Use existing building language:
research facility as "leech," where the labs are the "brain", the visitor's center/cafeteria the "heart" and the offices as the "lungs."

1 comments:

October 23, 2008 at 8:00 PM Michael Kubo said...

Seth,

I am glad to see that you are developing geometrical controls that have more to do with the pragmatic requirements of the program than the procedural methods you had started with (which relied on a series of intermediary "objects" that were manipulated to produce the eventual geometry of the building, rather than working directly on the matter of the program itself). What is not clear is why this shift—into working directly with the matter of the building—leads the forms to now become much more regularized than they were before. Particularly in the research block, but also in the T-shaped tunnels connecting the research block to the other wings, etc, the building is now organized much more conventionally than what you had before (the research block is now symmetrical and basically organized as two linear volumes connected by cross-pieces at right angles in plan). I am not sure why modifying the geometry of the exterior forms makes you change the organization so radically. What you have now in the research block is in fact a very conventional organization that is softened by a few exterior curves (rather than, say, being made out of boxes), to the point where it is not clear whether the curvature even helps at this point. In any case the research labs probably want a much more continuous/extensive area in plan rather than a series of very thin linear pieces.

The office wing is still much more in keeping with your original system, and here the geometry still has the potential to create a new or different organization for the building. It does seem like the forms are a little more controlled now than in the previous version. I still question what the transparent domes are doing for you—are they still only registering (making "legible") the cutting plane your procedural methods relied on, or do they come out of the programmatic requirements of the building? We would need to see a section of what happens in these bubbles, and plans of the different levels, interior renderings etc., to know what they produce. All we have to go on at this point are the exerior forms, so it is difficult to judge how the building is working internally.

You need to do detailed plan and section drawings of the building (and especially the office block) so we can judge, plus renderings or perspectives of the interiors that would tell us how these actually perform. What is the circulation? How does it relate to the domes or the voids in the plan? How many levels are there, what is the scale of the enclosure? What happens in the T-connector tunnel? Right now there is too little information.