A generous interface for Roman stonecarving
Posted: February 6, 2013
The work on the Art of Making website has been gathering a head of steam of late–we rapidly moved from wireframes and proof of concept mockups to functioning prototypes following the appointment of a talented developer.
I’ve just carried out some usability testing on the prototypes and thought it might be an interesting point to reflect and consider what we’ve been trying to achieve.
My main inspiration for the interface for the project came from Mitchell Whitelaw’s recent work on generous interfaces for cultural collections. In short the ideas are primarily to present content to the user up-front, not to hide it behind search boxes or esoteric lists of links; and to combine some elements of data visualisation into the presentation of content. In some ways the former element isn’t so original–e-commerce sites have been doing this kind of thing for a long time now, although it’s been rarer in the area of online digital cultural collections. The latter is particularly interesting however–check out Mitchell’s TEDx talk for an introduction to some of his work to see ways in which the use of data viz techniques might present ideas for research paths, e.g. by highlighting commonly occurring words and their relationships with other words in online archival collections.
I was keen to use this principal, partly because I personally find search/text interfaces rather dry and also because the project is based around a digitised collection of slides–the Roman stonecarving collection of the sculptor Peter Rockwell. The collection is visually very interesting, from photographs of entire Roman monuments down to more detailed views showing textures and evidence of particular carving techniques. It seemed a shame to hide these behind traditional “stingy” faceted browse and search tools. I wanted to present selections of these up front to users to encourage engagement with the collection and serendipity (whilst of course still preserving the ability for more advanced filtering of images).
Although we haven’t gone too far down the data viz route yet, by presenting selections of content up-front to users (e.g in a randomly selected group of images on the main “Explore” page and in groupings of related images on more detailed pages) I hope we may be encouraging greater exploration of the collection.
I carried out typical task-based, think aloud usability testing on the site which was of course useful and has provided us with plenty of feedback for the next design iteration. However I’ve often wondered about how this kind of testing can really answer the questions we may have about how engaging this kind of interface is. After all, the idea of generous interfaces could go beyond the usable more into the realm of emotional design, something which is more difficult to evaluate. It’s a question I asked Mitchell himself in an email a while back and was interested to learn he’d also been thinking of suitable methodological approaches to testing. In this reply he stated,
“…with a mixture of different modes – eg web analytics, ethnographic style observation, interviews with domain experts and lay users, as well as “real world” impact and influence, etc., we could build up an impression of how these systems work.”
Maybe you could call it selective data gathering and it’s certainly anecdotal, but I was pleased to hear some of the testers in our usability tests being pleased with some of the results they were seeing on screen, even if it was by way of just a “that’s really cool” type comment. Also, in the post-testing feedback questionnaire asking participants about their favourite aspect of the site there were favourable comments about the selection of images. I don’t have any answers to how we can rigorously evaluate these types of interfaces but I suspect it will mainly need to be qualitative and done most likely via some kind of contextual enquiry, observing how users create their own navigation paths, and possibly the kind of language they use in response to what they’re presented.
No comments? The comments haven't exactly been coming thick and fast recently so I've disabled them for the time being. Instead, why not drop me a Tweet at @littlednet or email me at email@example.com?