PUCK UI 2

Posted by J. on Thursday, April 23, 2009

Media and Literature Review
1.Brief
Its ultimate goal is to bring the reader up to date with current knowledge on a topic and to form a basis for informing subsequent research. The aim of your review is to provide the platform for studio based research within your research cluster.
A media and literature review is a body of text that aims to review current knowledge on a particular topic or area of practice.
Write a media and literature review of between 1500 and 2000 words.
Should include between 8 and 20 relevant sources.
Use appropriate academic language.
Double space lines.
Use APA referencing throughout.
Include a full Bibliography.
2.Review Synopsis
Usability and interface design in the context of multi-touch technology.

Overview
1.Introduction
i.Approach
ii.Essay Goals
2.The Research
i.Multi-Touch
ii.Tangible Interface Technologies
iii.Visual Psychology
3.The How
i.FTIR Touch Table
ii.Mixed-Reality
iii.Software
4.The Why
i.Parameters for Good Interface Design
ii.Exploring possible applications of multi-touch design
5.Conclusion
i.Identifying Flaws
ii.Future Developments
6.Bibliography
i.Literature Sources
ii.Web-based Sources

Review
9.
10.Introduction
This media and literature review intends to cover a number of related topics with a view to further research and finding solutions to problems relating to human-machine interfaces using multi-touch technology. Initially the review will cover the history behind multi-touch hardware, software and various related aspects within the field. Then I will cover the practical aspects of the multi-touch including related technologies and which approach is best suited to future research. Next I will look at the theory behind user interface design in terms of tangible user interfaces. Finally I intend to look at potential applications. If successful this review will inform the reader of the current state of multi-touch research as well as related technologies and the theory behind interface design.
11.The research
Multi-touch interfaces have actually been around for a while; in 1984 bell labs developed a touch screen that could change images with more than one hand. In 1991 Pierre Wilner published a paper about his “Digital Desk” that accepted multi-touch and pinching motions (Buxton, 2009). In the following 16 or so years not much happened to bring these technologies to the mainstream. There have been many reasons cited for this but they all faded into insignificance when Apple brought out the iPhone. Suddenly here was a phone that the average consumer could use, one that used multi-touch and could be developed (more or less) by an ordinary programmer. Also important in the field of multi touch is Jeff Han, whose development of a simple, inexpensive and scalable technique for multi-touch sensing has shown an international audience what touch interfaces are capable of. Until now these developments have only been available to researchers with big budgets. However the development of a touch interface that can be made using a projector, a webcam and some readily available hardware in conjunction with free software means that the DIY developer community as a whole has access to all the possibilities of a much more expensive commercial solution for potentially very little cost. It is important to remember that touch or even multi-touch solutions are essentially just another replacement for the common mouse pointer. The addition of such things as gestures, tangible interface elements and the larger form factor available to a touch table versus an LCD screen makes the strengths of this type of interface more apparent.
The software GUI interface has been around in many forms since its birth in 1973 with the Xerox Alto (Buxton, 2009). In that time many interface elements have been tried but surprisingly few have made the transition into mainstream computing. These include icon based file browsing, the alphabetical list, buttons, check boxes, scroll bars and drop-down boxes. Much of current GUI design theory is centred not on alternatives to these, but on when and where to use them within the interface’s individual information architecture.
The limitations and failures of touch technologies have been well documented over the past few years. In 1991 many computer companies appeared to have a touch screen computer in development. None of them were successfully released. More recently the iPhone has enjoyed success as a commercial multi-touch product, but criticism of its lack of tactile feedback is one that the technology has not yet been overcome. Tactile feedback is important because it helps with spatial memory; it is one of the reasons why touch typists do not need to look at their keyboards. Other related technologies include mixed reality or tangible user interfaces (TUI’s). These take the inherent flexibility of a screen based presentation format and include much needed tactile feedback solutions. Some of these solutions even feed back into the display to produce a virtual object whose position is extrapolated from printed markers. This is known as augmented reality.
Of course all these solutions are only part of a human-machine interaction. The design of a new interface should therefore be based on principles that combine the natural active vision and tactile processing systems of the human mind in conjunction with the most effective and intuitive TUI interface possible. In Colin Ware’s Visual Thinking Design (2008), various elements of design are linked with the cognitive processes that make them effective or otherwise. Without making too much of a leap from current solutions (applying the ‘principle of least astonishment’ (Hix, 1993)) a successful application of some of these myriad principles should result in an interface that is attractive, fun and efficient to use. The spatial cognition improvements demonstrated by using physical markers in a tangible user interface (Kim & Maher, 2007) provides a major step forward in multi-touch interface technology.
12.The How
For the purposes of this project, a Frustrated Total Internal Refraction (FTIR) solution (Han, 2005) appears to be the most practical solution so far. This requires a projector, and a semi transparent material for rear projection of the graphical interface. The multi-touch design requires a sheet of Perspex, some infra-red LEDs, and a high resolution webcam with some processing software. If successful this design can create a reasonably sized and flexible solution to begin TUI development.
The TUI hardware can be connected to the interface using a range of options, from a Bluetooth connection to a printed reactable barcode on the bottom of a static object and a number of simple yet effective solutions for tactile feedback are demonstrated at http://www.youtube.com/watch?v=0h-RhyopUmc. The software which interfaces with the chosen technology will need to be quick to develop for and suited to the highly visual application at hand. In addition, the multi-touch software (Han, 2005) provides co-ordinates, but reactables, gestures and other features will also need to be easily added, preferably without running too much software simultaneously. Ideally, C++ would be the most powerful tool for this but Processing and Flash are almost as capable and have a much lower development time from concept to working example.
13.The Why
Current interface design has not changed significantly since the advent of Windows 95 over 10 years ago. Since then almost every user has become accustomed to such ubiquitous things as menus, scroll bars and the OK/cancel buttons. It would be fair to say that this lack of change is a direct result of refining an approach that works very well. However there are significant disadvantages to attempting to use a touch based solution over an interface designed for a keyboard and mouse. In a study by Otago University ()it was shown that a touch interface is not suitable for any target smaller than 4mm in size. However, given that a touch table has roughly 5-10 times the desktop real estate of a computer monitor it seems feasible that any given UI element can be designed several times larger without making the interface less efficient.
Another difference between a GUI for a monitor and that of a touch table is viewing angle. A table by its nature can be viewed from any angle above its surface, whereas a monitor sits in front of its user. For this reason, rotation is an important factor to consider. A clever solution by The University of Tokyo 2 NTT Communication Science Labs (Kakehi, Hosomi, Naemura & Matsushita, 2006) involved four different projectors and a Fresnel lense that made all viewing angles equally feasible. However a more flexible solution would be to make individual software UI’s dynamically easy to rotate using the TUI and using interface elements that are easy-to-read from the widest possible viewing angle.
For a long time the computer desktop has been seen as a metaphor (for a desk), (Saffer, 2005) yet there are a number of arguments that suggest that this is not the best way to look at interface design. However this has been the status quo because it makes it easier for the majority of people to understand what a desktop actually is for and how it can be used. In the case of a touch table it is obvious that a metaphor for a desk is implicit in its design. Nevertheless, by moving away from the limitations of a desk by using the inherent advantages of a software interface (easy selection, quickly viewing a selection in many different ways (Agrawala & Balakrishnan, 2006)) it becomes apparent that the chronologically ordered piling system of a real desk is not the only way to create a personal information management interface.
Every individual works differently: some remember where their information is visually, while others use a system of careful organisation (Ware, 2008). A good TUI should therefore allow for these different cognitive systems while at the same time being simple enough to encourage greater productivity over traditional systems. For those that like custom interfaces, interface elements should be customisable without simultaneously rendering it less useful (Saffer, 2005).
A very important consideration in using a large space in such a ‘hands on’ way is the ease of reach. A typical human has a maximum reach in any direction of approximately a metre (Toney & Thomas, 2006). Thus any interface elements outside that range is either permanent or useless. For this reason interface elements in this space should display information that is related to the system more than the user actions. Software that runs in the background can also use this space. Inside this region is an area which is possible to reach from a sitting or standing position but which is not necessarily comfortable to access repeatedly. In a study by the University of South Australia (Toney & Thomas, 2006) this area’s suggested use is storage space. Shortcuts and resources can reside in this area. Finally the area closest to the user is known as ‘personal space’. This space can be used for the currently running application(s) and enable the user to use the interface with minimal physical strain.
The human mind is a very capable pattern recognition device. By providing visual cues such as sharpness, colour, orientation, scale, etc. (Ware, 2008) the interface should bring attention to information when it is needed without being irritating or frustrating for the user. In experiments using virtual interfaces with ‘real’ objects in a TUI environment it became clear that spatial cognition is greatly aided by having a physical object to relate to mental processes. Compared to a purely virtual environment creative thinking is greatly improved by such a mixed reality system (Kim & Maher, 2007). For these reasons this interface will be more suited to spatially orientated people for creative applications although it may also help non-spatial people think in a different way.
14.Conclusion
Despite any real or perceived limitations, multi-touch is an attractive technology. This is why there is so much public interest every time a multi-touch technology (such as the iPhone or Microsoft surface) is announced (Atkinson, 2008). Multi-touch and TUI is a powerful solution but it only works at its best when used in the right way. For instance using a touch interface for writing a novel is definitely not going to be a successful exercise, whereas for a group of people to use the same interface to create a mind-map of a project overview is likely to take considerably less time than using email and other current solutions.
Because of the excitement and fast moving nature of current multi-touch developments, it is difficult to find a solution that has not already been discussed. However, by expanding our knowledge of current developments in all aspects of the field, and applying this to a unique task that develops from it could lead to a solution that is both innovative and successful.

Bibliography
Agrawala, A. & Balakrishnan, B. (2006), Keepin’ it real: Pushing the desktop metaphor with physics, piles and the pen. CHI 2006, April 22-27, Montréal, Québec, Canada.
Apted, T., Kay, J. & Quigley, A. (2006), Tabletop Sharing of digital photographs for the elderly. CHI 2006 Proceedings • Collecting and Editing Photos. Montréal, Québec, Canada
Atkinson, P. (2008), A bitter pill to swallow: The rise and fall of the tablet computer. Design Issues. 24:4 Autumn.
Bimber, O., & Raskar, R. (2005), Spatial augmented reality. Wellesley: A K Peters, Ltd.
Buxton, B., (2009). Multi-Touch Systems that I Have Known and Loved. Retrieved March 30, 2009, from http://www.billbuxton.com/multitouchOverview.html.
Ghinea, G., & Y., S. (2006), Digital multimedia perception and design. Hershey: Idea Group Publishing.
Han, J.Y. (2005), Low-cost multi-touch sensing through frustrated total internal reflection. In Proceedings of the 18th Annual ACM Symposium on User Interface Software and Technology
Hix, D. & Hartson, H.R. (1993) Developing user interfaces: Ensuring usability through product and process. NY: Wiley. Chap. 2
Izadi, S., Agarwal, A., Criminisi, A., Winn, J., Balke, A. & Fitzgibbon, A. (2007), C-Slate: A multi-touch and object recognition system for remote collaboration using horizontal surfaces. Proceedings of the Second IEEE International Workshop on Horizontal Interactive Human-Computer Systems (TABLETOP ’07). IEEE Computer Society.
Kakehi, Y., Hosomi, T., Iida, M., Naemura, T. & Matsushita, M. Transparent tabletop interface for multiple users on lumisight table. Proceedings of the First IEEE International Workshop on Horizontal Interactive Human-Computer Systems (TABLETOP ’06).
Kim, M.J. and Maher, M.L. (2007), Collaborative design in a tabletop system employing tangible user interfaces. Proceedings of the 11th International Conference on Computer Supported Cooperative Work in Design
Lee, J., Lee, J., Kim, H. & Kim, J-I. (2007), Believable interaction with a quasi-tangible tabletop interface. Comp. Anim. Virtual Worlds 07; 18: 121–132
Liarokapis, F. (2007) An augmented reality interface for visualizing and interacting with virtual content. Virtual Reality. 11:23–43 .
Patten, J., Ishii, H., Hines, J. & Pangaro, G. (2001), Sensetable: A wireless object tracking platform for tangible user interfaces. Proceedings of CHI 2001, March 31 - April 5, ACM Press.
Pattison, S., (2008), Seeing things. London: SCM Press.
Saffer, Dan. (2005), The role of metaphor in interaction design, Carnegie Mellon University Thesis.
Scott, S. D., Carpendale, M. S. T., & Habelski, S. (2005), Storage bins: mobile storage for collaborative tabletop displays. Applications of Large Displays. IEEE Computer Society. June/July.
Toney, A. & Thomas, B. H. (2006), Considering reach in tangible and table top design. Proceedings of the First IEEE International Workshop on Horizontal Interactive Human-Computer Systems (TABLETOP ’06). IEEE Computer Society.
Ware, C., (2008). Visual thinking for design. San Diego: Morgan Kaufmann.

0 comments:

Subscribe to: Post Comments (Atom)