Romancing the Interface
I’ve spent the last week or so with Brian Eno and Peter Chilvers’ new iPad app Scape, and I have to say that I’m impressed – both by the application itself, and also by the experience of using it.
The video explains what’s going on quite well and there are a couple of good pieces about the app and its creators here and here, but I’ve found myself thinking a lot about the interface and what is not explained, or by the relationship between the iconic representation of elements in Scape, their function, and the process by which I’ve come to some understanding of what’s going on (My thinking here probably has some of its origins in our being focused on the idea of discoverability during the luge ride that was Max 6 development, but I’ve been fascinated with how trying to make sense of the program has slowed me down in a rewarding way).
On one level, the app certainly is discoverable in the sense that there are intuitive models from other software threaded through the design that make it “easy” to use. But I’m interested in the parts that aren’t efficient in the “Let’s get this interface stuff out of the way so that we can get down to producing things” sense.
I think that this application shares some features with slow food in that there’s an admirable sense of reward that originates in process – interacting with the application over time, and learning by going where to go.
In the case of Scape, that winds up being about looking and listening in real time. There are no pop-up quickie hints, no chatty paperclips. You will only figure out how things work by working with them – watching and listening. The elements themselves invite scrutiny about representation – both in how they look and in the way that they’re animated onscreen while they’re “running.”
From the very first, you start wondering about what the relationship between the little animation you see and what you hear is, and about what the graphic space in which you’re working represents. If you skootch similar elements so that they’re in close proximity, their size alters sometimes. What does that mean? When elements approach the edges of the screen, they shrink or vanish. Huh?
All that might lead you to believe that there’s a simple one-to-one relationship between what you see and what you hear. But that’s not quite true. Some elements, when placed, appear to “stay out of the way” of similar elements. As you watch and listen, it becomes clear that there are other more subtle rules that govern the interaction between different elements – but it’s something you’re going to hear rather than see – you need to trust your ears. While inefficient, I’ve really enjoyed the process.
The pleasure of the relationship continues as you use the application: the next time you launch the app to create a scape, the interface may open to show you that a new icon has been added to your palette. What does it do? Only one way to find out – drop it into a blank scape and listen. Add another element and see if things change. Choose a new palette on the right and see how that effects things. You can leverage what you already know for this exploration, but it’s the same iterative slow-time activity. And the new element is always a small surprise to open and explore.
I expect that there some users who’ll be driven to howlin’, fist-shaking fury at the way that Scape doesn’t explain itself (or the way it isn’t documented). I’m sure they’re dedicated and clever individuals for whom the time necessary to live with the app and to develop some kind of personal and inner map of what they think it does (acts which I’d say live at the corner of Idiosyncracy and Virtuosity) might seem a waste of valuable time if they’re focused on doing the pragmatic thing and wanting to just “get on with making Enoesque audio” (The good news for them is that I expect the “random” feature of the app will be as good as they’d be without the investment of time and attention and the effort of listening. Maybe better. Or it’ll at least save a lot of time – simply punch the button and listen for 20 seconds, and then maybe toggle the moods on the right and stardom will certainly follow).
But I’m seriously entertaining the thought that this application is interesting and reasonably unique precisely because of the way it’s set up to encourage developing a relationship with its interface that’s reinforced by the temporally bound experience of listening, and because of the way the application unpacks itself over time in a way that encourages the continuation of that relationship. I’ve never run into anything quite like it.
By now, my version of Scape has what I think might be a full palette (although I’d love to be surprised again), but the thing still engages me – for example, are my intuitions about how the elements interact anything more than personal ones? I expect that the answer is either “Not at this point,” or that I’ll eventually develop enough facility that my idiosyncratic readings will constitute my “style” when working with Scape. Maybe that personal or internal map of things is somehow the point – instead of something defined for me, I’m writing my own internal manual for the interface, and reworking the documentation with each new piece I get as I work with it (Some things have and will probably completely escape me – I note that Eno says that some elements play more sparsely based on time of day, which I would probably have been the last person in the world to figure out – I have spent my time with it in the evenings only, so far).
And, as an Oblique Strategies afficianado from waaaaay back, how can I not like the Scape Strategies?
To finish on a note more directly related to Max, I don’t think it’d be particularly difficult to add some of what I think I’m hearing in action to the Max patching I already do, honestly [another reason I’ve found working with it to be a salutary experience – the ideas aren’t that difficult or subtle; their implementation is]. Obviously, as a programmer I am going to be less successful at surprising myself – but I do think that considering probabilistic interactions outside of the boundaries of an individual bpatcher – between what I think of as the pieces of what I’m using to create the larger whole – may yield some interesting results.
And, since it might be that I’ve learned an interesting lesson, that process may… um… take a while. A good while.