Trust, it is sometimes said, is almost more important to the economy and society than money. Unlike money, trust is not a neutral unit of value and comparison, but rather a relationship quality. When you give someone your trust, you grant them a certain degree of latitude to act and in return expect certain benefits. In this respect, trust facilitates communication and the acquisition of information.
Where trust is regularly disappointed or even abused, the law generally comes into play. That is why, says Stefan Bechtold, who researches at the intersection of legal and behavioural sciences, societies have legal institutions in place: ‘When people encounter each other anonymously in society and the markets, they have to be able to trust what the other person says,’ explains the ETH Professor of Intellectual Property. ‘The law therefore has the mechanism of the contract, for example.’ Car buyers can sue if they do not get what they paid for: this is guaranteed by the state.
As a phenomenon, trust is so multifaceted that it not only applies to personal relationships between people, but also virtual ones on the internet, both on a private and business level. Because geographical distance plays only a minor role on the internet, we now have more relationships with people, companies and institutions from other countries and continents. ‘The theme of ’internet and trust? is important,’ says Bechtold, ‘since on the internet we increasingly have to deal with people we don’t know personally, but with whom we exchange data and information, and even enter into contracts.’
Technical question with a human touch
Bechtold is interested in the question of whether users trust the assurances of people with whom they are in contact online - and what they need to consider the internet as a whole trustworthy and to be able to surf safely. Typical questions of trust emerge on the internet when it comes to data traffic: after all, the content of a data package transmitted over the internet should arrive in an unaltered state. Encryption technologies are a feature offered by computer science that can generate trust.
Such technologies enable people to exchange messages via the internet without their being read or changed by third parties. With a digital signature, recipients can trust that the data they receive is exactly as it was sent by the sender. To prevent signed messages from being manipulated, there are hierarchies of keys known as public-key infrastructures. ‘Ultimately, users have to trust the top-level authority in such infrastructures,’ says Bechtold, adding that a distributed system in which we trust various authorities is also conceivable.
With this question, the legal academic meets the computer scientist: as a Professor of Network Security, Adrian Perrig investigates how we can make the internet safer. His work looks at a new, alternative internet architecture called SCION. Bechtold wonders to what extent such architecture can make the internet more trustworthy and what steps are necessary beyond the actual network architecture. ‘Even in ’future internet architectures’, trust is the key to efficient data communication,’ says Bechtold. ‘This applies not only to the hardware and software, but also to the institutions, companies and people active on the internet.’
In order to examine how a network architecture can increase or compromise trust on the internet, the researchers must consider both technical factors (speed, efficiency, reliability and security) and social aspects (surveillance, anonymity, privacy, freedom of expression and competition). For this reason, Bechtold maintains an interdisciplinary exchange with computer scientists such as Perrig and Timothy Roscoe, a professor at ETH’s Institute for Pervasive Computing. Bechtold and Perrig have also published a joint paper on responsibility in future internet architectures.
The societal side of the internet
Their collaboration also encompasses teaching and knowledge transfer: on 9 May 2017, Bechtold, Perrig, Roscoe and other experts will take part in a panel discussion on the ‘Internet and Trust’. The event is the first in a series of talks based on the ‘Science in Perspective’ (SiP) course programme launched by ETH’s Department of Humanities, Social and Political Sciences (D-GESS) in autumn 2016, in which ETH students examine the normative, historical and cultural perspectives of the natural and engineering sciences. The SiP Talks complement the course and are organised by D-GESS in collaboration with lecturers from other ETH departments and guests (see box).
Through his exchange with the computer scientists, Bechtold gains a more concrete understanding of when mistrust on the internet calls for a legal remedy and when a technical solution is more appropriate. ‘If there is a technical mechanism that can create social trust on the societal level more cheaply and simply than a legal contract, why shouldn‘t we try it’’
Science in Perspective Talks
The SiP Talks address current topics in which science and technology have a social dimension. Examples include internet security, monetisation of higher education and the criticism of scientific experts by politicians. They are directed at students, scholars and staff of ETH Zurich, as well as interested members of the public.
Internet and Trust - SiP Talk #1
Tuesday, 9 May 2017, 5.30 p.m. - 7.00 p.m., refreshments to follow
ETH Zurich, Main Building, Semper Aula (HG G60)
Welcome and opening remarks
Michael Hampe, head of the Department of Humanities, Social and Political Sciences
Members of the panel discussion
- Stefan Bechtold, Professor of Intellectual Property
- Jovan Kurbalija, Head of the Geneva Internet Platform
- Adrian Perrig, Professor in the Network Security Group
- Timothy Roscoe, Professor at the Institute for Pervasive Computing
- Brian Trammell, Researcher in the Networked Systems Group and member of the Internet Architecture Board
The discussion and event will be held in English.
Bechtold S., Perrig A. ‘Law and Technology: Accountability in Future Internet Architectures.’ Communications of the ACM, September 2014, Vol. 57, No. 9, doi: 10.1145/2644146