Jim was having breakfast recently with a network vendor who kept making a big deal out of the fact that they support standards. At one level, the vendor's argument seemed to be empty of meaning as it is not possible to find a vendor that does not support standards. At another level, the vendor's diatribe raised some interesting questions in Jim's mind such as what is a standard? What does it mean to support a standard? At what point do standards become useful? Is the standards process too lengthy? Too political? Was more coffee available? In the next few newsletters we will delve into all of these questions, except the one about more coffee.

Today when we think of standards there is a tendency to think of standards committees such as the ITU (International Telecommunications Union), which was known prior to 1992 as the International Telephone and Telegraph Consultative Committee or CCITT, the IEEE (Institute of Electrical and Electronics Engineers), the W3C (Worldwide Web Consortium) or the IETF (Internet Engineering Task Force). However, long before we had standards committees we had standards. One of the best examples of that is the Bell System. 

Throughout the last century what we referred to as the telephone network worked well. Of course for most of the century there was no concern about interoperability. For example, up until a 1968 Federal Communications Commission standard referred to as the Carterfone decision it was not possible to attach a device to the network that was not produced by Western Electric – the manufacturing arm of the Bell System. The Carterfone decision (13 F.C.C.2d 420) created the possibility of selling devices that could connect to the phone system using a protective coupler, and opened the market to customer owned equipment. Later the Bell System began to deploy central office switches from companies such as Nortel. There were not, however, lengthy standards meetings. The Bell System published the standards and everybody that wanted to connect to the network followed along.

Another example of a standard that came to life without standards committees was IBM’s Systems Network Architecture (SNA), which was a complete protocol stack for interconnecting computers and their resources. IBM made the specifications for SNA available to others, which spawned an era of plug compatible mainframe computer vendors including Amdahl, Fujitsu and Hitachi. SNA was often referred to as a de facto standard. In part that meant that IBM controlled the evolution of SNA but that SNA was the networking architecture used for communicating with mainframe computers whether or not they were manufactured by IBM.

Steve Taylor is president of Distributed Networking Associates and publisher/editor-in-chief of Webtorials. Jim Metzler is vice president of Ashton, Metzler & Associates.