On 27 September 2013 16:15, meekerdb <meeke...@verizon.net> wrote:

>  On 9/26/2013 8:56 PM, LizR wrote:
>
>  On 27 September 2013 14:50, meekerdb <meeke...@verizon.net> wrote:
>
>> But I think this is a confusion.  Because computations have states and
>> nothing corresponding to transition times between states people are tempted
>> to identify those states with states of consciousness and make an analogy
>> with frames of film in a movie (hence 'the movie graph argument').  But
>> there's a huge mismatch here.  A conscious thought has a lot of duration,
>> I'd estimate around 0.02sec.  The underlying computation that sustains the
>> quasi-classical brain at the quantum level has a time constant on the order
>> of the Planck time 10^-43sec. And even if it isn't the quantum level that's
>> relevant, it's obvious that most thinking is unconscious and a computer
>> emulating your brain would have to go through many billions or trillions of
>> states to instantiate one moment of consciousness.  That means that at the
>> fundamental level (of say the UD) there can be huge overlap between one
>> conscious thought and the next and so they can form a chain, a stream of
>> consciousness.
>>
>
>  Forgive me, but I think this is a level confusion. The "moments" I'm
> talking about are whatever state transitions are required in the underlying
> computation (assuming there is one), not consciously perceived moments. If
> the comp hypothesis is correct there must be such a series of transitions,
> otherwise, there isn't (or need not be).
>
>
> But then they aren't moments of consciousness and it's no longer clear how
> the closeness of another state can be measured if not by the content of
> consciousness.
>

I'm not sure that it's clear using the contents of consciousness, either.
The thing is, if comp is right then there are definite computational steps
that can be talked about, analysed and so on, but thoughts might be a long
way above them. Thoughts may be huge constructs relative to the
computational underpinnings, each one (perhaps) an ocean full of
computational fish. But it isn't easy to imagine or discuss the
computational steps down there at the Planck length (or whevever)... one
can't get one's head around it.

But you'd have to measure the states, and what counts as a closest
continuation, using those steps, ultimately - wouldn't you?

       So there's a certain amount of "mini-death-and-mini-rebirth" going
> on every second in the normal process of consciousness (in this view).
> Deciding what counts as a continuation and what doesn't seems a bit ...
> problematic. (And of course there are many continuations from any given
> moment.)
>
>
>  Not if there's nothing to overlap.  Sure there is, by some measure, a
> closest next continuation.  But when you're eighty years old and fading out
> on the operating table, it's going to be another eighty year old fading out
> on some other operating table.  I think someone has suggested that if you
> fade out completely then the next closest continuation could be a newborn
> infant who is just 'fading in'.  Which is a nice thought - but is it you?
>
>  Yes, that has been suggested, I forget who by. I had a problem with
whether it would be me or not.

Otherwise, the "problem" for me with QTI is, it's a ghastly thought that
you might go on forever as a 80/800/8million year old, forever fading on
operating tables.


It's worse than that.  Read "Divided by Infinity"
> http://www.tor.com/stories/2010/08/divided-by-infinity
>
> Ah yes, someone sent me that before. I'm not sure that's so bad, but both
times I failed to read it fully, just skimmed, so perhaps it is.

-- 
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 everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to