I study plausible deniability, when we deliberately make our actions ambiguous in such a way so that other people can't know our true motivations for doing something. How do we design software systems to support or deny certain types of plausible deniability? What do people choose to hide and how do people use software to do it?
Plausible deniability is a fascinating topic but not one very conducive to research because if people don't want to reveal their intentions to anyone, they definitely don't want to reveal it to researchers. However, sometimes people can slip up and reveal more than they intend to, especially if their understanding of how software works is at odds with reality.
On Facebook, people can write notes and "tag" certain friends to appear alongside the note. When viewing the notes, the order which people appear on the tagged friends list is the order in which they were added. They shouldn't be, and nobody is aware they are, but thats how the system was programmed. Looking at the order in which people are tagged reveals some fascinating internal motivations. In many cases, 2 or 3 people will be tagged first, then the rest of the list will proceed in alphabetical order of last name. Those 2 or 3 people, they were the ones who the author really want to read the note, but they didn't want those people to know they were the ones the note was intended for. So how do you plausibly deny this? You go back through your friends list and add a whole bunch of other people and now your intention is hidden. Except, in this case it wasn't, the system was not well designed for plausible deniability.
The question I want to ask is, as researchers, how do we go ahead and study plausible deniability? Once you become aware of it, you start to see plausible deniability everywhere. It's such an important part of normal human interaction that we must account for it in software design. But the frustrating thing is that plausible deniability is, well... plausible. They *could* not want to pretend not to have seen your email and claim the spam filter ate it, or maybe the spam filter really did eat it. It's only when plausible deniability goes wrong that you can conclusively proves that it exists. So what is a poor researcher to do?
Subscribe to:
Post Comments (Atom)
1 comment:
This is an interesting question, indeed, - how to measure, study and experiment with phenomena like plausible deniability, ambiguity etc.
I've read this paper of Erin Bradner and Gloria Mark "Why distance matters: effects on cooperation, persuasion and deception" and she is using Paulhus scale for measuring deception. But this tool seems too limited to really capture deception and ambiguity in people's behaviours.
Post a Comment