, now 62 years old and commonly referred to as one of the "fathers" of the Internet, said Thursday that were he 30 again, he would be focused on networking communications in space. In 1973, as an assistant professor at Stanford University and a DARPA scientist, Cerf co-designed the Net's underlying architecture, TCP/IP (Transmission Control Protocol/Internet Protocol), for the U.S. military.
"Now, I would be working on interplanetary stuff," said Cerf, vice president of technology strategy at MCI, while giving a keynote speech during a one-day Internet conference here at his old stomping grounds.
He was referring to a project focused on developing a new set of protocols for communication in deep space, where continuous connectivity can't be assumed.
Cerf said that he would also like to work in the field of bioelectronics because he would like to help develop life-transforming devices for humans. The concept is of particular interest to Cerf because his wife is deaf; and her life was dramatically changed for the better when she was fitted with cochlear implants. Bioelectronics is a field in which scientists develop electronics that can be wired to neurons in the brain in order to help the handicapped, for example.
During a question and answer period at the "Internet: Today and Tomorrow" confab, sponsored by Stanford's engineering department, Cerf was asked about his early work on TCP/IP and whether he was influenced by work called PUP (PARC Universal Packet), another smaller networking protocol, developed at Xerox's Palo Alto Research Center. Cerf said that because Xerox was proprietary about its work, he wasn't influenced directly by the group's developments. Rather, a couple members of PARC gave guidance to him and co-designer Bob Kahn on what things could go wrong in high-speed packet transfers.
"PARC gave us hints as to what they experienced," Cerf said, while talking to actual members of the early team at PARC who were sitting in the audience.
He went on to say that he and Kahn developed TCP/IP intentionally to be an open, international standard so that it would be widely adopted. At the time, he did not know that it would become a commercial service and go on to harness a vast amount of creativity in new applications.
"Probably 99 percent of all the applications of the Internet haven't been invented yet, and I can't wait to see the next ones," Cerf said.
However, the openness of TCP/IP points to the viability of such an approach.
That's why Cerf said he's "very unhappy" about the notion of patenting software, because it naturally constricts creativity. In recent years, technology companies have raced to secure intellectual property rights to a range of commonly used technologies that are poorly understood by the governmental patent attorneys that approve the patent applications, Cerf said.
"I'm glad we didn't try to patent it," he said.