An essence of being an SAS is to perceive lawlike behavior in the
universe it inhabits. Laws are essentially ways an SAS organizes what it
observes about the world around it. In this way an SAS is able to learn
from its surroundings, generalize event patterns into laws and constants
that characterize its universe, and predict the likelihood of future
events. An SAS with finite knowledge will allow for rare, apparent
violations of laws, but assumes that an explanation exists for that
apparent violation, which fits in with the laws it has knowledge of.

So it is no surprise, that, for example, a dragon event is never
observed. If some SAS reported the sighting of a dragon event, its
fellow SAS's would pass it off as a hallucination. That would be the
'natural' (i.e., fits in with the known laws) explanation. Similarly, if
a group of SASs reported the dragon event, it would be passed off as a
group hallucination, or possibly, a hoax. If there is an 'objectively'
confirmed string of dragon sightings by SAS scientists, either the known
laws would be updated while maintaining their simplicity (if that were
possible) or else there would be great effort by the SASs to find a
'natural' explanation (e.g., the SAS's passed through a wormhole into
another universe).

Similarly, the Strong Anthropic Principle is really a statement that any
SAS will find itself will observe itself in a universe that allows it to
exist. The SAS should be aware which conditions that allow SAS's to
exist and conditions which do not allow SAS's to exist. The appearance
of fine-tuning really becomes rather subjective - how narrow should the
range of values of allowed parameters should be so that there appears to
be fine-tuning?

Interpreting MWI is very difficult - I prefer to think of "many
histories" myself. There are countless universe-histories which follow
the same laws of physics that we have come to know and love. I am not
sure if it is possible to talk about the actual probability of a
particular history of our universe having occurred, or the probabilities
of possible futures of our universe. Given the history of the universe
up to now, the algorithmic complexity will presumably be the same for
each member of a large set of possible consistent futures. The number of
members of this set is unknown. For another set of equally less
consistent futures, the algorithmic complexity is probably higher, but
again the number of members of this set is unknown (though we expect it
to be smaller).

Reply via email to