Excellent conference; nothing really surprising, scientifically, but there are a lot of relevant things in search:
- Aula, A., "How Does Search Behavior Change as Search Becomes More Difficult?"
Not exciting, but quantitative work that shows that users will get more experimental and use advanced features when they are available and the search task grows harder. I'm not sure we're supporting this kind of nebulous search, but I'll learn more as I meet more people.
- Bateman, S., "Useful Junk? The Effects of Visual Embellishment on Comprehension and Memorability of Charts"
I'm only including this one because Bateman challenges Tufte's assertion about usability of chart visualizations. Interesting read, makes some good points, probably not going to change how I do things.
- Bohoj, M., "Timeline Collaboration"
This paper describes an excellent visual language for collaborative timeline description/generation. The specifics don't relate to us, but with Neil's work on time, this is something we should consider (generating a journalized view of a filesystem).
- Chetty, M., "Who's Hogging the Bandwidth?: The Consequences of Revealing the Invisible in the Home"
This paper describes in-home bandwidth monitoring, as a way of fairly sharing resources, but there's definitely a privacy issue here, but they only get into it in the end.
- Hecht, Brent, "The Tower of Babel Meets Web 2.0: User-Generated Content and Its Applications in a Multilingual Context"
This is a powerhouse; the authors developed a tool for cross-language concept alignment, allowing Wikipedia to be used as a sort of Rosetta Stone, not at the pure linguistic level, but at a knowledge theory level. Could be applied to cross-language concept alignment, and may be of help in our growing non-English corpus.
- Inglesant, Phillip, Sasse, M. Angela, "The True Cost of Unusable Password Policies: Password Use in the Wild"
Excellent study based upon field work involving two organizations with different password policies. Participants kept passwords long. Lots of qualitative data on complaints and problems, but no quantitative data (although is it necessary). No actual economic model, either. Would love to see both.
- Jensen, C., "The Life and Times of Files and Information: A Study of Desktop Provenance"
An excellent analysis of how information is created, moved, and copied, and who is doing these actions. Some quant., some qual. The focus is on finding related documents, which is of course of interest to us. Some things that could be made into algorithms, but the work focuses on real-time monitoring, not after-the fact. This work extends on TaskTracker, which was run behind the scenes. TaskTracker is one of the key tools in personal informatics.
- Kärkkäinen, Tuula, "I Don’t Mind Being Logged, but Want to Remain in Control: A Field Study of Mobile Activity and Context Logging"
Good analysis of the privacy implications of long-term logging of personal information (SMS, phone, GPS, photos). The study showed that the privacy concerns of even aware individuals tended to be subsumed by enjoyment of the technology.
- Lawrance, J., "Reactive Information Foraging for Evolving Goals"
This presents an excellent mathematical/algorithmic model to enable search on a moving target (in this case, a code base). Also, a longitudinal study of search behavior.
- Leskovec, Jure, "Signed Networks in Social Media"
Signed networks are networks where the edges (arcs) have positive or negative valence. The authors use positive/negative votes in ePinions, Slashdot, and Wikipedia for data. The authors are interested in what predicts positive/negative links, but what's really interesting is that this is the first paper I've seen use the signed network approach. The first few pages introduce the concept of signed networks; it's a good and fast read.
- Li, I., "A Stage-Based Model of Personal Informatics Systems"
Describes a five-stage recursive model of personal information generation, revision, and review. Worth a read; I think it's going to become an important paper in the field of personal informatics. Of course, we're dealing with other people's data, and after the fact, but along with a journal view, there's interesting ideas to be found.
One of the co-authors, Rob Reeder, is an excellent person in HCI-SEC. This work does a side-by-side comparison of policies for comprehension. It's probably primarily of interest to me and Simson. The conclusion is that users want a lot of context for information when it is presented, but they want it presented compactly. Can't have your cake and eat it to...
- Mancini, Clare, "Contravision: Exploring Users' Reactions to Futuristic Technology"
Used different techniques to expose people to new (proposed technology), and found that multiple forms of presentation are effective for different situations. What is interesting is that the proposed scenarios get into privacy issues, but that didn't seem to be a big concern. For example, using a mobile phone to alert the user to, say, give a diet alert. This is presented in a series of scenarios to viewers, and it elicits some comments about concerns. Ultimately, you have to dig to find value (w/r/t our work and HCISEC in general), but it's there.
- Rantanen, Matti J., "Indexicality of Language and the Art of Creating Treasures"
This is a fun one, but the meat is that the generation of lexical cues is based upon context and language play, such as puns. Could be useful if it could be turned into a search algorithm. I need to think about this more. The paper is ethnomethodological and observational; it's not CS, it's sociology. I don't have a problem with that.
- Sheng, Steve, "Who Falls for Phish? A Demographic Analysis of Phishing Susceptibility and Effectiveness of Interventions"
Set up an email-reading scenario involving phishing and spam. Used Mechanical Turk to recruit participants. Demographics are unsurprising, as 18-25 age range is known to be risk-prone, but the fact that gender had a significant (but not strong) correlation with bad behavior was surprising (to me); it was women who were more susceptible.
- Stutzman, F., "Friends Only: Examining a Privacy-Enhancing Behavior in Facebook"
Looks for reasons that people make their information on Facebook friends-only. Gender and network size are obvious, but "weak tie [privacy] expectancy violations" surprised me; it denotes strategic thinking in privacy that I didn't expect. Also, makes reference to a book I really need to read: Petronio, S. Boundaries of Privacy: Dialectics of Disclosure. SUNY, Albany, NY, 2002
- Villamarín-Salomón, Ricardo M., "Using Reinforcement to Strengthen Users' Secure Behaviors"
These guys used operant conditioning (really!) to reinforce good security behaviors. The experimental reward was a cash payment based upon successfully following security prompts. Very intrusive, probably impractical, but at least someone has done it.