On the weekend I was fortunate enough to visit the National Museum in Canberra. Didn’t get to see any of the exhibits, I was too busy being bowled over by the building. It was a delight. Playful, imaginative, energetic. Downright funky. Melbourne’s ARM architects (the guys behind Storey Hall and the Shrine extension) have done us proud again. If you are in Canberra it’s well worth visiting.
One of the many features is the Garden of Australian Dreams – a central interpretative courtyard representing our multi-facited history and aspirations.
You can see in the picture the wading pool. The edge is actually the top end of Australia. Yep, layers of symbolism. And I’m sure the designers were really proud of how much they layered into this area.
But then this happens:
The design is built and launched and then sometime later someone either loses their nerve in the face of litigious fear or little Caleb falls into the water and a complaint is made. The result is this awful sign that detracts from the overall design, sticking out like a sore thumb.
It’s sad, but being realistic and pragmatic, we have to expect this sort of thing as designers. And design for it up-front. And sometimes that means compromising the integrity of the “design vision” to ensure it survives real life wear and tear. They could have just moved some of the “islands” a little closer to the mainland so they were within stepping distance. No need for a sign then. Sure, it wouldn’t have been technically correct. But it probably doesn’t matter. And it would certainly be less noticeable than the over-protective signs.
What’s are our over-spaced islands? We face them all the time. Great design will anticipate them and design the compromises rather than letting compromises just happen to the design once we step away.
Every time we add a feature or an option to our designs there is a cost. The cognitive (plain-talk: thinking) load for the user goes up. Sometimes the benefits of that feature or option make it worth it. Often they don’t. And yes, as smart designers we know some of the best ways of representing those features so they have a minimal mental footprint, but it is still a trade-off we need to consciously make.
And unfortunately (or just realistically), we are often the only ones carrying the Let’s Keep It Simple can. Product people can just be myopically focussed on their particular feature and want it get it into the product at whatever cost to the overall user experience. But that’s partly their job and their perspective. What we bring to the table is that overall perspective for how best to arrange (or remove) the pieces to make the whole experience sing.
But it’s not just “evil” Product trying to load up our pages with all manner of things. We have to fight that inclination ourselves. One of the cold realities of design is that it is easier to add than subtract. It has to be our mantra to be always saying “how can we simplify, how can we simplify, how can we simplify…?” It takes huge discipline and a whole lot of thought.
And guess what? We don’t have to give users every option under the sun – something for every possible situation. Sometimes (and often) that just costs too much – too much time to develop, too much time to user test, too much time to specify, but definitely too much time to comprehend.
And interestingly, we are a lot better at seeing that in the real world than online. There is a point where there are more Chiodo shirts than can fit in my cupboard, no matter how much denial I want to wrap around that reality. The real world has a way of enforcing “what? are you crazy?” constraints. In the virtual world it is all too easy to just keep extending the page, adding another tab, designing another dialog, adding another level to the page structure.
37signals, defenders of simple, not surprisingly have some good things to say about this topic in their recent “Outsourcing Choice” post.
One of the great things about design is how diverse the field is. Yes, in our online design world, we think directly about interaction design and visual design, with bits of information architecture thrown in at times, but our influences can be so varied. Things like industrial design, architecture, fashion, set and lighting design. The stimulating common thread is there is creativity and expertise being thrown at tricky problems to find deeply satisfying solutions.
So its exciting to see the intersection of various design professions that allow us fresh insights into how best to solve problems. Take the design for the JetBlue passenger terminal at JKF Airport in the US. In designing the “interior experience”, architect and set designer (David Rockwell) hired a choreographer (Jerry Mitchell) to arrive at new ways of understand the movement of people to ensure the design did in fact produce a satisfying experience.
What was also interesting was the way they analysed public spaces in New York to determine exactly what did and didn’t work. It’s something we can always be mindful of. Often people come to us and say “let’s build one of these exciting thingies, just like ”. Rather than taking that at face value, what we can bring to this is the abstracted analysis to draws out the elements that work, why they work in that context and the principles that could be applied to the design challenges we face.
Here’s the article and look for the “audio slide show” link, which gives you the story with pictures.
I’ve never been a big fan of the Print-Friendly version link or button. They clutter the page. You get another version in another window and guess what? You have to select your browser print function anyhow.
Better to cut out the middle step and just let users press the print button. Do the “hard” work for them. Have a print-specific style sheet for the page that does it all for them.
I’ve been printing from Basecamp recently (ToDo lists specifically) and that’s the approach taken there. Super simple. Great format.
Here’s an article that starts a thoughtful exploration of something I’ve been pondering over the last few months: that great design and particularly providing design solutions to Sensis cannot be done without participating in a full discussion about business factors (eg. revenue and business models) that will play a big part of a site’s (and the design’s) success.
There is a tension between what designers can affect and what we have no control over. Because we can’t always change the ultimate success or failure of a project, we often dismiss the success or failure as completely outside the work we’ve done. I think this is a short-sighted, dangerous way to approach design. If we consider our role as designers outside of the success of the project, then we’ve discounted the value that we could potentially bring to the table.
We can’t just offer up a beautiful design and think our job’s done. This article ponders such designer conundrums as:
Designers have a tough time dealing with ugly design that works well…They say things like “MySpace succeeds despite its design“. (same thing happens with Google). However, this doesn’t address the success of the project: nobody can argue that MySpace isn’t an all-out success. Same with Amazon. Same with Google. [BC: Not to mention eBay] These sites don’t even write valid HTML, for gosh sake. But they’re wildly successful…the most profitable and beloved sites on the Web. I added that “beloved” part because it is absolutely critical. You can’t talk about Google and Amazon (and probably MySpace) without people saying they love them. How can we then go on and say they’re not well-designed? I think being beloved might be the very definition of great design…at least that’s what I aspire to.
I liked this blow-by-blow about how to explain the different ways to enter a date and time in Backpack.
It may look like a long and laborious process. Expensive even. Sure, it is when you writing it out in full like this. But it doesn’t have to be if it is just part of how you think. If everything has some thought behind it and a reason for being, then it is just part of the design process. You notice when the words look too loud and noisy on the page. And as this example shows, so often there’s a real balance between clarity and brevity. Cracking it and judging it are the key.
And my theory is when all those elements are well thought-out and just right, the interface sings. Users glide through their task. You hit the illusive “intuitive”. It feel just right in your hand.
Here’s an article about my favourite designer, Jonathan Ive, Snr VP of Industrial Design at Apple.
What I like most, besides the perfectionist streaks, is the glimpses into their design team process and culture:
The man who, after Jobs, is most responsible for Apple’s amazing ability to dazzle and delight with its famous products, chose instead to talk about process—what he called “the craft of design.” He spoke passionately about his small team and how they work together. He talked about focusing on only what is important and limiting the number of projects. He spoke about having a deep understanding of how a product is made: its materials, its tooling, its purpose. Mostly, he focused on the need to care deeply about the work.
“One of the hallmarks of the team I think is this sense of looking to be wrong,” said Ive at Radical Craft. “It’s the inquisitiveness, the sense of exploration. It’s about being excited to be wrong because then you’ve discovered something new.”
Here’s my vote for worst identity management (IdM) implementation: Medicare
I’ve recently moved house and have endeavoured to do as much of my change of address notification online as possible. Medicare offers some online services, but you have to register to do this. Registration isn’t too hard. You already have a unique username – your Medicare number, but to prove your identity they snail-mail you your temporary password to your registered address. Not great if the reason you are registering is to change your address, but I can live with that.
Problem #1: Change Your Password
A few weeks and a mail redirection later and I have my temporary password. I open up the Medicare site and see the following options (this post has lots of pictorial examples – if you are reading this from the email, you’ll have to click into the posting on the website):
I’m thinking “yep, change my password because I certainly don’t want to keep this gobbledygook one they sent me”. But before I know it, Change My Password initiates the whole send-me-a-temporary-password-to-my-previous-address loop and it’s another 2 weeks until I’m ready to do the “right” thing.
Problem #2: 5 Secret Questions
Secret question and answer solutions are pretty common these days, but Medicare has gone for both the excessive and the open-ended. As part of setting up your account, you have to provide 5, yes 5, secret question/answer combos. But wait, there’s more: you have to make up the secret questions as well as the answers. Exhibit A:
Problem #3: Ts&Cs Every Time You Log In
OK, I’ve got my login, and correctly reset my password. Now I get the standard 2 pages of terms and conditions. Only I get this every time I log in. Sigh.
Problem #4: Last Access Info Yet Another Click To My Goal
I scroll though the long Ts and Cs, click OK, only to get another page that tells me my last access dates and times and yet another click to get to the things I actually want to do.
Problem #5: Higher Access to do Basic Things
Finally I’ve logged in and am at the main menu. I want to change my address. But those options are not links. To do that I have to gain higher access. What the?
Clicking on Higher Access Level take you to another screen, where you have to answer 2 of your 5 secret questions:
Then finally you have to enter your current address as further authentication:
Wonder what their take-up is for online services??
A great quality for a designer is that way of looking at something that’s become totally accepted and assumed and being able to ask “but how could we make it better?”
If you think about an electrical plug, you could think “they work – what’s still to do?”.
Well, for a start: which plug is which, and how do I know if it is on?
“Just make it simple!” It’s what many of us do for a living. But what design rules are there to help direct the path to simple.
There are some crude ones. “Everything on a site should be just 3 clicks away” has been around for a while, and has recently been given a Telstra make-over as “One Click”.
Steve Krug, in his wonderful book “Don’t Make Me Think!”, refutes this simplistic rule of thumb, saying that users would prefer 5 or more super simple and obvious clicks rather than 3 clicks that each take a whole lot of thinking. Sure, all things being equal, reduce the clicks, but not at the cost of added thinking and complexity (aka “cognitive load”).
John Gruber, of the insightful Mac site Daring Fireball, recently posted Deal with it, in which he echoes and expands Tantek Celik’s statement that, all things being equal, less user interface controls means a simpler interface.
His refinement goes like this: It isn’t so much the number of controls, but how many “things” on an interface you have to “deal with” to get your task done. Again, it’s about cognitive load. Don’t make me think. It’s not just keystokes or fields.
His example is the event entering interface of Mac OS X’s iCal (think: lots of separate fields for every element) and 37signals’s Backpack (a single type-anything-and-we’ll-work-it-out-for-you field).
He hints at it in his article, but one thing we have to be mindful of is the trade-offs between Efficiency and Learnability. For a commonly-used productivity application like a calendar program, you turn up the dial on efficiency and provide ways for a newbie to progress to competent intermediate as fast as possible.
(You can go too far on the single entry field, of course. You can solve world hunger and peace if you know the right words and symbols to type into a Google search field. Most of its extended use has almost zero learnability.)
But it’s a bit tricker with our websites - it’s more of a balancing act. Demote learnability too much and you’ve lost the newbies. Spell out everything every time, and your familiar user starts puckering up to the single field simplicity of a Google map. Would I rather spell out every field individually, or just type in “bunnings fitzroy” into a single field on White Pages and have it find the closest one to me (with an option to disambiguate and remember which Fitzroy I was actually meaning on the results pages)?
But back to “Deal with it”. Count how many things you are asking a user to deal with to complete their task. And how hard is it to deal with each one - how much thinking’s involved? What’s the room for error? And have we consciously made the trade-off between learnability and efficiency that’s appropriate for the situation?
If only simple was easy!
