The Quest for a Perfect Interface
Chris McAndrew posted on January 04, 2017 |
Bentley University graduate program intent on creating a better user experience.

As an engineer, you can probably be relied upon to explain how to use computer software to friends, relatives—even coworkers. Helping out friends who are less techsavvy can be a hassle. This is why teaching engineering CAD software in schools is so important.

Talk to an expert in user interactions though and you quickly learn that it does not have to be this way. These days, a number of universities offer programs dedicated to improving human-machine interaction. I had the opportunity to discuss the field with the director of one of these programs.

William Gribbons is the founder of the User Experience Center and User Experience Studio and the director of the Human Factors in Information Design Program at Bentley University in Waltham, Mass. I spoke to Gribbons by phone to discuss the state of the industry and teachings that he focuses on with Master's students in the program.

Chris ( You have some experience opening the User Experience Center and User Experience Studio prior to the dotcom bubble. How has the approach to user experience (UX) changed over time? Has the user experience or customer experience become more or less of a focus in academic studies and in the field?

Gribbons: I've been doing this for 30 years. We've been talking about this for a long time, but it's funny how the whole world has started to notice. When I came into the field, technology just needed to work—that was the only measure of success.

In the 80s and 90s, the user really suffered.They endured unnecessary work. They had to work harder at it because they had to take on the workload. They got away with it because technology users were fairly sophisticated, and in the 90s, it opened up to everybody and it was a shock. It was NOT a heterogeneous population like that of earlier years.

In the early days, what they were doing was picking the low-hanging fruit—making things more efficient and embedding learning in the application.

Our readers may have used a lot of CAD tools that certainly live up to the suffering you mention. You have worked with some of them. What can you point out?

In the CAD world, the early users (of PTC and Autodesk software, for example) operated systems that were very powerful but very difficult to use.

A core theme that we address, from a psychological front,is how much are we asking the user to do? How much past learning can we use? From one product to another, or within a product line, what do they have to work differently?

Companies want to differentiate themselves, but that imposes a load to the user base. We think about the internal consistency within a product, the transferal from generation to generation of a product. Do you have transferability across multiple products, like SOLIDWORKS and the Dassault Systèmes suite of tools? Should there be consistency and transfer of learning, and should there be an industry consistency even it is counter to the business approach?

The UI of Solidworks 95. (Image courtesy of Solidworks)
The UI of Solidworks 95. (Image courtesy of Solidworks)

There are people who learned a CAD package 25 years ago and still look for a job that uses the same package.

You look at a lot of these companies—they hire from among their users. That perpetuates this. When you want to innovate, that is problematic. When SOLIDWORKS came on the scene it had a blank slate.

What about newer CAD companies that come on the scene? Are they at the same point SOLIDWORKS was when it launched with a blank slate to innovate something new?

It will be interesting to see if new CAD companies are adopted by new customers. It'll be tough going to acquire existing customers.

One of the great untapped opportunities in this field is how do we build in support and intelligent support. When we leave the user on their own to deal with changes and inconsistencies, it exceeds the capabilities of the user. Can we build intelligence in the systems (wizards and agents)?

All of the above tools have opportunities to improve, especially when thinking about new customers. The program Bentley runs, and others, pulls from a lot of areas. Gribbons’ background is in psychology.Much of your research is focused on software and human-computer interactions. How do you apply this or approach this in other fields (i.e. machine design, architectural design)? What can you apply UX studies to, and/or do you have an example of a client that was less traditional but that benefited a lot?

The nature of our program is mostly in mobile games, software applications and hardware. Increasingly, we have been introducing the theme of the “design of everything.” Anything that we design can be improved through a deep understanding of the customers. Buildings, hospitals and museums are applying these principles. We have one student who was an undergrad in urban planning, applying these principles to the communities that we live in.

Removing the C from HCI (human-computer interaction) does not require anything new. You are focused on the human interaction of anything, including software.

You pull from a variety of backgrounds, including engineering. Which background succeeds in the traditional UX field? Which engineering discipline do you work with the most? Are there other disciplines in which the work is more impactful, possibly because the UX experience is neglected there?

In most of the enterprise systems, they have a long way to go. The big CRMs, SAPs of the world, they grew up in this world. It was the user's job to figure this out.

The healthcare space, and electronic health records, they are absolutely miserable. People are dying because of these health records. I do believe there is low-hanging fruit in healthcare.

Semi-autonomous cars are interesting. You look at the horrific accidents. Is it is due to a lack of consideration for how a user operates in that environment?

Planes hit this point a long time ago when planes largely flew themselves. Humans do not operate well in this gray space. Either a system is fully autonomous or fully controlled—but humans struggle in between. They can put in interaction on a dashboard to say, “Put your hands on the wheel,” but humans will quickly switch to checking their newsfeed or phone.

When it becomes completely autonomous, we'll be okay. I think we’re going to regret not anticipating it correctly. Think about the way you anticipate the problem.Don't blame it on driver error.

Take the recent disaster in Chile. The plane ran out of fuel,and they blamed it on pilot error immediately. How do we design a plane where that does not happen. That plane should not have taken off. What we’ve done for the last 40 years is blame the user.

Computer interfaces used to be punchcards, then command-line keyboard entry and then the big UX change of a mouse. These days, there is full voice control with things like Alexa and Siri, wearable sensors that integrate across the Internet of Things and, of course, the hardware and input used for new headsets in virtual reality (VR). How do you approach new technologies, and what principles do you think must be kept in mind as new devices are developed?

About three to four years ago in the program, we started working on this. This is uncharted territory. We just hired a PhD who studies in VR. Most of the applications right now in augmented reality (AR) and VR are silly little headsets that will be popular for fun/amusement and games.

We are not at a point in which we are killing people with bad design. I think we’re going to be dealing with AR/VR in surgery, etc., and then there will be serious consequences.

Voice is interesting because there is no inherent memory in the system. Once you deliver a command it is gone. We have to think about how that memory of input is stored and shared.

The program you offer is one year and also has an online approach for those who are working or located far from Bentley's main campus. Can UX learning be done remotely? As a experience researcher, what is lost when doing things remotely and what is gained?

I have a number of students who work for publishers and education publishers. Education—and higher education, broadly—are poised for disruption. The notion of people coming to a physical place for four years and living there is evolving. What we’ve tried to build is a learning community not defined by time or place—to the point in which we have our alums continue to contribute. Students can choose to be online and choose where they are.

We’re flipping some of our classrooms, with students viewing things at their own time and own pace—those things that are best served in a virtual classroom. We’re making those decisions very deliberately.

There is a lot of inertia, but every time we spend another $30M on a building on campus, I think this is not going to work for people moving forward.

The advice used to be “Don't have multiple graduate degrees.” Now 25 percent to 30 percent of students come in with graduate degrees already. The concept of lifetime learning is growing.

To this day, good interface design is more honored in the breach than in the observance, if I may quote Shakespeare. With luck, the engineering and development tools of the future will learn from programs like those at Bentley University so as not to subject users to the same pain that many readers have had and continue to have.

To help improve your interfaces and learn about SOLIDWORKS education programs, follow this link. If you are a researcher, like Gribbons, looking for access to SOLIDWORKS, click here.

SOLIDWORKS has sponsored this post. It has provided no editorial input. For more information, go to

Recommended For You