History of the Internet Part 4

 The Establishment of a Broad Community 

The Internet is as much a collection of communities as it is a collection of technology, and its success is mainly due to both meeting fundamental community requirements and effectively leveraging the community to drive the infrastructure forward. This sense of community has a long history, dating back to the early ARPANET. The early ARPANET researchers collaborated as a close-knit group to complete the earlier-mentioned demonstrations of packet switching technology.

Similarly, the Packet Satellite, Packet Radio, and several other DARPA computer science research programs were multi-contractor collaborative activities that heavily relied on whatever available mechanisms were available to coordinate their efforts, beginning with electronic mail and eventually adding file sharing, remote access, and World Wide Web capabilities. Starting with the ARPANET Network Working Group, each of these initiatives created a working group.

Because of ARPANET's unique function as an infrastructure supporting multiple research initiatives, the Network Working Group developed into the Internet Working Group when the Internet began to evolve.

In the late 1970s, recognizing that the growth of the Internet was accompanied by an increase in the size of the interested research community and thus an increased need for coordination mechanisms, Vint Cerf, then manager of the Internet Program at DARPA, formed several coordination bodies – an International Cooperation Board (ICB), chaired by Peter Kirstein of UCL, to coordinate activities with other organizations. The ICCB was formed as an invited organization to help Cerf manage the increasing Internet activities.

When Barry Leiner took over the administration of DARPA's Internet research program in 1983, he and Clark saw that the Internet's continued expansion necessitated a redesign of the coordinating mechanisms. The ICCB was abolished, and in its stead, a system of Task Forces was established, each focusing on a certain area of technology (e.g. routers, end-to-end protocols, etc.). The heads of the Task Forces created the Internet Activities Board (IAB).

The fact that the Task Force chairmen were the same persons who were members of the previous ICCB was, of course, coincidental, and Dave Clark continued to serve as chair. Phill Gross became chair of a reinvigorated Internet Engineering Task Force (IETF), which was simply one of the IAB Task Forces at the time. As previously stated, by 1985, there had been a huge increase in the more practical/engineering aspect of the Internet. As a result of this expansion, attendance at IETF meetings skyrocketed, and Gross was forced to build a substructure for the IETF in the form of working groups.

This development was accompanied by a significant increase in the community. DARPA was no longer the only important participant in Internet financing. Aside from NSFNet and the other US and international government-funded programs, there was growing interest in the business sector. In 1985, both Kahn and Leiner departed DARPA, and Internet activities at DARPA decreased significantly. As a result, the IAB was left without a major sponsor and was more called upon to lead.

The expansion proceeded, resulting in even more substructures inside the IAB and IETF. Working Groups were merged into Areas by the IETF, and Area Directors were appointed. The Area Directors created an Internet Engineering Steering Group (IESG). Recognizing the IETF's growing prominence, the IAB revised the standards process to expressly acknowledge the IESG as the primary review body for standards. The IAB also reorganized the remainder of the Task Forces (other than the IETF) into an Internet Research Task Force (IRTF) led by Postel, with the former task forces renamed research groups.

With the expansion of the business, sector came greater worry about the standards process itself. Beginning in the early 1980s and continuing to the present day, the Internet expanded beyond its largely scientific beginnings to encompass a large user population as well as growing commercial activity. A greater emphasis was placed on making the process open and fair. This, along with a recognized need for community support of the Internet, resulted in the establishment of the Internet Society in 1991, under the auspices of Kahn's Corporation for National Research Initiatives (CNRI) and Cerf's leadership at the time.

Another restructuring occurred in 1992. The Internet Activities Board was reorganized and renamed the Internet Architecture Board in 1992, and it now operates under the aegis of the Internet Society. The new IAB and IESG established a more “peer” relationship, with the IETF and IESG taking on greater responsibility for standard approval. Finally, a cooperative and mutually beneficial connection was created between the IAB, IETF, and Internet Society, with the Internet Society taking on the objective of providing service and other measures that would aid the IETF's work.

The recent development and extensive deployment of the World Wide Web have resulted in the formation of a new community, as many persons working on the WWW did not consider themselves to be mainly network researchers and developers. The World Wide Web Consortium was established as a new coordinating body (W3C). Initially directed by Tim Berners-Lee (the creator of the WWW) and Al Vezza from MIT's Laboratory for Computer Science, the W3C has taken on the duty of creating the many protocols and standards connected with the Web.

Thus, we have witnessed a continuous evolution of organizational structures meant to support and assist an ever-increasing community working cooperatively on Internet challenges throughout the past two decades of Internet activity.

The Commercialization of Technology

Commercialization of the Internet entailed not only the creation of competing, private network services, but also the creation of commercial goods based on Internet technology. Because they found consumers for that method to networking, hundreds of companies included TCP/IP into their devices in the early 1980s. Unfortunately, they lacked genuine knowledge about how the technology was meant to function as well as how the customers intended to use this method to networking. Many regarded it as an inconvenient add-on that had to be pasted onto their proprietary networking solutions: SNA, DECNet, Netware, and NetBios. The DoD enforced the use of TCP/IP in many of its acquisitions but provided little assistance to vendors in developing viable TCP/IP products.

Recognizing the dearth of available knowledge and proper training, Dan Lynch, in collaboration with the IAB, organized a three-day session for ALL suppliers to come to learn about how TCP/IP operated and what it still couldn't do effectively. The presenters were primarily from the DARPA research community, who had both designed and used these protocols in their daily work. About 250 vendor representatives attended to hear 50 innovators and experimenters. The results surprised both parties: the vendors were surprised to realize that the inventors were so open about how things worked (and what still didn't function), and the inventors were happy to listen to new difficulties they hadn't anticipated but were being found in the field by the vendors. As a result, a two-way conversation began that has lasted for almost a decade.

After two years of conferences, tutorials, design meetings, and workshops, a special event was arranged in which vendors whose products ran TCP/IP well enough to come together in one room for three days to demonstrate how well they all worked together and also operated across the Internet were invited. The inaugural Interop trade exhibition was held in September of 1988. The list was limited to 50 firms. 5,000 engineers from prospective client companies came to see if everything worked as it was supposed to. Yes, it did. Why? Because the vendors worked incredibly hard to guarantee that everyone's goods interacted with everyone else's - including those of their competitors. Since then, the Interop trade show has grown tremendously, and it is now hosted in 7 cities around the world each year to an audience of over 250,000 people who come to understand which products interact seamlessly with one other, hear about the latest innovations, and debate the latest technology.

Parallel to the commercialization efforts highlighted by Interop events, companies began to attend IETF meetings, which were conducted three or four times a year to explore new proposals for additions to the TCP/IP protocol suite. These events, which began with a few hundred people, primarily from academics and paid for by the government, now regularly surpass a thousand attendees, mostly from the vendor community and paid for by the attendees themselves. This self-selected group works together to improve the TCP/IP suite. Its usefulness stems from the fact that it includes all stakeholders: researchers, end-users, and suppliers.

Network management exemplifies the interaction between the scientific and business communities. The emphasis at the start of the Internet was on developing and implementing protocols that enabled interoperability.

As the network expanded in size, it became obvious that the ad hoc processes employed to administer it would not scale. The manual table setup was replaced by distributed automated methods, and improved tools for fault isolation were developed. In 1987, it became obvious that a protocol was required to allow network parts, such as routers, to be remotely operated in a consistent manner. Several protocols were suggested for this purpose, including Simple Network Management Protocol or SNMP (intended for simplicity, as the name implies, and evolved from a previous proposal called SGMP), HEMS (a more complicated design from the research community), and CMIP (from the OSI community). A series of meetings resulted in the decision to remove HEMS as a candidate for standardization in order to assist resolve the dispute, but work on both SNMP and CMIP would continue, with the notion that SNMP might be a more immediate answer and CMIP a longer-term strategy. The market might select the one it deemed more appropriate. SNMP is now nearly commonly used for network administration.

In recent years, we have witnessed a new era of commercialization. Initially, commercial activities consisted mostly of vendors supplying basic networking devices and service providers providing connection and rudimentary Internet services. The Internet has virtually become a "commodity" service, with much of the recent focus on the use of this global information infrastructure to support other commercial services. There are products available to help with the delivery of such information, and many of the most recent technological advances have been focused at delivering increasingly sophisticated information services on top of the fundamental Internet data exchanges.

History of the Internet End Series