14.4.08

Drawing test

...two posts in one day? I must be busy. : D

I've been playing around with python, pygame, and cairo (through pycairo) the last few days. I needed an environment to quickly prototype ideas (after discarding processing for various reasons). With the above mentioned combination I can quickly whip up interactive vector-graphic driven systems. The following video demonstrates a trivial algorithm for generating shapes that takes two parameters which are changed over time. It's nothing special but took a trivial amount of time to implement with my new toys. Something interesting to look out for are the emergent shapes and waves as the parameters are changed.

video

Virtual ecosystems

As part of the Design After Nature project, CEMA has been looking at natural ecosystems in the context of creativity. Our primary assumption is that the diversity of organisms, complex interactions and energy networks that arise within ecosystems is the result of a set of ecosystemic principles that can be applied in a creative and artistic domain. This assumption is not all that far-fetched, aesthetic evolution demonstrates that the concept of natural selection can be applied powerfully to creativity (e.g., Genetic Images, The Electric Sheep.) CEMA has been involved in this idea for quite some time through various artworks, such as Plague (Alan Dorin) and Eden (Jon McCormack).

My own research focuses on biological development as a creative force, however I have been involved in this project and have been playing around with various ideas and prototypes. A very literal interpretation (from a computer scientists p.o.v!) is demonstrated in the following video. The virtual world (the white plane) contains various growing plants, which according to an internal genome grow leaves (the green parts), branches, and drop seeds. A seed inherits its genome from its parent, the genome is mutated slightly to provide a simple evolutionary capability. Initially the world is populated with seeds with randomly generated genomes, most of these aren't capable of reproducing and so their species die out. The ones who can reproduce start to populate the space. Eventually a small set of species remains as the system enters a homogeneous state (where the most efficient species has survived and cannot improve anymore). A simple energy model is incorporated (energy is obtained via photosynthesis through leaves, growth of new parts costs energy, and reproduction costs energy).

video

The system results in plants which reproduce quickly and are efficient at filling space. The ecosystem abstraction used is quite naive but demonstrates competition between species (for the same niche: space), evolution, and interesting dynamics over short and medium term timescales. The system is interesting visually, primarily because of emergent shades: there are only two colours in the system but different plants fill space in different ways resulting in different shades of colour.



3.4.08

Confirmation over. Caek time.

I have now finished all required work for my PhD probationary confirmation, and have successfully been confirmed. This means that the confirmation panel are convinced that I have a valid research proposal and a plausible schedule over the next 2 years.

I enjoyed my caek last night. It was a dodgy banana cake from safeway. It wasn't bad.

My research involves the design and implementation of a creative system that grows geometry using principles from biological development. The shapes (organisms) consist of cells that grow, divide, move and die, following a genetic program specified by the user/artist. This general methodology has been successfully demonstrated with Lindenmayer Systems through Prusinkiewicz' (and others) great work. My work looks at different structures and is directed at modelling organisms such as the Siphonophorae illustrated by the 19th century naturalist and artist Ernst Haeckel. One such illustration (my replication) is shown below.And on other news I ran my first fun run on Sunday. The Run for the Kids fun run attracted nearly 30,000 runners and supports the Royal Children's Hospital. I made it the entire 14km in under 80 minutes --- which was my target. It wasn't a fast time (1hr17m) but it's a starting point and I hope to knock at least ten minutes off next year. It was an awesome experience and I think I'll be running a few more over the years.