Wednesday, 7 January 2009

Getting away from switches

Seminar from Paul Blenkhorn who is a visiting professor with LSL in ECS at Southampton got me doing a little thinking...

Titled "Getting away from switches" it included discussion of previous interfaces - now discontinued. Paul is working in conjunction with a Japanese University and various specialist hospitals who support people with physical disabilities.

The solutions offered are software virtual keyboards with a range of interface options. Be provided various example of how you select characters move around screen with switches by
The interfaces are to some extent user defined, but there is also constraint of affordance from the input devices. It seemed quite timely in the light of the onion video of the 'new' apple keyboardless computer ;-)

One option which Paul was giving the thumbs up was a Mouse wheel which can select column on keyboard then traverse individual keys. This was a bit like the jog wheel which Sony introduced a few years ago. This example moved away from pixel control, rather taking a functional cut on the screen, with the mouse wheel successively accessing usable areas of the the screen real estate

Another option he was very keen on was a ten key mouse, designed originally for numeric input I think.

We had a bit of discussion about the emergence of user interfaces (and their demise) and there were a few views offered on the persistence power of texting via mobile phones. People persist in using them, and they master them. My view is that the interfaces have been developed iteratively with millions of users and hours of user trials, effectively driving a rapid development cycle, spurred on by competition for market share. There was some discussion of possible artefacts in solutions and behaviours which arise from the differences between Japanese data input and english language user input.

Seems to me to be an issue about the difference between individual personalised solutions and generic solutions. There was some discussion of how user's needs might be assessed. I wonder if a recommender system might work?

Some kind of quick test and then you get served a particular interface with training, then a re-evaluation subsequently to fine tune, or recommend alternatives. There is also an issue about how long does training take.

I think that mobile phone expertise is a combination of personal muscle memory learning with intuitive design which enables the user to learn or remember how to use the system

I think it would be really interesting for Paul to do a head to head with Jon Maber. Jon has been working on his own specialist input device and it would be good to see how this is progressing. I also think that the two of them could stimulate some interesting discussion. I would also like to see some kind of symposium which brought together a range of interested parties and got some good discussion going in this area.

No comments: