Identity Implies Trust in Distributed Systems – A Novel Approach

Abstract

Distributed software systems comprise a decentralized network topography wherein a collection of autonomous computers communicate with each other by exchanging messages [1]. The internet (Web 2.0) and social networks and immersive virtual worlds (VW) are the focus of our research. Distributed systems embed a variety of applications that seek to disclose, partially or fully obfuscate an individual’s identity ranging from “role play” VW, social networking to various forms of B2C, B2B, on-line auctions (EBay) and ubiquitous B2B e-commerce transaction via electronic payment. Trust and identity management are inherent phenomena of most if not all forms of such distributed web-based systems. This research seeks to investigate how the role of identity impacts upon trust and e-trust in distributed systems.

Trust is seen as a heterogeous phenomenon affecting users in such environments. According to Taddeo [2], e-trust is understood as a relation that holds when the trustor relies on the trustee to perform a given action to a given level of quality. This research will help to comprehend how trust is formed between two strangers from the identity portrayed in distributed systems. Preliminary results show that there is a relationship between an individual’s identity and their trustworthiness with another user. Should users trust each other? In the context of cyber-stalking (June 2008), a middle aged woman was charged for cyber bullying a 13 year old girl who later went on and took her own life [3]. Could this have been avoided? Other types of cyber criminal activity commonly operating within in such environment includes: pedophilia, identity theft, blackmailing and many more. This shows that individuals are abusing web 2.0 social networks, hence raises issues of serious concern in relation to trust and identity misuse.

To fully explicate the relationship between identity and trust, we need to thoroughly understand the nature of identity, self-perceptions and expression in a world increasingly dominated by human-computer-device communication of many kinds. Indeed a world in which a hybrid form of reality and virtual reality is emerging (Google glasses, Second Life avatars, social networks). Identity and identity management can be split into two broad categories, “Self” and “social identity”. Self identity is how one perceives themselves and social identity is seen as others perception of us. Misinterpretation of identity and trustworthiness is seen as a serious concern. Communication between strangers is common even in real life. However, in distributed systems, limited controls are in place. Does having better controls contribute towards the establishment of trust?

By taking all the mentioned risks into consideration, it is important to find a solution. The main purpose of this research is to provide a framework on identity management in Distributed Software Systems and to provide practical guidelines and processes for managing identity and trust based on a philosophical and ethical underpinning.

Experiments were designed to investigate the relationship between identity and trust on social networking websites (Facebook) and VW (such as Second Life). An OpenSim based experiment comprised of participants being interrogated on the portrayed identity in the avatar created. Questions such as, “Are you portraying yourself?” or “Why have you chosen this avatar?” were asked. The results were analyzed and show the privacy concerns in these VW’s. This experiment also shows the % of fabricated identities is higher than replicated identities. Another experiment will be conducted using Facebook profiles to investigate how identity and trust are related as the first experiment was inconclusive in this matter. This experiment should also investigate the main attributes that formulate an individual’s identity and show how those attributes help to form trust.