Sunday, April 26, 1998
By SCOTT MORITZ
If AT&T Corp. had listened to its own scientist, maybe its future would be clearer.
As a top AT&T researcher, David Isenberg had a simple yet radical proposition: Why have a network manage where calls and data go over millions of individual phone lines, when calls and data could more efficiently tell a vast, open network where they need to go?
The notion, outlined last year in an essay titled "Rise of the Stupid Network," is heralded by some as a manifesto for a new era of communications entrepreneurs. Isenberg's theory basically removed the old phone company from a new communication equation, and coming from within AT&T, the idea shook the foundation of a 100-year-old industry.
The essay, viewed by AT&T as something best contained within its own labs, leaked out via the Internet, where it was passed around among futurists and Internet developers.
After the essay thrust Isenberg into an important, albeit small, spotlight, AT&T soon made Isenberg feel unwelcome, he said.
In January, hurt by AT&T's refusal to let him speak about the essay at a national technology forum, and with other opportunities knocking, Isenberg resigned after 12 years with the company.
"He was not asked to leave, but given the complexity of the circumstances, no one was going to try hard to keep him here, either," said Greg Blonder, a former director of research at AT&T and Isenberg's boss at the time.
Now, with his own business, isen.com Inc., he offers advice to companies looking for direction in the new Internet-based communications order.
Isenberg, 48, of Westfield, started his career with AT&T at the research and development arm of the company, Bell Labs, which is now the independent entity Lucent Technologies. When he wrote the essay, he was working in AT&T's Opportunity Discovery Department, commonly shortened to ODD, where, he said, his role was that of the fool, to provoke thought, create debate, and rile the self-satisfied giant.
And rile he did. By attacking what he called AT&T's "obsolete assumptions" of how information should be handled, he hit the main nerve of AT&T operations, and brought the wrath of fellow researchers who had invested their careers in those "obsolete assumptions."
His deliberately inflammatory message was intended as a warning: Despite valiant technical efforts, the increasingly outmoded voice network was losing ground to a swarm of Internet innovations that was likely to to determine the future of communications.
Isenberg himself had worked on AT&T's True Voice project to improve the sound quality of the network. There, he got a clear view of the built-in limitations of a century-old network run by "intelligent" devices that operate on single format -- like a radio tuned to only one frequency.
The experience, he said, highlighted the advantages of a so-called "stupid" network, where users could send all types of information -- including conversation -- through a fat pipeline in packets that would not have to be specially formatted.
Isenberg says the business of a conventional phone network is to manage what is assumed to be a limited resource: bandwidth, or individual lines in the pipeline. The old phone system reserves the equivalent of a highway lane for each driver, and filters the size, speed, and type of information that travels over it.
The digital era, in which information was sent in compressed streams of 0s and 1s -- known as digital packets -- started the revolution. But it was the widespread use of the Internet that brought the packet technology to the masses.
Isenberg dropped the assumed limits and proposed an open network of abundant bandwidth, where packets of all types of information -- words, images, sounds -- stream independently over the quickest available routes. Instead of devices inside the network controlling traffic, the users' devices, such as computer modems at the outsides of the network, would do all the routing.
A fundamental rift between researchers and management seems to be at the heart of Isenberg's perspective.
It is unlikely the conventional telephone companies could take the necessary technological leap, said Isenberg in the essay, "as long as their senior managers prefer to talk with lawyers, regulators, consultants, and financiers more than with experts in their own employ."
Isenberg wrote the essay over Memorial Day weekend last year. "It came out very fluently. By the second day, I was thinking: Wow, this is like everything I have learned in the last 12 years at AT&T," said Isenberg recently at his home.
"I had no idea it would change my career," he said.
Isenberg's white split-level house, amid dozens of mirror-image split-level homes, contradicts the image of an iconoclast. A self-described nerd, Isenberg is a sailor, a pilot, and has a doctorate in biology. He grew up in Woods Hole, Mass., a coastal village on Cape Cod, where his father was a molecular biologist doing DNA research with the Marine Biological Labs.
These ideas in his essay had been swirling around the industry for a while; his contribution was to thread the ideas together and show how things are interrelated, he said.
"Once you read it, you immediately knew it was right," said Tom Evslin, a former head of AT&T's WorldNet Internet service. "It pulled together a lot of things we had been thinking about, very crisply," he said.
"It really helped reinforce to me which way the trends were going and where the little things we saw fit into a larger trend," said Evslin, who left AT&T in the fall to start ITXC Corp., a North Brunswick-based Internet telephony service.
The debate over a "stupid" vs. "intelligent" network continues both inside and outside AT&T. Isenberg preaches his gospel of the communications future at speaking engagements and through his Web site, www.isen.com.
People inside AT&T said reactions to the essay veered from intrigue to annoyance.
"There was a variety of responses," said Blonder, who, as Isenberg's boss, was among the first to read the essay. "Everything from 'You ought to be fired,' to 'I'm glad someone is bringing this up,' to 'I disagree, but thank God we are having this discussion,' " Blonder said.
"In retrospect, it would have been better if we had sponsored a debate within the company rather than having it come back at us from an outside source," he said.
It is Isenberg's either/or position that most bothers Daniel Scheinbein, who, as AT&T's vice president of network technology and development, has worked three decades with the voice network.
"Let me give you a different view of the world and ignore his for a moment," Scheinbein said recently at his Bedminster office. "There is always a debate on whether you want a dumb network with pipes or a smart network. The answer is always the same, and in my opinion it will always be the same: You always want the combination of the two."
Trains survived the rise of the automobile, and radio survived the rise of the television, VCR, and movies, Isenberg acknowledges, and in the multi-flavored world of the "stupid" network, some people will still choose the vanilla of the voice network.
Even though he is a full-time expert for hire, with his first fee check framed on his home-office wall, Isenberg still holds strong feelings for his former employer.
"I wish AT&T could have a better perception of the future and a better response to what is going on," he said. "I still have a sense of loyalty, having been part of something for 12 years. I'd still like to see it succeed."
But if AT&T wants a bit of Isenberg's wisdom on how to succeed, it won't come cheap. "If they want to hire me for a day and get my advice, my fee is going to go up triple for AT&T." [It's too bad that newspapers don't use :-) (smileys) yet . . . David I]
Copyright © 1998 Bergen Record Corp.
There is an accompanying article: "Can AT&T Hear the Call? Upstarts threaten giant in race towards the future" Also in The Bergen Record, Sunday, April 26, 1998.
Date last modified: 26 April 1998