Blog: Biology and Engineering

Usually engineers feel no special need to imitate nature's solutions to their problems (assuming there are any). Transport engineers do not think much about legs, nor structual engineers about tree trunks.

However, we live at an unusual moment in the history of engineering: today biology is sitting on solutions to problems that are really interesting and important, and on which traditional engineering has been blunting its tools for decades: fabrication on the nanoscale, learning, and parallel processing and programming. So at the moment the intersection of biology and engineering is very hot.

But this will pass. Once biology gives us a rough idea how to handle these problems we will take off from there, developing our own solutions. Those solutions will almost certainly be better than those we see around us. From a distance you can see that whatever strengths Natural Selection might enjoy, it also comes with serious weaknesses, weaknesses that as technological creatures we share. For one, it is restricted to a tiny subset of chemical elements. For a second, natural computation runs almost a billion times more slowly than "artificial computation" does.

Finally, worst of all, while nature is great at climbing up adaptive gradients, it is terrible at going down them, and it is hard to execute a comprehensive exploration of an adaptive landscape without both. Imagine that genes A and B in combination gave a pretty good feature but that the trait coded for by C and D would be even better. If organisms without either combination, AB or CD, were selected against, which is easy to imagine, it would be difficult for an AB to drift through solution space to CD by first shedding an A and then a B and then gradually picking up a C and then a D. All these intermediate stages would be selected against -- would not reproduce. The species as a whole would be stuck on AB -- on a "local maximum" -- for a very long time. Nature cannot take a design apart and improve it, bit by bit. This means that as slick as nature's solutions look, in fact probably all of them are stuck on top of a suboptimal (local) solution.

Add these three points up, and you see how much room is left. At some point we will know how to improve on every one of the hundreds of thousands of proteins in our cells. Indeed, we will find ways of improving on the use of proteins at all (like an oxygen carrying medium that is far more effective than hemoglobin).

Somewhere up ahead is moment when we will look at Nature as now we do dioramas of Egyptian slaves dragging stones up a ramp. "How amazing," we will think, "that nature is able to accomplish as much as it does with technology that is so weak and and slow and inefficient."

The range of possible effects of this change on the popular culture is quite wide. Right now the culture lays great emphasis on the wisdom of the "Natural". The assumption that Nature knows what it is doing underlies much contemporary thinking about eating, environmental management, waste management, and so on. Perhaps the general culture will just ignore the new point of view in engineering, but that is hard to imagine. Possibly the underlying rationale for environmentalism will shift to something like taking care of a retarded member of the family. Other outcomes are certainly possible.

One seems quite likely: Intelligent Design will probably lose much of its appeal.

Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

• Bookmark this page
• Share this article
  •  
  • facebook
  • slashdot
  • digg
  • Reddit
  • stumbleupon
  • linkedin
  • twitter
• Got more on this story? Email CIO
• Follow CIO on twitter

• Share
  Share this article

  •  google+
  • facebookfacebook
  • slashdotslashdot
  • diggdigg
  • RedditReddit
  • stumbleuponstumbleupon
  • linkedinlinkedin
  • twittertwitter
• print
• email
• Bookmark