On Sun, Jun 27, 2021 at 10:18 PM Tomas Pales <[email protected]> wrote:
> On Sunday, June 27, 2021 at 1:21:32 PM UTC+2 Bruce wrote: > >> On Sun, Jun 27, 2021 at 7:38 PM Tomas Pales <[email protected]> wrote: >> >>> On Sunday, June 27, 2021 at 2:36:38 AM UTC+2 Bruce wrote: >>> >>> Much as I respect Russell, his book is not an authoritative source for >>>> anything. It is all rampant speculation. >>>> On the matter of the stability of laws and the connection with >>>> simplicity, I refer you to the 'grue/bleen' paradox introduced by Nelson >>>> Goodman. That shows that the idea of simplicity as an explanation for >>>> anything is misplaced. >>>> >>> >>> Note that the relation between simplicity and frequency is not Russell's >>> speculation but a fact following from Kolmogorov's definition of >>> complexity: simpler objects are more frequent than more complex objects >>> because the same simpler object is contained in less simple objects. >>> >> >> This is only if everything is considered to be a bit string. There is no >> reason to suppose that this is true. >> > > No, atoms are more simple than ducks, and atoms are also more frequent > than ducks because there are atoms in every duck but there is no duck in an > atom. However, it seems that every object can be represented as a binary > string, which is a useful representation in computer science. > The problem with that is that it is dependent on the language in which you express things. The string 'amcjdhapihrib;f' is quite comples. But I can define Z = amcjdhapihrib;f', and Z is algorithmically much simpler. Kolmogorov complexity is a useful concept only if you compare things in the same language. And there is no unique language in which to describe nature. What a load of garbage! Science is not a matter of induction from observed >> data. >> > > What is science a matter of then? > Maybe it is a matter of finding laws. And laws are not just empirical generalizations obtained by induction. Goodman's grue/bleen paradox puts paid to that idea. Algorithmic simplicity >> has nothing to do with real world data. >> > > But the property of "grue" is more complex than the property of "green". > "Grue" means "green before time t (for example year 2030) and blue after > time t". > But that is the fundamental mistake. 'green and 'blue' are not simpler than 'grue' and 'bleen'. If grue and bleen are the common language, then green is defined as grue before time t and bleen thereafter. The complexity of the terms depends on the language, so complexity does not distinguiah between them. More complex properties are less frequent than simpler properties. > Only in your dreams. Because complexity is language dependent, relative frequencies cannot be uniquely defined. Bruce > For this reason, and given the way the world has been until now, objects > that have been observed as remaining green in the past are more likely to > remain green in the future, instead of becoming blue at some time. > -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAFxXSLR_xUpitZE%3DLp7zp-vLzkQo%2Bq9k4grgEzBdCqSJhPVx4w%40mail.gmail.com.
