Version classiqueVersion mobile

L’Internet des objets

Pierre-Jean Benghozi
Sylvain Bureau
Françoise Massit-Folléa

The Internet of Things

Chapter 5

Dissemination and implementation

Texte intégral

1The Internet of Things both draws on and generates added complexity to previously developed technologies. It carries the hope of increased rationalization and economic growth, but also raises new questions. Here, through an examination of public perception of the vagaries of technology and markets, we aim to reassert the notion that questions of acceptability are linked principally to the twin concepts of “security” and “liberty” ; and that clear answers are expected and must be given. Reluctance or even resistance on the part of members of civil society reveal real anxieties experienced by individuals and associations. There is a need not only to improve information about and understanding of what the IoT is, but also to testify that guarantees regarding the use of these applications will be sought.

Understanding and managing the risks of the IoT

2The prevailing view is that the evolution of technology is cumulative, and that its power inevitably leads to a headlong pursuit into an environment that is not only anonymous but also uncontrollable. Where companies and public authorities speak of benefits in terms of “growth and security”, public opinion often puts forward the “risks and threats” for individuals.

Managing environmental risks

3Currently, the issue of computer waste has not been satisfactorily resolved, but it will also be necessary to find additional solutions to manage new pollution generated by RFID chips. In addition, extra care will have to be taken in the case of standards relating to the radio-electrical environment due to the increase in the number of transmitters. Similarly, the risk of interaction between RFID chips and their environment will have to be considered when using implantable sensors that may interfere with medical apparatus. These implantable chip technologies will have to prove their harmlessness in long-term usage.

Preventing ethical and legal risks

4First, it will be necessary to ensure that those who refuse to be fitted with a device do not risk becoming victims of discrimination. Then, rules on responsibility may have to be revised to take into account the intermediation of electronic equipment. Lastly, the possibility of having “various identities” in the digital world could entail a dissipation of the holder’s rights ; this will have to be controlled.

Preventing economic risks

5As far as the economy is concerned, some indirect risks are linked to the occurrence of incidents in the activities structured and organized around the IoT, be these errors, accidental physical loss of tags, middleware and software bugs, or faults in terminals and readers. Furthermore, the very high level of interdependence between components for production and distribution will require organizational and security expertise, not only for individuals, but also for the prevention of industrial espionage.

6The advantages expected must therefore be evaluated. Companies assess these according to the cost of the reengineering processes, which they are sometimes reluctant to implement.

7Members of society are not completely unaware of them – the monitoring of Alzheimer patients, the traceability of animals and the tracking of young children are no longer in an experimental phase. However, the general public voices fear rather than impatience, specifically in the protection of personal data.

Maintaining a high level of protection in Europe

8European citizens and businesses benefit from one of the highest levels of protection in the world. In this context, it is essential that the development of the IoT does not adversely affect this, but rather that it becomes a means of improving it. Thus Europe could create new international quality standards on the protection of individuals within the framework of the IoT.

Listening to and informing public opinion on the IoT

9Almost all of the many surveys carried out in the past few years endeavour to assess citizens’ perception of risks related to the implementation of RFID. We present two noteworthy examples here.

The fear of the improper use of data in the IoT

10The first survey, carried out by CapGemini in 2005, reveals that the major cause for concern is the appropriation of personal data by undisclosed users and for undisclosed purposes. Second is the fear of being permanently tracked. Next come health concerns. The comparison between Europe and the United States highlights lower sensitivity to these risks in the latter. This may be accounted for by greater familiarity with the technologies, or a stronger involvement of civil associations. However, this could change very quickly. The survey also compared the “privacy impact” of various related technologies (bank cards, mobile phones, discount cards, access badges, video-surveillance systems, etc.) : where RFID is concerned, it appears that the impact is at least equal to or greater than other technologies. In 2006, the Humboldt University of Berlin carried out a similar study among its better-informed students on the major privacy risks perceived by them. Unauthorized access to information or skimming came first, the traceability of objects through data use came second, followed by the retrieval of information on social networks, technological “paternalism” and the transfer to users of responsibility for objects.

The fear of citizens for their privacy

  • 1 A survey carried out in France in December 2007 by the CREDOC reveals that it was the first cause f (...)

11A threat to individual privacy is the risk most frequently mentioned in opinion polls and public debates1, no doubt because it is the simplest to comprehend, and also because, for the general public, RFID is a more recent technology. As stated above, the perception of risk includes retrieval, processing and use of personal data by governments as well as companies. The nature of perceived risks is very varied : unwarranted use (forbidden or unauthorized use of the given application) of available personal information (on medical cards for instance) ; the general or selective traceability of individuals (via badges in the workplace, in libraries, etc.) or of their actions (handling of associated objects) ; and the aggregation of personal information for police surveillance or the development of consumer profiling. It leads to questioning on several points, such as the legal status of personal data, free consent, functional deactivation and the commoditization of data security.

12These questions are beginning to be addressed. In France, the Forum des Droits sur Internet (FDI), in partnership with the Commission Nationale Informatique et Libertés (CNIL – the French Data Protection Authority), has just launched a public consultation which is open until 15 January 2009. The aim is to “collect the opinions of French people” on topics such as image and identity control on the Web, private data and the State, and the protection of privacy in the future digital world.

Protecting privacy and sensitive data

13Two types of question must be distinguished : the first pertaining to applications for the general public (protection of anonymity and personal data) and the second to industrial applications (economic intelligence, vertical competition between producers and distributors, and the capturing of added value).

Protecting privacy

14The first meaning of the term privacy refers to respect for intimacy, confidentiality, private information and/or respect for the physical and moral integrity of individuals. When translated into French, the concept extends to include the “legal protection” of personal data and concerns freedom of expression and movement. The threats which loom over these aspects are connected to all information and communication technologies (this is already the case for search engines, trade platforms and social networks), but they could increase if the development of RFID is poorly regulated, due to the miniaturization of devices, the interconnection of servers and position determining technology.

Differentiating open and closed loop RFID applications

15A second distinction must be made : between open and closed loop RFID applications, that is, between the systems in which data does not exit (in the supply chain of manufactured goods for example) and those which communicate and exchange personal data with other systems (as in the case of a public transportation firm which draws up user identity profiles and collaborates with chain stores). In both cases, a technical choice must be made between the minimization of data quantity and the maximization of guarantees. This is where the debate currently stands between Privacy Enhancing Technologies and Trust Enhancing Technologies (see below).

Making the “silence of chips” possible

16Various technological solutions are now produced by laboratories, although none is yet used extensively. Below are some examples which involve citizens’ and consumers’ rights : these relate to compulsory or optional deactivation, opt-in configurations, memory block or even scrambling.

17In the United States, the “Faraday Cage passport” was designed following a surge in public opinion calling for a system where the chip could only be read when the passport is open. In distribution, the CASPIAN association (Consumer Against Supermarket Privacy Invasion) demands that tags be deactivated once they are taken out of shops.

18In 2003 RSA security created a blocker tag that prevents the unwarranted reading of a chip. The Open Business Innovation in Denmark developed the “Zero Knowledge” approach, where the bearer can choose to activate the “privacy” mode or not. In addition the Distributed Systems research group of the Swiss Federal Institute of Technology in Zurich created a watchdog tag which gives the bearer of the chip information on the chip reader, its location, etc. on a screen or a mobile phone. This principle is called “user-controlled identity management” : the user has control over his partial identities depending on the context he is in (in a surgery, with a friend, etc.) via a specific code for each partial identity, such as a pseudo name and different password.

19In this type of system, memorization for the user is quite difficult, and other devices resort to greater automation. In fact the VIP (Virtual Identity and Privacy Research Center) in Berne University developed the concept of “virtual persons” in 2003. This is a virtual mask protecting identities, overlain by a larger front covering several people, and giving the possibility for one person to use several masks configured to carry out a variety of actions in his or her name. RFID Guardian is a collaborative firewall project that enables interference with RFID waves via a small battery integrated into telephones and laptops. This tool, which has been available since 2004, targets the general public. About a dozen researchers (Vrije Universiteit, Technical University de Delft, Hogeschool Inholland) and firms (Riscure, Beyond Security) are working on its development.

Protecting sensitive company data

20Beyond “individual privacy”, new issues raised by the development of the IoT concern the identification and components of products and services, logistics data, consumption and movements, and the contextualization of data. However, the normal tension between security and quality of service, on the one hand, and respect for privacy, on the other, is very slight for some applications – in the case of protection of vehicles, luxury items, domestic and pedigree animals, for example.

Technical and other solutions

21Solutions created to protect privacy may be implemented simultaneously with or as a complement to other solutions using technology. However, they may also be designed into the initial architecture and functionalities selected, the procedures for use, and the legislation and the governance of the IoT.

The limits of Privacy Enhancing Technologies

22The protection of privacy, along with security, has become a market for specialist companies.

23PETs, or Privacy Enhancing Technologies, offer a range of technical solutions for individual users ; some of these designs are highly advanced (particularly in the fields of anonymization and authentication).

24However, none is used extensively, due, perhaps, to lack of information or public awareness, or as a result of the high cost of complex chips.

25It may be necessary to envisage a means of integrating PETs when designing IoT services so that they figure in initial costs, and also to avoid the disadvantages of adding solutions at a later stage.

A new market : trust

26The trust market includes not only consumers’ trust towards suppliers, but also the confidence that suppliers feel towards demand, and users’ trust in regulation procedures and institutions.

27These factors may drive innovation towards Privacy by Design or Trust Enhancing Technologies. These would be a means of increasing the acceptability of PETs through a reduction in the asymmetry of information exchanges, that is to say by improving user access to the data that concerns them.

28An additional difficulty arises from the differing definitions of “personal data”. Conceptions of society, identity and privacy are specific to each culture, so that the degree of acceptability for PETs varies from one community to another. However, at the moment we have no international comparisons on these points.

Managing multiple individual identities

29The various uses of the Internet have led to the emergence of atypical behaviour throughout the world. This is exemplified by a surge in the number of electronic addresses, recourse to pseudo-names in blogs and the existence of avatars in virtual worlds, etc. These “multiple identities” are more difficult to capture than a passport number, a social security reference, or bank account number. This also leads to diverse claims, such as the “right to oblivion”, the right to anonymity wherever identification is not essential, the right to scrambling or the right to lie to protect another from danger – all extreme forms of the right to “the silence of chips” which companies must put into production and authorities guarantee.

30Yet security problems and issues of trust must not be overestimated. Let us recall that the same disquiet was felt regarding the Internet at the beginning of the new millennium when e-commerce began to develop. Experience shows that the fears then voiced had less concern for the effective security of payments than for the discovery and the exploration of new patterns of consumption.

  • 2 According to a report published by the think tank “Identités actives” from the Fondation Internet N (...)

31Finally, the problem often originates from the “privacy paradox”, which is : “the fact that, whatever may be the values or fears they voice publicly, some individuals are more than willing to disclose a lot of information on themselves, and all the more so if they believe they will get something in return. However, they are not unaware of potential risks, but they manage these and weigh up the pros and cons. We may be shifting from a legislation centred on protection of individuals to a legislation centred on control by individuals of their personal data, their actions, and their image...”2


1 A survey carried out in France in December 2007 by the CREDOC reveals that it was the first cause for concern for 27% of those polled.

2 According to a report published by the think tank “Identités actives” from the Fondation Internet Nouvelle Génération (Paris).

© Éditions de la Maison des sciences de l’homme, 2009

Conditions d’utilisation :

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search