At any given moment we are all surrounded by interactivity. From using the toaster, to setting up your DVR to record your favorite show, interaction with physical objects is part of our everyday life. Coming from a web centric outlook, I typically associate words such as interactivity to the multimedia and computer software, however interaction far exceeds the spatial confines of our monitors.
Breaking free of the monitor, and starting my voyage into the physical realms of physical computing, I decided to step-back and analyze some of the physical objects that I regularly interact with. It wasn’t long before I noticed the perfect subject.
The Metro Card Vending Machine is a common place for many New Yorkers. As a matter of fact, I thought it had a rather intuitive design, until this morning. They have incorporated a sexy touch screen design, with large buttons so clumsy fingers don’t press the wrong thing. In my opinion the interface is quite intuitive and rather pleasant experience, or as pleasant as refilling your Metro Card could possibly be.
This morning, on the machine next to the one that i was using, a tourist proved me wrong. It turns out that the touchscreen is just one of several interfaces on the vending machine, and although the tourist easily navigated the metro card touchscreen buying experience, she still needed to interact with other parts of the machine that obviously didn’t receive as much love.
As the tourist selected her card type, amount, and payment type everything seemed great. She then dipped her credit card into the credit card slot, entered in her pin or zip depending on payment type and BOOM! The screen asked her to press “Enter”. Since I was there observing her interaction with the machine, I let her argue with the machine a little while to see how long it would take her to figure it out… At least a minute passed and she started getting visible upset, her friend yelled out from the other side of the turnstile, “Just forget it and go to the booth!” Finally, I stepped in before she became completely ballistic and let her know that the enter is actually on the credit card keypad labeled ENT over the pound sign (#). She thanked me and went on her way.
At that moment the tourist may have felt stupid, but is she really stupid? Could the design be flawed? Stepping back and examining the interface which i have now identified as the “entire machine”, and not just the sexy touch screen, a couple things stand out. The first thing I noticed was that there are two separate inputs, the multi-touch screen and credit card keypad. This inherently splits the users attention jumping back and forth from screen to keypad, possibly causing confusion on what they should be looking at next. Although, moving the keypad to a digital version on the touch screen may seem like a viable answer, this would causes an added level of confusion by interrupting the spatial congruity that the credit card slot and the keypad have. So what is the answer? In the book, The Art of Interactive Design: A Euphonious and Illuminating Guide to Building Successful Software
interaction: a cyclic process in which two actors alternately listen, think, and speak.
As the tourist was interacting with the Metro Card vending machine she was essentially communicating with the physical object.
“Please enter your zipcode and press Enter”
Obviously, the tourist can’t speak to the machine vocally, but she could communicate her selection through the buttons that are presented. Since the tourist could not find the Enter key she essentially was stripped of the ability to communicate, placing a muzzle on her interactive experience.
A possible solutions that may have worked better by simply rewording the communication, may have looked like this:
“Please enter your zipcode and press # on the keypad bellow”
If you aren’t going to write out ‘ENTER’ and simply use ‘ENT’ on the pound key (#) that now serves a multiple function, then why bother having an enter key at all. Either add a physical key that serves the singular purpose of being ‘ENTER’, or make sure to properly communicate the design flaw to the user to avoid any confusion.
This brought up a couple great points that designers of physical computing should be aware of:
- Don’t assume a task is obvious, although it may be obvious to you it may be confusing to someone else.
- Sometimes you may be forced to use components that are not ideal, but meet the need in other ways such as budget, time and convenience. Do your best to make sure that any confusing interaction is properly scaffolded to minimize error and avoid frustration
Popularity: 1% [?]
If you assume the designers are intelligent, trained designers, and this wasn't an oversight, what other reasons might there be for this choice?
Before I started my career as a programmer, i was a graphic designer, so I am all to aware of the external influences that may override some of your original design decisions. The decision on selecting that specific keypad could have been due to a variety of issues, such as budget constraints, manufacturer agreements (politics), compatibility issues between sensors, or available physical space for the component.
Another possibility is that the MTA contracted out the design to 2 separate designers or companies and there was a lack of communication between both parties. The touch screen interaction may have been contracted out to an interactive agency who specialize on screen based interaction design, and the casing may have been contracted to a company with experience creating physical interactions. Individually both solve their specific problems, but once integrated there is some disconnect.