Following Patterns and Innovating
Published: April 22, 2013
For the past few years, it has been fashionable to point out faults in things like TV interfaces and interactions, while positing that innovative new hardware like smartphones and tablets would fix everything any minute now—but it hasn’t yet.
For example, I chose my television programming service provider largely based on the service’s usability and usefulness. Dish has a rather good remote and, for years, has had many of the features that popular articles insist we need to break out of our current mode of interacting with television programming. For instance, users can perform keyword searches. Our service provider even offers to connect the system to the Internet, so you can control it with any smartphone, tablet, or via the Web—and even watch TV on those platforms from anywhere in your house.
But there’s a problem: There are issues of consistency that make it difficult to switch between the user interfaces. So, it’s hard to pop open the app on my phone to find out what my DVR is recording while the TV is playing a show. And it’s difficult to open my computer and tie a keyword search to my current viewing behavior, using its easier-to-type-on keyboard.
In a recent interview, Alan Kay spoke about providing devices like iPads in every classroom.
“I’ve used the analogy of what would happen if you put a piano in every classroom. If there is no other context, you will get a ‘chopsticks’ culture, and maybe even a pop culture. And this is pretty much what is happening.
“In other words, ‘the music is not in the piano.’”—Alan Kay
The same is true in any context. Simply adding user interfaces such as touchscreens will not magically solve our problems with TV. Existing products on mobile phones and iPad haven’t solved these problems. Adding any widget, feature, interface, interaction, or piece of hardware never automatically solves your problems with any information service or application. So how can we move the state of the art forward?
As UX designers, a set of best practices that we’ve inherited from prior work guides our work. Best practices have actually been proven to work. You should not confuse them with what I call common practice, which may be trendy and popular, but is still unproven—and may even be counterproductive or worst practices.
We generally organize best practices into broad categories by their applicability, for example:
Principle > Heuristic > Pattern > Template > Component
But there are other categories of best practices, and the bounds between them are open to much interpretation and argument. I won’t try to define any of these here, because that would be a whole essay in itself. In fact, for simplicity, I’m going to use the word pattern a bit too broadly, so just assume that I mean to encompass principles and heuristics as well.
Patterns and Innovation
As far as I can tell, our reliance on using new types of hardware to solve our problems is tied inexorably to the belief that change is always good. Perhaps this arises from the way growth has become ingrained in modern business practices. I’ve even seen some people go so far as to assume that, if change is good, all that has gone before is bad. Or at least, that everything old has become pointless and irrelevant.
Last year, Rian van der Merwe wrote an article that outlined his argument for breaking design patterns. He began with this definition:
“A pattern—whether in architecture, Web design, or another field—always has two components: first, it describes a common problem; secondly, it offers a standard solution to that problem.”
Then, he discusses breaking these standard solutions to come up with new, innovative ideas.
Taking another example, Stephen Turbek, writing for Boxes and Arrows, questions the value of pattern libraries as a whole:
“You cannot copy a UX pattern like you can copy a sewing pattern. Having someone read a pattern library will not make them a competent user experience designer. It would be akin to teaching writing by reading the dictionary—the whys are not answered.”
These writings all seem to allude to using the wrong sort of pattern libraries. Indeed, if you assume that patterns are a specific set of solutions, they won’t work. Patterns are not solutions like dress patterns, but constraints, guidelines, and examples. You use only the parts that apply, discarding others. All good pattern libraries provide variations on patterns and detailed explanations of when to use each variant.
Innovating Through Understanding
Most important, patterns provide an understanding of not just what the result should be, but also why. To stretch past current design best practices, you do not need to break patterns, but expand on them. This is something that comes up a lot as new technologies take over. Working in mobile, I see a lot of people who claim that it’s all so new that we have to set everything aside—despite 40 years of mobile telephony, 25 years of PDAs, and at least a decade of smartphones.
I often rail against working in the old ways and assuming that all new things are scary. Scott Jenson said this fairly well in a Quora discussion in 2011:
“Mobile, of course, isn’t really that hard. What makes it harder is that most people are coming from a desktop background, and they bring so much baggage to the table. The exact same thing happened when the Mac first came out, all sorts of DOS apps started to sprout ‘gooeys’ (GUIs), and it was comical. It just takes a while for an industry to shed its old ways.”
As Bill Buxton said while channeling Brian Arthur, “There is a precedent for all new technologies. The importance and the creativity are reflected in the insight with which those precedents are explored, tested, and combined.”
Applying Old Patterns
What I think is perhaps the most important facet of patterns—or principles or heuristics—is that they are generalized and are applicable to many situations, audiences, and platforms. In any discussion that devolves to innovation versus patterns, we should understand that new interfaces do not sweep aside patterns. Nor do new technologies change everything. They just change the importance and value of patterns and the way we apply them.
For example, at my very first, minimum-wage job, my office phone didn’t ring. (Why they couldn’t fix it I’ll never know.) However, the phone did make a very subtle clicking sound. But for the first few weeks on the job—before I got used to that sound—I felt compelled to pick up the phone every minute or two just make sure no one was calling. This memory sometimes comes back to me when I think of annunciators and how our mobile devices notify us of status changes.
Doorbells, phone ringers, and the LEDs that tell us we have new text messages all fall under the same category of what we might consider ambient information. Our mobile devices inform us of states and status changes without our having to deliberately pay attention to them. You don’t have to look out the window when someone rings your doorbell to see if someone is there, pick up the phone to find out if someone is calling, or read your mobile handset screen to see if there’s a new message. Instead our world beeps, blinks, and rings to tell us something has happened.
We continue to stretch this pattern usefully to new situations. Today, my mobile device has differently colored LEDs for various notifications—encoding the annunciator with content and context. On my phone, blue is a tweet, purple is an SMS, and pale yellow means I’ve received a new work email. The LEDs attract my attention, then their color registers a moment later when I notice the type of notification and its importance to me, in the moment.
What goes around comes around. And very often we don’t even need to study the history of a pattern—we just expand its scope because the right answer is sitting in some corner we simply aren’t paying attention to.
Just a few weeks ago, I read an article that, in all earnestness, discussed how the new thing in mobile is using color. This is something you might miss only if you were paying attention to just a very narrow set of trends, because the use of color has always been a key principle in ordering and organizing information. While the visual hierarchies that designers use vary, this is the one I’ve always used:
Position > Size > Shape > Contrast > Color > Form
There it is. Humans have color vision, so why would a designer ever ignore that when trying to establish a complex hierarchy to communicate the priority of various pieces of information? Well, most designers wouldn’t do that. Even when many iOS sites and apps were going for a monochromatic look, there was certainly use of color.
Patterns Are About People
Remember, patterns derive from principles and heuristics that are based on human physiology and cognitive psychology. Even as new technologies, user interfaces, and interactions arise, we base our designs for those new things on old, familiar patterns because they work for people. In interactive systems, tabs work regardless of whether they look like the tabs on file folders, because the concept of categories of similarly important information is important and universally valid. It is valid even for the tabs on folders inside file cabinets, which is why the original pattern came to be.
Principles of cognitive psychology apply to everyone. Whether someone is a business customer or a consumer, American or North African, we are all human. Sure, there are class and regional variations in people, but from working on products around the world, I’ve found that our similarities are much more striking than any differences.
This is why patterns exist and why they will continue to exist. We should all read and understand pattern libraries, even if we don’t believe in strictly following the patterns themselves.
Patterns Are Universal
When I went to college at the dawn of the Interne—partly because no one told me about human factors and cognitive psychology until a few years later—I ended up with an art degree. I focused on printmaking—etching and engraving mostly—but also had to take painting and sculpture. In foundations classes for all of these media, I learned principles and what amounted to patterns.
I also took some design classes—and on graduating, actually made money as a graphic designer before moving into CD and Web design—in the days before we knew what to call it. I immediately noticed that the art-school patterns that have been passed down through generations and the design-school patterns, which are based on observation and scientific principles, are the same.
When I realized that we needed to codify the best practices and principles that guide whole teams of Web designers in the workplace, I found that practice areas like information design and human factors also had their own good guidelines. While our existing best practices and principles did not change at their heart, those from these new areas of practice meshed nicely with what we were already doing—and we added more as the practices of user experience, information architecture, and interaction design became more codified.
As we continue to look into other fields, we’ll find that there are standard approaches—principles and patterns—to landscaping and literary criticism and everything else. I strongly encourage you to explore other fields, because when people create anything for human consumption or interaction, they strive to achieve many of the same goals as we do when designing a Web site or mobile app. To give you a slightly far out example: A favorite thing of mine is to read the accident reports that the National Transportation Safety Board publishes. They often go into great detail on human-factors issues, and I quite frequently learn new things about how people think and react to systems and situations.
Making the Best Experience
Breaking away from trends and stretching patterns can certainly lead to innovative experiences. But entirely breaking away from patterns and principles just leads to broken experiences.
I tend to think that design is actually a bit more like science than many believe. In both disciplines, we learn from the past and build on what we’ve learned to create new works in new spaces. In my view, UX works at the intersection of science and design—and in many ways, the two are not at odds with one another.
Patterns are not design trends. At their core, patterns are based on how human beings interact with things. In a discussion that I had with some other designers after a workshop on no-UI solutions, Xianhang Zhang said, “The art of predicting the future is knowing what hasn’t changed.” People don’t change in fundamental ways, so our job as designers is to consider how humans might use new or evolving technologies, while never losing sight of the fundamental ways in which people work.
Though innovation involves some risk taking, it comes about through continuing evolution, not from setting aside and ignoring everything that has gone before.
Change and the Future
It is important to embrace change. As ideas like embodied cognition gain currency, even the very underpinnings of cognitive science may be about to change. But until we prove that idea or any principle, thereby necessitating change, it’s best to employ our current models and patterns. There’s no reason to simply ignore them in favor of your intuition.
Please, challenge the status quo and try new ideas. But do try them first. When you’re innovating, you need to put your ideas in front of users and gather good data about them. Ideas and experiments are fine. But implementing innovative ideas or creating new patterns requires validation. This is the other side of building on prior work and is just as important. We all need to stretch the boundaries of what we know today to discover what we’ll all learn tomorrow.
You should not just decide that you like a branched pattern or new pattern more than an old solution. You have to prove that it works well by doing some form of UX research. Usability testing or, at least, well-designed, analytical A/B tests must indicate that the new design adequately meets users’ needs. Preferably, it meets those needs better than the solution that current best practice dictates.
One common complaint about pattern libraries is that they become stagnant. If so, it’s our fault—and the fault of those who make us sign NDAs that prevent us from sharing best practices universally. The key to scientific advancement is that, when we prove a better idea, we must share what we’ve learned with everyone.
Ulaby, Neda. “Why Are TV Remotes So Terrible?” All Tech Considered, Technology News from NPR, March 26, 2013. Retrieved April 7, 2013.
Greelish, David. “An Interview with Computing Pioneer Alan Kay.” Time, April 2, 2013. Retrieved April 7, 2013—“The music is not the piano” relates to questions that Alan Kay has gotten regularly for decades, every time a new computing device appears. Today, it’s the iPad. He argues that the hardware doesn’t matter as much as the interactivity, services, and availability of information across devices.
Van der Merwe, Rian. “Design Patterns: When Breaking the Rules Is OK.” Smashing Magazine, June 6, 2012. Retrieved April 7, 2013.
Turbek, Stephen. “Are Design Patterns an Anti-pattern?” Boxes and Arrows, January 26th, 2012. Retrieved April 7, 2013.
Quora. “What Are the ‘Standards’ in Mobile UX Design?” Quora. Retrieved April 7, 2013—Though this is an interesting discussion in general, Scott Jenson’s response—much more than I’ve quoted—is most illuminating.
Buxton, William A. “Innovative Interaction: From Concept to the Wild: Technical Perspective.” Communications of the ACM, Volume 55, Issue 9, September 2012. Retrieved April 7, 2013—In addition to the quotation about innovation, this brief perspective also offers useful historical examples that are not unlike those that I’ve outlined in my article. If you have an ACM membership—and you should—it is worth reading and is only one page long.
Arthur, W. Brian. The Nature of Technology: What It Is and How It Evolves. New York, New York: Free Press, 2009—Buxton’s quotation in this article comes from this book, which I must admit, I have not yet read. But now that I’ve heard of it, I’ve added it to my reading list.
Pontin, Jason. “On Innovation and Disruption.” MIT Technology Review, February 20, 2013. Retrieved April 7, 2013.
Krishna, Golden. “The Best Interface is No Interface.” Cooper Journal, August 29, 2012. Retrieved April 7, 2013.
Hoober, Steven, and Eric Berkman. “Display of Information.” 4ourth Mobile Patterns Wiki, December 13, 2011. Retrieved March 18, 2013—This chapter introduction reviews several principles that you can use in applying information architecture to interaction design and user interface design.
Borowska, Paula. “The New Mobile Design Trend—Colour.” Designmodo, March 26, 2013. Retrieved April 7, 2013—Maybe I am becoming frail in my old age, but I was shocked when I read this. Not just that she said this so matter of factly, but that people seem to believe it. It’s barely even true that designers don’t use color in apps. Maybe not so much in iOS, which every screenshot in the article shows, but it’s not that uncommon to use color even there. I see articles quite often that, like this one, confuse trends in narrow domains with broader principles or patterns.
National Transportation Safety Board. “Accident Reports.” NTSB.gov, undated. Retrieved April 7, 2013—If you won’t be overwhelmed by the fact that people die in almost all of the reported incidents, this is a good source for keeping up on human factors and seeing great analyses of how things that people use can go very wrong. Maritime reports tend to be pretty thin, but the others are quite long and well researched, so include references to academic research on attention, sleep patterns, vision, distraction, and things that I find I can apply in my everyday work, but haven’t encountered elsewhere.
Madrigal, Alexis C. “iPhone 5? Yawn. What Will the ‘Phone’ of 2022 Look Like?” The Atlantic, September 13, 2012. Retrieved April 7, 2013.
Cooper. “What Good Is a Screen? A Cooper Parlor on Wednesday, September 19th.” Cooper Journal, August 27, 2012. Retrieved April 7, 2013—See the article by Golden Krishna above. I attended this event, at which we discussed principles, and participated in a workshop to explore some of the ideas. I spoke with several people, including Don Norman, and got some good, quotable ideas.
Wilson, Andrew D., and Sabrina Golonka. “Embodied Cognition Is Not What You Think It Is.” Frontiers in Cognitive Science, February 12, 2013. Retrieved April 8, 2013.