I’ve been writing a few (pending) pieces on user interfaces and social networks, and where they seem to be going. There’s been a lot of buzz recently about Schulze & Webb’s Olinda social radio project, which is a online listening software realised in hardware. It takes cues about your interest in a station by looking at volume levels; the idea being that the volume you listen at reflects your interest in that station.
Last.fm started out with a similar concept; listening to the entirety of a song was also a ‘preference’ vote (or that you were away from your computer, how can you tell the difference?), and that skipping a song indicated you didn’t like it. These are quite harsh metrics; there could be other reasons for skipping a song (you’re not in the right mood, you have a twitchy mouse finger) so it’s not necessarily an analogue for voting.
However, I’m particularly enamoured of two things happening here; user-generated metrics and passive data gathering. User-generated metrics is the idea that users and their behaviour in an electronic space provide very tangible feedback about their relationship to those spaces. The classic example is Amazon’s recommendation engine – people who bought that also bought this – but there are many abstracts ways this information can be gathered, and what it can mean. Passive data gathering is the idea that, like the example above, the way people interact with devices (and interfaces) provide a great deal of extraneous objective information, which could not be gathered should the same information was expressed in explicit terms. ‘Skip this song’ says so much more than a form which assessed your liking of the current song, and whether you’d like to hear it again in a week, month or year. Just skip it; if I skip it again, I don’t want to hear it.
It’s only recently that the Wii has allowed physical movement and real post-digital controls into video games. You don’t just move left at this point in time in the game, you move leftish quite quickly. For years, promotional videos of gaming systems showed players swaying, weaving, moving their joysticks through space, all the while just pressing buttons on a block – yet nothing capitalised on this. The body was making subconscious hints that everyone in the room was picking up on – apart from the game itself, the mediator of the experience. I think this is one of the main reasons the Wii has cut through such a broad target demographic – it’s attempting to use this subconscious cueing, rather than impose an abstract (buttons and joystick) interface on the player. It’s working the way your lizard brain works, rather than the way someone who has spent years learning, studying and practicing button configurations, as well as their repeat rates, give, tension and other inherent variables affecting performance would work. The lizard knows how to spin your arm 100 different ways already.
I spent some time, unsuccessfully, trying to introduce the idea of insistence into interfaces. People press a key, or click a GUI button, and the software (as best it can) responds. People don’t work this way, though; they act like software is another person or intelligent agent with emotions and judgement of it’s own. We press that ‘call lift’ button ten times more often if we need the toilet, if there’s an axe murderer behind us, or we’re generally in a rush, even though we know the metal and current behind the lift couldn’t care less about what’s happening outside it’s immediate world; which is a a binary world, a world of lift, shaft, and electric.
When I post a tweet, or send an email, if there’s a pause between clicking a releasing the send button, why is the recipient not made aware of this? Could we introduce an x-apprehension-level: high header to our mail clients, or x-decisiveness-level:HELLYEAH to those people who whack the send button repeatedly until the window closes? We see capitals as shouting these days, but what subtle nuances of typing are lost when the action becomes captured; did we hold shift with one hand and aggressively peck each character, an insistent finger-jabbing shout? Or did we carefully engage caps lock, and type normally, the shouting emotionless and for emphasis?
There’s so many places we can hook these cues and gestures in; Apple’s Newton had a lovely ’scrub out’ feature to erase items, engaging the same behaviour we use with paper and destructible surfaces to indicate deletion. Apple has started taking a leaf from their own work here, and are using flicks, squeezes and pinches to control their iPhones. But we’re still missing the adjective in our otherwise simple noun-verb HCI discourse; to a degree, who cares? People don’t really want to know how long I lingered over the ‘post’ button before you got to see this entry. At the same time, it’s an interesting metric of my relationship to this article, and taken over time, this posts’ relationship to other posts, and on a global scale, their relationship to everything else; you can suddenly start asking which content is the author most confident of? which authors are the least self-assured? Which day of the week does the world find the least confident when posting? That tiny little piece of extra data introduces magnitudes of extra meaning, raising the intelligence available from the standard metrics by an extra dimension. It may not be information I want to make public, but for my own reference, it could be very valuable; and anonymised, combined with other blind datasets, could provide a means by which I can asses my own progress and confidence with my writing.
Usability squads employ iris-trackers to understand where people focus the most attention on web sites; this feeds back into the design, nipping and tucking layouts to better serve expectations. Sites could do this automatically themselves though, by using the mouse cursor as a cheap iris tracker. Places my mouse hover a lot express interest, as do clicks. Those areas should become larger, more obvious. The iGoogle homepage and Facebook are examples of information-heavy sites that could benefit from this approach. This could be handled at the individual or community level. Games could benefit hugely from reading these extra cues and providing less binary experiences; Street Fighter, the original, features strength-sensitive buttons which would allow a direct correspondence between the player and their gaming character’s attack strengths. Even Vista’s start menu defaults to showing a list of programmes, from top to bottom detailing the most used over a long period, to the most recently used.
With his woolly ideas gathered and spell-checked, he clicks publish, decisively.
Post a Comment