Apparently, the media have decided to celebrate September 2 as the 40th anniversary of the Internet. This seems to be because September 2, 1969 was the date on which two computers exchanged data in the first test under the ARPANET project. This took place at the University of California, Los Angeles (UCLA); and I find it interesting in the context of our general cultural ignorance of history that there is no mention of this date in the Wikipedia entry for ARPANET. The first specific date documented by the wisdom of the Wikipedia crowd is October 29, 1969 in the caption of a photograph taken from the UCLA "IMP [Interface Message Processor] Log" for that date. The portion of the log reproduced describes, as the caption says, "setting up a message transmission to go from the UCLA SDS Sigma 7 Host computer to the SRI [Stanford Research Institute] SDS 940 Host computer." No mention is made in either the image itself or its caption of the fact, cited in an Associated Press release this morning, that "the network crashes after the first two letters of the word 'logon.'"
Since I have now reached an age where I think about my own birthday as little as possible, I am not particularly interested in how either our pioneers or their acolytes decide to peg down a specific date of origin. What does annoy me, however, is the overall Associated Press story, since it was released under the title "Key milestones in the development of Internet." More important than whether history began with the connection of two computers within the proximity of the same university campus or with a connection across California from Los Angeles to Menlo Park is the origin of the actual Internet concept. This is captured in the Associate Press chronology as follows:
1974: Vint Cerf and Bob Kahn develop communications technique called TCP, allowing multiple networks to understand one another, creating a true Internet. Concept later splits into TCP/IP before formal adoption on Jan. 1, 1983.
What is missing from the chronology is any mention of how those multiple networks arose in the first place. Once the ARPANET concept had been proven, it was only a matter of time before similar networks (without ties to the Department of Defense, which had sponsored the ARPANET project) emerged and proliferated. Wikipedia lists several of the key emergents, including the British JANET and American public access services, such as CompuServe. Most important, however, was the emergence of Usenet, which provided Unix developers with discussion groups through which technical problems could be addressed and resolved without face-to-face meetings. Usenet was supported by UUCPnet, a network of Unix-based hosts.
The value of computer-based discussion groups caught on quickly; and it became readily apparent that many of the discussions would benefit from the participation of those on other networks. Thus, a major shot in the arm for the promotion of TCP emerged with the establishment of gateway processors through which users on all those other networks could both enter Usenet discussions and exchange electronic mail with the discussants in more private settings. Mail exchange was particularly tricky. Basically, you had to send your mail to the right gateway with a destination that involved your formatting the address of the receiver in a form which that gateway could understand. Many of us heavily involved with Usenet kept a copy of a page from the Communications of the ACM with a chart that provided all of the gateway addresses and the formats they required. Getting to a UUCP user was particularly tricky, since you had to define a path from the gateway to that user's actual host that might run through several computers.
The most important contribution of the Internet as we now know it, which is never mentioned in the Associated Press chronology, is that it simplified this mess. From a user's point of view, the Internet became a reality on the day when we could all throw away that Communications of the ACM chart and reach everyone through the format of an address consisting of a user identification followed by @ followed by a domain-based host. Perhaps the Associated Press chose to ignore this event because it was too complicated to explain. History is like that; and we have become too lazy to have the will to negotiate such complexities, which is probably why we have dispensed with any account of history except those that are over-trivialized by sources like the Associated Press!