A Case Study on the Electronic Global Village

Availability and Cost of Technology

As such, it is an intermediate period that is common to major new inventions, when the new technology is mainly applied to reducing the cost or speeding up existing ways of working. This kind of usage usually hides the primary value and utility of new inventions and can be expected to last for about two generations. Think how the automobile was regarded for many years as a horseless carriage. It took over forty years for modern highways, more reliable tires, and roomier and more comfortable cars to revolutionalize transportation and our style of living along with it. Consider the telephone. The telephone was invented in 1876, and, for the next thirty years, it was marketed throughout Europe as a "broadcast" medium. In 1891, E. Bellamy wrote: "You stay at home and send your eyes and ears abroad for you. Wherever the electric connection is carried, it is possible... for the dweller to takes his choice of the public entertainments given that day in every city of the earth." It was not until 1929 that the first telephone appeared in the Oval Office of the President of the United States. (Doherty and Pope, 1986)

The best innovation comes from the marriage of new technology and innovative ideas about public need. (Branscomb, 1979)

The reason why electronic conferencing has become viable has been due to the improvements in the range and cost of the technology upon which they are based. In this section I will be providing examples of this new technology and the manner in which prices have dropped to such an extent that conferencing is now accessible to the masses.

New Technology

Advances in technology have encompassed both the hardware used to provide the service and the software used to create the networking environment. It is interesting to note the types of technology that were emerging around that time and how many of them failed to take off, how many have 'kicked on' and how many are still waiting upon further technical developments.

Table 1. Examples of emerging technology (1979).

Hardware

The two major areas of hardware advance are the improvements in processing speeds and the memory mechanisms. It is part of the technological folk-lore of today how computers have changed in terms of speed, size (both physical and logical), and price. The analogy most commonly drawn to illustrate this improvement goes something like this: if a car were to have improved in price and performance in the same way as a computer then it would be possible to buy a Rolls-Royce for $100 and get 800 miles to the gallon.

The analogy may be trite but it does indicate the fundamental change for an installation that once regarded the hardware as something that must be optimally used and human labour as something that could be thrown around. Today the reverse is true. Each installation now has a mass of computing power to do things with. The cost per MIPS (millions of instructions per second) has dropped dramatically, whereas the cost per unit labour has increased. Figure 3 illustrates the drop in cost per instruction.

At the same time the packing density of memory has also increased owing to the same types of advances responsible for CPU performance increases. VLSI has allowed computer designers to build memory devices that can now carry up to 4 million bits of information on a single chip. Experimental chips carrying 16 million bits are also being developed. During the period under study 65 thousand bits was a common size and the introduction of 256 thousand bit chips was not far away.

Similarly, the increase in price/performance of external storage devices such as DASD has also impacted on the use and availability of computer technology. DASD especially has undergone massive changes in packing densities and read/write mechanisms. The IBM 3380L model that is currently available is capable of storing 7.5 giga-bytes of data on a 14 inch platter. The read/write head is extremely thin and is designed to actually fly over the disk surface. One of the computer industries favourite comparisons is that the precision of this read/write mechanism is the equivalent of a 747 jet flying over the surface of a lake at a height of one-twentieth of an inch!

With the advances of VLSI came the introduction of micro-processors and the personal computer. This advance has meant that a device that can access, store and manipulate data on a computer-mediated conferencing system, the PC, has become so cheap that it is well in the reach of most working people. The ability of the PC to connect to a modem and then to one of the many public networks has given the individual the opportunity to break free of geographical boundaries. Computer servers can now be accessed at an extremely cheap rate. For example, the connect cost to the OTC MIDAS service (an X.25 based service) is $11 per hour and $11 per kilosegment of data (a segment being the amount of data contained within the X.25 packet).

Coupled with the massive improvements in processor and memory advances has been the improvements in communications technology. The Bell Laboratory Record (1979) recorded that the cost per circuit mile for new terrestrial transmission systems as a function of the number of circuits carried (i.e. the bandwidth of the technology) was $200 for paired cables with ten circuits, whereas for wave guide transmission that carry 100,000 or more circuits the cost was under a dollar per mile. We see today that fibre optics is taking this cost down by the same type of factor whilst increasing the bandwidth. Figure 1 shows a comparison between metallic cables and optical fibres.

Figure 1. Comparison of some existing or proposed metallic cable and. This diagram was taken from the IBM Systems Journal: Potential technology implications for computers and telecommunications in the 1980s W. D. Frazer Volume 18 Number 2, 1979 which was originally published in the magazine 'Telecommunications' April, 1977.

possible optic fibre digital systems.

The increases in the performance of existing technology and the introduction of new technology have a two-fold affect: They make existing tasks complete much quicker and open the doors on new applications. Such advances also provide the optimum conditions for encouraging innovation.

Software

This innovation can be recognized through the developments of software that harness the power of the hardware. The development of high-level languages and the improved programmer productivity tools has meant that a whole new world of applications has been opened.

In the case of VMSHARE the system was originally written in EXEC2 (an interpreted procedural language for the VM operating system) and later (for performance reasons) was re-written in PL/I. The availability of these languages and the functions available within them is a result of the improvements in software engineering that have been facilitated by the advances in hardware.

The implementation of communications software has led to the emergence of many networking services that can be accessed by the public. VMSHARE was originally developed at TYMSHARE (a part of the McDonald Douglas group). TYMNET was one of the earliest public data switch services available and was provided by TYMSHARE. In the years between the commencement of TYMSHARE, the commencement of VMSHARE and 1983 there has been an explosion of Packet-switched public data networks (PDN). Figure 2 shows the services available in different countries throughout the world for the period 1971-1983.

Figure 2. Packet-switched PDNs. This diagram was taken from the IBM Systems Journal: Public data networks: Their evolution, interfaces and status R. Halsey, L. E. Hardy, and L. F. Powning. Volume 18 Number 2, 1979.

Costs

Along with the improvements in performance illustrated above there has been an equally important drop in price. Figure 3, Figure 4 and Figure 6 show how great this drop in unit costs was for the period when VMSHARE was taking off.

Figure 3. History of IBM computer cost/performance. for large systems (upper curve) and systems of similar function. This diagram was taken from the IBM Systems Journal: Computing and communications - A perspective of the evolving environment. M. Branscomb Volume 18 Number 2, 1979.

Figure 4. Projected improvement in cost per bit of random access memory. This diagram was taken from the IBM Systems Journal: Computing and communications - A perspective of the evolving environment. M. Branscomb Volume 18 Number 2, 1979 which was originally published in the article 'Microelectronics' by Robert N. Noyce - Scientific American 1977.

These price drops bring the technology into the reach of thousands of ordinary citizens of developed and developing countries. For small to medium businesses the possibilities of subscribing and contributing to an electronic-conferencing system (as well as getting access to networks in general) becomes very cheap. Figure 5 illustrates the fall in communications costs.

Figure 5. Approximate capital cost per circuit. relationship for a range of optical fibre and metallic cable digital systems (including cable plus repeaters, based on routes of 50 to 100km with no multiplexing included. This diagram was taken from the IBM Systems Journal: Potential technology implications for computers and telecommunications in the 1980s W. D. Frazer Volume 18 Number 2, 1979 which was originally published in the magazine 'Telecommunications' April, 1977.

Figure 6. Storage technology costs. This diagram was taken from the IBM Systems Journal: Potential technology implications for computers and telecommunications in the 1980s W. D. Frazer Volume 18 Number 2, 1979 which was originally published in the magazine 'Electronics' October, 1978.

Why Tele-Conferencing?

Scientific advance is dependent on the efficient communication of ideas. The communications system then is is the nervous system of science; the system that receives and transmits stimuli to its various parts. (Cole and Cole, 1971, quoted in Hiltz and Kerr, 1982)

Given the developments described in the preceeding sections it is clear how the cheapness and availability of the technology has allowed people to get involved with computer-mediated conferencing. However, why should people use this medium rather than the telephone or meeting face-to-face?

To provide an answer to this question it is useful to compare each of the media according to standard criteria. In this instance the media will be examined under the following titles:

1. Control - written word versus verbal. The written word allows the individual control over the rate and duration of the interaction. Verbal communication places the control in the hands of the group.

2. Time and Space - does the group need to coincide either temporally or geographically in order for the communication to occur.

3. Speed - speech cannot occur any faster than the average talking speed of the group whereas the written word moves at the average reading speed of the group.

4. System delays - the medium can influence the time it takes a response to an individual's query or comment to be returned. The medium may or may not enforce synchronization of participants.

5. Memory - the medium may or may not provide a mechanism to allow users to modify, append, reorganize or reclassify what has transpired and inform the other members that such changes have occurred.

Table 2. A comparison of communications media.

Thus, computerized conferencing provides a mechanism that facilitates communication that is distance independent and keeping a conversation going over long periods of time. Its price/performance effectiveness makes it an economically sound choice over the other methodologies.

In summary, the technology is now available and accessible. The VM community of SHARE had a need and found the technology that would best satisfy that need.

Design Principles

Hiltz and Turoff (1978) discuss the principles adopted by conference designers in order to ease the transition to the computer based medium. Amongst these principles are:

1. It is not necessary for the users to be computer literate;

2. The user can become productive after only a short period of practice (or in my experience - playing);

3. The system is segmented: i.e. there are advanced features that the user can choose to use as they become more experienced;

4. The system is forgiving - there is almost nothing that the user can do that will either hurt the system or end up costing the user excessive time or money.

Bureaucracy

Hiltz and Turoff (1978) detail the type of administrative service that is required to service one computer-mediated conference. It is interesting how innovation is often (always?) coupled with the creation of new institutions. In the case of the conference I will be studying I don't believe that all the offices suggested by Hiltz et al are inevitable (or more to the point - necessary):

1. Bulletin editor - solicits entries, appoints reviewers, writes/edits contents;

2. Conference moderator - invites participants, sets the agenda;

3. Group coordinator - provide assistance to members, acts as an interface to other groups;

4. Console operator;

5. User consultant - answers questions about how to use the system;

6. System monitor - add and delete members, send general announcements.

You will find when we examine the VMSHARE system that many of these functions are performed by the same people or are not relevant to the conference.