Marvels of engineering

During the summer, my apartment building gets festooned with spiderwebs on the outside. Sometimes I think the spiders are kinda stupid - some hardy soul occasionally clears off the spiderwebs, yet the spiders come back time and time again. *shrugs*

Anyways, for the last few days, a spider had had free reign over the top foot or so of the front door. All of the tenants have been either too busy, lazy, or arachnophobic to clear off the web.

What's been fascinating about this, though, is the amazing pattern of the ensuing web. I've observed that it has amazing symmetry, fairly even "web density per unit area", for lack of a better term, etc, etc. That got me thinking about how spiders would make good engineers - they face a fairly complex design problem, making a trap that conserves material yet is strong, is anchored well, and catches adequate numbers of insects.

When I looked closer at the web, I realized that the structures and patterns closely mimicked some structures found in human-engineering buildings. The base of the web formed an arch, and the rest rather remarkably matched some truss designs I've seen.

Upon further thought, I realized that spider web strands could be modelled as fairly perfect rods - next to no strength in torsion or bending, but lots and lots of strength in axial tension or compression, and very little mass of their own to factor into calculations. I'm sure that given a selection of anchor points (say, various locations on a door), a given "free" area where those big pink bags of mostly water (ie, humans) are not likely to break the web, and various other material/density constraints, one could develop a computer program to develop the best spiderweb design.

I wonder, would it closely match a spider's natural tendencies? I suspect it would.