It’s easy to look at a piece of technology and say it’s got lots of features. Deciding whether it will be usable is harder. We could get a bunch of volunteers and try it – but think of the problems: getting the volunteers is hard enough, but even when you’ve done that, all you’ll find out will be the worst problems. Even worse, you’ll have to actually build the technology before you can test it.
So people have developed a number of predictive approaches. They vary greatly but they all try to predict whether technology will be usable, preferably without needing real volunteers and if possible without needing to build the real technology.
Some approaches focus on one type of technology. That’s not enough; we need an approach that works for all types of technology, including non-interactive systems (notations) as well as the more high-profile interactive systems.
You’re still not out of the wood. So you discover that your technology is problematic. Is it problematic in the same way as some other piece of technology? Will the same solution be relevant to both of them? We need a standardised vocabulary, a few words describing familiar good/bad points.
OK, nearly out of the wood, the trees are thinning out, but we’re still not quite there …. When you’ve got a vocabulary, what of it? Will a potential usability problem be relevant to the particular use you envisage? We need a standardised, high-level account of what technology can be used for, and we need to relate our vocabulary of good/bad points to the potential uses.
The Cognitive Dimensions framework does that, and is still unique in doing so, we believe.
The CDs framework provides
• a standardised framework of ‘cognitive dimensions’, each relating to usability experience,
• where each dimension is (more or less) independent of each of the others,
• applying to every kind of information artefact (i.e. anything used for storing or manipulating or accessing information – that’s most non-natural things);
• a background in the experimental literature of cognitive psychology relating to each of the dimensions;
• a high-level classification of the activities people get up to with technology;
• and an attempt to say, for each activity, which cognitive dimensions are critical to making that activity successful.
The last point is crucial. No feature is good or bad in itself. Until you know what it’s to be used for, you can’t say whether a device is usably designed or not.
One other point. The dimensions are not brand new discoveries. On the contrary, they will all sound very familiar from your own experience – but you probably haven’t named them before, let alone built them into a system. (If you have, let’s hear about it, please!)
Instructors:
Luke Church (Cambridge University)
Thomas Green (University of Leeds)
