There was a paper at the CHI2008 conference in Florence Italy that addressed several aspects of adaptive UIs including predictability and accuracy. The reference is:
Gajos, K. Z., Everitt, K., Tan, D. S., Czerwinski, M., and Weld, D. S. 2008. Predictability and accuracy in adaptive user interfaces. In Proceeding of the Twenty-Sixth Annual SIGCHI Conference on Human Factors in Computing Systems (Florence, Italy, April 05 - 10, 2008). CHI '08. ACM, New York, NY, 1271-1274. The article focuses on adaptive toolbars and not on content, but it does discuss some of the issues around predictability and accuracy of the adaptive user interface. Here is a portion of the abstract that summarizes some of the results. "We present a study that examines the relative effects of predictability and accuracy on the usability of adaptive UIs. Our results show that increasing predictability and accuracy led to strongly improved satisfaction. Increasing accuracy also resulted in improved performance and higher utilization of the adaptive interface. Contrary to our expectations, improvement in accuracy had a stronger effect on performance, utilization and some satisfaction ratings than the improvement in predictability." There was one other paper from Florence that deals with adaptive interfaces for small screen devices. Findlater, L. and McGrenere, J. (2008) Impact of Screen Size on Performance, Awareness, and User Satisfaction With Adaptive Graphical User Interfaces. Proc. CHI'08, ACM Press. The ACM Digital Library is filled with research and theorie about adpative user interfaces. One of the issues that comes out of the adaptive interface literature is the amount of data a system gathers before presenting the user with a change to content or the user interface. Clippy, the infamous style of user assistance in Windows a decade or more ago, was a system that was supposed to monitor user input and based on usage patterns (and a Bayesian algorithm is memory serves me well) present the user with tips or hints on how to do better. The problem with clippy was that it's algorithm threshold was not conservative enough -- it should have waited longer and gathered more data before presenting it's mostly useless suggestions. There was an article in The Economist that explained why Clippy was undone by a poorly tuned Bayesian algorithm. Chauncey On Mon, Jun 9, 2008 at 12:40 PM, Jeff Howard <[EMAIL PROTECTED]> wrote: > Has anyone actually used this design? My guess is that it's > impossible to accurately speculate about how good or bad this is > without trying it and without being tainted by pre-knowledge of its > adaptive behavior. It all depends on the execution. > > There are plenty of poor examples of adaptive UIs, but there are > great examples too. > > Quicksilver is an adaptive example that I love. It's guessing and > learning all the time. And right out of the box it's more than a > little dumb. But over time it has learned what I tend to search for > and serves it up practically the moment my fingers hit the keyboard. > > Another example that's closer to the MIT example was the BBC > redesign described a few years ago. > > http://www.liamdelahunty.com/blog/media/theglasswall.pdf > > Instead of shifting the location of content, it highlighted > particular paths through the content, based on past behavior so that > frequently clicked areas grew more prominent over time. Like a > well-trampled path across a lawn. > > // jeff > > > . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . > Posted from the new ixda.org > http://www.ixda.org/discuss?post=30025 > > > ________________________________________________________________ > Welcome to the Interaction Design Association (IxDA)! > To post to this list ....... [EMAIL PROTECTED] > Unsubscribe ................ http://www.ixda.org/unsubscribe > List Guidelines ............ http://www.ixda.org/guidelines > List Help .................. http://www.ixda.org/help > ________________________________________________________________ Welcome to the Interaction Design Association (IxDA)! To post to this list ....... [EMAIL PROTECTED] Unsubscribe ................ http://www.ixda.org/unsubscribe List Guidelines ............ http://www.ixda.org/guidelines List Help .................. http://www.ixda.org/help
