Imaginary Interfaces

Reference Information:
Imaginary Interfaces by: Sean Gustafson, Daniel Bierwirth and Patrick Baudisch


Authors:
All the authors currently conduct research as the Hasso Plattner Institute in Germany.


Summary:

• Hypothesis: This paper has one quantitative hypothesis. This was that "participants would perform fewer Graffiti recognition errors than reported by Ni and Baudisch.  The main point of this is that the user does not need a screen in order to interact with a device. Therefore we could have imaginary screens that represent the screen to a user which would allow for interaction without any physical device being touched and that we could do this was a relatively low error rate. 

• Methods: They constructed various imaginary interfaces and chose various students and people from off of campus. The three chosen tests were Graffiti, Repeated Drawing and Multi-Stroke Drawing. While the initial test setup was not enough to correctly capture user input gestures, after adjusting various things they were able to analyze their results. The tests were things such as a user needing to draw graffiti or finding points on a grid. Each test lined up very well with the hypothesis.
• Results: Ultimately the team concluded that they were pretty close to correct with their hypotheses. They had about a 5.5% error in recognizing user input. Much lower than other methods. Fingertips were concluded to be the most accurate locations to use. And there is a very large different between rotation and stay as far as conditions go in the experiment.
• Contents: The removal of a screen entirely from the whole computer system creates some unique issues. When you do not have a tactile surface on which to conduct your user input/output many different issues arise. Environmental issues as well as other users could very negatively impact this type of interface. Ultimately this desires to create a technology that is not only useful but also very unintrusive to the user on a daily basis.

Discussion:
This whole article was amazing. The concept of being able to interact with a device without the physical touching of it is something that has been immortalized in science fiction in various forms for years. While this is still seemingly in its early developmental stages it can have some great uses in the future. One of the faults i believe is that at this time this is a far less practical system that the current use of surfaces to interact. While this may even evolve to become cheaper and more practical than what we use currently, will it evolve at a rate at which it can outpace surface interaction becoming even cheaper and convenient? It seems like it is a race of sorts. We also have a loss of that tactile feedback that users seem to love so much. Without the avaliablity to alert the user of something via a vibration or something of that sort we would have to result to various visual cues. Far less effective than something tactile. Ultimately this is going to come down to getting a device that can consistently receive and correctly interpret user input, as well as being cheap enough to distribute to the masses should this be applied to some kind of mobile device.