Feeds

Father of internet warns against Net Neutrality

No Kahn do

5 things you didn’t know about cloud backup

Robert Kahn, the most senior figure in the development of the internet, has delivered a strong warning against "Net Neutrality" legislation.

Speaking to an audience at the Computer History Museum in Mountain View, California at an event held in his honour, Kahn warned against legislation that inhibited experimentation and innovation where it was needed.

Kahn rejected the term "Net Neutrality", calling it "a slogan". He cautioned against dogmatic views of network architecture, saying the need for experimentation at the edges shouldn't come at the expense of improvements elsewhere in the network.

(Kahn gently reminded his audience that the internet was really about interconnecting networks, a point often lost today).

"If the goal is to encourage people to build new capabilities, then the party that takes the lead is probably only going to have it on their net to start with and it's not going to be on anyone else's net. You want to incentivize people to innovate, and they're going to innovate on their own nets or a few other nets,"

"I am totally opposed to mandating that nothing interesting can happen inside the net," he said.

So called "Neutrality" legislation posed more of a danger than fragmentation, he concluded.

With the exception of Google's man in Washington DC, Vint Cerf (with whom Kahn developed TCP/IP), most of the senior engineers responsible for developing the packet switched internetworking of today oppose "Neutrality" legislation. Dave Farber, often called the grandfather of the internet, has been the most prominent critic.

Engineers fear rash legislation would inhibit the ability of systems engineers to improve latency and jitter issues needed to move data at speed.

"The internet is still pretty fragile today," said Kahn.

Life of Kahn

Kahn's history as protocol designer is a minor note, compared to his role as a politically astute manager and advocate at key moments in the development of the technologies responsible for the internet.

When he embarked on a career in networking, peers and seniors tried to talk him out of what was then considered a crazy choice.

"People thought I was throwing my career away. People thought time sharing wouldn't take off and if it did it would only be in a few palce, so wouldn't have commercial values," he said. "If I had listened to many people in the field I would not have gone into networking."

Working on colour TV, automatic game control loops, information theory, and even microwaves were all considered "cooler" than networking.

Later, he found DARPA was a reluctant sponsor. The US Department of Defense's research agency didn't have many computers when Kahn arrived in 1972 and couldn't see much of a use for them.

Technical history is often seen as an inevitable progression, punctuated by moments of individual genius, but the gentle backroom cajoling rarely gets mentioned.

It was certainly needed. Ironically, when Kahn arrived at DARPA it was to take a break from networks, and work on factory automation research. But the hype du jour, Artificial Intelligence, was sweeping the land and Congress cut the budget for his project. Kahn began to reassemble a team of packet switching veterans.

In the early 80s he managed the gradual, and awkward transition from private defense project to public network, fighting off the cumbersome, bureaucratically-devised OSI model of internetworking.

"I fought a ten year battle to protect the name 'Internet'", he says. "It cost a million dollars and eventually the name prevailed - but we could have lost the internet."

The CRNI (Corporation for National Research Initiatives) was really essential to winning that one, he said.

Kahn urged today's engineers to "Think Big... we are at the very early stage of a revolution that's going to take most of the 21st century". He rejects any labelling as the "father of the internet", saying credit for its growth can be shared by the entire industry.

You can find a video of his talk here. It's a 230MB download in Windows Media format only, and there's no transcript yet.

Thanks to veteran internet engineer Richard Bennett (who has plenty to say about the subject) for the heads up. ®

Gartner critical capabilities for enterprise endpoint backup

More from The Register

next story
6 Obvious Reasons Why Facebook Will Ban This Article (Thank God)
Clampdown on clickbait ... and El Reg is OK with this
So, Apple won't sell cheap kit? Prepare the iOS garden wall WRECKING BALL
It can throw the low cost race if it looks to the cloud
EE accused of silencing customer gripes on social media pages
Hello. HELLO. Can EVERYTHING EVERYWHERE HEAR ME?!
Time Warner Cable customers SQUEAL as US network goes offline
A rude awakening: North Americans greeted with outage drama
Shoot-em-up: Sony Online Entertainment hit by 'large scale DDoS attack'
Games disrupted as firm struggles to control network
BT customers face broadband and landline price hikes
Poor punters won't be affected, telecoms giant claims
Broadband slow and expensive? Blame Telstra says CloudFlare
Won't peer, will gouge for Internet transit
prev story

Whitepapers

Best practices for enterprise data
Discussing how technology providers have innovated in order to solve new challenges, creating a new framework for enterprise data.
Implementing global e-invoicing with guaranteed legal certainty
Explaining the role local tax compliance plays in successful supply chain management and e-business and how leading global brands are addressing this.
Advanced data protection for your virtualized environments
Find a natural fit for optimizing protection for the often resource-constrained data protection process found in virtual environments.
How modern custom applications can spur business growth
Learn how to create, deploy and manage custom applications without consuming or expanding the need for scarce, expensive IT resources.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?