It has been argued that self-replicating robotic probes could spread to all stars of our galaxy within a timespan that is tiny on cosmological scales, even if they travel well below the speed of light. The apparent absence of such von Neumann probes in our own solar system then needs an explanation that holds for all possible extraterrestrial civilisations. Here I propose such a solution, which is based on a runaway error propagation that can occur in any self-replicating system with finite accuracy of its components. Under universally applicable assumptions (finite resources and finite lifespans) it follows that an optimal probe design always leads to an error catastrophe and breakdown of the probes. Thus, there might be many advanced civilizations in our galaxy, each surrounded by their own small sphere of self-replicating probes. But unless our own solar system has the extraordinary luck to be close enough to one of these civilizations, none of these probes will ever reach us.
Without having read the rest of the paper, that's silly. Runaway error propagation is not a problem for organic self-replicating systems so there's no reason to expect it will be a problem for artificial ones.
http://arxiv.org/abs/1605.02169
After reading this some more, I still think it's very silly. Selected quotes :
ReplyDeleteIt might be helpful to look at a serious obstacle that biological cells are facing during their process of selfreplication... The problem is that ribosomes only work with a finite precision and sometimes introduce errors into the newly synthesised proteins. The question is then, what happens if ribosomes introduce errors in the next generation of ribosomal proteins?... If the ribosomal accuracy is below a critical threshold, the quality of the ribosomes deteriorates from generation to generation with the ribosomes becoming more and more error prone until protein biosynthesis collapses. This has been called “error catastrophe”. But if the accuracy is above the critical threshold, the ribosomal error rate actually converges (from generation to generation) towards a stable steady state.
In principle it is of course possible to construct a probe that does not suffer from an error catastrophe, in the same way that living cells can reproduce without running into a catastrophic error propagation. However, I am going to argue that firstly there is actually no motivation whatsoever to build such a perfect probe and secondly that rational probe design will lead to a probe that operates in the low accuracy regime.
The logic is simple, for very little effort (creating the necessary blueprints), one can generate multiple generations of probes, which visit and investigate a growing sphere of stars around the home system. However, there is a diminishing return in the sense that the further the systems are away the less important is a successful visit by the probe. This is because they are less of a military threat, because the number of scientific discoveries decreases with the number of investigated stellar systems [said no astronomer EVER !] and most importantly, because the time a probe needs to reach further stars counts in the millennia.
. Hosek (2007) quite rightly points out that individuals with a finite lifespan tend to prefer benefits/projects/results that can be realised within their lifespan over benefits that might only be experienced by future generations. Unfortunately, we don’t know the life expectancy of extraterrestrial lifeforms, but we can make an educated guess... If we thus take a few millennia as upper guess for the lifespan of intelligent extraterrestrials, we see that the inspection of more distant solar system via self-replicating probes quickly becomes less interesting because it is well beyond the individual time horizon. [Yeah, but an expanding civilization which expects to keep expanding would benefit from an error-free probe. This is a bizarre argument that civilizations are only interested in exploring nearby star systems, and anyway he postulates that the probes are used to prepare systems for colonization. So it solves absolutely nothing about why the Galaxy hasn't already been colonized.]
Agree. Ridiculous.
ReplyDeleteProbes are really, really smol?
ReplyDelete