The Irishman who built the internet — interview with Dennis Jennings (video)

20 Oct 2015

Dennis Jennings

In 1983, Irishman Dennis Jennings, who was working in the US at the time, made a fateful decision that set the technology world on course to the internet as we know it today.

Jennings was in charge of the supercomputer programme at the National Science Foundation in Arlington, Virginia, when the decision was taken to deploy a protocol known as TCP/IP in a network linking various university research departments across the country.

This was a pivotal decision that accelerated the development of the internet to what it is today, a vibrant, living internet of machines and people that is already being expanded to include objects and things.

“Dennis’s contributions to the creation of the National Science Foundation Network (NSFNET) are a perfect example of the global collaboration and innovation that have driven the growth of the internet since its inception,” said Michael Beckerman, Internet Association president and CEO, which represents the biggest internet companies in America.

“As the voice of internet companies that create jobs and facilitate the spread of knowledge around the world, we are proud to honour Dennis for his work in helping to create the foundation on which today’s internet economy is built.”

Jennings was honoured this week by the Internet Association at the Yahoo European headquarters in Dublin as the Irish capital became the location for the 54th meeting of the Internet Corporation for Assigned Names and Numbers (ICANN).

Jennings was inducted into the Internet Society’s Hall of Fame last year for his contribution to the development of the internet.

A pivotal moment in the internet’s inception

Jennings, who after a distinguished academic career at UCD has become a technology investor and start-up mentor, said the whole creation of the early internet was almost accidental.

He was in Washington working with friends when he interviewed for a job to run the Computer Science Network in the US. At the same time, a job opened up for a programme director with the National Science Foundation to build a network connecting all the supercomputers in universities across the US.

He opted for the National Science Foundation role because he felt the role would be more interesting and have more impact.

‘Today, the average smartphone has more power than the supercomputers we worked with in 1985’

The US Congress at the time had agreed to spend US$500m on new supercomputer centres to give US scientists access to this technology on a networked basis.

To make this happen, Jennings effectively made the decision between a standard known as OSI (open systems interconnection) and TCP/IP, which Vint Cerf and Bob Kahn created in the ’70s. Until that point, the internet existed in the form of Arpanet and was in the hands of the military and scientific community.

The crux of Jennings’ decision was whether to build a specific network for supercomputers or a general purpose network that all the science community in the US could access. “This was to be a network of networks and you needed a protocol to make that happen,” he explained.

It wasn’t an easy decision to make because TCP/IP implementation at the time was poor and connected researchers preferred other standards and protocols.

They relented when it was agreed that TCP/IP would be a temporary measure and OSI would be used a few years later.

But that never happened as it was clear that OSI would never replace TCP/IP because it wouldn’t enable the connections to work in the same way.

From the decision to implement TCP/IP things took off at a rapid pace. The backbone of the internet was agreed, initially at 56Kbps speeds, and the first web browser was invented in 1990 by Sir Tim Berners-Lee. The first commercial browser, Mosaic, was created by a team led by Marc Andreessen and this gave way to the arrival of Netscape.

“Within a few years the commercial internet was well underway and when companies realised this was a way of expressing and offering services there was the dot-com boom and subsequent crash,” explained Jennings.

‘Did we think we would build a network that would become the communications infrastructure of the world? No.’

“Today, the average smartphone has more power than the supercomputers we worked with in 1985.

“If you look at what’s happening we are now entering a world of infinite connectivity, bandwidth and computing power. It accelerates still.”

Back in 1985, Jennings admits that the team were tremendously ambitious about building a network for researchers in more than 394 universities across America to access supercomputing power.

He admits that if he had any idea their work would lead to the communications infrastructure for an entire planet, it mightn’t have happened.

“Did we think we would build a network that would become the communications infrastructure of the world? No.

“In fact, if I was told that what would happen, I couldn’t have done it. If I was told there would have been security issues, would I have focused on that? If I had been told there was only one backbone I would have built another.

“But had we any idea that we would be building the future communications infrastructure for the world? No. Absolutely not.”

Dennis Jennings interview (Part 1)

Dennis Jennings interview (Part 2)


Updated 9.20pm, 21 October 2015: This article was updated to reflect that the Internet Hall of Fame is managed by the Internet Society rather than ICANN.

John Kennedy is a journalist who served as editor of Silicon Republic for 17 years