The problem according to a NASA podcast I was listening to the other day is
micro-curcuitry is so small, being bombarded with particles of radiation in
the Solar wind, stray electrons etc, causes multiple errors in the chip
curcuits, which means you have to have say for example 3 curcuits all doing
the same job, or process the same data over and over, so the results can be
MO you can't have a self sustaining machine without some form of
consiousness. Because the machine would have to be "self aware" it woulds be
pretty useless at maintining itself if it couldn't even reconise it's own
reflection in the mirror to give itself a new paint job LOL

This is quite simply untrue. What about chlorrella? Paradoxically thev
first VN machines are likely to be more intelligent than more advanced
machines involving nano technology. The first VNs will use CAD/CAM
logic and would be "intelligent" using the word in the sense that PE is
"intelligent".

compared, and the data in effect cleaned to minimise errors. However unless
you have clean pre-processed contol data which of couse wouldn't exist
outside the laboratory in the void of space, AFAIK there is no absolute way
of ensuring against computer errors in space. What that means in practice is
the computers they can send into space are about 1/5 the power of state of
the art chips used in consumer electronics on earth.

This is true, there are a variety of ways round. The first point is
that a cosmic ray can strike even in a low radiation environment. One
answer would be to have a error correcting system built into the chips,
or to have 3 chips running in parallel with a voter (Smith's triplex
system).

I have also noted your views on religion. I feel I should point out.

1) That all religions do not support ID. Pope Benedict has excoriated
the notion.
2) What I was really saying is that if you are to have a naturalistic
view on the origin of life, you cannot then say that a self replicator
is going to be an encedibly complex piece of machinery.