Designing the User Interface, Ben Schneiderman & Catherine Plaisant
May 1st, 2005
Shneiderman, Ben., and Catherine Plaisant. Designing the User Interface, Fourth Edition: Strategies for Effective Human Computer Interaction, Boston: Pearson Addison Wesley, 2005
When I first picked up this book, I had the feeling I was onto a good thing. It’s a brand new 2005 fourth edition (over 19 years), it’s heavy with a well-packed 650 pages, all printed in full colour, and flicking through the book, it is obvious the authors have applied the subject matter of the book to it’s design and layout.
I have to admit at this stage that I have not tried to read the entire book, that said I have read several sections and it does appear to deliver everything my initial anticipation suggested. The book is well structured and aimed at practitioners, researchers and students in the field. It covers a broad range of subject matter, from basic theories and guidelines, through design processes, interface evaluations, detailed discussions and examples of various interaction styles, and specific design issues. Each section provides substantial further reading guides, and solid summaries of the content covered.
Overall this book is a combination text and working reference. It is comprehensive, well balanced and up to date. Every student, researcher or worker involved in the production and development of software or hardware that interacts with users should have a copy and be familiar with this book.
Chapter 1. Usability of Interactive Systems
This chapter is a good introduction, and introduces a range of goals and measures in the field. As someone who is working on becoming more involved in the practice of user interface design in a professional role, the section entitled Goals for Our Profession (p. 39) was of most interest. Shneiderman suggests three goals:
- Influencing academic and industrial researchers –
Shneiderman recommends a combination of scientific and observational ethnographic research to build the knowledge and understanding in the field. He suggests the following areas as starting points: reduced anxiety and fear of computer usage; graceful evolution (from novice to expert); specification and implementation of interaction; direct manipulation; input devices; online help; and information exploration. - Producing tools, techniques, and knowledge for commercial developers –
Improving the tools and knowledge of commercial developers would go a long way to improving the overall usability of the systems they produce. To me it seems obvious that if a developers system and environment is well designed and easy to work with, then the software they produce has a better chance of demonstrating the same traits. As well as this developers have other concerns than specific usability, and easily applied tools, specifications and guidelines is going to make it easier for them to focus on the task at hand. - Raising the computer consciousness of the general public –
Shneiderman’s point here is not to make the public more aware of the existence of computers, but to shift their view of computers as scary and unpredictable into anger at the developers and designers of such machines. I think we are starting to see the beginnings of this. People are becoming aware of better ways to do things, and of better systems and are rewarding the developers of such items by buying them. I think the Firefox web browser is a good example of this, it’s usage has grown from 2% to almost 30% in the six months it has been released, and almost all of this has come from Internet Explorer users. Apple’s dominance of the digital music market through the iPod and iTunes, and the growing “halo” effect from this has turned into growing market share in the Macintosh line.
Chapter 3. Managing Design Processes
One of the most important section of this chapter relates to Organisational Design. It highlights that in order to be most successful usability design and engineering needs to be part of an organisation, and designed into an organisations thinking and processes.
Design in in herently creative and unpredictable. Interactive system designers must blend a thorough knowledge of technical feasibility with a mystical esthetic sense of what attracts users. (p. 113)
Shneiderman puts forward Three Pillars of Design that “can help user-interface architects to turn good ideas in successful systems.” (p. 114) –
- Guidelines documents and processes
- User Interface Software Tools
- Expert Reviews & Usability Testing
In the section of Development Methodologies, Shneiderman focus’s on Cognetics Corporation’s Logical User-Centred Interactive Design Methodology (LUCID). This methodology identified six stages: Envision; Discovery; Design Foundation; Design Detail; Build; and, Release. (p. 119) “The thoroughness of the LUCID framework comes from its validation and refinement in multiple projects.” (p. 122)
Another interesting topic raised in this chapter is the social and legal ramifications of developing or redeveloping products. “A social impact statement, similar to an environmental impact statement, might help promote high-quality systems in government-related applications. … Early and widespread discussion can uncover concerns and enable stake-holders to state their positions openly.” (p. 130) I guess my work with Kazaa, and the many current digital copyright issues have made me very aware of legal and ethical issues, but not so much the broader social implications. I guess this is especially important in government systems, but equally applies to new technologies and systems and the widespread effects these technologies might create…
Chapter 11. Quality of Service
A large part of the perceived quality of service of a system is how long it takes for the system to produce a result – the longer it takes the worse the system is perceived to be. An interesting point raised in this section is to do with errors or unexpected results. The longer a system takes to produce a response the more anxious and cautious users become, as the cost of an error or unexpected result is much higher. Also caught up in this is the users “planning ahead” while they are waiting. Long waits with unexpected results are more wasteful because the user has usually done more planning of what they want to do next. They then loose this planning when they have to fix the problem, or try a different method. (p. 460)
Much of users perceived quality of service is to do with expectations and previous experience. Users who have always used a slower system don’t mind it. Users who move from faster to slower systems become frustrated very quickly. This has lead to some networked systems always providing a slower response time during lighter loads in order to prevent users from experiencing the faster responses, and becoming frustrated when the servers are under normal or heavy load. (p. 463)
Productivity increases usually come from short response times, but this is not always the case, as longer responses can slow down users thinking and reduce errors rates, or encourage them to find clever shortcuts. Understanding the task and the complexities involved can assist designers in developing systems that appropriately enhance productivity for the users. (p. 466)
Chapter 12. Balancing Function and Fashion
One of the more interesting sections in this chapter address’s a need for non-anthropomorphic design of user interfaces. The reasons given for this include: it increases users expectations of what the computer is actually capable of, and therefore their disappointment when they realise the computer can’t actually do the task expected; to clarify the difference between people and computers – users and designers must accept the responsibility for the computers errors rather than blaming the machine; reducing the level of anxiety some people feel about computers making them feel inferior or dumb. (p. 485)
Shneiderman presents a range of studies and examples where anthropomorphic designs were used, but provided no eventual benefit to the user, or in many cases reduced productivity. The only exception to this rule appears to be in children’s games and educational software, where children’s expectations and imaginations seem to be more accepting of the characters and their obvious flaws.