Context - site visit 2

Context Recognition by User Situation Data Analysis (Context), a joint project between the University of Helsinki/Basic Research Unit of HIIT and the Advanced Research Unit of HIIT.

Site visit in Helsinki on 16 June 2004

The project studies the context of humans and computers, how to construct the context and how to use it.

The project has just completed its first prototype of an enhanced phone book for mobile phones. Current phone books show only name and number for each entry, but does not say anything if the intended receiver of the call is available and can be interrupted to accept the call. As a matter of fact, studies show that many (perhaps most) calls fail to reach their intended receiver (the receiver's phone may be switched off, the receiver may be occupied with more important things or talking on the phone to somebody else, etc.). Studies also show that in most cases (70%), people tell where they are when talking on a mobile phone, but seldom do so when they use landline phones.

The enhanced phone book has been designed and implemented overriding the standard Contact Book of a Nokia 6600/7650 phone. In addition to name and number of the intended receivers, it also shows the current locations of the receivers, how long time they have spent in that location, if their phones have been switched to silent and/or vibration mode, and if they have used their phones within the last fifteen minutes. It also shows the previous "important" place of the receivers; an important place is where people spend a considerable part of their time (home, office) compared to places where the only pass by.

The phone book is being tested during the summer of 2004. A family of four persons (mother and three children) each have been given a Nokia 7650 phone. The test is divided into three parts according to the pattern ABA. The family first uses the phone without the enhanced phonebook during three weeks, then the new phone book is added for three weeks, and then again removed for the last period of three weeks of the test. Especially, the test is expected to show the communication failures both without and with the enhanced phone book, and if the usage of the phone is changed when returning to the initial level without the help of the book in phase three. 

The presence data presented above will certainly be useful for the caller. The project also studies if the availability of the receiver could be automatically deduced from the available data. The presence data may also be used to tag other media, e.g., pictures taken by the phone. Metadata such as location and time but also who was present at the time the picture was taken and to whom should the picture be sent can also be added. Instead of tedious manual entering of such data, the mobile phone could add the metadata automatically, metadata that also could be used to enhance searching the user's picture databases.

Current mobile phones have constrained resources both when it comes to memory size and power. Preferably, algorithms should be efficient in both time and space. In addition, they should work with online data, adapting to new data, i.e. data streams, measured by the phone. Stream algorithms may discard redundant data, store only the most frequently encountered data, and compress the data to use available memory efficiently. The researchers expect to make some progress in this area.

The project will also continue to study what context is and what computers could do with it. In another field test, the researchers followed thirty students to study how they behave in different contexts. Four minicams where used to record the students, their faces, their mobile phones and their surroundings. Results show that users in a mobile context are much more exposed to disturbances. The researchers have among other things compared vigilance times, i.e., the average time spent for uninterrupted focusing on the phone. In laboratory conditions, a user has a vigilance time of 12.8 s, while in a busy street the vigilance time is only 3.8 s. Such information should be taken into account when designing user interfaces for mobile devices. 

The researchers are concerned about ethical questions related to the project and will help in organising a workshop on Location System Privacy and Control at the Mobile HCI 04 conference in Glasgow in September. The researchers will also help in organising a workshop on HCI Issues in Proactive Computing at the NordiCHI 2004 conference in Tampere in October.

Some recent publications related to the project are 

• Kankainen, A., Oulasvirta, A., Rantanen, M., Tiitta, S., Kankainen, T., & Mäntylä, M. (2003). Arkipäivän läsnä-älyä. Prosessori-lehden tutkimusnumero.
• Kankainen, A., & Tiitta, S. (2003). Exploring everyday needs of teenagers related to context-aware mobile services. Proceedings of HFT 2003, Berlin, Germany, 19-26.
• Kurvinen, E., & Oulasvirta, A. (2004). Towards socially aware pervasive computing: a turntaking approach. Proceedings of the 2nd International IEEE Conference on Pervasive Computing and Communications, Orlando, Florida, March 2004, pp. 346-351.
• Oulasvirta, A., & Salovaara, A. (2004).  A cognitive meta-analysis of design approaches to interruptions in intelligent environments. Proceedings of the 2004 Conference on Human Factors in Computing Systems, Austria, Vienna, ACM Press, pp. 1155-1158.
• Oulasvirta, A., & Tamminen, S. (2004). Temporal tensions in human-computer interaction. CHI2004 Workshop on Temporal Aspects of Work.
• Oulasvirta, A. (2004). Finding meaningful uses for context-aware technologies: The humanistic research strategy.  Proceedings of the 2004 Conference on Human Factors in Computing Systems, Austria, Vienna, ACM Press, pp. 247-254.
• Oulasvirta, A. (2004) Human-computer interaction in mobile context: a cognitive resources perspective. Licentiate Thesis, Cognitive Science, Department of Psychology, University of Helsinki.
• Oulasvirta, A. (to appear). Developing future technologies that support social practices and community building. Human Technology.
• Rantanen, M., & Oulasvirta, A. (in preparation). Simulation approach to ubiquitous computing.
• Tamminen, S., Oulasvirta, A., Toiskallio, K., & Kankainen, A. (2003). Understanding mobile contexts. Proceedings of Mobile HCI 2003, Udine, Italy, 18-35. A revised version accepted to Personal and Ubiquitous Computing.
• Rantanen, M., Oulasvirta, A., Blom, J., Tiitta, S., and Mäntylä, M. (2004). Group and public messaging in mobile context. Full paper accepted to NordiCHI 2004.
• K. Laasonen, M. Raento, H. Toivonen, Adaptive On-device Location Recognition, Pervasive Computing: Second International Conference, PERVASIVE 2004, LNCS 3001, Springer Verlag (2004), 287-304. (C) Springer-Verlag.

Research Consortium

• University of Helsinki, Department of Computer Science and Helsinki Institute for Information Technology HIIT, Basic Research Unit (BRU): Professor Hannu Toivonen, researchers Kari Laasonen, Mika Raento, Renaud Petit
• Helsinki Institute for Information Technology HIIT, Advanced Research Unit (ARU): Director Martti Mäntylä, researchers Antti Oulasvirta, Matti Rantanen, Sauli Tiitta