@article{Zukerman2009Back,
author = {M. Zukerman},
title = {Back to the future},
journal = {IEEE Communications Magazine},
volume = {47},
number = {11},
pages = {36-38},
month = {Nov.},
year = {2009},
abstract = {In the 1970s - before dense wavelength-division
multiplexing (DWDM), Google, Facebook, iPhone, and Skype - we had a telephone
network based on circuit-switching technology for primarily real-time services
(voice) and also for a very small traffic volume of data. This network provided
reasonable voice quality, reliability, availability, and accessibility to
customers. The telephone exchanges of the '70s somehow could maintain call state
information, which is considered "unscalable" by many in the 21st century
despite the tremendous technological advances in chip technology of the last 40
years. Furthermore, according to the recent Cisco White Paper [1], "The sum of
all forms of video (TV, VoD, Internet, and P2P) will account for close to 90
percent of consumer traffic by 2012." This immediately leads to the question: if
we are going to have fundamentally similar services, why shouldn't we consider
increasing the use of networking concepts and solutions we had in the '70s?}
}