Upgrade your account to a Community Supporter account and remove most of the site ads.

Reply to thread

Message: <blockquote data-quote="Umbran" data-source="post: 7766307" data-attributes="member: 177">It doesn't have to be.  As I already noted, on Earth, on the surface, with the protection of Earths' atmosphere and magnetic field, and the Sun's magnetic field, we still see one error per 256 MB of memory per month due to cosmic rays.  My work laptop, then, experiences nearly 770 such errors in a year.  Now, imagine the data storage such a machine needs.  Traveling at 0.1c, it will take that machine ~40 years to reach the nearest star.  How many errors do we expect?  If it were just ten of my laptop, it'd be on the order of 10,000 errors.  We are talking about needing at least a couple orders of magnitude more for this machine, yeah?  So... millions or tens of millions of errors?  We are no longer talking about having a single point of failure, we are talking about having an accumulated burden of errors.  How many of these can the thing tolerate before it stops working, or works oddly, in some way?  Um... that's why they are self-replicating.  You send out one, or a handful, and they reproduce.  But, that's the point - if you have a living creature that reproduces, over tens and hundreds of thousands of years, you get changes!  Species are not stable on million-year timescales. Consider that the species we think of as modern humans is not 500,000 years old - we have gone from stone knives and bearskins to the Space Shuttle in that time.  You figure the species that is these machines will go blithely along following its original directives for a million?</blockquote>

Verification