I'd disagree strongly. I think the point of the paper was the evolutionary algorithms. Whether or not the wheels could adequately replace a modern bicycle wheel is kind of a side note. The work was entirely about creating a program that would produce better and better results over time, while balancing multiple objectives to meet. As you can see in the videos, (and he even admits in the paper) most of the wheels would not work very well if they were produced.
That said, I think if you're going to be completely dismissive of a thesis that I'm sure represents thousands of hours of work, I might put a sentence or two explaining why you think the algorithm isn't the important part of this.
A quick scan of the paper (it's 105 pages) it looks like it was just simulations.
Basically until a simulated of a wheel is actually built, I'd dismiss all this.
I'd disagree strongly. I think the point of the paper was the evolutionary algorithms. Whether or not the wheels could adequately replace a modern bicycle wheel is kind of a side note. The work was entirely about creating a program that would produce better and better results over time, while balancing multiple objectives to meet. As you can see in the videos, (and he even admits in the paper) most of the wheels would not work very well if they were produced.
That said, I think if you're going to be completely dismissive of a thesis that I'm sure represents thousands of hours of work, I might put a sentence or two explaining why you think the algorithm isn't the important part of this.
I agree. It's really cool that he managed to find the standard 3-cross based on a pretty rough model and an optimization step.