-------------------- thinking on halting problem, after poking at wikipedia and turing machines a smidge i know little about turing machines so i am likely to be wrong of course but it seems to me that the halting problem is solveable so long as the analysing system is sufficiently advantaged in regard to the analysee system. if, for example, it can enumerate every program that the analysee system can run, the problem is trivially solveable. relies on the analysee being limited in what programs it can run. then when it is not limited, we enter a cool space of trying to make things undecidable in ways that are harder and harder to figure out, probably something studied in computer science somwhere, some kind of execution cryptography, i don't really know. the wooden machine and wikipedia note a very tiny turing machine with 2 states and 3 symbols. being so tiny makes it seem more reasonable to think of modeling behavior. note: the tape is the linear RAM. A B 0 P1,R,B P2,L,A 1 P2,L,A P2,R,B 2 P1,L,A P0,R,A The states the machine can be in are A and B. this is like having a 1bit process register. The tape or ram can hold 3 different values: 0, 1, and 2. Each entry in the table shows the 3 things that happen when a value is encountered as an instruction. Something is written out, the tape or memory address pointer is incremented or decremented, and the machine's state changes or stays the same. So: Read 0, State A -> Write 1, Decrement pointer, change to state B Read 0, State B -> Write 2, Increment pointer, change to state A Read 1, State A -> Write 2, Increment pointer, change to state A etc etc The machine has a state consisting of 1 bit, an unbounded integer address, and its memory contents. The memory -- at this point i encountered some further accumulating issues and text was also lost. notably aligning with the soup kitchen opening.