Postcards from the Future
The famous science fiction author, Arthur C. Clarke, once famously said, “Any sufficiently advanced technology is indistinguishable from magic.” What I wanted to talk about today was the magic of designing those technologies. Designing technology has always been magic, but I think that now is a special time.
Not that long ago, when we would design software, we’d get it mostly done. We’d bring someone in to our garage, office or lab. I would watch them use the software. When they got stuck or swore, I would write that down. We’d fix that software usability bug. Then maybe we would bring someone else in and repeat the same thing. When they got stuck, we would fix that.
After a couple of times, when we ran out of money, time or patience, we would release the software into the world. You have no idea what happens when you release it out into the world. That’s changed dramatically in recent years. Today, if you log on to Google, Amazon, Facebook or Yahoo, you’re likely part of an experiment. Every day, designers are trying out different alternatives. They are seeing what things work the best.
I call this idea “design at large,” where it’s real-world at scale. We’re able to compare alternatives and learn from what we’re finding. What I wanted to share today was a couple of stories from my graduate students, some of the companies and designs that succeeded, and others that haven’t. I think that these successes are tremendously exciting. But they don’t happen all the time. Many startups fail. What we’re trying to do in the design lab at UC San Diego is to be able to build principles of what makes for effective design.
This is my former master’s student, Mike. In early 2010, Mike and his friend Kevin had a startup that was called Burbn. Burbn was an application that let you check in at a place and share a photo. They had a bunch of things in the design that they knew just had to be true. You only wanted to be able to share photos with a few close friends, because who else would want to see these things?
It was all place-based. You would check in to a spot and it would tally points. There were these game mechanics to get you up on a leaderboard. Burbn didn’t go very far, so Mike and Kevin tried again. This one you may know. Their next application was called Instagram. Instagram has more than 400 million users worldwide. They had more signups in the first hour than Burbn had in its entire history. What’s interesting about this is that Mike and Kevin were roughly right the first time. There is something about place, photos and mobile.
But they got a lot of the details wrong. When they started over, they tried a bunch of different things. One of my favorite examples of serendipity is, when they were building Instagram, they assumed like with Burbn, that close friends would be very important. You’d only want to share stuff with close friends.
They thought, “We’ll build that later. We’ll release the first version without any authentication,” so that anyone could see anyone else’s photo. As you know from Instagram today, the magic of being able to see celebrities, great photographers and random people in different locations using photos to check the weather, all of that is part of the magic of what makes Instagram, Instagram.
This came about not because Mike and Kevin were smarter than anyone else. They’re pretty darn smart. But it was because they tried lots of things. For a while, Mike had this thing where, every Friday, he would try out a new design by bringing people in and testing it out. He got more chances to try stuff than lots of other people did.
The second example is another former master’s student of mine, Akshay Kothari. This is the Pulse news reader. If you saw the early announcements of the iPad, you may have seen Steve Jobs demonstrate the Pulse news reader on stage. What’s great about Pulse is the Akshay and Ankit tried about 25 different prototypes early on.
They had no idea what was going to work. What does news reading on an iPad mean? They found that people really gravitated towards the much more visual ones. I talked to Akshay on the phone recently. He said, “It’s weird. Something that was literally a class project in six weeks is now something that serves news every day to 450 million people around the globe.” Pulse was acquired by LinkedIn and is now LinkedIn’s news source. Again, we see the same thing with Akshay that we saw with Mike. Trying all these different prototypes really fast enabled him to get feedback and figure out the thing that would work.
Here’s our third example. Many of you may know and love this. This is the Palm Pilot. In 2000, Palm sold nearly eight million of these. It was the first handheld computer that people really loved. How did they figure out how to get the right functionality in a small package with a useable interface? It turns out that the journey to eight million users begins with a block of wood. Jeff Hawkins, the founder of Palm, cut out a little piece of plywood in his garage. It was about the size that he thought the Palm might be. He brought it with him everywhere.
This is the magic of design. He imagined that a piece of plywood was actually a functional Palm Pilot. Jeff would go to meetings. He would take notes. If he needed to call someone, he would look up their information. The advantage of driving design by prototyping, as opposed to by making a list in the boardroom, is that if you made a list in the boardroom, you’d have way too much stuff. It would be slow, unwieldy and never work. What Jeff was able to figure out by trying and living with the design was the few key things that were especially important.
If any of you were in chemistry, you may remember Boyle’s Law. Design has Boyle’s Law also. The Boyle’s Law of design is, never go to a meeting without a prototype. Here, you can see when Microsoft first released Windows with a mouse, they worked with a design team to build more than 100 different prototypes.
You might look at this picture and say, “That looks like what a designer does.” There are 100 different forms that the designer made. I look at this picture and think, “That’s 100 different questions that the designer asked.” Each prototype is really a question. Should it be round or square? Should it have one button or two? Should it be the same size or different sizes? Ergonomic or stylish? All of these different questions that a designer might ask. When I look at this, I see that prototypes are questions. The way that we can all be like designers is to ask lots of them.
We’re in the middle of a campaign so I thought this would be a fun example from politics. In the 2008 campaign, Dan Siroker left his job at Google to help out a young politician who was running for president. One of the things that Dan and his team did was to try out comparing alternatives of effective designs with the Obama campaign. Some of the prototypes that we saw at the beginning were vastly different designs. I wanted to use this example to underscore that it can work for the small stuff, too.
Their original home page had about an 8% click through rate on the signup button. They wondered whether they could get an increased rate of donors, which is critical if you are a grassroots-driven political campaign, simply by changing the text on that button. They came up with three alternatives. You can take a guess as to which one you think worked. What’s notable is that everyone in this campaign was super smart. Not everyone was able to guess which was most effective. It turns out that the “learn more” button vastly outperformed the other three. This brings me to a point that, often we can use principles to be able to understand why design is effective. We can sometimes even use those to base our designs.
There is a principle that psychologists know called Commitment Escalation. You start with a small ask and then ratchet up from there. The “learn more” button is a small ask. You’re not saying, “I want to give money,” yet. You’re just saying, “I’d like to learn a little bit more.” Then once they have their foot in the door, they think, “Maybe I’ll donate.”
I’ve had the same experience with an online class that I teach on Coursera. We’ve had almost 300,000 people sign up for it. What’s really fun about teaching it large scale online is that you get so much more feedback about your instructional materials. The same experiences that Mike, Akshay, Jeff and Dan had, we’ve been able to do with our course materials. You put 300,000 eyeballs on an assignment, and you’re able to work out the bugs and answer lots of useful questions.
For those of you who are interested, here are some of the student prototypes. People make all sorts of cool stuff. Again, you’re seeing the magic in these rapid prototypes, like Jeff’s block of wood. We pretend that these are interactive systems to be able to try out ideas early, get feedback, and figure out the real problem to solve.
I’ve shown mostly examples of professional design. I wanted to close with something closer to home. On social media recently, I asked friends, “How have you used prototyping in your own everyday lives?” People came back with all sorts of stuff.
When I posted this, I had no idea what anyone would come back with, if anything. Every two or three minutes, something new started coming in. It was really cool. The first one that came in was that Maria prototyped her living room in painter’s tape and marked where things might go.
Uri prototyped a new iPhone holder. Eric had this really ornate prototype of a home renovation. Alexandra even prototyped her tattoos by wearing temporary tattoos for a month to figure out which one she liked best. That’s something you definitely want to prototype before committing to it.
To close, the prototypes that we’ve seen today, the magic of them is that they instantiate a future that doesn’t exist yet. As designers, we time travel just a little bit into the future with our prototypes. Once we know what the world is like there, we can send back these postcards that give information about what to make. Like the folks on the last slide, I think this is something that all of us can do in our everyday lives. Happy designing. Thank you.