Slashdot Mirror
Celebrating The Origins of Packet Switching
XaN-ASMoDi writes with an interesting historical piece at the BBC on the early history of packet switching, excerpting: "It has often been said that change is the only constant in the 21st Century. And there is little doubt that the restless tone of these times is something that the web has helped to accelerate, but the only reason that [...] the web can cope with that punishing pace is thanks to work done four decades ago by British mathematician Donald Davies at the UK's National Physical Laboratory (NPL).
On 5 August 1968 Dr Davies gave the first public presentation of work he had been doing on a method of moving data around computer networks called 'packet switching.'"
It has often been said that change is the only constant in the 21st Century.
I've never heard that applied specifically to the 21st century. It's an oddly-specific use of the phrase: does that imply that change has not been constant in other times? Empires grow and fall, cultures collapse or are swept away or conquered by the next Big Empire, customs change, ethnic identities change, etc etc. The only unique thing about the 21st century is that we've inherited a tradition of rapid technological change. Technology is important but it's hardly the only thing that changes over time and it strikes me as fairly myopic to single out the 21st century as a time of change.
made me realize the intarnet is not tubes at all
It's an oddly-specific use of the phrase: does that imply that change has not been constant in other times?
Re-read it. It says that the only thing that doesn't change in the 21st century is that things change. Arguably, things always change, in which case the phrase is vapid. At the same time, the rate at which things change is accelerating. Compare life in Europe, circa 1000 with 1100. Then compare life in Europe 1900 and 2000 - the differences are much greater!
You can lead a horse to water, but you can't make it dissolve.
Paul A. Baran and Leonard Kleinrock also get credit for packet switching. It appears that multiple inventors independently (or semi-independently) came up with the idea and helped make it usable.
Table-ized A.I.
A lot of people made contributions around that time. As a then undergraduate flunky in a physics lab, I remember playing with an early network that linked the Illiac II with a bunch of other computers. Since we weren't supposed to be playing with it, all we could do was come up with the login prompts, but several Midwestern universities were linked already. ARPANET sputtered to life in 1969, too soon for the touted theoretical contributions to be seminal. I don't doubt Davies did important work, but the article is over hyped.
Well ... it seems, Donald Davies was a little bit late on his work. According to Prof. Leonard Kleinrock (http://www.lk.cs.ucla.edu/), he had actually filed his Phd Thesis at MIT in 1961 creating the underpinnings for packet switching. (Kleinrock, L., "Information Flow in Large Communication Nets", Ph.D. Thesis Proposal, Massachusetts Institute of Technology, July 1961). About 7 years before Davies' talk. Although, they could have worked simultaneously on the same thing, it took years at that time to get something published distributed widely. Nowadays, we have slashdot!
My Dad worked with Donald Davies. That's him on the far right in the pic on the BBC. He would have been 85 tomorrow.
by Paul Baran in the early 1960's.
http://en.wikipedia.org/wiki/Packet_switching
Baran developed the concept of packet switching during his research at the RAND Corporation for the US Air Force into survivable communications networks, first presented to the Air Force in the summer of 1961 as briefing B-265 [1] then published as RAND Paper P-2626 in 1962 [1], and then including and expanding somewhat within a series of eleven papers titled On Distributed Communications in 1964 [2]. Baran's P-2626 paper described a general architecture for a large-scale, distributed, survivable communications network. The paper focuses on three key ideas: first, use of a decentralized network with multiple paths between any two points; and second, dividing complete user messages into what he called message blocks (later called packets); then third, delivery of these messages by store and forward switching.
"To those who are overly cautious, everything is impossible. "
Interesting that this event just came up. I just finished reading Where Wizards Stay Up Late: The Origins of the Internet today. For those interested in the history of ARPA and internetworking development (and the people behind these developments - Davies, Kleinrock, Licklider, et. al.), I highly recommend this book.
"The purpose of learning is growth, and our minds, unlike our bodies, can continue growing as long as we live." - M.J. A
Compare life in Europe, circa 1000 with 1100. Then compare life in Europe 1900 and 2000 - the differences are much greater!
But has it really changed that much in the last 30-40 years?
The majority of the largest changes in the last hundred years would be: electrification ('10s), radio ('20s), telephone ('30s).
Secondary changes would be automobile growth ('40s), the mainstreaming of the automobile ('50s) and television ('50s). The most recent change since the 1950s would the mainstreaming of computers in the late '90s.
I think the delta from before World War I and to just after World War II would be the greatest amount of change (just look at the technology used in each of those). The Roaring Twenties would be the greatest change. IMHO it's been mostly incremental technologically, but perhaps a bit more revolutionary from a societal point of view, as technology has been absorbed into our psyche.
Even with something like the mobile phone, the basic idea hasn't changed since the early '80s, even though they've become more sophisticated. It's the social effects that have become more prevalent in the last five years. But someone who's used a touchtone (DMTF) phone in the '60s would be perfectly at home with a mobile (if they just wanted to dial, and not have to deal with most of the suck-ass address book UIs).
If you want to talk about easy digital, packet based networking, then the telegraph is worth a mention. Messages for electronic telegraphs were sent as packets that are later relayed to its final destination. It was also digital in the sense that each character was transmitted as discrete values. Another predecessor to TCP/IP based networking would be the mail system. While it wasn't digital, it was also packet based and as anyone who've used the postal system knows, it's only best effort. For added cost, you can get ack-like behavior with return receipt.
EvilCON - Made Famous by
I was a Telecommunications Specialist in the military Autodin System (Automatic Digital Network) in the DCS from 1975 until 1979. I was a network troubleshooter - fault isolation and correction. This was a worldwide packet switching system for military communications and supported the NSA need to funnel intercept data back to the US from Menwith Hill and other bases. I could be wrong, but something seems wrong with the given timeline as there was a full-blown worldwide system in 1975 when I entered the service. I had approximately 150 circuits to bases that I was responsible for keeping up - modems, crypto gear, tropospheric scatter and microwave links carried our data (had HF radio too). Was there enough time to set that up? Was there a parallel development that is not showing up in technical history? Just a thought. We used a Philco-Ford, then Aeroneutronics Ford computer for routing. We had two Procs that swapped on error, and drum memory. I viewed the circuits by dialing them up on an "MTC" Monitor and Test Console, to watch the bit traffice on a dual trace oscilliscope. Things have changed so dramatically that I can go onto Google Earth to view my old base and note the changes that have taken place in the interim. I thought the advent of advanced technology would have caused my small base to shut, but that's not the case. It has been security hardened and 4 new large satellite dishes added.
E Proelio Veritas.