Tuesday, May 24, 2011

Rebuilding Iberian Motorways with Slime Mould

Wet machines and Soft Computers planning road routes organically. 

Place your 'problem' in bag and shake to get the answer!

Although done on a simple flat map/surface there is no reason why this couldn't be a 3d model with variable temperatures/variables throughout. The modelling of landscapes can be more accurate organically in order to find target map paths that are efficient from a biologic standpoint.

These types of navigational problems* were among some of the first tackled by 'hard' machines (computers like you are reading this with) when they were first developed and it's nice to see the initially parallel development of bio computing.


* Travelling Salesman problem : What is the shortest route visiting each city exactly once and then returns to the starting city? See more classic computing problems here 

Thursday, May 19, 2011

Wearable and Always On Computing

I'll be surprised if 2011 doesn't see something further happen around the wearable computing space.  We need to stop tinkering with metal boxes and facilitate direct interaction with the world a bit more.

There are two social dynamics to this kind of interfacing:

1. Broadcast the display externally on walls, tables, car bonnets or bodies (not private) OR
2. Broadcast internally on glasses or hidden earpieces (i.e. privately).

I think both approaches are more favourable to the current head down into a mobile neck stretch. Mobiles are private devices and Tablets/iPads a bit less so but they are both metal objects you have to put in front of your face and carry around.  The world is only there in periphery when using devices like these. 

Directly communicating with others and including the web as a 'third voice' is still not an elegant flow when taken out of presentation theatres and onto buses and high streets.

Pervasive and wearable computing will see an always-on environment for audio and video. The machines will listen to you 24/7 and parse what you say. The video components will continually record and pattern match the objects around you. Forget Amazon recommends when the data you can input is your whole day! We don't need to key the data about us like monkeys with typewriters. Spines everywhere will rejoice as we lift our heads to look back at the world once more.

The demos from MIT Wearable Computing Team in 2009 still look fantastic and the prototype only cost around $300 back then.

The TED talk - Pattie Maes' lab at MIT, spearheaded by Pranav Mistry

The interface ideas

The evolution of Steve Manns private eye glass display