Dr. Dave Godfrey, University of Victoria, Department of Writing
Dr. Michael Levy, University of Victoria, Department of Computer Science
Equity on the Internet will not arrive without strong involvement by citizens working with their governments. Technologically, the Internet is allied to the dominant trend of the last few decades: globalization. Sociologically, it is allied to the counter-trends of local control and local content.
British Columbia offers a useful example in that it provides a constituency which is large enough to have some economic clout and yet small enough so that its members can expect to know one another and to act with some degree of collectivity. This paper examines some general principles that define goals of equity on the Internet and provides some specific recommendations for this particular constituency. It was originally presented to government officials in the fall of 1994 and revised for this conference. Some of the recommendations have been accepted and some have not. However, we feel that the general principles remain very valid and that the overall goals and targets provide a model for any constituency, large or small, which seeks to obtain equitable access to Internet-type services for its citizens.
The current government in B.C. has made determined efforts to ensure that the citizens of B.C. have full and equitable access to the new telematic technologies: that is, combinations of computer and telecommunications functions. These technologies can be divided roughly into two kinds: various proprietary systems and the protocols used by the Internet. The Internet is a large, non-centralized co-operative whereby the owners of computers, whether servers or clients, share the costs of the tele-communications which link those computers. The Internet has been dramatically successful in recent years, and we believe that this Government can fruitfully apply some of the techniques that contributed to this success.
For governments, there are five special challenges when dealing with telematics. The chief of these is the rapid rate of change. When negotiating for railroad access to Canada, a ten-year de-lay was not a catastrophe; today, a government which dealt with the telematics of a decade ago would be a laughing stock. Al-though in fact the general outline of changes is relatively predictable from decade to decade, the precise nature of the implementations is not predictable and regulations can cause more harm than good.
The second challenge lies with the combination of technologies that drive the Internet: computers and their applications, which are almost wholly unregulated versus telecommunications which is extensively regulated (but at the Federal level). As a result, it is easy for firms that start out on one side to slip over a boundary, as happens when BCSC buys telephone switches to link its computers or when BC Tel’s Advanced Communications buy large servers to host content for its Ubiquity network.
The third challenge is finding a mechanism that allows participation at all levels, including: access to material, providing content and even development of new related technologies (such as novel Client/Server protocols that address specific needs).
The fourth conceptual challenge is with the levels of the Internet. What does it mean to "be on the Internet." Technically, someone with a 2400 baud modem in Fernie dialing out to Compuserve can then "get on the net" but this is a far cry in terms of capacity from a firm in Victoria with a 10Mbps connection. Conceptually, the comparison is of a single lane gravel road with a 5,000 lane highway.
A fifth, more easily resolved challenge, is the level of involvement of government in providing data application services to itself and to the other sectors. BCSC is a large institution with a well-developed base of staff and services. Some say it should be privatized; some say it should expand into and compete with the private sector.
Despite these challenges, the province is well placed to provide equitable access for its citizens. First of all, as clearly evidenced by the Victoria Freenet and by the rapid growth of Internet Providers, the citizens are well aware of their own needs and are willing to work hard and with innovation to meet those needs. Secondly, the province has a strong set of institutions and firms which are technically competent and have provided many of the building blocks for an exemplary telematic environment.
This short paper looks at a set of solutions for these challenges based on a set of fundamental principles and leading to a specific set of recommendations that would provide a coherent solution. If not accepted, this set of recommendations can at least provide a benchmark for completeness and benefits to the people against which other solutions can be evaluated.
1. Modern societies must face up to an old challenge in new garb: the avoidance of class division via knowledge access and transfer.
As the invention of the printing press led eventually to the development of the public library, so must the inventions of the telematics area lead to the development of new institutions, such as freenets, schoolnets and culture nets. Governments must monitor the gap between what the rich and powerful can "learn" and what the poorest and weakest have by comparison and move rapidly and effectively to reduce that gap to a mini-mum. Governments at all levels must accept that one criterion of their success or failure will be the size of that gap.
2. The X-factor is the major threat to rapid deployment of telematic power.
Neither large corporations nor large governments seem capable of dealing effectively with these rapid changes in the field of telematics. Small business and local institutions tend to move more rapidly to the required "culture of adaptability" which the nature of telematics makes mandatory.
It is here that Governments can benefit from an examination of the mechanisms that were successful in the Internet. Even though the internet started its life as a military research project, it was designed using principles of distribution and devolution. Specifically, the following were a crucial part of the organizational design of the internet:
A. All protocols were non-proprietary. This committment went beyond the more common and more bureaucratic approach to "Open Systems" where we see consortia of powerful special interests (such as dominant technology corporations) form essentially by-invitation "open" committees. Instead, the Internet protocols were all published on-line, accessible to anyone in the world who had access to a University of Research Institute computer account. (For an interesting rant against OSI, an "Open Systems" competitor to the Internet protocol suite, see the paper "The Future of OSI: A Modest Prediction" by Marshall T. Rose, in Proc IFIP International Conference on Upper Layer Protocol, which was held in Vancouver in May 1992.)
B. The goals of the project were aimed squarely at infrastructure rather than content. Thus the earliest funded projects worked on gateways (now called routers) and network management protocols.
C. The sample protocols developed by the Internet pioneers can be characterized as simple and accessible. Tools to use these protocols were freely disseminated.
D. No attempt was made by the Internet committees to select, prohibit or promote specific protocols. Protocols lived or died strictly on the basis of their merits. Furthermore, this merit was determined by a complex "marketplace" of use, rather than by some sanctioned committee of experts.
Perhaps the most notable example of a successful Protocol development is the HTTP protocol. This protocol was invented not by a member of the Internet Society, but by an Internet user (Tim Berners-Lee) working for CERN in Switzerland. The protocol was influenced by "gopher" - an earlier university-developed protocol. The true utility of HTTP became apparent with the development of Mosaic (and then Netscape), a client program that combined HTTP with the well established notions of SGML and Hypertext. Mosiac was developed initially by a group with the US government funded Nationall Center for Supercomputer Applications. Although HTTP was developed independently of the Internet Society, it has all the usual hall-marks of the Inter-net protocols: it is simple (ASCII based, readable); server source code was made available; and it is easily integrated with other software. Its adoption rate has been a true phenomenon. It is very likely that the NCSA spin-off company (Netscape Communications Corporation) will be extremely profitable [this statemate made in November, 1994].
The key point we wish to make about this example is that it came about because of a policy of devolved control and infrastructure support rather than because the central organization had identified a problem and solicited a solution.
On the other hand, the telematic infrastructure is large and growing and some aspects are large-scale and will continue to require regulation.
Where there is to be regulation, it should be at a level closest to the people wherever possible. Portland, by regulating cable, does a far better job of providing services to its citizens than the CRTC does for the citizens of Victoria.
Wherever a choice is possible, governments and citizens should favour the small and the local over the large and the cumber-some. In those limited areas where there is a true benefit of scale, such as in the provisioning of backbone services, this should not be allowed to serve as an excuse for monopoly or oligopoly in related areas where there is no benefit of scale.
3. The market is not "sufficient unto itself."
Although there are some who claim that there is no role for government in this at all, history tells us otherwise. Left to itself, the market distributes wealth and knowledge in an in-equitable fashion, which tends to lead to major class divisions, which tends to lead to revolution and violent change. To maximize the benefits of telematics for all of its citizens, those citizens must be informed and governments at various levels, especially the local, must act rapidly and effectively to enhance local production of knowledge and avoid excessive dependence on foreign content, whether informational or entertainment.
Competition is an effective force, when there is true competition. In the early eighties, software development flourished because there were few large players and thousands of small ones. That does not mean, however, that history will reverse itself and Bill Gates will forever avoid the temptation to take undue advantage of his dominant position in certain areas of software applications.
4. We can afford multiple backbones.
In earlier decades, those who wrote on this subject often spoke of the necessity for clear division among the highway and the truckers, that is the carrier versus content providers. Technically, this was based on the very high capital costs of building the backbone; it was assumed that there could only be a single backbone for the country and that therefore the provider of the backbone would have an unnatural monopoly if allowed to also be a content provider.
Provided that all carriers must charge posted and equitable rates, however, it is now clear that a province such as BC could afford to support a number of backbones at least in terms of connecting all of its major urban centres. If monopoly is thus clearly avoided, then it might be possible to maintain the required levels of content production in BC, together with low-cost, equitable access to the highway, and still have firms such as BC Tel and Rogers compete with one another both at the carrier and the content provision levels,
5. The province should maintain a window of expertise in the field.
No matter what experts it hires, the CRTC will always lack the level of expertise which comes from being an active player in the field. Since the backbone level is crucial to equity, the government must maintain an effective Crown Corporation at this level, one capable of competing in an adaptable and effective fashion with the other backbone provisioners.
Taking together these challenges and these principles, one can arrive at least one scenario for the second half of the 1990’s in BC. Taken together, this set of recommendations will meet these challenges in an effective and adaptable fashion.
1.1 The provincial government should seek full control over the regulation of communications within the province and share that power with local communities.
Once this control is gained, it should licence backbone providers to operate in the province and licence fees should be used to support content production and community access.
1.2. By 1998 at the latest, BCTel should be allowed to deliver all forms of content to the home and cable companies and wire-less companies should also be allowed to deliver telephony services. In both cases, however, firms should adhere to current CRTC regulations in regard to Canadian ownership and should also have at least 30% of their ownership within the province.
1.3 A single Crown corporation (such as Westel or BCSC) should also have 30% of its shares sold to BC shareholders (widely distributed) and should be charged with the role of providing a crown-owned backbone. Other crowns, such as BC Hydro, should be restricted from this field and should be able to purchase communications services from the designated or any other licenced backbone provider. Existing government telecommunications assets should be sold or transferred to the designated Crown.
The designated Crown would thus act as a window of expertise and a counterweight to the "market forces" of competition which would otherwise tend to provide excellent service in the urban centres and very limited service in the hinterlands.
2.1 Backbone providers should not be allowed to sell or manage Internet provisioning in competition with small firms.
2.2 BCSC should be limited to the provisioning of data application services to direct government ministries. All of these contracts should be made competitive.
2.3 BCSC should not be allowed to both evaluate and bid on RFP’s. A new, small and independent body should be established to oversee these bids. This body should have no ties to BCSC.
2.4 A new, small body should be established to provide general guidelines to ministries in regard to applications, telecommunications, hardware etc. Ministries should be free to issue RFP’s for all goods and services and to decide whether to purchase from BCSC or other bidders.
2.5 Crowns, universities, colleges, school boards, etc. should purchase services from the private sector.
The key mechanism of these recommendations is one which reduces knowledge differentials. As part of its backbone licencing process, a fee should be levied based on quantity of data transferred. Whatever the source, or direction, this fee would be constant and could be a percentage of the rate-fee currently charged by BCTel, Westel and other backbone providers,
However, these fees should not go into general revenue. Half of them should be allocated to each of the following areas.
3.1 Content Production Support
The SRED model should be followed so that BC-based firms who produce content for the telematics field should be allocated a provincial tax-credit which supports their content production.
3.2 Support for Community Access
The second half of these licencing fees should be used to sup-port community-owned organizations, such as freenets, which have strong local ownership and which facilitate local access to the Internet. Freenets, in turn, should not compete with private Internet providers but should concentrate on the development of local content and the provision of terminals to those who cannot afford computers in their homes.
The advantages of this model are that it can be easily managed and that the greater the flow of entertainment and information over the backbone the more funding available for reducing the knowledge deficiency among regions and individuals.
Rather than a complex Arts Council, or Telefilm, or a Government Agency such as the NFB, the goal should be to have lean and non-discriminatory mechanisms which facilitate these two goals. No more than 7% of the revenue stream should be allocated to ad-ministration and the revenue stream should pass directly to these two fields rather than being trapped within general revenue.
The universities should be charged with defining and monitoring the nature of the knowledge domain and the gaps and deficiencies which exist within BC.
No new funds or officials need to be allocated to this activity. The Universities are already well-funded and simply need to redirect some of their research/analysis activities to this domain from other more esoteric domains.
Taken together, these principles and recommendations provide a framework for reducing inequity in a specific constituency, one which has considerable leverage in terms of purchasing and legislative powers. Other constituencies need to start from the principles and define sets activities and regulations which would help them obtain their goals. Monitoring bodies should be kept as independent as possible and should co-operate with one another, through forums such as this conference, to share expertise on methods, policies and evaluation techniques which will enable the Internet to continue to flourish as means avoiding dichotomies of the knowledge-rich versus the knowledge-poor.
For more information, please contact:
Dr. Dave Godfrey at dave@pinc.com
Dr. Michael Levy at merlin@pinc.com