Lawrence Roberts, Who Helped Design Internet's Precursor, Dies at 81 (nytimes.com) 46
In late 1966, a 29-year-old computer scientist drew a series of abstract figures on tracing paper and a quadrille pad. Some resembled a game of cat's cradle; others looked like heavenly constellations; still others like dress patterns. Those curious drawings were the earliest topological maps of what we now know as the internet. The doodler, Lawrence G. Roberts, died on Dec. 26 at his home in Redwood City, Calif. He was 81.
The New York Times: The cause was a heart attack, said his son Pasha. As a manager at the Pentagon's Advanced Research Projects Agency, or ARPA, Dr. Roberts designed much of the Arpanet -- the internet's precursor -- and oversaw its implementation in 1969. Dr. Roberts called upon a circle of colleagues who shared his interest in computer networking for help in creating the technical underpinnings of the Arpanet, integrating and refining many ideas for how data should flow. Dr. Roberts was considered the decisive force behind packet switching, the technology that breaks data into discrete bundles that are then sent along various paths around a network and reassembled at their destination. He decided to use packet switching as the underlying technology of the Arpanet; it remains central to the function of the internet.
And it was Dr. Roberts's decision to build a network that distributed control of the network across multiple computers. Distributed networking remains another foundation of today's internet. Dr. Roberts's interest in computer networking began when he was a graduate student at the Massachusetts Institute of Technology in the early 1960s. He paid close attention to the work of his longtime colleague, Leonard Kleinrock, who had done research on theoretical aspects of computer networks, analyzing the problem of data flow. Dr. Roberts also followed the ideas of J.C.R. Licklider, a prominent psychologist and predecessor of Dr. Roberts's at ARPA, who envisioned what he called an "intergalactic computer network."
And it was Dr. Roberts's decision to build a network that distributed control of the network across multiple computers. Distributed networking remains another foundation of today's internet. Dr. Roberts's interest in computer networking began when he was a graduate student at the Massachusetts Institute of Technology in the early 1960s. He paid close attention to the work of his longtime colleague, Leonard Kleinrock, who had done research on theoretical aspects of computer networks, analyzing the problem of data flow. Dr. Roberts also followed the ideas of J.C.R. Licklider, a prominent psychologist and predecessor of Dr. Roberts's at ARPA, who envisioned what he called an "intergalactic computer network."
Re: End of an era. (Score:1)
Better than Orwellian unaccountable shadowy NGOs. At least corporations have to answer to shareholders, as flawed as that can be, at least they answer to somebody
Re: (Score:2)
Re: (Score:2)
With the benevolent creators of the internet dieing off, the future looks bleak and in the hands of orwelian corporations.
I sometimes hear sentiments like this and it always leaves me scratching my head.
Given cost, capability and availability of technology it's never been easier or cheaper to act.
It's like living in the matrix, a swarm of sentinels coming your way and you keep bitching about how many of them there are and how hard it is to reach over and press the big red EMP button.
Re: (Score:2)
It might be easier and cheaper in financial and technical terms, but you would run afoul of dozen of laws and have your life ruined forever.
Only if you live in North Korea (or Australia).
Re:I call Sean Hannigan! (Score:5, Informative)
Al Gore didn't invent the Internet.
He did however lead the way in creating and passing legislation that lead to the expansion of the NSFNet and the creation of the Internet as the true "network of networks" that we know it as today.
Re: (Score:3)
expansion of the NSFNet
Read that as NSFWNet. Which actually is more fitting.
That was his boss, Bill (Score:1)
NSFWnet was Gore's one-time boss, Bill Clinton.
Re: (Score:2)
As any fule kno, it was Al Gore what invented teh interwebs.
Actually, the main thing Al Gore did was support legalization of general access to, and commercial use of, the Internet by the general public, rather than leaving it restricted to a few universities, the military, and their contractors.
So what he did was legalize spam.
Re: Nuclear Bombs, Communism, and the Cold war (Score:2)
Fight for Net Neutrality - For Roberts! (Score:2)
Just because a few selfish board-members rammed one of their agents into the FCC, it does mean that one of humanity's greatest inventions has to be corrupted and sold to the highest bidder.
Fight to save the internet!
Misconfiguration (Score:3)
He should've set his ttl higher...
His heartbeat hit a lag spike and dropped out. No carrier.
We should've listened to Ted Stevens -- the arteries of the internet are too clogged.
I'll see myself out.
Origins of packet switching (Score:3, Informative)
The article says "Dr. Roberts was considered the decisive force behind packet switching", which is misleading - packet switching was already used in a commercial service in the UK at the time discussed. He was perhaps the decisive force behind using packet switching in ARPANET. The wikipedia article https://en.wikipedia.org/wiki/... [wikipedia.org] is informative.
A real contributor (Score:5, Interesting)
I was a Telecommunications Specialist in the military (AF) Autodin system from 1975 to 1979. It was a worldwide military packet switching message system using Philco-Ford computers and microwave and tropospheric scatter communications links as well as undersea cables back to the states. We had one satellite dish at RAF Croughton England at that time. I could actually see the packet communications on an O-scope and read the attempts of cryptographic equipment to synchronize. The "procs" were big cabinets and generated a lot of heat. We had two for fail-over and they used drum memory. All model T - ish, but it worked. It got the job done. I had approximately 150 to 200 circuits to monitor and keep up. Some of those circuits had Colonels in deep bunkers at the other end. It always amused me as a two-striper to have a Colonel call me up to troubleshoot his circuit outage. Of course, I'd have to order him around a bit to accomplish that. I could see both the black and red sides of a message and that, of course, is a severe no no in encryption. We had a very high level of Top Secret clearance as a result. The point is that there was a full-blown, worldwide military packet switching network in place and working when I joined the Air Force in 1975. Amazingly short time from conception to implementation - unless the story is more complex than we hear.
P.S. It used a programming language that sounded like "shall-A." I wasn't a programmer, so I'm not exactly sure what it was. We had a captain programmer who only rarely interacted with us- usually when I screwed up what he was doing by my habit of pushing the circuit alarm button repeatedly to get a fault report absent mindedly as I thought about the problem. He had programmed the computer locally to recycle the computer when the button was pushed rather than have us call each time to operations on the phone (in the next room) to recycle the computer manually. But for some reason he failed to tell us what he was doing. Understandable because enlisted ran the Switching center. He was the only officer and we never saw him. I think he was afraid of us. (laughs)
Does anyone know what "shall-A" may have been?
Re: (Score:2, Interesting)
Quoting from http://www.computerculture.org/2012/02/recalling-the-autodin-part-i/
"The original Overseas coding was SHAL-A assembler for all functions until 1981 when we installed PDP-11’s and moving head Disks to replace Tape Drives. The PDP-11 code was Macro-11 assembler for the On-Line PDP-11 functions and Fortran for the Off-Line functions (Table loader/Retrieval, etc.)"
So your recollection seems about right. The comments have some more technical info.