I am writing from the ‘Information Interaction in Context Symposium‘ in New Brunswick (the one in New Jersey, not the one in Canada), the home of Rutgers University. Usually I would wait until a conference is over and the dust is settled before blogging about an event, but in this case I’ll make an exception. Specifically, I would like to share some of the highlights from the keynote speaker while it’s still fresh in my mind.

This year the keynote speaker is Dr. Tefko Saracevic from the School of Communication & Information at Rutgers University. Dr. Saracevic is a well-known figure in the field of Information Behaviour. He was one of the first scholars in the field to recognize the importance and influence of context (e.g., social, cultural, cognitive, etc.) on user’s information seeking and searching behaviour (see Saracevic’s 1996/97 stratified model of Information Retrieval interaction).

Dr. Saracevic started with a great overview of the field from the theoretical perspective. He then outlined some of the shortcomings within current Human-Computer Interaction theories, specifically how HCI theories are not strong on ‘context’ and criticized that the majority of Information Retrieval (IR) tests rely too heavily on the ‘traditional’ recall/precision model (including TREC). As a solution, he proposed five axioms related to ‘context’ in information interaction. I especially like the first axiom “One cannot not have a context in information interaction” which was inspired by Paul Watzlawick ‘s axiom in the Communication field – “One Cannot Not Communicate”. You can find the other four axioms (which are also very insightful) on Saracevic’s website (in the abtract of the talk or presentation slides).

In sum, I agree with Dr. Saracevic that the understanding of ‘context’ is crucial; especially for designing effective IR systems. But understanding the importance of ‘context’ and actually capturing & being able to make use of the information is two different things. Nowadays it is very easy and cheap to capture different kinds of contextual information such as user’s geographic location (via GPS-enabled phones) or user’s social ‘location’ (via online social networking sites); the harder question is how to take full advantage of this information without actually overwhelming the user with a tsunami of data and/or compromising user’s privacy. Going forward, these are some of the concerns that the IR field will have to address.

[Note: Add the latter to my to-do list once I'm back to Halifax ]

Saracevic, T. (2010). The Notion of Context in "Information Interaction in Context." Inivited keynote at the conference Information Interaction in Context, New Brunswick, NJ.

Saracevic, T. (1997). The stratified model of information retrieval interaction: Extension and applications Proceedings of the American Society for Information Science, 34, 313-327