Open source software has been a great success in recent years. Businesses rely on it; the Internet depends on it. Can the communications infrastructure, historically the preserve of blue-chip and monopoly telecommunications companies, follow in the footsteps of the Internet and open source? I will compare the constraints and opportunities available for a wireless commons from technical, regulatory, and open organizational perspectives.
For much of the past century, the communications infrastructure has become increasingly centralized. The limitations of spectrum, the cost of building phone lines, cable systems, and radio transmission facilities, and the the economic benefit of interoperability have conspired to favor a few large companies to build and operate the infrastructure.
The end-to-end architecture of the Internet has demonstrated that standards and computer technology can reduce infrastructure costs and make interoperability compelling in an environment with many suppliers. Technological advances – such as spread spectrum, agile radios, and mesh networks, all designed to operate in a less hierarchical environment – may provide solutions to limitations of spectrum.
There is a great deal of skepticism about the new technologies. A recent article in The Economist (19 March 2005), for example, argued against public involvement and the readiness of the technology:
. . . should governments be getting into the telecoms business at all? . . . Critics complain that public networks distort the market by competing with private firms. . . . Kevin Warton, a communications-policy expert at the University of Pennsylvania . . . [says that Wi-Fi] is not a reliable substitute for broadband services such as cable or DSL. . . . the limitations of the technology seems to have been overlooked.
There are indeed technical challenges with using Wi-Fi for community wireless. The effective range of most Wi-Fi transmitters is less than 100 meters, and the signal is easily obstructed. Nevertheless, the City of Cork in Ireland recently launched a Wi-Fi mesh network covering a 1.5 square kilometer network in the city's core (ElectricNews.net, 2005). Such projects are relatively recent and rare: the BC Wireless Project in British Columbia, for example, does not list any hot spots on their home page (BC Wireless, 2005). Private industry, meanwhile, does not appear prepared to filling the gap.
A focus on Wi-Fi technology, however, ignores the larger issue. The range and reliability of Wi-Fi is limited by the high frequency and low power specified by government regulation for the license-exempt 2.4GHz and 5GHz bands in which it operates. Furthermore, the aforementioned new technologies could reduce or eliminate the problem of spectrum scarcity with a technology like Wi-Fi. The question is not technical possibility, but – as The Economist hints – political will.
The incumbents in the telecommunications industry are generally averse to new technologies. As Lessig notes (2002, p. 6), Machiavelli (1972) explained this tendency centuries ago:
The innovator makes enemies of all those who prospered under the old order, and only lukewarm support is forthcoming from those who would prosper under the new. Their support is lukewarm partly from fear of their adversaries, who have the existing laws on their side, and partly because men are generally incredulous, never really trusting new things unless they have tested them by experience. (p.51, emphasis added)
Law, then, can become the critical battleground for innovation.
For example, a number of cities in the United States are investigating municipal broadband, in most cases using Wi-Fi technology. Philadelphia's effort is the largest, but there are also initiatives in Chicago, several cities in Florida, and elsewhere. (Lipman, 2005; Scott, 2005). In response, there are efforts (e.g. Cox, Giovanetti, McClure, Titch, Rizzuto, and Tuerck, 2005) – largely by major telecom companies (as shown by Ellison, 2005; Fleishman, 2005) – to ban such endeavors, arguing that they are not in the public interest and should not compete with private business.
The telecommunications industry has succeeded in limiting innovation in the past. For example, FM radio was crippled by U.S. Federal Communications Commission (FCC) regulations imposed at the bequest of RCA and others who did not want competition with their AM infrastructure (Lessig, 2004, pp. 3-7, 196). Earlier FCC intervention in 1927 led to an American radio market in which two companies supplied 97% of night-time broadcasts (Lessig, 2002, p. 74)1.
There is some regulatory movement towards allocating spectrum for wireless broadband: on 10 March the FCC allocated 50MHz, in the 3650MHz band for to provide broadband services (FCC, 2005). However, it is important to note that this is licensed bandwidth, unsuitable for license-free applications such as mesh networks.
The regulatory terrain remains uncertain. Is there an alternative route to community wireless under the current regulatory regime, even without the support of government?
In 1984, Richard Stallman was concerned that software source code at the MIT AI Lab, which had previously been open to all, was becoming proprietary (Stallman, 2002). Leveraging existing copyright law and the efforts of himself and others, he was able to kick-start the development of a software commons of free and open source software (FOSS)2 which since become essential to the Internet at large, and to many important companies in particular (e.g. IBM and Apple). Could something similar be done with wireless?
Community wireless and FOSS would appear to be similar. Both are built around a commons, one for software, the other for bandwidth. Both have arisen as distributed alternatives to hierarchical incumbents. Both software and bandwidth are intangible resources. Can the wireless commons repeat the success of open source, or is the comparison misplaced?
Weber (2004, p. 271) lists a number of characteristics of tasks likely to be amenable to an open source approach.
First, he requires that individual contributions can be based on non-proprietary knowledge. Both the knowledge and license-exempt spectrum available for community wireless fit this definition.
Second, the product must be valuable to a critical mass of users; the efforts of cities like Philadelphia certainly suggest that it is.
Third, the product should benefit from a strong network effect. This is a property of open source that Weber describes as anti-rival: its value increases the more that it is shared (2004, pp. 149-156). Software is by its nature non-rival: one person having more does not mean anyone else has less. Radio, on the other hand, is naturally rival. Spectrum can be “used up” and is vulnerably to abuse by powerful transmissions; only certain architectures, such as mesh networks, allow it to be non- or anti-rival. Also, the network effect in wireless is diminished by the cost of additional hardware required for growth – a limitation that does not apply to software, which can be duplicated at zero cost.
Fourth, there should be benefits to widespread peer review. This applies much less to a physical network, and so may not be a competitive advantage for an open process.
Finally, a small group should be able to take the lead and develop something useful on its own, and a voluntary community should be able to make iterative efforts to improve the product. It is certainly possible a community like Philadelphia could such a role. But if they do, the limited range of the available spectrum mean that the network effect is largely local.
Weber also argues for the importance of motivation. Open source and free software are the products of a scholarly and culture and liberal ideology extending back to the 1950s, predating commercial software. This ideology has played a central role for Stallman and many of the developers who followed. Compared to even the commercial software industry (which shares many of the values of the open source movement), the business of telecommunications is hierarchical and conservative; its incumbents dominate in a way that no software company – even Microsoft or IBM – ever has.
In its favor, wireless shares with software the potential for a commons. Just as with the Internet and FOSS, a distributed architecture is possible – especially using anti-rival mesh technology. The current license-exempt bands provides the necessary raw spectrum; existing open source software enables mesh networks right now (Cringley, 2004). However, current deployments are marginal. Standing against wider success is the might of an industry opposed and alerted by the history of the Internet, the rival and excludable nature of spectrum, and the lack of a culture of openness. Most difficult of all is the local nature of community networks: success in one community may not translate into gains elsewhere. Compared to the engineering required to build an operating system, the barrier to entry may be low, but the network benefits are also limited. Individual communities may find the model beneficial, but any shift towards community networks on a large scale is likely to be much slower than the rapid progress of open source. Regulatory support could make all the difference.
1Interoperability and economies of scale for equipment mean that Canada generally follows the U.S. lead.
2The differences between free and open source software are largely ideological; while significant, the distiction is beyond the scope of this paper.
BC Wireless Project. (2005). Retrieved March 27, 2005 from http://www.bcwireless.net/
Cox, B., T. Giovanetti, D. McClure, S. Titch, R. Rizzuto and D. Tuerck. (2005). Not in the public interest – The myth of municipal wi-fi networks. New Millenium Research Council. Retrieved from http://newmillenniumresearch.org/archive/wifireport2305.pdf
Cringely, R. (2004). The little engine that could: How Linux is inadvertently poised to remake the Telephone and Internet Markets. PBS. Retrieved March 20, 2005 from http://www.pbs.org/cringely/pulpit/pulpit20040527.html
The Economist. (2005). Wi-Fi pie in the sky. The Economist March 19, p. 68.
ElectricNews.net. (2005). Wi-Fi network covers Cork City. The Register March 25. Retrieved March 27, 2005 from http://www.theregister.co.uk/2005/03/25/cork_citiy_wifi_network/
Ellison, C. (2005). Municipal wi-fi: Let's keep it local. eWeek February 3. Retrieved March 17, 2005 from http://www.eweek.com/article2/0,1759,1759677,00.asp
Federal Communications Commission. (2005). FCC opens access to new spectrum for wireless broadband in the 3650mhz band. Retrieved March 17, 2005 from http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-257309A1.pdf
Fleishman, G. (2005). Sock puppets of industry. Wi-Fi Net News Februrary 1. Retrieved March 30, 2005 from http://wifinetnews.com/archives/004765.html
Lessig, L. (2004). Free culture: How big media uses technology and the low to lock down culture and control creativity. New York: Penguin Press.
––. (2002). The future of ideas: The fate of the commons in a connected world. New York: Vintage Books.
Lipman, J. (2005). A wireless city? Chicago may give it a try. Daily Southtown March 14. Retrieved March 27, 2005 from http://www.dailysouthtown.com/southtown/dsindex/14-ds5.htm
Machiavelli, N. (1972). The Prince. George Bull, trans. London: The Folio Society.
Scott, R. (2005). Bill targets public broadband. Talahassee Democrat February 19. Retrieved March 30, 2005 from http://www.tallahassee.com/mld/tallahassee/business/10937628.htm
Stallman, R. (2002). The GNU Project. Free software free society: Selected essays of Richard M. Stallman. Joshua Gay, ed. US: GNU Press.
Weber, S. (2004). The Success of Open Source. US: Harvard University Press.