buying, Sharing, and
Renting Information Goods," _Journal_of_Industrial_Economics_ 48 (2000): 473. On Mi-
crosoft's attitude toward a neutral HTML, see Bank, 74 ("If Microsoft doesn't get to do
anything 'proprietary' with HTML in the browser," Gates wrote, then "we have to stop
viewing HTML as central to our strategy and get onto another strategy."). See also ibid.,
73 (regarding Office). See also Moody, 200. Moody believes that because the Apache
server was a dominant server for HTTP, and uncontrolled by any commercial venture, it
did not fracture in the same way. See ibid., 149.

[4-58] For articles identifying scenarios in which this proposition fails, see Douglas Licht-
man et al., "Shared Information Goods," _Journal_of_Law_&_Economics_ 42 (1999): 117;
Hal R. Varian, "Buying, Sharing, and Renting Information Goods," _Journal_of_Industrial_
_Economics_ 48 (2000): 473; Douglas Lichtman, "Property Rights in Emerging Platform
Technologies," _Journal_of_Legal_Studies_ 29 (2000): 615.

[[284]]

p283 _ -chap- _ toc-1 _ p284w _ toc-2 _ +chap+ _ p285

v?