Monday, November 29, 2004


Limits of offshore user experience research

Offshore usability is a hot topic. Aaron Marcus has a editorial on the subject in the current User Experience magazine, and there are increasing numbers of articles profiling the growing capabilities of India and China.

Is usability, like programming, "easy" to outsource to foreign countries? Only at the most basic level, examining key-stroke level usability. As soon as one factors in words, colors, and pictures, culture starts to become a factor. One needs a local presence to understand issues. In my experience with international testing, some things turn out as uniform results across countries, other things vary by country. But often one doesn't know which will be which before hand. Even when you can guess there are differences, you aren't expert to say what's appropriate. I've shown the US auto site to British users who are puzzled by the phrase "preowned."

On the Core77 site, Steve Portigal highlights an interesting interview with Samsung's design chief in Business Week. Samsung is increasing user centered research and opening more design offices around the world to understand local markets, and local tastes. Samsung is a company that is moving away from 1980s style globalism to more current localism. It is a rough process. Another profile of Samsung design, in the Tokyo Axis magazine, notes how Samsung had to scale back its global design outreach following the Asian financial crisis in the late 1990s. In a profile of Samsung Design Europe in London, the German magazine form (Jan 2004) notes the cultural gap between the hip Clerkenwell studio and corporate HQ in Seoul. One can imagine some accountants at Samsung seeing the overseas design centers as a source red ink, not profits.

Samsung's European studio has done interesting user research on kitchen environments, and has shaken up design thinking at Samsung. Consumers are positive to the new offerings. Samsung products are seen as more innovative, user friendly, and cool than in the past.

If user research is seen as a cost, instead of as a source of value, some businesses will be tempted to cut corners by doing research at home for foreign markets, or outsourcing usability to countries seen as less costly. But it is a false economy. You can't know your customers unless you talk to them. Talking to some foreigners you want to pretend are like your customers is not true user research. The fact that Samsung, a firm historically known for competing on cost, has decided to localize user research to meet the needs of specific markets, is a sign that outsourcing is not the way forward.

Saturday, November 27, 2004


Reader's Digest books: a conflicted user experience

For some years I have been a fan of certain glossy reference books published by Reader's Digest. Covering topics ranging from workshop practice to cooking, the books show excellent information design: clear layout and great line drawings. When I tell my friends how good these books are, I often get a puzzled look of disbelief (What? Reader's Digest?), or a snicker, as though I admitted to liking the decor of dentists' offices. It's about the most uncool brand around.

I find these books in used bookshops or as remainders, and their content rarely dates. I was excited to see an advert in my Sunday newspaper that Reader's Digest has just published a driver's atlas to New Zealand, which looked like another example of great information design. As the book is not available in stores, I filled out a postcard and sent it off. Yesterday my atlas arrived, and it indeed has a very thoughtful layout. I was feeling very happy until I opened the envelop on the box, which had my bill. Pages of leaflets and computer printouts spilled out. I have coupons to pay for my book in installments, coupons to register for prizes, stickers to put on coupons to register for prizes, order forms to buy more books, stickers to put on order forms, and confusing numbers referring to my customer account or membership.

I'm not sure this postal relationship with RD is going to work for me. How can one company do information design so well and so poorly at the same time?

Friday, November 26, 2004


Demise of slide projectors

The Washington Post reports today that Eastman Kodak has stopped making slide projectors. With the rise of digital cameras, it is small wonder Kodak wasn't selling many projectors. But slide projectors offered features that aren't easily matched by digital media, and we are losing something interesting in the race to digital.

Slides excel at allowing socialability. Although slides were the butt of jokes about being trapped into sitting through someone else's vacation slides, slides do allow everyone in group to see images together. Sure, one can watch one's digital photos on TV, even a very wide screen TV costs tens of thousands of yen, but it still isn't as big or crisp as a slide. One can project digital photos on a wall with a digital projector, but few ordinary people own those due to the cost. And really bright digital projectors cost a small fortune.

The other endangered dimension is the multiprojector slide presentation. I've seen museum and corporate slide presentations using dozens of projectors at once at blow away any digital multimedia I've seen. One can build massive image mosaics from many slides, incredibly rich in detail, or juxtapose different images together to create emotive and provocative material. Anyone curious about how sophisticated slide presentation can be should consult Robert Simpson's Effective Audio-Visual (Focal Press, 3rd ed).



I came across a reference to the new X3D standard, which has recently gotten ISO endorsement. I'm having trouble finding any plain English description of it. It would seem to be a 3D equivalent of SVG, an XML-based description of 3D content that extends VRML. What's possibly exciting, if I'm understanding this correctly, is that it could pave the way for more immersive applications on the web in ways VRML or proprietary formats (Shockwave) did not. Because the format is interoperable, things like CAD files can be accessed and manipulated by people other than the authors of the file, and possibly reused for other purposes. It could be the equivalent of digital music sampling.

Sunday, November 21, 2004


Fake filters

I'm getting concerned about the emergence of "fake filters" on sites that appear to be user enhancements but actually hurt the user experience.

Users increasingly want to see content only relating to their needs, and no one else's. So web developers increasingly offer users different options to filter content on their sites: things like profile questionnaires and product selectors. So far so good. But this information has some valuable market intelligence for the site owner. They can see not only what page was visited, but how users classify themselves, and their needs. That information can be useful to improving the content for users, to see the users' priorities. The information can be valuable for other marketing reasons, and if not tied to user content, can degrade usability.

A simple example. I was looking this week at the web site of a US telecom. It asks the user to first indicate what size firm it is. The user might reasonably expect to see information tailored to companies its size in return for the extra effort of answering the question. But the question turns out to be bogus. Regardless what the user clicks, he is taken to the same content. Perhaps it was just a stupid gimmick to make the user feel special. Or perhaps the telecom wanted to find out the number of employees of firms visiting the site. Either way, the user gets nothing for his effort.

Friday, November 19, 2004


Sadly, design won't change the world

There is much I love about what might be called the "design mentality": its focus on doing, on seeking opportunity, on improving things, on trying things out. The design dimension is overlooked in many contexts, and efforts to expand the scope of issues addressed by design is most welcome.

But just because design can improve things does not mean it will improve things. Sometimes designers take their "master of the universe" role too far. I'm reminded of a story told by Ralph Caplan. A designer told him, "Once you get their confidence, clients ask for advice on anything -- like, can you recommend a good dentist? You find yourself saying, 'Well I'm a pretty good dentist. Why go outside?'"

One design organization that seems to offer advice on anything is the UK Design Council. It's a great organization, one of the most active and organized champions of design going. But I'm constantly perplexed how they want to venture into social issues by claiming design is the solution.

For example, there is the "Design Against Crime" initiative. "How do you beat crime? More policemen and longer prison sentences say some. But good design can actually prevent crime" say the Design Council. In recent weeks the Design Council released its "Touching the State" report which looks at "how design can increase our sense of citizenship" and promises to reverse plummeting election turnouts.

Ouch. Next there will be a report on how good design will solve conflict in the Middle East.

Let's not forget the wise council of architect William Pena: "To put it positively, a social problem calls for a social solution. After there is a social solution then it can be part of a design problem for which there will be a design solution. You cannot solve a social problem with an architectural solution."

Wednesday, November 17, 2004


Expectation shifts

I read reviews of books (and less helpfully, of music) on sites like Amazon. I can't help but notice how a generally high scored item will get a low score from someone who will write something like "I read all the rave reviews and was eagerly anticipating enjoying this enormously. Instead, I found it was a piece of crap. What are these other people smoking?"

Less commonly, I find a review like this: "At first I hated this, but as I forced myself to read/listen to it more, I found I convinced myself I actually liked it."

I find these sentiments wordy, and repetitive. For those familiar with the notion of design patterns, I wonder if we could identity common patterns in reviews, and give them short snappy names, so we can scan reviews more quickly? Or, maybe it's been done already?

Tuesday, November 16, 2004


Give me a keyboard

I've been looking at a look of small devices lately and have been lamenting the absence of proper keyboards. I love touch screens, but find handwriting recognition wanting, and have no optimism about any easy-to-learn and efficient system emerging that relies entirely on a stylus for entry. I want a smallish device, quick to turn on, easy to add substantial notes on. Psion was about the only firm that got the mix right, and they exited the consumer market sometime ago. I checked their site to see what I'm missing as a non-enterprise computer user: Psion Teklogix - NETBOOK PRO Hand-Held Computer.

Sunday, November 14, 2004


The superstitious user

Several years ago I bought an Apple computer that acted as though it was possessed. I would type letters on the keyboard, and the screen would show gibberish characters that looked like the curses one finds in comic strips. What's more, the computer had another strange behavior: it would turn itself on in the middle of the night, and alternatively turn itself off whilst I was in the middle of using it. It would have been funny were it not for the attitude of Apple. The helpful people at the help desk told me that since the computer was not dead on arrival -- my problem was intermittent -- I couldn't return my new, crazy computer to them. Perhaps they were afraid of tangling with "unexplained" phenomena.

I don't know what was wrong with this computer. My uneducated guess is that it was something to do with a short in the circuitry. If I bought into the work of Robert Jahn at the Princeton University's Princeton Engineering Anomalies Research (PEAR) for the Scientific Study of Consciousness-Related Physical Phenomena, I might have questioned if my bad attitude had something to do with rogue behavior of the Apple. PEAR notes that "the emotions of human operators may intensify their interactions with the controlling devices and processes" and adds, "protection against such consciousness-related interference could become essential to the design and operation of many future information acquisition and processing systems." In other words, we need to exorcise our machinery against the bad vibes of malcontent users.

I'm not a superstitious person, but I have to work at it. The human mind has a an amazing ability to try to make sense of random noise. The classic case is astrology, where amazingly elaborate systems are constructed to explain the most mundane random phenomenon. Psychologist Stuart Vyse has noted that superstitious thinking manifests in many information technology applications. Bad usability of one sort or another is often at the core of this superstition.

Even though computers seem like the are, and should be, predictable, in fact they are often very mysterious and erratic. To the everyday user, computers are a black box. (A black box is a secret. In the Second World War, black boxes were secret bomb targeting instrumentation on aircraft, sealed in a black box so that even the crew didn't know how they worked or how to fix them.)

Far from being predictable, computers often seem random in their behavior. Thousands, or even millions, of lines of code can yield unpredictable behavior. When computers act erratic, due to some bug in the code, or some oversight in the use case, the user can adopt unhelpful strategies. They can develop an elaborate theory to explain the behavior of the errant program, much like the astrology enthusiast does to understand why the bus was late today. This makes the program seem even more complicated than it is. Alternatively, the user can fail to recognize the program is acting randomly, and assume they are doing something wrong. This can lead to the unfortunate situation of the user doing something obsessive, like repeatedly entering the same information again and again.

Consider the common case of a failed password. The user suffers the indignity of having his password rejected. He might think immediately: did I type it incorrectly? A second attempt fails. Now the user is seriously questioning his motor skills, or his memory. But maybe the server isn't working. Or maybe the network is faulty. Separating what's under the control of the user is not easy, so it's not easy to understand what is a random circumstance.

Some users will blame the computer for anything that doesn't work out to their liking. Others will blame themselves for the same. Until software gets better at illuminating the black box, user frustration will be with us.

Friday, November 12, 2004


"It seemed like a good idea at the time"

During a business trip this week, I had a chance to read a delightful novel by Magnus Mills, The Scheme for Full Employment. While hardly a story about usability, Mills somehow manages to poke fun at many of our field's sacred cows: slavish adherence to process and efficiency, cultish enthusiasm for "design classics," doing things by the book. A wonderful satire.

Saturday, November 06, 2004


Situation Design and Situational Control

"The future of design lies in situation design and not in product design; products merely implement the situations." So wrote design patron Edgar Kaufmann in 1966. Kaufmann saw situation design, the antecedent of today's experience design, as a path to "intense personalism" to balance the "great impersonality of mass production."

Personalism -- letting the user choose his or her experience -- is what is lacking in discussion of experience design. Though attitudes are slowly changing, one still hears designers claiming to "design" experiences for people, as though users were passive and powerless.

Sometimes users are passive and powerless, but that doesn't make for good design. In a brilliant article,
Roller Coasters vs. Driver's Seats: Design and the Concept of Situational Control, Rashmi Sinha notes the importance that situational control has on users. If users have little control, they can feel burned, or just not get what the designer is intending. Rashmi notes that sometimes designers can work to win (rather than coerce) the attention of users, but other times designers need to work with the attention the user is prepared to offer.

This is an important topic, and I hope Rashmi will revisit it in the future.


Usage-Centered Design

I'm currently skimming Constantine and Lockwood's Software for Use after hearing favorable comment on it from Nigel Bevan, and recently finding a used copy for myself. Constantine and Lockwood advocate an approach called usage-centered design that gives precedence to how a product is used, as opposed to focusing by whom it used (personas) or in what situations it is used (scenarios). The usage approach is much more methodical than Alan Cooper's persona 'n scenario narrative approach, and is stronger and weaker for that. Cooper writes with wit, and the narrative approach animates discussion of design, an attractive attribute when talking with clients. In contrast, Constantine and Lockwood's discussion reads like a Java program.

Though I don't like its techy attitude, the usage centered approach seems far more robust in describing complex interaction than the narrative approach. I'm impressed how the usage approach can transcend personas, allowing one to see common user needs between two disparate users who are in dissimilar contexts.

Usage centered design seems like a useful tool to untangling the spaghetti of complex software, for example, the unfortunate complexity of mobile phone account administration and tariffs.

Friday, November 05, 2004


New Uses for AOL disks

Unwanted AOL registration disks have been a bane for Americans for some time. It turns out that some people think they must be important after all. In a recent Economist article, an AOL executive confesses that a high portion of calls to the AOL help desk are from prospective AOL customers who don't have computers. These would-be customers are inserting the disks in their stereo and hoping to connect to the Internet. AOL considers these people technology laggards, but perhaps they are actually techno pioneers, expecting Internet enabled entertainment appliances.

Another humorous anecdote from Genevieve Bell, Intel's resident anthropologist. Seems connecting to a WiFi in a cafe in Australia is considered pretentious, just not laid back.

Thursday, November 04, 2004


Iterative Design as Creative Act

I get annoyed when I hear Designers (the kind with a capital "D", who are most often graphic designers) criticize usability as uncreative, or worse, as anti-creative. It's an easy smear. It's true some usability practitioners lack imagination (as sadly do some graphic designers). And even many imaginative usability practitioners seem keen to dress-up the scientific respectability of the field. Hence some people call themselves usability engineers, as if people were robots. But most usability practitioners I know like people enormously, and have great empathy for their day to day lives that is filled with the detritus of designed artifacts. The same can't always be said of the Designers, who can perversely love things (and themselves) more than people in general.

The other aspect of the criticism of usability relates to the poor user. The Designer views him or her as uncreative as well. From the Designer's perspective, usability is a case of the blind leading the blind, and no big ideas can emerge through usability. The mantra of the criticism is: "You can't really innovate through usability." In this view, usability is just a form of editing, involving no creative writing.

A cornerstone of good user centered design is iterative design. Iterative design takes a rough design, gets users to react to it, tweaks the design, gets more user reaction, and continues until a happy design emerges (or until an immovable deadline comes up.) The essence of iterative design is that every design is tentative. As such, each design is alive, full of potential.

But iterative design is a messy process. Each group of users tested will give feedback on a design that may not always point the way forward. Often, there is too much happening in a design to isolate what features are good. And the number of users involved in an iterative test is often too few to draw solid generalizations. Sometimes iterative testing results in an idealized passage from rough draft to ever more refined version with each cycle. In practice, each cycle can introduce big unknowns where there hasn't been sufficient user data to decide how to proceed. Finding what doesn't work is only the first step toward finding what might work. Here is where creativity enters the picture in a big way. There are several options. If how to tweak the problem isn't obvious, one can always ask users for suggestions, either verbally or through a participatory exercise. More likely, the team reviewing the results will brainstorm the issue and come up with some possible solutions. The alternative solutions will need to be tested with users, since they are speculations at this point.

Exploring alternatives can yield surprises, and suggest new paths to develop. Such exploration is known in philosophical terms as abduction. This fancy sounding word refers to the process of divergent thinking, generating possibilities, in contrast to convergent thinking, zeroing in on a solution. Design theorist Nigel Cross, among many others, has argued that design is essentially abductive, speculative, conjectural. Iterative design is firmly part of this tradition. Some people mistakenly see only the convergent side of iterative design, and miss the role played by that divergent thinking during the team discussion after a testing cycle.

Wednesday, November 03, 2004


Seeing multiple dimensions

I have often encountered issues surrounding how to capture different aspects of user's behaviors that are occurring at once. For example, how to get video of users working with a prototype or doing a exercise like card sorting or participatory design. The issue is especially acute for group sessions. The Stanford Design Observatory is an interesting attempt to create a flexible, dedicated participation lab analogous to a usability lab. Very cool indeed.


Another Blog?

As a user centered design consultant, I am in the business of questioning the purpose of designs and content. I've pondered blogs for a while now, but hesitated to create my own. The question in my mind has always been: is it necessarily? Would it be useful to anyone? To me?

Far too many blogs are exercises in self congratulation: look what I know! Look what I found! I'm part of the insider group! The temptation is enormous. One can blog things with little concern for the negative feedback that might inhibit us from saying these same things face to face. Social inhibition gives way to narcissism. Face it: hardly anyone comments on blog entries. Blogs are broadcasts, not dialogues. Broadcasts without an editorial staff.

That said, there are many interesting blogs, even if they represent a minority of all blogs "published." Blogs enable some interesting minds to reveal their necessarily messy thought processes. To show what random raw material is reaching their eyes and percolating in their minds. Apart from the social dimension of "sharing", blogs succeed in allowing people to externalize, and refine, their thinking. Blogs are certainly a less interactive means of externalizing and sharing ideas than the mythic conversational salons of the past. But blogs provide a means of linking people and ideas that's compatible with the global dispersion and specialization of knowledge and interests.

I hope this blog will allow me to:
-- develop and share ideas;
-- keep friends in the US and UK abreast of what I'm up to now that I've moved to New Zealand;
-- make contact with a few kindred people who might have interest similar to me.

I hope you can gain value as well. Let me know about your blog if you sense it covers similar themes.

This page is powered by Blogger. Isn't yours?