Why Trust Management?

To answer this, let’s first listen to a story.

A long time ago, I was a young Ph.D. student trying to find a research problem in not one, but two fruitful areas. It was in mixing multi-agent systems (MAS) with user interfaces and observing how people saw things in their everyday lives, like reading menus and allowing self-organization.

I also took to reading others’ Ph.D. theses because it seemed like a good idea—and I think it still is now. I found inspiration in one such thesis, which I thought was excellent. Jeffrey Rosenschein discussed agent collaboration and coordination, if I recall correctly. A throw away comment appeared before my eyes, which I paraphrase here: It is essential that the agents trust each other in such situations. I wondered what that meant, and at least within the field of MAS, nobody had answered that question yet.

So, I tried to.

There are plenty of journals devoted to AI, or computer security and so on, but there was no multi-disciplinary journal dedicated to the science of trust.

I formalized, as best I could at the time, trust from the point of view of the agents. I thought about how it worked for humans and used that to inform the (pseudo-) formal model. The result was something small I could investigate even further into, across a wide spectrum of combined thoughts from different disciplines: computer science, Artificial Intelligence (AI), psychology, philosophy and sociology, to name a few!

It got me that Ph.D., but along the way, I ran into several obstacles: it wasn’t computer science to some people, and it was too technical for others. It also didn’t do a good enough job of describing or differentiating between trust types or uses for others. People saw absolutely no use for (psycho-social) Trust in computational fields.

I’d also like to note that up till this day, trust within the field of security has not managed to answer my questions, even though it was in this area that this issue was first looked into.

What did I learn?

Over the next several years, the Internet as we now know, has flourished.

People saw the money in it, and much of the rest is history—it’s certainly true that, all of a sudden, trust became rather more important in the grand scheme of things once people realized that it can be made profitable in various ways.

Fast forward to now, and we see a myriad of different places it can be made to work, or already is working: ad hoc networking, computer-user interaction, robotics, agents (yes, agents, phew!), eCommerce, sociotechnical systems… The list is, of course, long and exciting.

Moreover, take a look at any security-related conference and you’ll find a session dedicated to trust. Much more interestingly, you’ll find outstanding multi-disciplinary trust conferences like IFIPTM and PST. Exciting times indeed.

The Journal of Trust Management

It’s founded on a problem, similar to that I had faced while working on my Ph.D. There was a lack of a more advanced space for trust-related research in the field. There are plenty of journals devoted to AI, or computer security and so on, but there was no multi-disciplinary journal dedicated to the science of trust.

The Journal of Trust Management was created as an open access journal to fill this gap. It provides a platform for researchers to publish their work which some consider too social, while some regard as too technical. It is a truly multidisciplinary look at how trust affects the world, both computational and human, and how they all interrelate.

It’s still the early days, but we have had some success and continue to grow, and in the next blog I’ll explore some of the papers we have published to give you an idea of what’s out there, whilst also suggesting a few areas we can grow into.