Content

Margit Vanberg, The TCP/IP protocol in:

Margit Vanberg

Competition and Cooperation Among Internet Service Providers, page 46 - 47

A Network Economic Analysis

1. Edition 2009, ISBN print: 978-3-8329-4163-5, ISBN online: 978-3-8452-1290-6 https://doi.org/10.5771/9783845212906

Series: Freiburger Studien zur Netzökonomie, vol. 14

Bibliographic information
46 or node in the network can make communication between many sites impossible. Research at NPL was motivated by a search for a communications protocol that would allow remote machines and other terminal devices (such as printers) to communicate with a time-sharing computer via the public telephone system (Naughton, 2000: 123). Circuit-switching is uneconomical for computer communication, because the traffic generated by computers is “bursty,” by which is meant that the data stream tends to be very irregular, with a lot of data transmitted in a short time interval, followed by a long pause before more data is transmitted. A dedicated line would be wasted for much of the communication. The new communications technologies developed independently at RAND and at NPL were both based on the concept of packet-switching. With packet-switching, the data stream to be transmitted is decomposed into smaller segments, called “pakkets” or “datagrams,” which can be sent on a variety of different possible routes between A and B, before being reassembled in the correct order at their destination. Packet-switching uses capacity only when data is transmitted. Therefore, packetswitching allows a more cost-efficient use of the underlying resources. The network can handle more than one communication per link at a time by placing packets in the pauses of packets originating from unrelated data transmission. A packet-switched network therefore requires fewer communication lines. A further advantage of pakket-switching is that since every packet can potentially travel a different route, a defect or lost cable segment or node does not disrupt the transmission. Packetswitching therefore also guarantees a better performance. 3.4.2 The TCP/IP protocol ARPA reverted to this research on packet-switching for building the initial ARPA- NET and the packet-switched terrestrial radio network and satellite network operated by ARPA (Naughton, 2000: 132ff.).33 In the 1970s, Robert Kahn at ARPA pursued the idea of establishing an open network architecture, which would enable communications between the diverse ARPA networks that were using packetswitching technology. His vision was to make possible the inter-working of these independent networks by a “Meta-level Internetworking Architecture” that would allow the connected but otherwise independent networks to maintain differing internal communication designs (Leiner et al., 2003: 4). In 1973, Kahn, together with Vinton Cerf, then a computer scientist at Stanford University, began work on a standard for communications across networks using packet-switching technologies (Kahn, 1995: 3). Their research resulted in the Transmission Control Proto- 33 The specific protocols for the applications run on the ARPANET were essentially written by graduate research students of the computer science departments hosting the initial ARPANET nodes. These graduate students started the Network Working Group and communicated their ideas by writing “Requests for Comments” (RFCs). The RFCs remain to this day the way standards related to the Internet are discussed and disseminated. 47 col/Internet Protocol (TCP/IP), which became very influential for the development of the Internet because of its property of functioning in very different kinds of networks and guaranteeing the compatibility of these diverse networks. In 1978 the first functioning version of the TCP/IP protocol, TCP/IPv4, was released. The principle idea of the TCP/IP protocol suite is to connect independent packet-switched networks by so-called gateways or routers. The routers are computers specialized in the forwarding of data packages. They understand the protocols by which the hosts in their home network communicate and follow a common standard for the communication with routers of neighboring networks. TCP/IP is a whole protocol family. The two main protocols of this protocol suite are the Transmission Control Protocol (TCP) and the Internet Protocol (IP). Among other things, the TCP protocol is responsible for opening and closing a connection, for providing flow control, i.e. how fast to send data, and for checking that transmission was successful (Comer, 2006: 187ff.). The IP software is responsible for forwarding data packages in the Internet (Comer, 2006: 73ff.). It divides the data stream into packets and provides each package with information such as the source, the destination, an indication which packages are part of one message, whether there are more packages to come, and the order of the packages belonging to one data stream. With this information the packages are recompiled in the order of the original message once they reach their destination. TCP/IP became the official standard of the ARPANET in 1983; and the NSFNET later also adopted TCP/IP as its standard communications protocol. TCP/IP has since remained the standard of communication in the Internet; in fact, the Internet is often set equal to computer-networking via TCP/IP.34 The following sections will explain in more detail the addressing system and the method of data transmission in a TCP/IP network. 3.4.3 IP addresses, autonomous system numbers and the domain name system Communication in networks presupposes that each user can be uniquely identified. In the Internet, this function is fulfilled by the so-called IP address, which encodes information on both the home network (in the network prefix) of a host as well as the host itself (Comer, 2006: 42). The Internet Corporation for Assigned Names and Numbers (ICANN) grants IP addresses to regional Internet registries, which in turn assign blocks of IP addresses to large ISPs.35 These ISPs finally assign IP-addresses to large customers (other ISPs or corporations). For private customers ISPs often use dynamic address allocation, assigning a public IP address only for the duration of an online session (see Comer, 2006: 355). This practice preserves IP addresses when an ISP has been assigned only a limited address block. A small ISP with many dial-in- 34 See, for instance, the definition of the term “Internet” by the Federal Networking Council, cited in chapter 2 above. 35 See www.iana.org for details; site last visited on Feb. 15, 2008.

Chapter Preview

References

Zusammenfassung

Die Konvergenz der Netztechnologien, die dem Internet, der Telekommunikation und dem Kabelfernsehen zu Grunde liegen, wird die Regulierung dieser Märkte grundlegend verändern. In den sogenannten Next Generation Networks werden auch Sprache und Fernsehinhalte über die IP-Technologie des Internets transportiert. Mit den Methoden der angewandten Mikroökonomie untersucht die vorliegende Arbeit, ob eine ex-ante sektorspezifische Regulierung auf den Märkten für Internetdienste wettbewerbsökonomisch begründet ist. Im Mittelpunkt der Analyse stehen die Größen- und Verbundvorteile, die beim Aufbau von Netzinfrastrukturen entstehen, sowie die Netzexternalitäten, die im Internet eine bedeutende Rolle spielen. Die Autorin kommt zu dem Ergebnis, dass in den Kernmärkten der Internet Service Provider keine monopolistischen Engpassbereiche vorliegen, welche eine sektor-spezifische Regulierung notwendig machen würden. Der funktionsfähige Wettbewerb zwischen den ISP setzt jedoch regulierten, diskriminierungsfreien Zugang zu den verbleibenden monopolistischen Engpassbereichen im vorgelagerten Markt für lokale Netzinfrastruktur voraus. Die Untersuchung zeigt den notwendigen Regulierungsumfang in der Internet-Peripherie auf und vergleicht diesen mit der aktuellen Regulierungspraxis auf den Telekommunikationsmärkten in den Vereinigten Staaten und in Europa. Sie richtet sich sowohl an die Praxis (Netzbetreiber, Regulierer und Kartellämter) als auch an die Wissenschaft.