I am currently taking a seminar taught by Dean Michael Smith and Prof. Jim Waldo called “What is the Internet, and What Will It Become?”. Let me start off by saying that this in itself is extremely cool. I can’t believe that, as a freshman, in my first semester, I’d have two amazing professors, both of whom have years of experience in Computer Science. As a potential CS Concentrator, I can’t imagine having a better start to my education. But, hey, it’s Harvard, am I right?

For the seminar, we’ve had to read the book Where the Wizard’s Stay Up Late: The Origins of the Internet by Katie Hefner and Matthew Lyon. This book explores the history of the internet, starting with the room-sized computers with batch processing and moving across time to now. While Al Gore may have (unintentionally) claimed to have invented the internet in an interview with Wolf Blitzer, the book compares the creation of the internet to the construction of a medieval cathedral: everyone puts down a few bricks and anyone can claim credit for putting down the most important bricks, but ultimately it’s a slow process driven by motivated people.

That’s exactly what the ARPAnet was: a system slowly designed and built by individuals and teams who were passionate about their work. First commissioned by ARPA (Advanced Research Projects Agency), the ARPAnet was the original “internet”, as a network of computers around the nation. The idea was to have a real-time, reliable, and efficient system of connecting computers across the nation. While this seems like a natural step in the evolution of computers, it, in fact, took a lot of convincing.

At the 1967 meeting called by Bob Taylor, Director of ARPA’s Information Processing Techniques Office (IPTO), in Ann Arbor, scientists viewed a national network of computers as frivolous and unnecessary. The biggest issue on their mind was the idea that all the processing for the network would be conducted on the host’s computer. This meant that the already limited processing power would be eaten into. That’s when Wes Clark, from Washington University who didn’t even really like the idea of a network, realized that a sub-network could be created that would be dedicated to routing. This idea took off. And, at the next symposium in Gatlinburg, Tennessee, the idea of a network began to gather enthusiasm. Small steps, such as this one, slowly built up to the development of the ARPAnet, and the eventual contracting of the system to Bolt Beranek and Newman.

This process was driven by a few key individuals in particular. Bob Taylor was one of the leaders in managing the creation of ARPAnet. He was the first to pitch it to Herzfeld, who headed ARPA, and hired all the key players. He wanted to increase efficiency, bring down costs, and, simply, do something cool.

Joseph Licklider, a former Harvard professor, pioneered the idea that humans and computers would be at a point of intersection, with close interactions and a sort of “symbiosis”. He also led much of the development and recruitment of the team that ultimately put together the ARPAnet.

Donald Davies and Paul Baran, a Brit and American respectively, coincidentally produced the same sort of work: the idea of packet-switching. Packet-switching allowed for information to be essentially broken up and reassembled Ikea-style down a network. This increases efficiency both in costs and in usage of cables; cables don’t have to be held open for the information. While both hit several roadblocks in the implementation of their ideas, their research revolutionized networking and routing.

The development of the internet is characterized by a lot of people putting together a lot of small parts of a giant puzzle. In fact, this sort of mimics the culture of the internet today with crowd-sourcing and the burgeoning sharing economy. It’s fascinating to see how organically the internet was constructed, less as a product that was to be perfected and sold, and more as an ongoing, developing system that keeps growing and improving.

Be Sociable, Share!

• 
•