Musings on design & other stuff jen's place

Back in December, I asked a question on the design@sun blog, and then posted a summary of the responses. In that summary, I said that I'd write a presentation with more details, which I did. In February, I presented the detailed findings as a "half-baked" talk to SunLabs East. At the end of the talk, an intern or a new employee (someone young, who I didn't recognize) asked me this question: "How can user research help you to innovate, rather than just make incremental improvements?"

I was taken aback. ... What a great question. Nicole jumped in and gave a really good answer about how user research allows us to get at the root problems that the users are trying to solve, and that we're not asking them to tell us what's wrong with the software, per se, but figuring out their goals and how to better meet them. Of course! Why didn't I say that?

But there was a bigger question, which is why I'm pretty sure that I didn't come back with the answer that Nicole did immediately ... and that bigger question is, "When faced with challenges, how do we leap-frog over just solving the symptoms of a problem and do something really innovative? something fundamentally different?"

I have an answer to that question, but before I tell you what it is, I have to put in a plug for the very-much undervalued "incremental improvement". If we continually made incremental improvements to the user's experience, we'd be far better off than we are now. Quite often, the designer says something like, "This [site | application | device | experience] needs a complete re-design", at which point engineering management rolls its eyes at what they think a complete re-design is going to cost. As a result, nothing gets done, and the user experience stays just as bad as it is. So, for goodness sake, make the incremental improvements! They are cheap, fast, and will have an immediate benefit for your users. Now, back to the question at hand...

There are probably a million answers to this question (how to innovate), but I have one, and here it is. Imagine what the solution would look like, behave like, or provide, if there were no constraints. If money were no object. If you had an unlimited engineering team to develop it. If the existing standards that everyone used were thrown out. If you were in charge ... then what would the answer be?

It's only by putting myself in that frame of mind that I have been able to truly innovate. To have a problem and then imagine the ideal solution, without boundaries placed upon it. Believe me, the boundaries will come. There will be functional requirements and business requirements, and of course, user requirements. ... but before then, dream big. Live outside the box. Read Plato's Republic. Think differently.

To borrow a line from Steve Jobs, "innovation is the ability to see change as an opportunity, not a threat."

I can't believe how long it's been since I've posted here (there were some interim posts here), but that's life. The holidays were great, the New Year has begun, and I'm starting fresh ... with work, with my diet, and, hopefully, with my phone. A few weeks ago, I asked, "What do you really want from your mobile device?" (and then I posted the trends in the answers).

I know what I want, but it doesn't exist yet. Not even on the iPhone. Now don't get me wrong, I've been following the iPhone since it was first announced, and I really want one. But what if I buy one and then they release the next cooler version that has something that I really need? ... sigh.

I've even bought the ringtone that I want to use (yes. I spent the 99 cents, much to the chagrin of many colleagues -- it's a dollar, for Christ's sake, not a war of business ethics). I can't wait to use it ... it's the first eleven seconds of the title track to turn.

It was rumored that the next iPhone would be announced before Christmas, so I waited and watched ... Nothing. Maybe they were referring to this. Of course, we still did our share to buoy the Apple stock price over the holidays, but the MacWorld Conference and Expo is only a week away, so I wait ... hopeful.

When we last talked, I was describing the reasoning behind the McGinn design model, having shared that this explanatory text was developed with my colleague, Jen Hocko.

But, you ask, how can such a model be meaningful to the practitioner on the street? I'm a practitioner, which is why I'm keen on keeping anything I write grounded in research, but applicable to our everyday work.

Takeaways for Designers

First, at the lowest levels, what we know about human sensory perception should be codified in our user interface guidelines or standards — not be discovered during our usability tests. We have decades of research on what humans do naturally, which we should be careful not to forget when we’re caught up in the latest cool project, technology or deadline. For example, a recent article highlighted how a marketing department used usability testing to "discover" that bigger targets are easier to find, and that red sends a stronger signal than blue, when users are searching for something specific on a web page. No one needed to run a usability test to find these things out — they just needed to know how the eye perceives light, and have read up on Fitts’ law, first published in the early 1950's.

Next, one’s perception of what is intuitive and innovative is affected by:

• Variations in individual experience
• Variations introduced as a result of context
• Variations in expectations based on culture

Imagine for a moment that you were born on a remote island. What would your expectations be for the behavior of a switch, a dial, or a push button? You’d have no expectations, because you would never have seen them before. Once you were shown that a light switch could result in a room becoming illuminated, you might be impressed with how innovative it was, but the interaction might not have been intuitive at all.

So as result, creating truly intuitive and innovative user interfaces requires attention to several layers of the McGinn model, not just one or two. Many of us start designing for the user group, either blissfully unaware that we need to pay attention to anything below that layer, or by making broad assumptions that all the lower layers are the same for everyone in that user group.

By asking ourselves with each new design who we are designing for and where in the model we might find variations, we will uncover opportunities for making our products and services more intuitive and more innovative. By evaluating our interfaces against each layer of the McGinn model, we can uncover opportunities for improvement and alignment with the people who will eventually, hopefully, be using it.

Okay, I know, I've been busy. I promised you something that I haven't delivered. So here I am, making good on my threat. Back in part 1 of this thread, I shared with you my frustration at trying to get this new model published. In part 2, I showed you the design hierarchy that comprises my new model. And, now, today, I'm going to put some more words around it.

The only thing that I need to preface this discussion with is a little bit of credit. This model was originally presented as a talk that was given by me and a fellow graduate of my MS program, Jen Hocko, who is a Senior Usability Specialist at the MathWorks. While the model is mine, Jen and I worked closely over several months to craft the explanatory text below, so I'm not sure that I could tell you immediately where my words end, and hers begin. In addition, we had several reviewers of this work: Beth Loring, Chauncey Wilson, Joe Dumas, and Bill Gribbons. As a result, this discussion has been influenced by their review comments. That said, here it is.

A Model for Creating Intuitive and Innovative Designs

One of the aspects that we (Jen Hocko and Jen McGinn) had discussed privately, is that prior knowledge and experience play a significant role in whether users perceive designs as intuitive or innovative. Any useful model would need to account for these factors.

Like the model proposed by Hancock, McGinn’s model advocates an approach to design that incorporates attentiveness to the many factors influencing a user's perception of a design as intuitive and innovative. The old phrase "perception is everything" is the basis for this model: essentially, what affects our perception of any design?

Starting at the base, we believe that designers should rely on the physiological bits of HUMAN SENSORY PERCEPTION, which are reasonably consistent despite changes in subsequent layers. These physiological responses include how humans perceive light and sound, as well as the Gestalt principles of visual organization. Our designs should take advantage of what is common across humans, and those similarities should be codified in all our design guidelines.

The bottom layer is pre-cognitive, and accounts for what is physically similar across humans, while each of the subsequent layers describes what we have learned based on our experience.

The next layer of the McGinn model has to do with the user's CULTURE. Hofstede defines four dimensions of national culture: power distance, uncertainty avoidance, individualism, and masculinity. In addition, other culture-related factors that should be taken into account when attempting to design any interface include reading patterns, cultural aesthetic preferences, and concepts of time. The mental models that we have for what is "normal" come from our experiences living in a particular culture.

For example, suppose we are working for an airline company. The passenger from Israel will have a set of expectations of how that experience will go, from passing through security, to the timeliness of the flight, to the ultimate experience of the flight itself. Compare that to the Russian traveler on Aeroflot, which was notorious for taking off within a day or two of the scheduled flight time. And then, compare those to the expectations of the American traveler, who believes that a flight should be on time within a few minutes, and that security checks should be minimally invasive.

CONTEXT is also an important consideration when designing intuitive and innovative user interfaces. The context in which a person operates is rarely static. For example, when a person who is an expert Windows XP user tries to accomplish the same things on another operating system, he can instantly become a novice. His age and economic status haven't changed, but some aspects of the task at hand have. It is these task-based characteristics that determine a person’s user group; his needs for information and support have just changed, because the context in which he tries to accomplish his goals has changed.

As a result, context and USER GROUP (the next layer) are very tightly coupled. Given the context, the same person may fit several different user profiles: she may be a homeowner with a mortgage, a wife with three children, or a technical professional with an advanced degree. The user group a person belongs to depends on the context in which they are being categorized. A person’s goals and resulting information needs will change with their user group, depending on the context that person is in. Take, for example, the expert user of the Windows XP operating system. When he needs to eject a CD from a system that runs the Mac OS or Linux, he requires different mental models and knowledge for how that task will be accomplished.

Lastly, we can design for the INDIVIDUAL. We don't have to, but there are certainly markets to be exploited this way. Gateway and Dell built successful business models by customizing systems for each person who ordered their products. The Bentley College masters program in Human Factors in Information Design (which both authors completed) is similar in that regard. Virtually no two people take the same ten classes. Some get credit for work study, some for having completed the certificate program, some for taking a negotiating class as an elective, and some for taking a course in statistics. Each course of study is designed to fit the needs of the individual.

But what does all of this mean for the designer or usability practitioner? How can it be applied to what we do? ... Hang on ... I'll post that on Friday :)

I was thrilled to hear today's design review, which was a redux of Jakub Franc's World Usability Day talk in Prague. I found it particularly interesting that the personas developed by Jakub's team of graduate students at the Czech Technical University were very similar to the ones that I developed with two classmates, to prototype a "smart" pill dispenser, for a graduate class :)

The college from which I received my graduate degree ran an annual 2-day conference called "Aging by Design" for three years (2004 - 2006). I am fortunate to have been able to attend in 2005 and 2006. Designing for aging populations has been an area of research and design interest for me, since seeing my father's cognitive decline and subsequent interment in a nursing home, before his death in 2005.

If you are interested in reading more on the challenges of how to create good designs for older adults, you can start with the small book, Designing for Older Adults. Other interesting books include The Handbook of Aging and Cognition, Cognitive Aging, and Handbook of the Psychology of Aging.

It started like this ... a question came across an email list on usability: what makes a design intuitive?

After many other people responded, I took a stab at it:

So here is the nut of the problem, as I see it. Prior knowledge varies from human to human and culture to culture. The things that don't vary across cultures, the physiological bits of sensory perception, like how a sound reaches our ear drum or how color is received by the cones in our eyes don't help us a lot with design, except for things like size, contrast, and color choices that are limited to red & green for fine discrimination versus blue for peripheral vision. Other things that we could count on across cultural boundaries include the Gestalt laws of proximity, closure, continuation, and similarity.

So, understanding what works (and doesn't) at the level of human sensory perception should be codified in our visual style guides (or smell, if you're designing for a puppy :) . Beyond that, the next level is culture, then the user groups that we are used to designing for. We need to view the intuitive-ness of a thing as having layers of effectiveness (pardon my text pyramid) -- there's some magic point between User Group and Individual that we are always striving to design for.

       
                    /\
                   /  \
                 /      \
                /        \
              / Individual \
             /--------------\
            /   User Group   \
           /------------------\
          /      Culture       \
         /----------------------\
        /Human Sensory Perception\   <- or species, in the case of the puppy
       ---------------------------

There's also a modified Maslow's hierarchy in the paper on individuation by Hancock, et al (Hedonomics: The power of positive and pleasurable ergonomics. Ergonomics In Design.) At the bottom is pain avoidance, and at the top is individuation (personal perfection). In between are functional, usable, and pleasurable.

Later in my discussions with Jen, I added another layer, Context, in the middle, so my hierarchy now looked like this:

But what does this model have to do with creating intuitive and innovative designs? That's what I'll discuss in subsequent blog posts.

It's time. Last May, a colleague and I gave a presentation at the Boston UPA Mini Conference on User Experience and Usability. Since that time, she and I have written a paper on the same topic, and have twice tried to get it published. No luck so far.

It was from a flash of insight, just an answer to a question on an email alias, that I came up with a new theoretical model for designing innovative and intuitive products and services. And the potential publishers, like I tell my daughter with her math problems, don't just want the answer, they want to see all the work.

I'm thinking that I just finished an Masters degree, and the amount of research backing up my model could be a dissertation. But here's as much as I will say now: the McGinn Model, as my colleague Jen Hocko named it, is based on Hancock's Model, which is based on Maslow's Hierarchy of needs.

And starting tomorrow, I'll be talking about the model here. If I can talk it through with you here in this medium, and take my time to research one layer at a time, maybe I'll eventually find the right venue in which to publish it.

Until then, viva la blog!

I know, I said that I didn't get Facebook. But now, I do. It's been over a month since I've started really playing with it, and I'll tell you what I've learned, from a design perspective.

First, Facebook is really "sticky" -- I get this term from Malcolm Gladwell's book, The Tipping Point. He defines sticky as a marketing term for infectious. How long does a product or brand or a visual stay with you? That's one measure of stickiness. With Facebook, what it means is how often do you visit the site, and how much time to you spend there? For me, it's a couple times a day, and I spend a few minutes to a half hour. But why, you ask.

One element of Facebook is that it sends me friend requests and notifications in email, so I am taken out of the task I'm doing (interrupted) and motivated to go see what gift I've received, what's written on my wall, or who wants to be my friend. It's like the little rush I get, when I see the postal truck coming down the street toward my house :)

Anther element of it's stickiness is the notion of time. I like to view my friend's status updates, which gives me a good idea of who is or was doing what when. As a result, I feel more connected to them on a regular, real-time basis.

Another thing that I enjoy are the applications on Facebook. They are fun, silly, and mildly informative. I have famous quotes, a fortune cookie, a friend wheel, a map of the world that shows where I've been, favorite music, books, ... I could go on. I can view my friends applications and information and they can view mine. We can learn more about each other than we normally would in person. And when I find a friend who has an application that I like, I can click one button to add the application to my page. Very sticky.

Lastly, the meta-finding I have is that Facebook gives me a distributed community of friends -- this access is important, because I don't have physical access to them. I can't walk down the hall and knock on their doors. Several people I work with -- okay, most of the people I work with -- live in other geographies like California and India. What Facebook gives us is a virtual water cooler to gather around. A way to relate to one another that is not just about the task at hand in a meeting. But it also gives me that ability to feel closer to the people I don't take classes with anymore, and people who have left Sun.

I get Facebook. It's about feeling connected to people in a technological world that often leaves us feeling lonely or isolated.

Barry Schwartz, the author of the Paradox of Choice, describes our anticipation of our experience with a thing as "expected utility" and the actual use of the thing as the "experienced utility". I'm really tired of trying to buy things that I think will yield a good user experience, only to find that regardless of how much research I've done, there's no way to evaluate the darn things until I've bought and used them. I'll give you some examples of what I mean by that, from my kitchen.

In the manufacturer's photo of my kitchen faucet, the handle is on the right. My husband and I are both right handed, so we thought, "Here's a really beautiful faucet, that will work well for us" (expected utility). Once we had it installed, we realized that right-handed people hold pots and pans and glasses with their right hands, and turn on the faucet with their left hands (at least we do). So we had to have the plumber back out to turn the faucet around, which is not really the same, because there's a black button on the "back" of the faucet, which now faces front. Every time I see that black dot, I grimace. It's interfering with my enjoyment of the beauty of the faucet (experienced utility).

Another example: my coffee pot. When we installed granite counter tops in our kitchen last summer, I decided to treat myself to a new coffee pot. The old one wasn't broken, but it was 3 or 4 years old, and pre-dated our getting the whole-house water filter, so it made God-awful noises when the coffee was brewing. I've been a die-hard fan of one particular brand of coffee makers, for about 20 years. They have all sorts of models, but I prefer the ones with clocks, so we can set up the coffee the night before, and then have it brew the coffee in the morning, right before we get up. On this new model, it does those things, but the little dial that they used to have to set the clock, has been replaced by a sequence of 7 button presses (across 3 separate buttons, none of which are dedicated to setting the time). I can never remember the key sequence, so I keep the manual close by. When the power goes out, I dread having to re-program the thing. Does it hamper my enthusiasm for the product and the manufacturer? You bet.

Last example: we recently replaced our dishwasher with a new one. We spent a LOT of time selecting it: we chose it for the completely configurable racks, the fact that it's very quiet, the top rack can be moved up and down, the stainless steel interior, and lots of other features. What we didn't know until it was installed and we used it, is that we need to use special soap that is less environmentally friendly the the soap we were using before -- otherwise the dishes don't get clean. And some things will never get dry, no matter how long you leave them in the washer. And the icons on the display aren't documented anywhere -- not in the manual or anywhere online. And it takes 30 - 60 minutes longer to wash a load of dishes than our old model. So, the features that we bought it for are all there, and it looks great. But are we really happy with it? Not so much. Because the user experience was not what we had expected.

My friend Manya was quoted today in the local business journal, Mass Hight Tech: "To stay competitive, businesses need more sophisticated websites, she said. IAs are designing them and the products they promote to make an "emotional connection" with consumers".

I agree, and this emotional connection is why Hancock's hierarchy of hedonomics makes so much sense. Emotional connection to our users, our consumers, moves us beyond just pain-free, functional, and usable, and on to pleasurable. What that means for us as businesses, initially, is having a competitive advantage. But over time, what were once delighters become satisficers -- we expect all things to be not only usable but emotionally and aesthetically pleasing. A few years back, we were just happy to get web search results, but now that Google has raised the bar, we expect all search engines not only to give us valuable results, but in a fraction of a second.

Of course, Don Norman wrote a whole book on Emotional Design, but he focuses more on tactile things that we use as consumers (like my kitchen faucet, pictured), rather than web site or software design. So Manya asked me last week, what web sites I felt really connected with me emotionally. I replied that I couldn't come up with one, but that if I were designing a web site that I wanted to connect with people on an emotional level, I'd ensure a few things: that the imagery was appropriate for my audience (many web sites use pictures of smiling people to try to make that connection), that the vocabulary and tone of the language used on the site was appropriately formal or informal, and that the visual design was not only usable, but aesthetically pleasing (like the new Solaris installation application). Really, just adhering to good design principles ... in hindsight, I should have pointed her to sun.com, which does a very good job of all three of the things I'd recommended. And I know that they are easier said than done.

I was disappointed in a recent article that came across an email alias that I belong to. The article was about some amazing and unexpected results of usability testing e-commerce pages.

"One office supply retailer wanted real answers, so they tested five specific color and button size design elements. The results didn't go as expected -- one change lifted conversions 44.11%."

As I read through the article, I found that their amazing "discovery" was that people completed the purchase check out when the designers made the checkout button larger. Wow. If only someone had figured that out scientifically before now. Say, in 1953.

Another big finding of this test was that users were more apt to find the button if it was red, rather than blue. Really? The cones that detect the color red are in the front of the retina, and the ones that detect blue are on the periphery. As a result, green and red are most easily perceived during fine discrimination, and that's why blue makes such a great background color.

So, while I'm happy that they got findings that they could act on, I'm really disappointed that a usability test was used as the vehicle to "discover" what was already known.

Yesterday, a number of us had a party in honor of Joe Dumas, in thanks for his 8 years of teaching at Bentley College. At the end of June, Joe decided to take another route, still working, but no longer working at Bentley. Joe has not only been a beloved teacher to many of the 200-or-so of us who have gone through the program, but a mentor, and a friend. It's strange to think of the Human Factors in Information Design (HFID) program without him.

Joe has been instumental in our understanding of the field, has offered introductions and letters of recommendation, and has been an enthusiastic supporter of our work. I feel very lucky to be one of a dozen people who got to take a second class with him.

I know that Joe will be teaching from time to time as part of his new job, but I asked him (selfishly), "Just don't teach them everything, okay?"

There's a lot of talk about how to create software that is usable. Usable by humans. So what do we know about humans that aplies to usability? To start, we have Maslow's work (circa 1943) on how humans are motivated. Maslow's hierarchy of needs is a pyramid that describes a number of physical and emotional needs that must be satisfied before a person can reach peak perfomance.

Based on Maslow's work, Dr. Peter Hancock came up with a model of what it means for a product or service to be "usable", in his June 2005 paper, Hedonomics: The science of enjoyable human-technology interaction. Like Maslow, Hancock comes up with a hierarchical model. At the bottom is "safety/pain avoidance", the next layer "functionality", then "usability", followed by "pleasurable experience" and, at the top, "individuation/personal perfection".

What Hancock suggests is that in order to be considered usable, a thing (or service) must be both pain free and do what it's intended to do. For example, if a lamp gives me an electric shock every time I turn it on, Hancock would not consider it to be usable. Or if the lamp was beautiful and pain free, but didn't work, that would also not be usable.

But Hancock's model gives us something more -- it gives us two more layers beyond usability that were, until now, implicit. What that means for us as designers and developers is this: neither functionality nor usability should be where we stop designing or implementing. The overused example of the iPod is over-used for a reason: it's a fantastic example of Hancock's fourth layer, pleasurable experience.