Came across this recent 3-part article by David Talbot in the MIT Technology Review.
- What ails the current Internet? Why what is being done is not ok?
- What can be and should be done?
- What is being planned?
Answers to each of these form a separate part.
The crux: Current Internet architecture uses the dumb-network, smart-device paradigm. Specifically, security is left to the user. Not only has this cost a lot of money, it has led to increased security threats and the possibility of an e-Armageddon. (Infact, also mentioned in the article is that, for 10 years, people have predicted a digital “Pearl Harbor” in which a security attack crashes the Internet leading to a power grid failure blanketing the entire US of A. Hmm… Nice Hollywood material.)
As a person with some basic knowledge in Computer Networks and the operation of the Internet, one knows that the Internet has a lot of scope for optimization and restructuring.
With adhoc networks coming in, (such as home networks, sensor networks) and given the promise of greater throughput by MIMO (Multiple Input Multiple Output) systems and Cognitive Radio, the future Internet should not only include security issues but take into account the disparate and mobile nature of the network devices.
This page which concludes the article contains some universities that have received related grants from the NSF, in case you are interested.