Reviewed by:
Jan-Pascal van Best
Information and Communication Technology
Faculty of Technology, Policy and Management
Delft University of Technology
The Netherlands
Email: j.p.vanbest@tbm.tudelft.nl
Hubs and Spokes: A TeleGeography Internet Reader is a report by TeleGeography, Inc., an independent research and publishing group based in Washington, DC. The report has two main parts: the first part, Global Internet Primer, discusses the historical growth of the Internet and its related institutions, together with technological, economical and policy-related causes for the developments. The remainder of the report gives otherwise hard to come by information about the current state of affairs in the worldwide internet architecture.
The global Internet primer starts with a description of the development of the Internet architecture, from a domestic US network, via a US-centric network, with almost all international connections from or to the US, to a more diffuse model, with Western Europe, among others, as an important Internet connectivity hub, next to the US. The reasons for the original US-centric network are sought in mostly in economic and policy factors, like the relatively early liberalisation of the US telecom market and first-mover advantages of US carriers.
The primer continues with a discussion of the financial structure of the Internet. Originally, Sender-Keeps-All (SKA) was the prevailing form: since most relevant Internet service providers were either universities or government agencies, with comparable data flow patterns, no financial settlements were necessary. The report proposes a breakdown of the ISP industry into four classes: Backbone ISP's, Web hosting providers, Downstream ISP's and Online service providers. This breakdown shows the asymmetry in traffic interchange that occurs between ISP's and so helps explain the financial compensation strategies of ISP's.
Internet governance is the next section of the Internet primer. It discusses the history, from an ARPA work group to the current ICANN (Internet Corporation for Assigned Named and Numbers) and ISOC (Internet Society). The problems surrounding the shortage of IP addresses, and the slow transition to IPv6, take a large part of this section. It is a pity that the political aspects of Internet governance, like the ICANN mandate, influence of large companies on ICANN, and Internet democracy, are hardly discussed in the report.
In the section on demand, the current and expected growth of the Internet is discussed, with much attention to access networks. Several factors for Internet growth are discussed, like the growing number of Internet users, interactive applications, streaming audio and video, Internet appliances and Internet-enabled mobile phones. The end-user factors that cause this growth in demand are not discussed in detail.
The last section discusses IP telephony and Voice over IP (VoIP). This section seems a bit out of place, but may be partly explained by a publication grant for this report from Global Transmedia Communications Corporation. It is discussed that the total volume of telephony traffic is dwarfed by Internet traffic already. There are however reasons, especially for telecommunications operators, to consider IP-telephony integration, notably for added extra value in the form of, e.g., web-enabled call centres. The different Quality of Service-demands of the Internet and the telephony world, and the hefty signalling, metering and billing required for interconnection with the PSTN, still are high hurdles to come by.
The second part of the report, in the chapters International Internet Bandwidth and International Internet Indicators, gives a fairly comprehensive list of international ISP's, a description of Internet bandwidth available between regions, countries and cities around the world, and a list of Internet exchange points. This is useful information to use as a starting point for all kinds of statistical analyses. If this information were made available also during the coming years, this could give a good picture the evolution of the Internet connectivity. The report closes with a list of Internet host counts, by TLD (Top-Level Domain name) and by country.
All in all, the first part of the Hubs and Spokes report is an interesting read for people interested in how the Internet came to be, with a strong accent on economic factors. The second part may provide researchers with the hard data that is often hard to come by in the Internet world.