4 Interaction design activities and methods
As a result of the increasing variety and complexity of interactions, and in order to address the kind of questions discussed in Section 3, designers need to employ an ever-expanding range of methods throughout the interaction design process. In this short course, we briefly introduce these methods and, using a few examples, we show you how they are used in establishing requirements, producing alternative designs, prototyping some of the designs and evaluating the prototypes.
In the video below, Sarah Wiseman discusses the design of the telephone interface. The video traces the history of phone dials and keypads, and how in the 1950s researchers used a variety of methods to select the configuration of keypad numbers still used today.
Activity 6 Why are keypads the way they are?
Timing: 1 hour
Watch ‘Phone buttons’ below and then answer the following questions:
Download this video clip.Video player: tm356_2016j_vwr003_640x360new.mp4
Transcript
SUBJECT
You know on your phone, the buttons are laid out one to three on the top. Have you ever thought about actually why they're like that? It's not a chance at all. A lot of work back in the '50s went into deciding how phones should look at. And it actually was all decided just in this paper, really. This is a technical report from Bell Labs. They were deciding what the phone should look like now that rotary dials were no longer the only way to dial a phone number.
As touch tone came into things, they had the opportunity to design what a push button phone should look like, because no one had done it before.
INTERVIEWER
Some of our younger viewers might not be familiar with it.
SUBJECT
I'm not familiar with it either.
INTERVIEWER
Didn't you ever live in a house with a rotary?
SUBJECT
No, of course I didn't.
INTERVIEWER
I did.
SUBJECT
Well, that tells us something about our ages, doesn't it? And you stick your finger in the hole and then you can-- you have to drag the dial all the way around and then let it go. And it-- buzz-- back. And you have to do that for every phone number. But in the '50s, they were realising they had the technology. You didn't have to do that anymore. And you could actually just press buttons to get tones to make the phone calls to dial out. And so they were making a phone with buttons on it.
The first thing they thought was that they could just put the buttons in the same place where all the dials were. Well, it's OK. People can use it. But there's no reason it has to be like that. So why keep it like that? Why not take the opportunity to have a bit of fun and be creative with things? Basically, they don't know if this is the best design, either. This is just a design that people are used to. What if there's another design out there that people are better at using?
So they thought, this is one design. But there are a whole host of other ways we can design the number layouts. And so here we have a set of the designs they tried out. So this one is the-- this is how calculators are laid out. Or in this paper, they refer to adding machines. Or laying them out like a brick wall, I guess. I don't know. Or like a bowling alley or something like that.
That one's quite insane, isn't it? Yeah, kind of heading up the machine. I mean, who knows what they were thinking. Like I said, they just had the opportunity to try anything out. But you do you have to remember at this point, the touch times phone wasn't a common thing. Like it may seem to you and I that obviously they should use a three by three grid. But only people who had used adding machines and calculators at that time had any experience of how numbers should be laid out. It's not a common thing.
They got loads of people in from the labs and they just got people in and made them dial phone numbers. It was as simple as that. They were very good at testing human factors at Bell Labs. I'll tell you a cool story.
They were trying to work out how long the phone cord should be and the copper in cord is quite expensive. So they wanted to try and work out how short they could make it. And the way they did that, the person in charge of the human factors department, went into everyone's rooms and started cutting the cable shorter and shorter and shorter until someone went, hang on, what's happening here? And then that point they were like, that's the shortest it can probably go.
They were really good at doing experiments to find the answers to these things. Yeah. So they got people in and they got people to just dial phone numbers on all of these different layouts. And they were testing for how quickly they could do it and how often they made errors when they were doing.
INTERVIEWER
What did they find? What happened?
SUBJECT
Well, I think you know the ending, because you know what a phone looks like now. But they kind of whittled these ones down and they came up with these five final designs. You see they've given them some interesting names. They've called this one here speedometer, which is quite funny. And so you can see there they're trying to test how quickly they go, how quickly people can type on them, the percentage of errors that people make on them and how people preferred them.
And you can see there's kind of a trade off here. So one of the slowest ones was actually one of the most accurate ones. And that's a common thing we know. If you go slower, you're more likely to be accurate. But actually there was no massive difference in the timing and the errors. And so what they did was they started to look at people's preference and which ones people hated the most.
And because there was not much of a difference, what they went with was the one that would just-- for engineering reasons, they say-- they went with the telephone layout. What's perhaps the most interesting thing, I think, is that they did compare the thing we're calling the telephone layout and the calculator layout. And they found that the calculator layout was slower than the telephone layout.
This paper-- this is from 1955, so five years prior to that one. And they're trying to work out-- they're starting to think about the buttons. They're not testing the actual machines yet. What they're doing in this paper is giving people blank layouts. So they're giving people this. They're just giving people a booklet that just has this on it. And then they say to people, you have to put the number nought to nine in that grid. How are you going to do it?
And again, remember people haven't seen a telephone before and few people have used a calculator before. And it turns out when people filled this in, 55% of people chose the phone layout. 55% of people went, well, obviously it should go like this. You start counting from the left hand side. Only 8% of people did the calculator layout. 7% of people did this fairly peculiar layout.
So they started with one. Yeah, obviously they start with one. But then what they did was they started to fill them in vertically like this, which just baffles me. I'm not sure why that's the most sensible way to do it. Like I said, you use a telephone so often. And you just don't really think about why it should be like that. Just because you've been, perhaps, brought up with it, you just like, obviously that's how the numbers are laid out. But there was a time when people didn't know how numbers should be laid out. And so they had to do all of this type of experimentation to work it out.
And it's a really good example of how human factors matter in design. They assumed this rotary would be the best. But it didn't up being that way. And because of what they learned from experiments, we now have a telephone that's optimally designed for us.
INTERVIEWER
We'd like to thank audible.com for supporting this video. Audible has a huge range I think of something like 150,000 audio books you can download. And you can download one for free if you go to audible.com/numberphile. After watching this video, you might want to check out a book about Alexander Graham Bell or something about inventions. Or check out one of my all time favourite books. It's called The Mote In God's Eye. And it's about sort of first contact with an alien civilisation. And I could talk about it for hours. It's really cool and really clever.
So thanks to audible.com for their support of us. And if you do you want to go and have a look, maybe sign up and get that free book, you're probably doing us a favour as well. That's all for now. See you next time.
SUBJECT
But you have to balance out accuracy with speed. The ideal number entry interface, in terms of making the most accurate one, might be one with the numbers spaced like a foot apart or something like that. So you have to physically move to press the buttons. That would be quite accurate, probably. But it would be incredibly slow and really annoying.
Interactive feature not available in single page view (see it in standard view).
- a.How did researchers test which keypad configuration was best and what did they pay attention to?
- b.How did researchers determine how long a phone cord should be and why did they set out to do that?
- c.What were the two aspects researchers were focusing on when testing for different number arrangements? Which of the usability goals we discussed in Section 2 would you say best describes those two aspects?
- d.Besides usability goals, did they focus on any user experience goals for the layout? If so, which one(s)?
Discussion
- a.In one experiment, they tried out different layouts and tested people on how fast they were on each layout, how many errors they made and also asked which one they preferred.
In a different experiment, people were given pieces of paper with layouts of empty places and were asked to fill in where each number should go. This showed that 55% of people felt it was most natural to go 1 2 3 on the top horizontal line.
- b.The researchers gradually cut shorter the cables of the phones of their colleagues until the colleagues started complaining. Then they knew they had found the minimum length. They needed to know this because copper is an expensive metal and was part of the cable.
- c.Researchers were looking at how fast people could enter numbers in the different arrangements, and how many errors they were making. The first aspect corresponds to the usability goal of efficiency, which is about how quickly a user can complete a task; the second aspect corresponds to the goal of safety, which is about preventing users from making errors.
- d.Yes, they also wanted to know which layouts people preferred. This can be described as a user experience goal focusing on qualities such as ‘what is the most natural’, or the ‘most intuitive’ layout.