Paper Reading #16: Performance Optimizations of Virtual Keyboards for Stroke-Based Text Entry on a Touch-Based Tabletop

Comments
Paola Garza 
Zack Henkel


Reference Information
Title: Performance Optimizations of Virtual Keyboards for Stroke-Based Text Entry on a Touch-Based Tabletop
Author: Jochen Rick
When/Where: UIST 2010

Summary
This paper discussed finding an efficient keyboard solution for table-top touch devices. The solution suggested is shape writing, which is where the user strokes through all of the letters in the word on a virtual keyboard without ever lifting his finger. The problem is that table-top touch devices are used differently than a normal computer, with the user generally standing up, looking down at the surface. Physical keyboards are impractical, and the QWERTY keyboard layout is inefficient, since the hands cannot be rested on the surface to type. Because of this, stroke-based virtual keyboards are a viable alternative.

Various keyboard layouts. Source: Paper

The author discussed different keyboard layouts, and then conducted a user study to examine how people would complete stroke sequences. From this, a mathematical model was created using Fitts's Law. Then, adjustments were made to existing keyboard layouts in order to evaluate the usage of each. This study showed that a layout optimized for one purpose might be very problematic when applied to a different purpose. From the data retrieved by evaluating each keyboard layout, a new and improved layout was designed using a simulated annealing process with hill climbing.

All in all, it was determined that getting people to adopt a new keyboard layout is difficult, since it requires a learning process, whereas most people are already familiar with the QWERTY layout. However, with the introduction of new technologies, such as interactive tabletops, there is an opportunity to introduce a new keyboard layout or even new typing methods to improve performance and have a better chance of being widely accepted.

Discussion
This paper was rather difficult to get through, since it was mostly mathematical evaluations of designs that are difficult to conceptualize if you have not seen them before. However, I believe that there were some very good points raised. For instance, while there are many keyboard layouts that improve upon the QWERTY layout, most people have become accustomed to using it and are therefore unwilling to switch to a new system. However, with new technologies, new methods can be introduced without requiring a large learning curve. Therefore, I like the paper's suggestion that we develop new typing methods for new technologies. I believe this should be kept in mind when designing new devices.