Oh, look -- it's Steve Case talking to Jeff Bezos. And there's Al Gore chatting with Bill Gates. And over there is Larry Ellison, dancing by himself. Aren't those the guys that invented the Internet?
Well, no, they're not. Neither, for that matter, is Linus -- the only Internet superstar the entire world knows by his first name -- nor is Marc, the original Internet supernova, (although they certainly have contributed more to its development than Gates, Ellison and Bezos put together.)
The folks who were really responsible for bringing this rough beast into existence have mainly toiled in obscurity, while better-known figures crafted marketing strategies and staged press conferences. They have, too often, been the unsung heroes of the Internet, the unknown soldiers of the information age.
You've heard about Vint Cerf, who spearheaded the development of TCP/IP, Bob Metcalfe, who invented the Ethernet protocol and Tim Berners-Lee, the father of the World Wide Web? Here are the stories of just a handful of their less-celebrated peers -- and there are many more such stories yet untold:
Douglas C. Engelbart
Tim Berners-Lee may have invented the World Wide Web, but he didn't invent hypermedia.
The man who did was legendary computer pioneer Douglas C. Engelbart, who also developed the mouse -- and many of the other fundamental computer technologies we now take for granted -- during his long career. The system he developed, called "Augment", featured considerably more powerful hyperlinking capabilities than does today's HTML -- among other things, all the links it established were fully bi-directional.
Engelbart demoed that capability -- along with shared-screen teleconferencing, full-featured email, interactive multiuser document editing with full version control and many other of what we now call "groupware" applications, all running on a network of workstations that empoyed a graphical, windowing, mouse-driven environment -- and he did it at the December 9, 1968 Fall Joint Computer Conference. It was for him, the culmination of an almost 20-year crusade.
Engelbart was then and today remains a visionary in every sense of the word. As a consultant for Ames Research Center in December, 1950, he experienced the epiphany that launched him upon a path that, 50 years later, he still follows today.
Inspired, in part, by Vannevar Bush's seminal July, 1945 Atlantic Monthly article, "As We May Think," Engelbart realized that, with the huge technological advances brought about by WWII and the steady stream of new discoveries, the world was swiftly becoming entirely too complex for unaided human minds to handle. At the same time, he forsaw that the steadily-increasing use of computers -- still little more than toys at the time -- would eventually drive major sociological changes.
Engelbart's experiences as a Navy radar technician convinced him that CRTs could be used to dynamically display and update graphics and textual information -- at a time when the standard computer display device was still paper tape. Unlike virtually every other computer scientist of his time, he also was interested in harnessing computers to process documents and images and as tools to enable collaboration and enhance productivity, instead of merely using them to crunch numbers.
In 1958, after four years spent on a failed high-tech startup venture, he joined the staff of Stanford Research Institute. By December, 1960, he felt secure enough in his position at SRI to formally propose to the U.S. Air Force that it fund a framework for developing the comprehensive computer environment he had first imagined ten years earlier, in a document diffidently titled "Augmented Human Intellect Study".
His proposal attracted sufficient funding from the Air Force to let him get started on building what would, by the mid-1960's, become the Augmentation Research Center at SRI. In 1963, Engelbart was waiting with a proposal in hand when J. C. R. Licklider first opened doors of the new Information Processing Techniques Office at the Advanced Research Projects Agency. In 1969, ARC would become the second node on the nascent ARPANET -- and Engelbart's Augment environment would be used to administer the network's first Network Information Center.
Unfortunately for Engelbart, his character flaws worked greatly to undermine his influence. He disliked the administrative side of managing ARC, but he could never convince SRI to divorce those duties from the more scientific and evangelical ones that came naturally to him. He also had chronic difficulty conveying to scientists outside of ARC what was so compelling about his revolutionary vision of a fully-integrated computer environment -- and specifically what it could help they, themselves to accomplish.
He had even greater trouble articulating his vision to management at SRI and his situation with them deteriorated until, in May of 1976, Engelbart was replaced as ARC's Director. His staff immediately began jumping overboard and 15 of them quickly wound up at Xerox's Palo Alto Research Center -- where, three years later, Steve Jobs would see Engelbart's mouse in use with the experimental Alto microcomputer and steal the idea for Apple's Lisa and Macintosh computers.
SRI fired Engelbart in July, 1977, sold the Augment system to TYMSHARE Corp. as a potential commercial software product and closed down ARC for good. Still hopeful that someone would recognize the value of his vision of human augmentation, Engelbart followed his brainchild to TYMSHARE. When McDonnell-Douglas, bought the company, it also acquired Augment and Engelbart -- and paid no more attention to them than SRI and TYMSHARE had done.
Engelbart contracted -- and survived -- lymphoma in the mid-1980's. He formally retired in 1989 and founded the Bootstrap Institute in affiliation with Stanford University. He now devotes his time to studying how to improve the performance and effectiveness of organizations by the use of both technology and changes in corporate culture and managment practices.
In lieu of royalties, he got a check for $10,000 from SRI for inventing the mouse -- and received nothing at all for having developed hypertext, email, word processing, structured programming, context-sensitive online help, GUIs, groupware or any of the other innovations he pioneered at ARC.
Robert W. Taylor
If one man can honestly claim to be the Father of the Internet, Texas native Robert W. Taylor is that man.
As a program manager for NASA in the early 1960's, he began funneling research dollars to Doug Engelbart at SRI, even before the legendary founding Director of ARPA's Information Processing Techniques Office, J. C. R. "Lick" Licklider, began supporting Engelbart's work. Conferring with the man whose seminal 1960 paper, "Man-Machine Symbiosis," had ignited his own interest in computers as something more than mere number-crunching machines so impressed Taylor that he promptly accepted a job working with Licklider at ARPA.
It was there, in late 1965, that Taylor had his inspiration for what would eventually become the Internet.
He had noticed that what would now be called "virtual communities" were evolving among the users of the three time-sharing, interactive computers that were being funded by ARPA, but he thought it was silly and wrong that each such computer was accessed via a physically and logically separate teletype terminal. Taylor thought they should be connected together, so that one terminal -- and one set of user commands -- would enable a user to access any of the three.
Even after he became Director of IPTO in February, 1966, it took Taylor a while to sell to the ARPA community the idea of hooking everyone into a common network. Some, like Engelbart and Licklider, quickly grasped the advantages of being able to share resources and data and to collaborate electronically. Others, however, were jealous of their still-costly CPU cycles and resisted sharing them for what they regarded as nebulous benefits.
But Taylor was persistent -- and he had the ear of Charles Herzfeld, the head of ARPA, who immediately budgeted the first million dollars in funding for the project. And Licklider was enthusiastic enough to co-author with Taylor a 1968 paper, "The Computer as a Communications Device," (it and Licklider's "Man-Machine Symbiosis" are combined in a single PDF document available from Memex.org) that laid out the benefits of virtual communities and networked computer systems in compelling and prescient detail. And so it was that, in 1969, the ARPANET was born.
And -- having stayed to father the ARPANET -- Taylor promptly left the government agency in favor of a stint as a researcher and consultant at the University of Utah. He would stay there only a year -- still long enough to establish it as the third node on the ARPAnet -- leaving in 1970 to become the first Director of Computer Research at Xerox Corporation's brand-new Palo Alto Research Center.
He remained at Xerox until 1983 and there, under his direction, PARC's scientists would sow the seeds of much of today's ubiquitous computer technology. They began by developing the Alto -- the world's first personal computer -- in 1972. The Alto featured a mouse-driven, user-friendly GUI that built on the groundbreaking work of Doug Engelbart's Augment environment at ARC -- work Taylor had helped to fund.
Taylor's staff at PARC included visionary Alan C. Kay, Robert M. Metcalfe, the inventor of Ethernet, and a host of others. Between them and their PARC peers, they would create such technologies as as the use of icons and the desktop as visual metaphor in user interfaces, applications such as desktop publishing and sophisticated word processing and peripherals such as the laser printer -- almost none of which would come into general use until a decade or more after their development at PARC.
Taylor -- who would go on to found Digital Equipment Corporation's Systems Research Center and remain its Director until DEC was swallowed by Compaq Computer Corporation in the late 1990's -- has always been puzzled as to why it took so long for the world to embrace the advances he saw as obvious.
As he confessed to John Markoff in a December, 1999 interview for the New York Times, "I was sure that, from the early 1970's, all the pieces were there at Xerox and at ARPA to put the Internet in the state in the early '80's that it is in today. It was all there. But it didn't happen for years."
Now retired, Taylor devotes his time to crusading for Internet access as a basic human right and to setting straight the often-inaccurate journalistic accounts of the early days of the Internet. He goes to particular pains to debunk the popular myth that the ARPANET was funded in order to develop a data network for the military that could survive a general nuclear war.
As he told me in 1995, "I was the one who presented the funding request to Charlie Herzfeld, who approved it. I made the case for it and it had nothing to do with that at all. It was just the two of us involved in the decision, so there's no question of there being any hidden agendas. I pitched it to Charlie as a pure research project and that's the basis on which he approved the funding for it.
"End of story."
John Gilmore was employee # 5 at Sun Microsystems. It made him a rich man.
Like many of today's dotcom millionaires, he could easily have chosen to use his wealth strictly to finance a life of total self-indulgence -- but Gilmore chose a different path.
In early 1986, despite a supposedly-binding vote in which the Usenet community overwhelmingly approved the creation of the rec.drugs newsgroup, the so-called Backbone Cabal -- the network administrators for the major nodes of Usenet's UUCP-based backbone network -- flatly refused to carry the new group. A passionate free speech advocate, Gilmore was outraged.
Gilmore had a T1 connection to his South San Francisco house, Toad Hall. He and Brian Reid, the admin for DEC's decwrl server, conspired with Gordon Moffett of Amdahl to start an alternate news backbone that would carry its own, separate hierarchy of groups. Gilmore financed the telecommunications costs for the entire alt. backbone out of his own pocket, until the solidarity of the Cabal eventually broke under the pressure and the alt groups went mainstream.
Gilmore's comment? "The Net treats censorship as damage and routes around it."
In 1990, the Secret Service began its infamous "Operation Sun Devil" crackdown -- a nationwide witch hunt for hackers so bizarrely misinformed, so incompetently conducted and so studiously wrongheaded that it would have been wholly laughable, were it not for the pall of fear and oppression it cast over the entire computing community.
In late spring, 1991, concerned about the civil liberties and First Amendment impacts of Sun Devil, John Perry Barlow and Mitchell Kapor set up the Electronic Frontier Foundation. Its mission was to educate the public -- and law enforcement -- about the issues, pay the legal costs of the Sun Devil defendants and work for the preservation and expansion of civil liberties at the intersection of the digital and physical worlds. Kapor provided seed funding.
Gilmore and Steve Wozniak immediately matched Kapor's financial contribution, penny for penny. Within a year, the EFF had a full-time staff and a full-time Washington lobbyiing operation. Within two years, it had won convincing -- and extremely expensive -- victories against the Secret Service in both major Sun Devil raid cases.
Gilmore also has a long-held interest in cryptography. In 1992, he helped found the Cypherpunks -- a loose-knit group that advocates for the elimination of legal restrictions on cryptography. He has also provided major backing for the Cryptorights Foundation, an international lobbying group dedicated to raising awareness of the importance of cryptography in human rights struggles around the globe.
Gilmore also financed and led the team that, in 1998, cracked the hallowed DES encryption algorithm in a mere three days -- and thereby demonstrated, as he puts it, that "anyone with $200,000 can break the code."
In the 1980's, Gilmore co-founded The Little Garden, the first private Internet service provider in the San Francisco Bay Area. In the 90's, he, Mike Tiemann and Henkel-Wallace founded Cygnus Solutions, the first commercial provider of technical support for free software.
John Gilmore has been a free software advocate and developer since 1981. He ported GNU Emacs to the Sun platform in 1983 and has been involved in numerous free software development projects ever since. Having "retired" in 1995, he remains a tireless advocate in the real world for the freedom of encryption, software and speech in the virtual world.
Steven E. Deering
Steve Deering is the prototypical computer nerd. Soft-spoken, mild-mannered, intensely private about personal matters, perpetually tieless and possessed of a moon face framed by a bushy beard, he dominates technical discussions not by the force of his personality, but by virtue of his sheer intellectual brilliance and his incredible capacity for work.
As a graduate student, Deering became interested in the possibilites of the Internet Protocol Suite, which was then still limited to unicast and broadcast modes. He became convinced that IP needed to include support for a third mode that would permit multiple machines to communicate as a self-selected group, without having to flood the network with redundant data packets in the process.
In 1985, he first presented his ideas for a set of multicasting extensions to the existing - and still-dominant - IPv4 in the form of RFC 966, "Host Groups: A Multicast Extension to the Internet Protocol," written with his thesis advisor at Stanford, David R. Cheriton. Deering alone authored RFCs 988 and 1054, which progressively revised and extended his original proposal.
In 1988, while finishing his doctorate, Deering presented a segment of a tutorial session on routing at that year's ACM SIGCOMM symposium. His presentation -- a 10-page paper entitled "Multicast Routing in Internetworks and Extended LANs" excited considerable interest and resulted in his being hired to continue his work on IP multicast at Xerox's PARC after he obtained his PhD in 1989.
In 1992, in partnership with the redoubtable Van Jacobson, Deering went on to develop the so-called mBone -- the IP Multicast Backbone. The mBone served as a scientific testbed for a wide variety of multicast-based applications, including global videoconferencing, software distribution, distance learning and one-to-many spectacles, such as live streaming video of space shuttle missions and rock concerts.
Also in 1992, the Internet community began to come to grips with what appeared to be an impending apocalypse -- the fast-approaching exhaustion of assignable IP address space. Deering became interested in crafting a definitive solution to the problem, so, while others concentrated on interim measures, such as Classless InterDomain Routing and Network Address Translation to extend the useful life of the existing IP protocol, he waded into the treacherous swamp of competing architectural proposals for a successor protocol -- IP: the Next Generation.
The Internet Engineering Task Force -- under whose auspices the issues were thrashed out -- took its standard tack of promulgating work groups to tackle various aspects of the problem. As might have been expected, each group saw a different aspect of the IPng elephant as the defining characteristic of the beast.
Some were primarily concerned with increasing the size of the address space, without greatly altering the existing core protocol in any other major respect. Others felt the current iteration -- version 4 -- had been shown to be woefully inadequate in a series of fundamental ways and advocated for increasing the functionality, robustness and scalability of the protocol's central architecture while expanding the address space. Some insisted it should be a strictly incremental upgrade of IPv4, while others argued for a clean break with the shackles of IPv4's architectural legacy.
It was one of the most contentious issues the IETF ever tackled -- and it fell to Steve Deering, in his role as the chair of the IPng Task Force, to play mediator, moderator, messenger and sympathetic ear as the contending camps wrangled over the issues, refined their respective proposals and returned to do battle again. And again. And again.
It took until the end of 1995 before Deering's seemingly-inexhaustible patience and ability to synthesize the best and most practical aspects of the competing models and the IETF's focus on "rough consensus and running code" as a bellwether of progress finally resulted in a compromise that was acceptable to a large enough number of the various working group members. That December, the proposed new standard -- now officially known as IPv6 -- was published as RFCs 1883 through 1888.
Shortly after striking the grand IPv6 compromise, Deering was hired as a Project Leader -- and later made a Fellow -- by Cisco Systems. His new job neatly distracted him from the cut-and-thrust of the still-contentious ongoing evolution of IPv6 by permitting him to return his focus to his first love, IP multicasting and routing problems -- where it has largely remaning ever since. He also keeps busy as a major presence on the Internet Engineering Steering Group and the Internet Architecture Board -- the two major technical policy-making bodies of the IETF -- and he is in constant demand as a speaker and instructor at computer conferences and symposia of all kinds.
Jun Murai is often called the "Godfather of the Japanese Internet." As grad students in computer science at Keio University in 1983, Murai and collaborators Mario Tokoro and Fumio Teraoka created the first Unix-based computer network on a Japanese campus, the Keio S&Tnet.
The next year -- despite a lack of cooperation from the notoriously-inflexible government telephone monopoly, NTT -- Murai and his collaborators extended that network via UUCP dial-up to create the first experimental, academic, wide-area data network in Japan: the Japan Unix NETwork. By 1985, JUNET was exchanging messages with USENET and Japan began its painstakingly-slow integration into the swiftly-expanding global Internet.
JUNET lasted until 1994, when the Ministry of Posts and Telecommunications finally authorized commercial ISPs to begin operating. In the meantime, Murai had gone on to organize the Widely Integrated Distributed Environment Project "to establish (a) large-scale distributed computing environment" in Japan -- a venture of which he remains the General Chair today.
Construction of the so-called "WIDE Internet" -- which, in the partially-deregulated Japanese telco environment of the late 1980's, could finally make use of high-speed leased lines -- began in 1989. Its role was to provide a private, commercial testbed for Internet technologies in Japan and to advocate for wider connectivity to the Net.
Because of his high profile as Japan's premiere Internet advocate -- as well as his stature as an engineer and scientist -- Murai very quickly rose to prominence in the Internet Engineering Task Force and its political ally, the Internet Society. In 1993 he was elected to a term on the Internet Architecture Board, the IETF's version of the Supreme Court and the final authority on Internet standards.
In 1996, Murai was appointed to the International Ad-Hoc Committee -- a group of Internet insiders and ISOC allies that tried and failed to craft a generally-acceptable expansion of the DNS namespace to supplement the increasingly-crowded .com domain. The following year, he was elected to a term on ISOC's Board of Directors and, in October, 1998, he was one of the first members of the Board of the Internet Corporation for Assigned Names And Numbers (ICANN) -- a post to which he, like his fellow members, was "elected" via a process that fell well short of democracy.
Through it all, the well-fed Murai has remained anonymous to all but the best-informed Internauts -- at leaast, to those who live outside of Japan. In his native country, however, he is a star. In addition to his Internet guru status, Murai is also a well-respected musician, whose friends and collaborators include Ryuuichi Sakamoto, founder of the legendary Yellow Magic Orchestra and undoubtably the most famous Japanese composer in the world.
(Copyright© 2001 by Thom Stark--all rights reserved)