The Internet Protocol Journal - Volume 5, Number 4

Book Reviews

Ruling the Root

Ruling the Root: Internet Governance and the Taming of Cyberspace , by Milton L. Mueller, ISBN 0-262-13412-8, The MIT Press, 2002, http://mitpress.mit.edu

"WASHINGTON, Apr. 1 /Governance Newswire/ — The organizations that create street names, assign addresses, and assign telephone numbers have issued a joint announcement: Henceforth any conversation not conducted in Bahasa Malayu will result in termination of the relevant address or telephone number assignment."
The above bit of fiction is not pure silliness. Fear of equivalent, Internet-related excesses is the essence of Milton Mueller's book, Ruling the Root. The Syracuse University professor believes that administration of Internet addresses and domain names provides a fulcrum for overall Internet governance. He says they create a "political economy" vulnerable to serious abuse. Domain name administration is equated with control over Internet content, because, "a domain name record [is] very much like an Internet driver's license" as if it provides permission to use the Net, and even authorizes the locations one may visit.

Organization
The book covers both IP address and domain name administration. The material on IP addresses is thin, perhaps because it is a well-managed area without significant controversy. This is in marked contrast to the recent history of debate on Domain Name System (DNS) oversight. So it might have been instructive to see a comparison between the two administrative models, beyond simply noting that domain names can be interesting.

Discussion covers Internet technology, the history and politics of DNS and IP administrative management structure, and the intellectual property aspects of name assignment conflicts. Mueller suggests a three-layer hierarchy: technical, economic, and policy. What is missing from this "architecture" and from the entire book is any concern for the pragmatic details of administration and operation of these global, mission-critical services. Yet such tasks are difficult to perform well, as Network Solutions repeatedly demonstrated over the years, by losing registrations and corrupting critical data files; and the effects of problems are large.

When Star Trek's Captain Picard commands, "make it so," we know that he fully appreciates the challenges in implementing his directive. However, for Ruling the Root, policy development is not concerned with the operational complexities.

Not surprisingly, the book often demonstrates a misunderstanding of constraints inherent in DNS technology, although the tutorial on basic Internet technology is adequate, in spite of making the common error about the "T" in TCP/IP. [1]

Differing Opinions
Other reviewers of the book have called it well written, insightful, and nuanced. Indeed the discussion of history that is fully documented and involves simple, clear, objective facts is quite good. The rest of the time Mueller presents biased and unfounded descriptions of Internet governance, motives, and decisions, while failing to distinguish between what is fact and what is his opinion.

Ruling the Root sees adversaries, conspiracies, and threats, and permits no balancing sense of diverse collaboration, constructive criticism, or productive compromise. The technical community is somewhat less suspect, but is deprecated with the usual cliche about its naivete. So Mueller misses the essential point that techies designed, built, operated, and grew this robust, survivable, equitable system for global operations and service governance.

Professor Mueller's treatment of the dominant DNS registry, Network Solutions (NSI), now VeriSign, is curiously superficial and soft. NSI benefited spectacularly from the National Science Foundation's decision to permit charging for domain names, and from the policies and delays in the formation of the Internet Corporation for Assigned Names and Numbers (ICANN), as well as ICANN's distraction away from its intended registry oversight function and toward abstract debates about Internet governance. Yet the book does not consider NSI's role in ICANN-related political processes.

Mueller fails to understand the history of the organization that managed the DNS from its inception, the Internet Assigned Numbers Authority (IANA) and Jon Postel's role in running it. IANA is incorrectly represented as a simple operations arm of the U.S. Government. The grass-roots basis for its real legitimacy is missed. Its policy role is missed. Its collaborative processes are denied. For example, Mueller tells us that the description of IANA in RFC 1083, published in 1988 meant, "a new world was being defined by the RFC." In reality it was simply documenting established practice, as is typical for operations RFCs.

Validation
Mueller's substantiation of his analyses is also problematic. The book must be read with careful attention to the actual authority of each source. Goals and agendas are often misstated. For example, he characterizes the pre-ICANN International Forum for the White Paper (IFWP) as "the real arena for arriving at a decision [about the details of the new organization]." Its actual goal was simply to be a forum for discussion. Discussion, not decision-making. [2]

The book claims that the pre-ICANN International Ad Hoc Committee (IAHC) was formed "to develop and implement a blueprint for a global governance structure for the domain name system." In fact, the IAHC was formed for "specifying and implementing policies and procedures relating to iTLDs (international top-level domains, now called 'generic' TLDs, or gTLDs)." [3] He claims, "They had asserted that the root was theirs to dispose of." To the contrary, the IAHC was explicitly subordinate to IANA, and had nothing at all to do with management of the DNS root or any non-gTLD part of the DNS. Interestingly, the endnote Mueller offers as substantiation disproves his characterization.

Ruling the Root is loaded with endnotes—27 pages of small print. However, even the formal citations are problematic. Note #55 cites a newspaper article as a primary source, as if it were definitive proof the person discussed in the article held a specific opinion. Mueller's Note #45 claims to substantiate that, "Postel himself... admitted...it is unclear who actually controls the name space." Yet the note is for Internet Architecture Board (IAB) minutes. Attributing it to Postel was a fabrication.

Back-room, deal-making, conspiracy explanations are offered without substantiation. Of changes to Internet Engineering Task Force (IETF) management, Mueller states: "The most important reason the IETF didn't institute voting was that Jon Postel and several other senior figures vowed that they would refuse to run for office." Postel never made such a vow, and the process to effect these IETF changes did not experience any such attempts at influence. Of Postel's instructing some root servers to retrieve copies of the DNS root from a non-NSI master, Mueller claims that Postel was "apparently concerned about the direction U.S. policy was taking."

No substantiation is offered, because the claim is false. Postel and others were concerned about NSI's reaction to its own loss of control. The switch was intended to see what it would take to move NSI out of the hierarchy. These are not small matters of nuance. They show a pattern of misrepresentation.

The Author
Professor Mueller's credibility would have been aided by disclosing his own affiliations. The only ICANN constituency (the Non Commercial Domain Name Holders Constituency) claiming to represent the noncommercial world focuses on the civil society concerns that dominate the public debate about ICANN. Professor Mueller's discussion of the group is quite thin and does not disclose the fact that he held a dominant management position in it. In his criticism of dispute-resolution activities, he neglects to mention that he is a paid arbitration panelist.

An important book should be read because it has factual detail and thoughtful insight. Ruling the Root is, instead, important because it so thoroughly embodies the difficulties that have emerged in discussing Internet policy. Because so many people take Ruling the Root seriously, it should be read. However, the serious problems of the book encourage borrowing it, rather than buying a copy. Based on the pattern noted in this review, a thorough audit of those problems would be appropriate for the relevant Syracuse University academic ethics committee.

Dave Crocker [4], Brandenburg InternetWorking
dcrocker@brandenburg.com

References
[1] The "T" stands for transmission, not transport or transfer.

[2] http://web.archive.org/web/19981206105122/
http://www.ifwp.org/

[3] http://www.iahc.org/iahc-charter.html

[4] Factual claims in the review that do not have citations are based on the reviewer's direct experience. Dave Crocker wrote the first Internet standard for domain name syntax (RFC 822). He also was the IETF area director for initial work on DNS security. More recently he was one of Jon Postel's appointees to the IAHC. He naively thought that its work should be conducted in the manner that had been typical for Internet administration. So the last few years of charged, global politicization have been an education. He must also note that he was once Jon Postel's officemate.



High-Speed Networks and Internets
High-Speed Networks and Internets: Performance and Quality of Service , 2nd ed., by William Stallings, ISBN 0-13-032221-0, Prentice Hall, 2002. http://www.prenhall.com/stallings

This thoroughly updated classic covers topics of traffic engineering, queuing, and traffic modeling. The book gives a complete look around the protocols of the next generation: Resource Reservation Protocol (RSVP), Multiprotocol Label Switching (MPLS), and Real-Time Transport Protocol (RTP). It gives the keys to understand the way Frame Relay, TCP, and ATM react to congestion and flow control. The book also deals with new trends and standards that will lead the telecommunications industry in the following years. A very useful book, from the same author of traditional titles such as: Data Communications, Cryptography, Computer Architecture, and many more.

Organization
High-Speed Networks is divided into seven parts. The first one discusses the basic background needed to understand the rest of the book. Following the introduction, the second chapter goes on with the classical: the Open System Interconnection (OSI) model and the TCP/IP suite.

Part II explains packet-switching technologies in detail. The forth chapter explains the architecture of Frame Relay, and the next one focuses on ATM, including its operation and the adaptation layers. Chapter 6 works on high speed LANs, covering Fast Ethernet and Gigabit Ethernet, with the different media supported by each.

The third part is one of the most important; chapter 7 presents an overview of probability and stochastic processes. Although it is a brief one, it is useful to make revision of some concepts. The next chapter works on queuing analysis, introducing the basic elements of a queuing model. It explains the topics with plenty of examples: M/M/1, multiserver queues, and networks of queues, presenting all the formulas. Chapter 9 is dedicated to self-similar traffic. As recent studies indicate, traffic on high speed networks does not have the characteristics needed for the queuing theory. It introduces and explains the concept of self-similarity. Then the author applies this concept to data traffic analysis and examines performance implications. Based on papers on this subject, Stallings explains this new approach to traffic modeling not analyzed before.

The forth part focuses on another main topic: congestion and traffic management. Chapter 10 explains the effects of congestion and the different ways to control and avoid it. In the following chapter the author discusses control mechanisms at the link level. He examines different ways used by protocols to handle flow control: Stop and Wait, Sliding Window, and Go back N-ARQ. An analysis of the performance gained by using Automatic Repeat Request (ARQ) techniques follows.

These chapters give a detailed description of the different ways that communications can be handled. Chapter 12 focuses on transport-level traffic management. It explains TCP flow control in detail, including the retransmission strategy. The way TCP avoids congestion is discussed thoroughly. The next chapter continues with congestion control in ATM networks. The framework for traffic control is explained in detail, with sections dedicated to Available-Bit-Rate (ABR) and Guaranteed-Frame-Rate (GFR) traffic management.

The next part of the book is about Internet routing. Chapter 14 presents the algorithms used to compute the minimum path, and introduces some elementary concepts in graph theory. Later the author concentrates on Interior routing protocols, analyzing the Routing Information Protocol (RIP) and Open Shortest Path First (OSPF), the most important ones. Next the book discusses exterior routing protocols and multicast. The author describes in a simple way these addressing schemes and the related protocols.

The following section is dedicated to Quality of Service (QoS) in IP networks. The first chapter discusses integrated services, with coverage of queuing disciplines such as Weighted Fair Queuing (WFQ). A review of the Differentiated Services architecture follows.

After discussing the concepts, the author examines the protocols that support QoS: RSVP, MPLS, and RTP. He explains the philosophy behind each protocol, its characteristics, and its implementation.

In the final part of the book, the author changes the subject to compression. In Chapter 19 he presents an overview of information theory, discussing typical areas such as entropy. The next chapter continues with loss-less compression, facsimile compression, and others. It discusses the Lempel-Ziv algorithm used in PKZIP. The final chapter reviews lossy compression, explaining the discrete cosine transform, a key component of the Joint Photographics Expert Group (JPEG) and Motion Picture Experts Group (MPEG) standards.

Two very interesting appendices end the book: one for Internet standards and the standardization process and the other one dedicated to sockets, containing source code. Although the book is not dedicated to programming, the inclusion of TCP sockets can be useful to understand its implementation.

A book worth reading
We are facing an essential book for networking professionals, designers, and engineers. It covers unusual topics such as self-similar traffic and data compression. It is the basement for the design of any high speed network. As Internet traffic continues to grow, the optimization of network resources becomes a critical topic. Also, more and more voice traffic is carried over packet networks, congestion being one of its worst enemies. The time-sensitive traffic needs attention, and this book provides the tools to manage it.

In addition to its solid coverage of topics, the book has plenty of bibliography and many links to the principal sites for each chapter. With no doubt this is a very useful book, from the well-known technical author William Stallings.

Rodrigo J. Plaza, Iplan Networks, Argentina
rplaza@iplan.com.ar



Would You Like to Review a Book for IPJ?
We receive numerous books on computer networking from all the major publishers. If you've got a specific book you are interested in reviewing, please contact us and we will make sure a copy is mailed to you. The book is yours to keep if you send us a review. We accept reviews of new titles, as well as some of the "networking classics." Contact us at ipj@cisco.com for more information.