Jia, Rachel, Jiaying, Alba, Chuck

From KokkugiaWiki

project development

COMMENTS NOV 5 the studies of subdivision are progressing well in the abstract yellow diagrams. as to are the subdivisions which respond to the attractor. i would consider two systems of decision making regarding the subdivision: 1. using attractors(within a generative distribution system) 2.using substitution rule sets. obviously a combination of these is possible also as long as the attractors doesn't adversely affect the legibility of the complex order that the substitution system should generate.

if you are going to develop option 1 i outlined above - developing a generative system for the distribution of points which respond to the current subdivision and cause further subdivision, then you will need to develop the logic for that. the first ways of thinking about this that i would propose are either recursive growth algorithms (branching for example) or agent based algorithms. however it is critical that you don't consider either of these systems to be generic, break them down to their fundamental behavior and build up from a response to a particular problem. don't simply take a swarm that aligns, coheres and separates - non of these behaviors might be applicable, instead think about what behaviors are in response to your architectural concern.

in addition to further developing the logic of the system i think you need to advance the geometry research. the regular polyhedra are interesting in helping you understand the nature of geometric subdivision, however these might not be appropriate to architecture. i think you should design your own components which subdivide in 3D. this doesn't mean they have to be precisely self-scaling (as this is very limiting), instead you can actually use the gaps in the geometry as designed outcomes in generating a particular architectural character.

COMMENTS 8TH DEC
It is very hard to see what is going on with your small images. However the differentiation within the grid is beginning to display some emergent characteristics. Although the differentiation is probably too periodic at this stage. I think the exploration in both the cubic and the surface studies are worth pursuing - it is useful to understand the divergent ways that the methodology you are developing can be deployed.

It will be important to present your work as the negotiation of two systems, the substrate (grid or surface) and the movement of the system which triggers the subdivision.

The circular fractal research looks as though it might be interesting, however the images are too small for me to give you any real critical feedback.