<Nature takes on peak oil (pun somewhat intended)The Oil DrumUsing NATO to fight peak gas>

204 comments on A further comment on "That's Oil, folks . . ."

Comments can no longer be added to this story.

GliderGuider on January 5, 2007 - 2:12pm Permalink | Subthread | Parent | Comments top

Thanks. I'd been thinking about the role of efficiency since I encountered the concept of adaptive loops a few weeks ago. The three-dimensional diagram of the adaptive loop that I first saw in Thomas Homer-Dixon's book "The Upside of Down" came from the work of Dr. Buzz Holling. It has as its axes Productivity, Connectedness and Resilience. It made intuitive sense to me that efficiency should be included, except I don't know how you could show nice printed pictures of a loop in four-dimensional space.

Here is how I've been thinking of the inverse relationship of efficiency and resilience.

Resilience implies that the system has the ability to redirect resources from elsewhere within the system to contain and heal the impact of a shock. Crucially, this reallocation must not affect the system's performance in such a way that the reallocation itself acts as a secondary system shock. If the resources are reallocated from a portion of the system that can't function without them, the act of reallocation may become the first event in a breakdown cascade.

One definition of an inefficient system is that system operations have more resources available to them than are actually needed to accomplish the tasks. In this case, redirection of resources away from a task will have less of an impact since it is more probable that the task's efficiency can be improved to accommodate the loss.

In a very efficient system, all resources are fully utilized. Any redirection of resources is done in a zero-sum context - the task from which the resources are taken can no longer function (or at least can't function as fully). As a result, very efficient systems are much more prone to cascades, as resources are sequentially redirected to try and cope with the breakdown caused by the previous reallocation.

It now seems to me that it would be helpful to consider adaptive loops with axes of Efficiency, Interconnectedness and Resilience. Of course the brains at the Resilience Alliance have probably already thought this through. It makes me less sanguine, though, when I consider just how efficient our socio-economic system is, and how little slack we have left to play with.

0
Comments can no longer be added to this story.
GliderGuider on January 5, 2007 - 3:00pm Permalink | Subthread | Parent | Comments top

Here's one additional thought. I just realized that every time a system reallocates resources to deal with a problem, it loses further efficiency. That happens both because the resources being reallocated may not be optimal for their new use, and some additional losses will occur due to something analogous to friction or thermodynamic effects. Not all of the resources you take away from the donor task make it to the target, and those that do may not be quite what's needed. Both effects result in the need to take away more resources from donor tasks than are needed for the repairs.

In a very efficient (ergo low resilience) system this spells mucho trouble, as it increases the effective damage of each cascading shock.

0
Comments can no longer be added to this story.
Leanan on January 5, 2007 - 3:02pm Permalink | Subthread | Parent | Comments top

I think Wal-Mart is finding that out.

0
Comments can no longer be added to this story.
slaphappy on January 5, 2007 - 7:05pm Permalink | Subthread | Parent | Comments top

That is a pretty broad statement to rest on such a thin reed.

EVERY time? really? Often when resources are allocated to deal with a problem (Hmmm, anti smoking adds and public smoking bans to combat smoking) the system gains efficiency. Often the problem is the inefficiency itself (how much farmland are we using so that people can die earlier and at great expense?), so allocating resources to deal with it does the exact opposite of what you are saying.

Same thing with complexity BTW. People talk a good game about systems having some inherent complexity limit, or some such. Pure bunk. As a general rule, complex systems tend to work better than simple ones, otherwise why would the complexity have been added at all? It's nice when a system can be both simple and good, but in my experience, it's a rarity. You can see this in science (General relativity bumps off Newtonian physics), computer science (quicksort bumps off bubble sort), mathematics (Complex analysis easily tackles problems that 100 pages of algebra would never solve), etc...

Such broad and sweeping statements (including my own, of course) help nobody.

0
Comments can no longer be added to this story.
toilforoil on January 5, 2007 - 3:37pm Permalink | Subthread | Parent | Comments top

It seems to me that in this discussion with D.S., you've adopted the notion of efficiency as employed by economics in its mechanistic analogue (mainstream). Resilience necessarily is inseparable from time, while efficiency, in the sense you are employing it, is with-out time, or time-less. If efficiency was a measure of the effect of work on the rate at which entropy rises, and a lower rate a sign of greater efficiency, then it would stand side by side with resilience and not in opposition to it.

I haven't read the work of the Resilience Alliance, though I have completed 'The Upside of Down'. I suspect the problem of definition in relation to time is one reason the word efficiency is not used in the adaptive loop.

0
Comments can no longer be added to this story.
GliderGuider on January 5, 2007 - 3:49pm Permalink | Subthread | Parent | Comments top

I think I see what you're saying, though my first thought was that "connectedness" is also time-less. I also don't see that resilience must necessarily have a temporal component - is "ability to recover" not as good a definition of resilience as "ability to recover within a given time period"? I'll give the hamster a poke and see if he can spin my mental gears a bit faster on this.

0
Comments can no longer be added to this story.
Don Sailorman on January 5, 2007 - 4:19pm Permalink | Subthread | Parent | Comments top

If you think of an airplane or a boat as a system that encounters rough weather through time, I think it becomes clear that the time dimension actually amplifies the tradeoff between efficiency and resilience.

(I prefer the term "robust," to "resilient" but let us not quibble over words.)

0
Comments can no longer be added to this story.