Five Questions With Bill Buxton

Bill Buxton is a Principal Researcher with Microsoft Research. He is rather famous in the human-computer interaction research world. His book "Sketching User Experiences: Getting the Design Right and the Right Design”, published a few months ago, is one result of his thirty-plus years of investigating how people use technology. As you might infer from the title of his book, Bill is big on using simple methods to quickly prototype UI and other ideas. For example, if you were faced with the task of demoing a concept for improving the visibility of the mouse cursor as a presenter roams around a conference room, would you think about using Windows Movie Maker to stitch together a series of screen shots, and then coordinate the resulting movie into a live "demo"? Simple, no? Yet quite effective.

I mentioned that Bill is well-known in the HCI world. As in: his work has garnered him The Canadian Human-Computer Communications Society Achievement Award. He is well regarded in other worlds too: he was named the New Media Visionary Of The Year at the 2000 Canadian New Media Awards, and the Hollywood Reporter has named him one of the ten most influential innovators in the North American film industry, and the Ontario Horse Trials Association named him the 1996 Veteran Rider of the Year. Not exactly your typical computer research geek is he? Here is what Bill has to say:

DDJ: What was your first introduction to testing? What did that leave you thinking about the act and/or concept of testing?
BB: The first product I bought. I thought I was buying a product. What I was really doing was paying for the “privilege” of testing version n so that the producer could fix the design flaws in version n+1. Things haven’t changed much.

DDJ: What has most surprised you as you have learned about testing/in your experiences with testing?
BB: That testers are still spending time finding flaws that are there because of lack of design. It would be wonderful if they were there as quality control to ensure that a great design was executed greatly. That is almost never the case. Hence, they scramble to try and weave the proverbial silk purse out of the sow’s ear.

DDJ: What is the most interesting bug you have seen?
BB: Perfectly executed code that did exactly the wrong thing. Civil engineers do not get awards just because their bridges don’t fall down, and rather last for generations. That should be taken for granted. So should it be for software. That should be the starting point, not the goal. The objective, and the thing worthy of award, is a great design, beautifully executed. I’m still waiting.

DDJ: How would you describe your testing philosophy?
BB: Don’t waste the tester’s time on bugs that should never have occurred in the first place. Let their considerable talents be put to good and appropriate use.

DDJ: What do you think is the most important thing for a tester to know? To do? For developers to know and do about testing?
BB: The importance of design. That it is about getting the right design, as well as the design right. Just as a builder will likely come up with better results if working from great plans, so could it be with software. It never is.

DDJ: Is there something which is typically emphasized as important regarding testing that you think can be ignored, is unimportant?
BB: Code reviews are not a substitute for what in the design world is known as the critique. We just try to refine code rather than seriously evaluate alternatives.

DDJ: What do you see as the biggest challenge for testers/the test discipline for the next five years?
BB: To move a key part of design to the front end, prior to engineering, so that we know what we are going to build, that it can be built, how long it will take to be built, how much it will cost, who should build it, how it will be built, etc.

DDJ: Going meta (to channel Jerry Weinberg), what else should I ask you? What would you answer?
BB: What we in the software industry should be asking ourselves is this: if our process is so great, and not needing a total overhaul, how is it that there is not a single company in the software industry that has a track record of producing new, as opposed to n+1 products? Our process sucks.

DDJ: Is there anything else you would like to say?
BB: Time to toss out the status quo. Current product development process does not work – extreme programming and agile development not withstanding.

Posted by The Braidy Tester at 07:30 AM  Permalink