Huge thanks for Bill Buxton (www.billbuxton.com) for his excellent talk this week, the second Cambridge UPA talk. We are setting a precedent for having excellent UPA talks here in Cambridge! Huge thanks too, to Bill and his colleagues at Microsoft for arranging everything including the location for the event (Microsoft Research) and refreshments. This talk drew around 50 people from the local usability community.
My talent for reviewing is rather unworthy when it comes to describing a talk of this quality but here's some of what I took away from it.
Well, firstly I took away a handful of extremely impressed work colleagues for whom Bill's talk has resonated strongly, one in particular will certainly never be the same again!
Bill is clearly a designer with a great passion for his subject and is a top rate speaker - well recommended for any after dinner speaking in my view. His background in music served to emphasise that 'right on' feeling when a user experience is just about perfect, as with his Orange Juicer with a neat camming action to effortlessly squeeze the last from each orange in a moment of almost musical perfection, an excellent story, well told (Juicer Article).
Like others, Bill makes the argument that companies need to take design more seriously. Design needs to be an executive position, with the CEO right behind it. Steve Jobs has proven that this can work, when the 2nd day in the job he announced that he would save Apple through Design. The latest range of Apple products, including the iPod, prove what Design can mean to a company, and yet in many companies, Design is still not an executive post.
Bill talked a lot of the software industries predisposition to n+1 products, we rarely radically innovate and some products here today have been here since the 1980s in a similar form. As we work on these n+1 products, complexity and costs go up but the market for these products goes down. Bill asked the audience how many people were on the latest version of Adobe Acrobat - few were. In fact I just upgraded from version 4, and only because a usability report I wanted to read demanded I upgrade before I could access it!
Bill explained that much software product development begins with a block of engineering, followed by sales (this needs a diagram to explain really and I've just found out, you can find these diagrams online Performance by Design: The Role of Design in Software Product Development). He argued that before the engineering, what is needed is a design stage. And a design stage where early on it is cheap to try out lots of ideas and easy, even preferable, to make mistakes (if you don't, you are not living close enough to the edge). He refined these ideas to suggest that the design phase breaks into 3 pieces, 'pick', 'cull' and 'evaluate'. And even by the time you get to the end of the 'evaluate' phase, your costs are still quite low and at that point, you can have a rational discussion on whether to proceed or not, before you have committed a huge investment. Bill contrasted what we do in software with the motor industry, where, by the end of their design phase, they build a full working model of a car BEFORE they decide to go ahead and build this car for real, and the model is so good, you wouldn't know it wasn't the real thing. Why don't we do this in software? Bill explained in the motor industry, 'forms' are used to capture the essense and shape of a new vehicle and believes this kind of living specification is the way to specify software. He explained that early Apple software Mac Write and Mac Paint were pretty hopeless as bits of software but as mini applications served to convey what other applications should be like.
Bill talked about the need for a "Green Light Process" for designers to work to. Find out up front how they are to be measured and be prepared for their design to be assessed against this Green Light Process, to become accountable for their designs.
Bill then talked about the radical evolution of mountain bike design (a favourite topic of mine too). He showed how engineers came up with the design for Trek's first rear suspension bike, and then how they hired a fairly junior industrial designer who really made it into something you might like to own! At this point in the talk, I think we were too polite to challenge Bill, since he had just said "Design is first" and now he is saying engineering happened first here. At that point he extended his software process description, explaining that R&D, including engineering and Ethnography, were important forerunners of the design phase - and this made a whole lot of sense.
Bill had some firm ideas on what usability people do, stuff involving statistics, and not really designing. Chatting later it was apparent that Bill's 'usability people' were meant as those focussing on usability testing. I explained that in many contexts the 'usability person' is a general term meaning a potential range of skills from ethnography, to design to usability testing - often, a 'usability person' is a software team's best chance as doing some decent design work. We agreed that 'Interaction Designer' or 'User Experience Designer' would be better terms for doing the kind of design work Bill was advocating.
In terms of expectations, Bill explained that when he walks over a bridge he doesn't feel the need to thank Brunel that the bridge didn't collapse, or when flying, doesn't see the need to thank Boeing that the plane didnt fall out of the sky. And similarly with usability, getting it right is a basic expectation it is not something worthy of any kind of praise at all. It's just sad that so often it is an aspect that is lacking. Instead, it was clear from the Juicer example, that what we need to get to are what in other circles are called 'delighters' or 'wow factors'.
Bill provided a good demonstration of the longevity of software and the importance of considering things like Moore's law when you are designing the next big bit of software - you need to predict the change in hardware over the time in which you are developing the software and make sure your architecture deals with it. He was keen to point out that huge, flexible, thin and cheap screen technology will be with us in the near future and we should be planning for it now (only 5-7 years away). Bill's discussion on plasma screens and other predictions for the future can be followed up in his TIME magazine article: "Forward into the Past - Here's a safe bet: the next big thing is already here".
During the questions at the end, Bill explained that back in his days as Chief Scientist at Alias, he thinks he should have spent more time on fixing the organisation to get it structured right to do good design, rather than focussing so much on design himself. He suggested spending 80% of his time on organisational aspects, shaping the people, culture and environment, may have served them better.
Bill is currently a visiting researcher at Microsoft but is returning to Toronto at the end of May, where he has his own consultancy, and can be reached through his web site. He's also writing a book right now, in which he will elaborate further on some of the ideas presented in his excellent talk.
Huge thanks again to Bill Buxton for supporting our Cambridge UPA group and providing us with some excellent thoughts, rants and ideas which will stay with many of us.
Best wishes in your return to Canada Bill!