Select Committee on Science and Technology Second Report



6.1  This Chapter considers the benefits of connecting computers into communication networks, and the consequences of networking for future applications. As computers become more pervasive and play an ever-increasingly central role in our lives, they will have to become more trustworthy, more reliable and more autonomous.



6.2  Individual computers are powerful and useful machines, but their potential can be greatly expanded by connection to other machines. The benefits of such networking increase with the number of computers that are connected together. The logical conclusion of this is the internet, which offers a single network[49] to connect all computers in the world.

6.3  The internet has its origins in the US in the late 1960s. The first significant attempt to employ computers as communications devices was ARPANET[50], the Advanced Research Projects Agency Network of the US Department of Defense. ARPANET was established as an open computer science research project with no specific defence objectives. Most of the networking protocols that underpin today's internet grew out of this work which engaged many of the best US computer scientists and benefited from debates on protocols among a wide community. It is one of the clearest demonstrations in the computer world of the potential benefits of government-supported open research.

6.4  In the United Kingdom, the early use of the internet was epitomised by JANET (the Joint Academic Network), established in the 1980s and making the UK academic community one of the first to enjoy full internet connectivity. The introduction of the world-wide web[51] in the 1990s has, for many users, made the communication capability of computers as important as their ability to perform calculations.

6.5  Networking connects the user not only to remote information sources, but also to remote computing resources. Increasingly, the vast set of computers connected to the internet — most of which are idle at any time — are being seen as an enormous computing resource looking for useful work to do.

6.6  One of the first projects to exploit this resource was the Search for Extra-Terrestrial Intelligence (SETI). Data from radio telescopes may contain the tell-tale traces of intelligent life elsewhere in the universe, but huge amounts of computing power are required to carry out the analysis of the data. Enthusiasts can register with SETI[52] to make their PCs available, when otherwise idle, to contribute to this communal effort, and one of them could then be involved in what would be one of the most important scientific discoveries of all time — that we are not alone in the universe.

6.7  A similar distributed computing project is folding@home[53] which enables volunteers to make their unused computing resource available for the study of protein folding. Proteins are large molecules that are the workhorse of all biological systems. They are assembled in the cells of animals and plants under genetic control but, before they can carry out their biochemical functions, fold into complex shapes. Protein misfolding is implicated in a number of serious diseases, such as Alzheimer's. However, the folding mechanisms are little understood, and simulations to improve that understanding (with potential for better treatments) require a massive computer resource.

6.8  With the recent introduction of broadband technologies, the majority of households in the developed world will soon have access to the mushrooming global information resource over the internet at communication speeds of at least half a million bits a second[54]. The regulation of this environment is a major challenge for government agencies, since international differences in what constitutes (politically) acceptable content are complicated by a communications network that knows no national boundaries.

Wireless networking

6.9  Increasing computer mobility has led to the demand for networking capabilities to be extended from fixed-location machines (that can be wired into the internet) to mobile systems that are with their user at all times. Wireless networking has only recently advanced to the point where this is becoming practical — "WiFi" networking keeps a lap-top computer on-line as it is carried around an office or airport; "Bluetooth" networking allows a personal digital assistant (PDA) to connect to a mobile phone in a nearby pocket or briefcase and from there, using the digital mobile phone network, to access email via the internet.

6.10  Future demands for "ambient computing" (see paragraph 6.18) and "smart dust" (see paragraph 6.19) will require significant further developments in wireless networking. These technologies assume vast populations of computing systems that are in communication with each other, largely unsupervised. As Commissioner Liikanen noted (Q 518), there are global regulatory issues about the radio wavelengths used for such networking, to which the EU is alert.

The Grid

6.11  A recent DTI initiative has been to encourage the development of "Grid" computing technology in the UK. Through the development of software to facilitate co-ordinated use of existing hardware, Grid technology aims to make computing resources available as though from a utility — any network connection will be able to provide access to powerful computing, database and other resources. This style of computing is seen as important for the development of many scientific and engineering disciplines; see, for example, the evidence from the Central Laboratory of the Research Councils (p 192). The development of Grid technology to support large-scale scientific collaborations is funded under the Research Councils' e-Science programme (see paragraph 7.20).

6.12  Grid computing will make use of internet computing resources (such as those mobilised by the SETI and folding@home projects described above) routinely and generally available on demand. The development of suitable standards will make the integration of remote networked resources more thorough and their functionality more robust, allowing the more sophisticated applications demanded by cutting-edge science to be supported. However, it should be remembered that the world-wide web was first developed to support scientific research. It may not be long before Grid technology advances from serving scientists to providing benefits available to every internet user.

The future of networking

6.13  While computer networking has made dramatic progress over the last decade, there is still considerable room for progress. Optical fibres have capacities that are much greater than those seen by typical end users and, as noted by Intellect (p 200), the major communications bottleneck is the "last mile" — the connection from the exchange to the home or office, or the wireless link from the base station to the handset.

Future applications

6.14  The technological advances outlined in the previous paragraphs offer the potential for a wide range of innovative computing applications. While predicting the products that will succeed in the marketplace in the future is always difficult, it is less risky to give an indication of the sorts of products that will be enabled by foreseeable developments in technology. Many of the innovations that are likely to affect our lives over the next decade or so will grow out of research already planned or under way.

Mobile phones

6.15  3G mobile telephones[55] are, at the time of writing, already entering the marketplace. However, few customers are yet aware of the capabilities these new systems will offer once the technology and services are fully developed.

6.16  A glimpse of the future can be gleaned from the current market push of current generation mobile phone handsets with the capability to take, send and display colour photographs. A few users also exploit the capability of current technology to offer continuous internet connection (through GPRS technology) and thereby access email and other internet services on the move. This can be done almost as readily as from a fixed machine, although data transfer rates are much slower and the costs much higher.

6.17  3G technology will take all this onto a higher plane, bringing the level of service available to the mobile user much closer to that in the office. Whether or not the customers will adopt all this new technology remains to be seen, and depends on several non-technical factors such as the way the products are marketed, the pricing structure, and so on. The stakes in this are very high. All across Europe, the service providers paid governments formidable sums for 3G licences (i.e. access to the allocated radio frequencies). They obviously aim to recover their investment.

"Ambient Computing"

6.18  Looking further into the future, the EU's 6th Framework Programme emphasises "Ambient Computing" as a major theme for the next few years of research funding (Q 517). Whereas mobile phones are used for communication at their users' request, Ambient Computing assumes an environment where computer systems are everywhere and get on with their tasks of sensing, controlling and communicating unaided by humans.

6.19  Taking Ambient Computing to its logical limit, the "Smart Dust" project about which we heard during our visit to Silicon Valley[56] is a vision of computing so cheap, robust and miniature that it can be embedded into anything — even paint and wallpaper — to monitor aspects of the local environment. Wherever it ends up, it finds neighbouring "dust" to communicate with, building an ad hoc network of communicating systems and sensors throughout the environment.

Biomedical systems

6.20  Intelligent toilet bowls are already in development[57]. These perform a number of biological analyses at every opportunity, providing regular feedback to their users on their state of health. This is only the beginning of consumer biomedical electronics. In the future, we may have intelligent sensors that can be swallowed (providing a detailed report on the state of our digestive systems, for example), injected, inserted, perhaps even inhaled.

6.21  The replacement of invasive, labour-intensive examination techniques by simple, off-the-shelf electronic sensors will make screening much more affordable, leading to earlier diagnosis via near-patient testing devices and, in many cases, a much better prognosis. Intelligent drug delivery systems will also give doctors much better control over a treatment regime, again improving the patient's condition and minimising adverse side-effects.

Extended networking

6.22  The examples cited above all illustrate the potential of near-future microelectronics to enhance the capabilities of SoCs — computing, as it were, in the small. There are also major opportunities for computing in the large. The continuing improvements in networking, particularly in the last mile of connection, will continue to enhance personal communications and accelerate progress towards the global village — all those with access to computers, whatever their geographical locations, will be near neighbours.

6.23  Technologies such as the Grid and e-Science will lead to capabilities that could, for example, give schoolchildren easy access to supercomputing and the ability to examine atoms in one laboratory[58] while controlling a radio telescope in another.


6.24  Computers have been a byword for unreliability — "to err is human, but to really foul things up you need a computer." The jibe is unfair. Computers are astonishingly reliable and accurate machines. A PC can faultlessly performs billions of operations a second for days on end. The frustrating problems that users experience are almost always a fault in what can be extremely complicated software. As has also been said, "computers are machines for making very fast and very accurate mistakes."

6.25  Problems with the reliability and usability of computers are being addressed through programmes such as IBM's vision of "Autonomic Computing" (Q 322)[59]. The human autonomic system is made up of a large number of largely independent units, each of which is self-adjusting, self-healing, and responsive to the needs of the other units. Can computers be built in a similar way, so that they look after themselves far better than they do at present?

6.26  Another, and more significant, question is the confidence that people have in computers. We generally accept embedded computer applications in, for example, cars and aeroplanes without thought. People tend to have more concerns when their interface with computers is more obvious, such as credit card transactions over the internet.

6.27  Developments such as the EU's ambient computing initiative described above would lead to many aspects of our lives being managed (or at least filtered) through machines. If such applications are to achieve critical mass, it will be essential for users to have complete trust in the outcomes. We were therefore both interested and pleased to learn that an important strand of the EU's research programme relates to establishing the conditions for necessary trust in the technology (Q 556).

Human-centred computing

6.28  Reaching the physical or financial limits of CMOS technology in 10 to 20 years time is likely to have a major effect on the chip fabrication industry. However, the existing capabilities of the technology — let alone those it will have at maturity — will enable the development of products with capabilities we can barely imagine today. Perhaps genuine machine intelligence will finally become a reality.

6.29  One thing is certain: most of us in the developed world will be surrounded by computers wherever we go. We have already passed the point where the personal computer — that is, one computer for each person — has become a reality. Most of us now have tens of computers in our PCs, mobile phones, PDAs, video recorders, microwave ovens, cars, and so on.

6.30  In the future these tens will become hundreds and then thousands. Instead of each of our computers demanding our personal attention and working in isolation, they will increasingly integrate with each other to form a community of computers working together to support us in our daily lives[60].

49   The internet is actually a series of interconnected networks that now jointly span the globe. Back

50   For a history of ARPANET, see Back

51   Based on the hypertext addressing protocols devised by the British physicist, Tim Berners-Lee, while he was working at CERN. Back

52 Back

53 Back

54   At such speeds, the entire text of the Bible could be transferred in about a minute. Back

55   See Appendix 10. Back

56   See Appendix 6. Back

57 Back

58   In case this sounds too far-fetched, during our visit to IBM's Almaden Research Centre in Silicon Valley (see Appendix 6), we were privileged to operate equipment that enabled the user, sitting at a PC, to move individual atoms using an atomic force microscope in an adjacent room. The researcher told us he hoped to provide internet access to this facility in the near future. Back

59   See Back

60   During our visit to Silicon Valley (see Appendix 6), Intel's researchers aptly described this vision as "human-centred computing". Back

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2002