[continued from previous message]   
      
   >>> The whole point of using genetic algorithms is that, in suitable   
   >>> domains, they greatly outperform brute force searches. They are   
   >>> vulnerable to hanging up on local maxima, but this can be in part   
   >>> addressed by annealing.   
   >>>   
   >>> I find the "smuggled information" response underwhelming. If a   
   >>> genetic algorithm can import information from an "artificial"   
   >>> environment set up by human experimenters then why can't it import   
   >>> information from a "natural" environment? At which point you're back   
   >>> at the cosmological argument again.   
   >>   
   >> In this case the device was programmed with (as I recall) 2 kilobits   
   >> of data, starting with a random sequence and ending with the refined   
   >> sequence. That is, it appears that a quantifiable amount of   
   >> information has been created.   
   >>   
   >> Hence my question to Dembski. I understand your underwhelm. It does   
   >> seem to warrant further study and explanation.   
   >>   
   >> One way to look at this would be to ask, can RM+NS potentially produce   
   >> say just 2 bits of information (e.g. a single advantageous point   
   >> mutation)? Regardless of one's overall position, I think you'd have to   
   >> concede that it could, even if you dismissed it as trivial. What about   
   >> 4 bits? 8 bits? At what point would we discomfirm the ID assertion   
   >> that information "can only come from a mind", or the "conservation of   
   >> information"?   
   >   
   > I find it self-evident that demonstrating that 1 bit comes from a source   
   > other than a mind that is be sufficient to disconfirm the ID assertion   
   > that information "can only come from a mind". ("conservation of   
   > information" is not equivalent, and is not sufficient to conclude design.)   
   >   
      
   You didn't consider my explanation below, i.e. why we need to think   
   stochastically.   
      
   >>   
   >> Is the resolution akin to entropy: entropy could potentially increase   
   >> spontaneously in a closed system with a very small number of (say) gas   
   >> molecules with different velocities, such that the hotter ones   
   >> randomly moved to one region and the colder ones to another. However,   
   >> as the number of particles increases, the probability of this   
   >> occurring decreases exponentially.   
   >>   
   >> Inference: macroevolution (a non-trivial increase in information) is   
   >> like entropy decrease in a stochastic ensemble (e.g. a spontaneous   
   >> non- trivial temperature gradient).   
   >>   
   >> The former is in an open system, but I'm not suggesting direct   
   >> equivalence.   
   >>   
   >>>>   
   >>>>>   
   >>>>> (AlphaFold broke the back of the protein folding problem, but in   
   >>>>> that case one could appeal to import from environment as the source   
   >>>>> of the information in the trained model.)   
   >>>>>   
   >>>>   
   >>>   
   >>   
   >   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)