Boxes and Arrows: Comments by Chris Roast

Chris Roast — Fri, 26 Jan 2007 14:52:54 GMT

We are currently using a tool designed to analyse the visual design of websites. The tool is not focused upon usage data but on whether or not the planned visual consistency of a site matches what users actually encounter. For a brief introduction have a look at www.shu.ac.uk/business/gist.

I hope to write an article about it one day!

Chris Roast — Fri, 26 Jan 2007 14:52:55 GMT

“Findability” for me is too focused upon a negative user experience – finding something implies having lost it or being lost. I asked my son whether he found his classroom every morning – his response was “no?”.

I’ve always liked the implied difference between tools that offer a “search” dialog and a “find…” dialog, the former is far more realistic and manages expectations, the latter implies both success and that you know what you are looking for (very high expectations of the user and the system).

Being rather idealistic, I would want to design a site so that people never have to find stuff (information or services), really they should experience getting the information or service – that is normally why they are there. So should we consider looking at “gettability” (the sub-concious cousin of “findability”!), or perhaps we return to a far more traditional concept of user satisfaction? Although it is rather un-innovative and very general it does capture most cases. Its generality is a problem, but that’s when we specialise to different qualities, such as whether the experience is more important than functionality. (We have to select based upon the high level objectives of the project.)

Chris Roast — Fri, 26 Jan 2007 14:53:05 GMT

Matt, I like the work, have you compared the M & P paths with other icon related work? There is work by Barnard and May on icon design that shows positive results for users interpreting icon structure (at the cognitive and not perceptual level). Their work seems to be concerned with user interpretation of icon after perceptual processing, however it is the next stage of recognition and discrimination.

Chris Roast — Fri, 26 Jan 2007 14:53:06 GMT

I like this idea, it seems to be a relatively efficient approach to “estimating” the level of real user involvement we’d like in most UCD. As in many cases it may not be the ideal approach but it is potentially far better than other alternatives.

One of our degree options concerns user centred design for technically adept students. For this we seek to ensure that the students encounter real users related to an assessed case study. Ensuring and maintaining an empathy with users can be particularly hard for the students. Roll playing is a good start but prehaps it should be seeded by an approach like “the method”. However in an educational context, the level of devotion required may just not be viable.

Chris Roast — Fri, 26 Jan 2007 14:53:13 GMT

Thanks Christina. Finding the right position, or role, for the GIST tool within a team has been hard, I try focusing upon design managers as you suggest.

Chris Roast — Tue, 14 Aug 2007 19:52:04 GMT

One comment on the interesting prototyping discussions. There is a merit to real paper-based lo-fi prototyping, especially early in the design process if we’re looking for user engagement and user experimentation. Actual paper prototypes facilitate and encourage design participation because of the naturalness of the medium. The client is often as much an expert with paper, tape and scissors as the designer. Prototyping with PP or PDF will rarely obtain the same level of client contribution, belief in and commitment to the product.

However there is clearly a significant step is when a paper based prototype is worked up to a digitial prototype (PP, PDF, HTML, or whatever). From the client’s perspective the scribbles and sketches suddenly become “real”. Although engaged in paper prototyping, once the client sees “their” ideas in a digital form they they then start to realise what user experience might genuinely be like. This suggests that early paper prototyping helps establish the key architecture details, language and expectations, however when considering the user experience the higher “fidelity” approaches are needed.