Leonardo’s Laptop, Ben Shneiderman
April 25th, 2005
Shneiderman, Ben. Leonardo’s Laptop: Human Needs and the New Computing Technologies, Cambridge: MIT Press, 2002
“The old computing is about what computers could do; the new computing is about what users can do.” (p. 2) So opens Ben Shneiderman’s book on the future possibilities for human computer interaction. Leonardo’s Laptop is a cast into the not to distant future to inspire both users and developers to improve the functionality of our computer systems and bring them more into line with what users are actually seeking to do in their lives.
When I started reading this book I was looking for a much stronger level of practice from the book, especially as it is loosely based around the works and practices of Leonardo da Vinci. However I found the book much more philosophical and inspirational in it’s overall tone. Where the book does venture into practice I didn’t find the leaps and bounds I expected, but did find solid well considered examples of how computers could (and probably already should) operate with us in our every day lives. As the first book I read in starting this unit I think this was actually a very good thing, as it refreshed my existing knowledge of the area, and opened up my mind to different ways of thinking and different perspectives without getting bogged down in practical theory and whether or not I thought the theory could be applied directly today.
Chapter 1. Inspiration for the New Computing – Sample Chapter
Shneiderman starts the chapter by highlighting scenarios that are technically possible today, but are not achievable due to the split between the few who can use the technology successfully and the many who can not, and even more so the very few who could afford the technology and the very many who can not. He suggests that the, “…Renaissance integration of disciplines…” exemplified by Leonardo da Vinci could assist us in repairing the split and envisioning the next generation of technologies. (p. 3)
A key element I picked from this chapter is Shneiderman’s assertion that there are two transformations that need to occur in the shift from the old computing to the new computing. The first is a shift in what users value about their computing experience, from gigabytes and megahertz, to how many emails they sent, how many discussions they contributed too, how much work they got done. (p. 12) This seems obvious, but is really important – any important revolution or transformation comes from the people who need it.
Shneiderman’s second transformation to new computing is from machine-centred automation, to user-centred tools and services. (p. 13) In other words technologies won’t aim to replicate or replace human tasks, but to assist humans (users) to do their tasks better, faster and more accurately. As an example Shneiderman points to the decline in research into medical diagnosis programs, but to a rise in medical assistance tools such as rapid access lab result analysis and medical documentation.
One of the strong focus’s of the book raised in the first chapter “…is to support creativity in many domains…”, and that the new computing “…can help make more people more creative more of the time.” (p. 17) I believe this is an incredibly important goal for not only technological, but also broader social change. Solving problems can be difficult and frustrating, but the success in solving them provides enormous satisfaction. Enabling broader creativity in our society provides more opportunities for growth, but also a greater possibility of personal satisfaction and self worth.
Chapter 2. Unusable at any bandwidth
The chapter starts by highlighting the raised public awareness of automobile safety, and the dangers of insecticide and pesticides during the 1960’s. These dangers were well known by certain professions, but their was an unwillingness to speak out against these products and their dangers. These same life threatening dangers can be present in poorly designed interfaces for computer systems. As an example Shneiderman highlights the Therac-25, and computer controlled radiation treatment device for cancer patients. Despite serious radiation injuries to several patients and a death, no one believed that the computer controlled system could be at fault. It took more than a few deaths before the systems were shut down and an investigation held. “Again, no one could believe the Therac-25 had design flaws; especially as the software logs of the equipment usage did not reveal the problem.” (p. 23). After more deadly overdoses the machines were finally shutdown and an extensive investigation carried out, recommending that:
- Documentation should not be an afterthought.
- Software quality assurance practices and standards should be established.
- Designs should be kept simple.
- Ways to get information about errors should be designed into the software from the begining.
- The software should be subject to extensive testing.
This is an extreme example, however the recommendations outlined seem obvious, and should have been in-place and carried out even in the 1980’s when this example took place. Even so much of our computing today is still riddled with errors and design flaws – illegible error messages, lack of documentation and support, crashes and unexpected/unexplained behaviour, and a lack of responsibility for these things from the developers.
This chapter points out that no matter how fast our computers and connections get, these issues will remain as they are designed into the systems, and it is not until the users of these flawed systems start to see these flaws as wrong (instead of part of using a computer) and start demanding improvements that the developers will see these issues as a priority and improve their design and design practices.
Chapter 3. The quest for universal usability
This chapter discusses the need for universal access to the new computing technologies. This covers divides often caused by economic ability, physical and mental ability, levels of knowledge and experience and the general diversity of users. Shneiderman advocates an approach in which designing to accommodate a wider range of users does not “dumb-down” or restrict the technology, but in which the design solutions benefit all users of the new computing technology.
Chapter 4. New Methods, New Goals
This chapter advocates “New design methods combined with a fresh set of guidelines and goals could accelerate movement to the new computing.” (p. 52)
“The first method to support user-centred design is a user needs assessment that determines the range of services needed by users.” (p. 53) This mostly involves observation of existing users going about a task, as well as user interviews, and application of ethnographic and anthropological theories and methods, that help the designer understand the nature and frequency of the tasks users are trying to achieve.
Shneiderman’s second method is usability testing, “…the key to rapid design evolution.” He describes the “heart” of usability testing as the tabulation of the problems or errors that users encountered during the testing, and the subsequent recommendations for improvement that follow from these. Shneiderman sites a report by Clare-Marie Karat (1994) which states that money spent on usability testing returned a 100 to 1 payoff in reduced costs over the life of a product. (p. 55)
The third method of user-centred design is customer feedback which happens after the product has been rolled out or distributed. The internet and new technologies have made direct feedback to developers much easier, but feedback from support and customer assistance professionals is very important here as well.
Shneiderman goes on to say that these are not new and have developed over the last 20 years, however they have still not been widely adopted. He sites Donald Norman book, The Invisible Computer (1998), in recommending that usability professionals get involved earlier in the process, and make recommendations that are sensitive to the development pressures that or facing a product. He also recommends that developers become more aware of human interface considerations, and how good human needs design practices can lead to much better products.
I think it’s important to note here that Shneiderman devotes a section to the discussion of AI machines, and the early perceptions of computers replacing human tasks. Shneiderman states that this is a decaying area of research, and that psychologically most humans want to be in control of their computers and decisions, not the other way around. Shneiderman argues that the sooner we let go of these 50’s, 70’s and 80’s fantasies the sooner we will have, “comprehensible products based on consistent, predictable, and controllable designs that genuinely serve the users” (p. 64)
At this point Shneiderman provides some high-level goals for user-interfaces rather than detailed guidelines: “comprehensible interfaces … generate feelings of mastery, satisfaction with accomplishment, and a sense of responsibility. … interfaces will be comprehensible because they are consistent, predictable and controllable.” (p. 65) Shneiderman restates his goal of universal usability “…in which everyone can be a successful computer user”, stating that it, “makes good business sense because it creates larger audiences…” (p. 67)
Chapter 5. Understanding human activities and relationships – Sample Chapter
Shneiderman draws on a variety of public minded and psychological thinkers as well as his own research to develop his four circles of relationships: starting at the middle with self, growing outward to family and friends, to colleagues and neighbors, to finally citizens and markets. Each of the circles is characterised by it’s size, level of interdependence, shared knowledge and trust. (p. 80-82) He then goes on to define four stages of human activities: collect (information); relate (communication); create (innovation); donate (dissemination). (p 83-86) These cover the broad spectrum of the types of activities we (can) carry out every day.
Shneiderman then takes these two and creates an activities and relationships table or ART, a “4 x 4 table shows what activities you can accomplish with members of each circle of relationship using one of the ” This “framework to accelerate technological innovation” is a tool to help users solve some of their problems in fresh ways, or to help developers spot new opportunities. (p. 87) Shneiderman demonstrates the table’s uses through examples of personal digital photography libraries and mobile devices such as the Palm range.
This is a very big step back from other HCI readings in that it steps back beyond the task or activity you may be looking at, and asks you to reconsider it in the broadest terms of human interaction, without looking at the computer or technology as it stands. As stated by Shneiderman this is tool that may help developers see new or different applications for technology, or for changing perspectives on a task or activity. I found this (and the following chapters) pushed my thinking greatly, but probably beyond the context or intention I had for this unit of study.
Chapters 6-9: The New Education: e-learning; The New Business: e-business; The New Medicine: e-healthcare; The New Politics: e-government
These chapters apply Shneiderman’s new computing and activities and relationships table to the four industries above. While they are valid and good reading, as previously stated they go beyond the intention I have for this unit, and really could not be easily summarised here.
Chapter 10. Mega-Creativity
Creativity can solve almost any problem. The creative act, the defeat of habit by originality, overcomes everything.
– George Lois
This chapter looks at how software tools may be developed and used to enable more people to be more creative more of the time. The term mega-creativity refers to the idea that millions of people could benefit from such software tools, (p. 208) rather than the increased creativity of an individual. After considering various approaches to creative processes (inspirationalists, structuralists, and situationalists), differing levels of creativity (everyday, evolutionary, and revolutionary), and his own framework (see ART tables above), Shneiderman proposes eight tasks that should help people be more creative more of the time: searching; visualising; consulting; thinking; exploring; composing; reviewing; and, disseminating. He suggests that these be used a checklist when looking for tools that can assist in creativity, or if you are interested in developing new ones. (p. 219)
I almost grouped this chapter in with the previous ones, but it had a strong resonance with me at this point in time. The concepts of creativity, design and art, and how they can be developed and enhanced (and even what they are) is a recurring theme across the board for me. Understanding more about my own creative processes and activities will hopefully help me to continue to be creative and to be more creative in the future, in particular in the areas of human computer interaction.
Chapter 11. Grander Goals – Sample Chapter
Shneiderman starts this last chapter with some recognition of the value of old computing, and recognises that continued growth in the area can bring positive change: “in 1990, it required twelve hours to produce a solution on a personal computer, but by 2000, with improved hardware and software, it required only 6 seconds.” Faster computers, better compression, higher bandwidth will continue to be valuable. (p. 234) However, he then goes on to a damning critique of technologists who continue with old computing thinking and focus on the mimicry and replacement of human tasks.
Shneiderman concludes this book with a brief look at a modern Leonardo (Leonardo II) and what this person might be like and how they might work. Of most interest to me, are the two principals to which Shneiderman asserts Leonardo II would be committed:
- Technical excellence must be in harmony with user needs.
- Great works of art and science are for everyone.