Samuel Morse opened the telegraph with a line he had chosen in advance. When the sent its first message in October 1969, nobody had prepared anything, because the task at hand was mundane: log in to a machine at Stanford Research Institute from one at UCLA. Leonard Kleinrock, who ran the receiving end, tells the story. The message is two letters long, and not by design.
It was lo, because the system crashed mid-word. Charley Kline at UCLA typed l and asked SRI by telephone whether it had arrived. It had. He typed o, confirmed again, typed g, asked, and at that point the system crashed. The word being typed was log, which the SRI host would have completed to login. Kleinrock notes they had no camera and no voice recorder, and the only record is an entry in the .
Larry Roberts submitted the ARPANET Program Plan to the US Defense Department’s Advanced Research Project Agency in early June 1968, and it was approved later that month. A request for quotation for the network interface message processors went to 140 potential bidders by the end of July, and 12 proposals came back. Bolt, Beranek and Newman won the contract a week before Christmas.
Four sites were chosen for what each could contribute. UCLA supplied Kleinrock’s and Stanford Research Institute supplied Doug Engelbart’s Human Intellect Augmentation System. UC Santa Barbara brought interactive graphics and the University of Utah advanced three-dimensional graphics.
The first host-to-host message went out on the night of October 29, 1969, when Charley Kline set about logging in to the SRI host from the UCLA host. The procedure was to type log, and the SRI system was clever enough to complete the command by adding in. Kline and Bill Duvall at the SRI end each had a telephone headset so they could confirm progress by voice. Kleinrock notes the irony, since they were using the telephone network to launch the packet switching that would displace it.
Kline typed the l and asked whether it had arrived. Got the l, came the reply. He typed the o, asked, and got the o. He typed the g, asked, and the system crashed. So the first message on the network was lo. Kleinrock observes that they had prepared nothing, unlike Morse with his chosen line, and that lo could hardly have been more succinct or more prophetic. They had no camera and no voice recorder, and the only record of the event is an entry in the IMP log.
wat. The first message between two computers was lo, because the machine crashed before the g of log arrived.
wat wat
The detail that makes it human is the telephone. Kline and Bill Duvall at SRI each wore a headset so they could confirm each character by voice, using the telephone network to launch the packet switching that would displace it.
How do we know?
Leonard Kleinrock of UCLA, who directed the Network Measurement Center at the receiving end, recounts the sequence in a guest blog for ICANN marking the fiftieth anniversary.
What is still unknown
A first-person recollection published fifty years after the event, by a participant, on an organization’s blog. Kleinrock himself notes the only contemporaneous record of the moment is the IMP log entry.
Learn something?
Where to next
What happened when a brainless organism designed a rail network?networks
Why can one failure take down the power for fifty million people?networks