The Domain Name System depends on hierarchical organization and authority. It was first successfully tested 25 years ago today. __1983: __ Paul Mockapetris and Jon Postel run the first successful test of the automated, distributed Domain Name System. DNS will lay the foundation for the massive expansion, popularization and commercialization of the internet.
The fledgling internet of the time (Arpanet and CSnet) relied on a bulky and exponentially growing "phonebook" of addresses called the "host tables." It was a text file maintained by SRI International in Menlo Park, California. You contacted another computer on the network by looking up its numerical address, and typing it in.
Craig Partridge, another DNS pioneer (.ppt), later called the host tables an "operational nightmare." Everyone on the network had to copy it nightly to get the latest version. There "were many opportunities for error," Partridge wrote, "and we experienced many of them."
"People had figured out that the old scheme wouldn't work forever," Mockapetris told Computerworld a few years ago. He worked at the University of Southern California Information Sciences Institute, and his manager, Jon Postel, assigned him to devise a new way of assigning and recording internet addresses.
Their solution was brilliant. It still used an underlying system of numerical designations, but allowed you to reach a computer by name as well. It was also hierarchical and distributed. Top-level domains would mark out various types of users, like .mil or .edu. Once a name like berkeley.edu got assigned to the University of California at Berkeley, its local network administrator could independently add computers within the domain, numbering and naming them. Or the Berkeley administrator could subdelegate areas of the domain.
After testing the new plan and tweaking it for a few months, Mockapetris, Postel and Partridge published their idea in a Request for Comments (RFC) memorandum in November 1983. The system gained gradual adoption over the next few years (with prodding from the Arpanet overlords at Darpa), first supplementing and then entirely supplanting the host tables.
The first generic, top-level domains weren't officially established until October 1984 (and implemented in January 1985), but they live on: .com, .edu, .gov, .mil, .net and .org. Though DNS was originally designed to handle 50 million-plus entries, it's been expanded and internationalized. There are now probably more than a billion entries, counting all the DNS names hidden behind firewalls.
Without the Domain Name System, it's doubtful the internet could have grown and flourished as it has. Would a dot-com boom (and bust) have been the same as a dot-22.33 boom (and bust)? If numbers were being used as addresses, would Web 2.0 have emerged as Web B? Would I be writing this? Would you be reading it?
Source: Various
March 23, 1983: Reagan Taunts the Russians With 'Star Wars' Plan
June 13, 1983: Pioneer 10 Reaches an End ... and a Beginning
June 18, 1983: Sally Ride, the First American Woman Into Space
Sept. 26, 1983: The Man Who Saved the World by Doing ... Nothing
