Pervasive computing means change. And change means resistance. Many people are disturbed about change, concerned about its effects. Pervasive computing can only be successful if change is managed in a proper way. People need to understand the value of pervasive computing and how their concerns are treated. Change also means a loss of power to the currently powerful. Change redistributes power; this is something welcomed by the powerless and feared by the powerful.
Part of the change process therefore needs to deal with the concerns of the people involved. Technology can only be successful if people are willing to use it. So, in this section I present a set of concerns that may arise when pervasive computing technologies are introduced.
Treat concerns seriously. One of the biggest problems with e-business on the Internet was that concerns weren't addressed at all. The Internet hype made many believe that everything is possible and that the new technologies will make everything better. This is one of the biggest mistakes you can make. Technology does not do anything better. It just amplifies existing processes. If the process is bad, technology will make it worse.
1.5.1 Strength of Traditional Links
Existing supply chains will not necessarily embrace the idea of pervasive computing. Pervasive computing means that new intermediaries will be able to play a significant role. This means that a next round of disintermediation is about to happen. E-business already created a first round of disintermedia-tion, but pervasive computing will be much more profound because content, for example, is channel independent. Content created for web browsers will be reused for television, radio, and newspapers. A web company can suddenly produce a newspaper or run a television station without too much trouble.
Many television and radio stations will therefore try to prevent others from entering the market. But as long as these current industry barriers are not torn down, these markets will remain closed stovepipe channels that other types of content and service provider will find expensive to access.
Solutions to this problem will break up many traditional industrial barriers. More and more companies will provide services and solutions in a cross-industrial manner while still addressing the personal needs of the single industry.
1.5.2 Privacy and Security
Through their pervasiveness, these new technologies can collect a lot of data about the users and their habits. Most people are concerned that this data will be misused by the companies collecting the information. To make the most out of pervasive computing technologies, information must be collected and shared among partners. Certain information about customers' online behavior, preferences, and so on will have to be shared between supply chains either explicitly or implicitly, to allow customers to navigate seamlessly around their information and services "universe." Without technology that can reassure customers about service providers' adherence to privacy policies, regulators may inhibit the sharing of such information. Many people will also be reluctant to share information with services they do not trust, so a trustworthy relationship must be built up. To ensure the privacy of the users, it may be useful to set up a trustworthy third party that collects all information on behalf of the services users want.
Pervasive computing will increase the number of service providers so sig-nificantly that the control of information flow between users and the service providers becomes impossible. A trusted third party could store personal information and provide the required information to preselected partners.
Security in general is also a sensitive area. On the Internet, people are afraid of giving away their credit card information. In a universal network, their houses and cars may be at risk if information is not properly secured. Application-level security in all networks must become at least as sophisticated as that available through the Internet to adequately address users' concerns. At the same time, these security barriers need to be easy to use. Today's fire-wall technologies are too complex for most people to configure. Future firewall technologies should be as easy to turn on as locking the front door of the house.
Part of the reason that supply chains are currently so "closed" is that media owners are paranoid about unauthorized copying and distribution of their assets. Television channels, for example, buy broadcasting rights for a certain country. A universal network makes it difficult to restrict broadcasting rights to a certain country.
The Olympic games,20 for example, cannot be broadcast live over the In-ternet at least until 2008. The International Olympic Committee (IOC) says that the Internet does not restrict the viewers to a certain region. And a major asset of the IOC is that it sells rights to certain countries and makes money selling the same content to different regions. An additional problem with advanced Internet technologies is that once you have converted the content, in this case the games, you can not redistribute the content via various channels, such as television sets, DVDs, radio, mobile phones, books, web sites, and so forth. Redistribution is not only cheap but, more importantly, very easy.
Another example is the MP3 hype, which makes record labels and some artists crazy, because suddenly they are no longer part of the supply chain and are unable to earn revenues. Well, at least not in the way they used to make money. Most record companies find it unthinkable to redesign their business models. But to survive, they need to.
Media owners, without significant advances in digital rights management technology and changes in their business models, will themselves inhibit their growth into multichannel media distribution.
Several companies work on solutions, but so far these are only point solutions. Some companies, such as DigiMarc, 21 work on solutions for adding digital watermarks to images. A digital watermark can be implemented in two ways. One way is to add invisible information to a picture: the images can retain their original format and can be viewed by appropriate viewing software. A JPEG image will be slightly modified but will remain a JPEG image. The advantage is that no additional software is required by the viewer. The major disadvantage is that the watermark can be removed rather easily in most cases; by resizing, for example. The other way is to create a format that can only be viewed with special software. This method ensures that nobody can remove the watermark, but makes it more difficult to use the content, since people need to have the right software installed on the right platform. In a universal network, this solution is not acceptable.
The Secure Digital Music Initiative (SDMI) 22 is a new format for music. It was created by the music and content industry to ensure that music distributed over the Internet can only be viewed by people who have paid for it. This is a nice idea, but as long as MP3 is out there, nobody will care about SDMI because it adds complexitynobody will go for more complex solutions they have to pay for. This idea already failed once, when the banking industry tried to introduce the SET (Secure Electronic Transactions) standard. It failed because it created a lot of overhead and was more expensive. Here is the rule of thumb: The more people you want to reach, the easier and cheaper the solution has to be. Therefore, the music industry has to accept that MP3 is available and cannot be replaced simply by another format. To make money, the music industry needs to create value-added services.
Another company is working on creating links between users and their region. InfoSplit23 (see Figure 1.5) is using a new technology to establish a relationship between a user and a geographic location. This technology would allow businesses to sell content to geographic regions. But of course, the technology is far from perfect. Right now it is fairly easy to fool this technology by faking IP addresses. But it can be expected that this technology will make advances and become more secure.
These are only point solutions and will not work perfectly in a universal network. A new standard to tag device-independent content needs to be created; instead of creating technology to identify the region of the user, industries need to rethink the whole concept of rights management. Just mapping the existing processes to the new realities is not enough. This is a lesson that many people did not learn from the e-business (r)evolution. Just replicating existing processes and ideas to a new technological platform is not enough. If the platform creates new possibilities, the processes and ideas need to be extended or rethought.
1.5.4 Disregard of Technology Standards
Software standards have not always been a success. Especially in highly public areas, the best proposed standard does not always become the most pervasive. Some technologies have successfully been widely adopted across the corporate information technology (IT) world. IP is a good example of such a standard. However, IP is still far from being universally supported in the telecommunications industry. And there are only a handful of other nonhardware standards that are models of successful standardization. Even if a given standard has been adopted by most participants, a big problem can arise if some companies create their own "flavors" of the standard, making interconnection of different components difficult.
The problem is that when competitive differentiation can be obtained, vendors will always try to circumvent standardization processes. Vendors may pay lip service to standards by partially implementing them; but in many cases they will also implement features that are against the spirit of the standard or that offer "richer" or "advanced" alternatives. Particularly in the hype-fuelled software industry, vendors have historically been keen to do precisely that. Just look at the HTML standard and its very similar versions. Unfortunately they are not identical and therefore are sometimes problematic in implementation.
Unless one infrastructure vendor maneuvers itself into a position of colossal strength, wide adoption of standard technologies will be the only cost-effective route to pervasive computing. History would seem to indicate that such a situation would not be easy to engineer. True pervasive computing will be independent of certain standards. Pervasive computing would mean that different technologies and software standards will be able to connect to each other flaw-lessly. Bridging standards will play an even bigger role than they do today. And it seems that XML will play an important role, too.
1.5.5 Capabilities of Hardware and Battery
Mobile devices often are not connected all the time to a power plug; today we already experience this with laptops, mobile phones, and PDAs. My first mobile phone would last for a day (a Motorola 24 in 1995); my new Nokia25 operates a week without recharging. My first laptop (a good old HP Omnibook 570026 ) allowed me to work an hour without recharging; the new Apple iBook 27 allows me to work up to eight hours without recharging.
Personal mobile devices present the most complex technological challenges to vendors as potential platforms for accessing content, applications, and services. Mobile devices without connection to power or network outlets are becoming ever more important in an always faster-paced world. Therefore, their batteries must become smaller and more powerful.
However, the current state of battery technology is such that "advanced" features like sophisticated user interfaces and audio playback severely impact the mobility of compact devices. Perhaps more seriously, the other problem that hardware manufacturers currently face is that the hardware required to drive a 3G-capable wireless handset is currently too power hungry and inefficient to fit into a production device. Improvements will undoubtedly be made; some innovative technologies are in the pipeline, but the speed with which they come to market will have a profound effect on the suitability of mobile devices as pervasive computing access terminals.
Another big problem is the power-hungry Intel 28 computer chip. Compared to a few years ago Intel has made improvements, but for today's needs they may not be enough. Several companies are working on improvements. Probably the best-known company without a product was Transmeta29 in 1999. It was famous because of one employeeLinus Torwalds, who headed the Linux development. In 2000 it produced its first set of products, a new chip called Crusoe, which emulates all of the major chip designs in software and saves battery power by running very efficiently. Some companies, such as Sony, 30 are shipping laptops with the Crusoe chip; other companies have created stylish new devices for connecting to the Internet. Acer, 31 for example, has created a web pad that has a 50-meter wireless range and up to eight hours of battery life.
In the future we will see further enhancements in battery life and better, more powerful, yet less power-consuming CPUs. They will enable instant In-ternet access anytime and anywhere.