Glen writes: ``Well, my specific problem is that I think atheists and theists are delusional. They think they know things they cannot know. So, if Nick's point is that the concept of "theist" (or "atheist") is too muddy to define validatable[*] tests for, then, as an agnostic, I would completely agree.''
Let's say I have a program and it does something that I don't expect. Someone says it is the witchcraft from a particular gal that works on the HVAC system. I ask "Is it because the computer got too hot? Did she fail to keep the air conditioning running properly?" They respond, "No, it's not the temperature, it's the witchcraft." I look in the index of the instruction set under W and under remote access protocols, interrupt mechanisms, etc. Nothing. I keep removing degrees of freedom from the code and environment and all of my suggestions are rejected by my peer as "Not an instance of witchcraft." I ask for suggestions on how to prove that witchcraft is at work and they just show me pictures of their witch suspects and give me a pamphlet on building big fires. Meanwhile, I discover a simple, mechanical, explanation for why the program isn't doing what I expect, fix it, and tell the growing mob of witch burners about what I discovered. (Of course, their explanation is that they were successful in intimidating the witch and she was forced to release me from her spell.) It should be possible to associate with any proposition a probability function that takes as arguments other routines that describe how to perform an experiment and the result of that experiment. The details of the experiment routines should be provided and should not include "call a friend" or reference anything that is already known or obvious. All functions and routines should be written down before doing the experiment. It should be possible that by sweeping over the space of unknowns (potential inputs) in the experiment routines to get some probabilities near zero and some near one. The refusal or inability to write these functions and routines is an indicator that the speaker is full of it and would rather talk about witches. When cornered on a question, can the believer justify or change their belief? Marcus ============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
