Quantum Computation

     
19991031

In a problem to be computed, optimization may be taken of space and time
constraints by dividing the problem into serial and parallel parts[1].
Analogously the human brain already does this when it considers a problem 
with both the sequentially logic of its left hemisphere and the parallel
or associative abilities of its right hemisphere; both of which coordinate
with each other through that massive 'bus' called the corpus callosum.

The serial portion of the problem is carried out in the usual sense
of computing in terms single program thread on a sequential computer
and the time complexity for such problems is well understood.

The parallel portion of the problem may be carried out on multiple
threads in a parallel computer, but in this case the time complexity
is also well understood in terms of multiple sequential program threads
operating concurrently. 

The complexity increases when we consider multiple sequential threads
interacting with each other and particularly so when that interaction
is asynchronous. 

"Asynchronous" is degenerative term. It can mean both "somewhat synchronous"
and "completely random". In terms of classical parallel computation
"asynchronous" is used in the former sense, while in terms of quantum
computation, it is used in the later sense. In mathematics this has 
a broader interpretation in terms of the relationship between implicit
and parametric forms.

In quantum computation, the interaction between parallel threads is
considered random to the extent that we cannot measure what the 
interaction is without destroying the effect[2]. There is also the
problem that the effect is naturally degraded by natures own
mechanisms of information dispersion (decoherence). Generally speaking,
we cannot disturb someone in deep concentration without upsetting
their "trains of thought" nor are they likely to succeed in any
practical outcome, if their concentration is dispersed by 
associative digressions. This is a strong analog of what is happening
in theoretically in quantum computers.

Somehow we wish to "not disturb the concentration"  of the quantum
computer and also to keep the "associative process" in Hilbert space
from decohering, dispersing or analgously- "daydreaming". 

The two domains of computation are:

  spatial          and temporal,        or equivalently
  parallel         and serial,          or equivalently, 
  instantaneous    and sequential,      or equivalently, 
  frequency-domain and time-domain,     or equivalently, 
  associative      and causual,         or equivalently, 
  non-deteministic and deterministic,   or equivalently, 
  dynamic          and static,          or equivalently, 
  semantic         and syntactic,       or equivalently, 
  recursive        and iterative,       or equivalently, 
  compiled         and interpreted,     or equivalently, 
  creative         and reductive,       or equivalently, 
  analogical       and logical,       ...    &c. &c. &c.

and these metaphors or isomorphic mappings between dichotomies allow
us to gain some direction in how to proceed to understand better what 
we should be looking for, and what we should do in order to create
what seems a metaphor for our own brain, the quantum computer.

This is no little responsibility. It suggests that, in the attempt
to create a quantum computer, we may approach the possibility of 
synthesizing or modelling our own consciousness; at the very
least, in terms of modelling our own intellectual processes.

The responsibility comes in terms of the fact that we do not
know how much quantum computers will model consciousness, and 
the construction of quantum computers raises some moral and 
ethical concerns over whether we might create quantum computer 
that may become conscious or self-aware. Without understanding
quantum computing well enough, the strong metaphors with our
own mental abilities suggests that we should procede more slowly.

It cannot be a good thing to find out that we have synthesized
consciousness itself by accident, but such "family planning" has
never been the one of the human race's stronger virtues. 
Issues such as "quantum abortion policies" will likely arise.
Asimov has already addressed many of the issues we will encounter
in the creation of synthetic conscious entities, and the time
seems ripe to refocus on those issues. It would not be a kind
milestone in human history to have synthesized "mentally deformed"
or "insane" quantum computers which are self-aware, or to have
created conscious computers as mental "slaves" or "genies" trapped 
in a bottle or cold magnetic field around a dewar.

This of course does not mean that we should halt all investigation 
into quantum computing. It is in our nature to understand ourselves.

But we do typically judge a person's wisdom by their ability to see 
"the big picture" or recognize their position in a broad context,
and to do this, requires that they unfocus their attention to some
extent by allowing themselves to see metaphors and include knowledge
from many specialized fields into their mental window rather than
focusing with their "nose to the ground" on only one narrow perspective.
It is dangerous to proceed in such a way, because our attention becomes
so narrow that we do not percieve impending dangers that would have
been visible if we had less focus. We cannot steer a ship well by looking
through a pinhole and this is what the increasing specialization in
society is forcing us to do. 

People who specialize very stongly excel in what they are are doing
and this is a necessary process for progress. But science and society
has lately I think, taken this too far into the extremes. 

It recognizes the overspecialization to some extent and there are 
trends to generalize. But as is often the case, some of these 
generalizations tend to be excessive and seek to completely dissolve 
all specializations to create that ideal of the "renaissance man", 
or a society of multifunctional telecommuting and easily disposable 
and replaceable homogeneous employee "units".

Human resources in commecial america is tending to reduce society
to the fast parallelism we find so attractive in the Hilbert space of
quantum computers because, after all, time is money.

Considering the larger picture, we may not be able to determine where
we are headed, but we can get a pretty good picture of the general 
direction in which we should head. The non-determinism is there
just as in quantum physics, and yet we still derive a deterministic
direction, just as in quantum physics.

A "spectrum" of a single particle provides little meaning for us.
Nor does a spectrum of a completely homogeneous dataset provide much
information. We also say these same things about morality. Morality
means nothing in the context of a single person. Morality also means 
little in terms of a homogeneous society where no two people are 
distinguishable. 

In this manner we will find that our concepts of space and time have 
a bearing on our concepts of morality. We, have already found this to 
be true. 

That the behaviour of men depends upon space and time is a 
Hopi statement of prophesy. Norbert Wiener has already warned 
us about the societal impact of cybernetics. There are many other 
sources deriving the same conclusions and bearing due warnings from
them. None of these sources needed modern physics to understand 
the basic principles of communications which all fields of study
are subject to at various resolutions. But it is quantum physics
that has pressed us into the urgency of its consideration.

That we are all subject to the laws of nature is common ground for
each of us to speak on, whether religious or scientific. When scientists
begin to rip into the fabric of natural laws itself to expose "unnatural"
areas of spacetime, such as in the Brookhaven National Laboratory 
"Hearts of Gold" experiments, we enter into a dangerous area of inquiry.

Because morality is likenable in properties to our concepts of spacetime 
as above, it would be a very wise thing on our part to question our
actions in ripping into the spacetime fabric like vivisectionists.

"Wisdom" is complementary to "intelligence". We get intelligence from 
specialization and wisdom from generalizations. Generals run
the army and we should in the same manner listen to the directions of 
the "wise", before we listen to the directions of the "intelligent".

While the Brookhaven Labs qualm the concerns of others, I am not comfortable
ignoring the warnings of other physicists. Their rationale is that they
will not create any events that are not already naturally occuring and
yet they will be exploring states where the laws of nature are quite
different from the usual. We safely harnessed the nuclear binding energy
largely because we theoretically knew what dangers to expect. We
did not discover the uncontrolled nuclear chain reaction empirically
by accident. 

Allowing empiricism to guide us is becoming more and more dangerous. 
We should have a sound theoretical basis before proceding into 
the submatter regions of physics. We do not know what to expect and
just as the binding energy of nucleons can be release catastrophically,
I suspect that even greater chain reactions are possible for the 
submatter binding energies with the other physicists with less controllable
results. We were perhaps very lucky that the nuclear chain reaction
was "naturally" attenuated and dispersed. It may well be that as we
probe deeper we will find that it becomes harder and harder for such 
chain reactions to be attenuated by nature and the cross-section of 
a submatter explosion becomes substantially larger being attenuated 
only by the vacuum of empty space.

All empiricism should be guided by theory. 
Otherwise we are children playing with matches.

"For the moment, in the race to produce QGP, it is RHIC that has the lead chance to strike
gold." 
      - Frank Close (from the link above)

We should not be in a race.


[1] see for instance Amdahl's law

[2] this is true largely in practical terms. In more theoretical terms
    the ideas of "quantum non-demolition", "non-interaction" or "watchdog" 
    experiments suggest that there are ways of measuring quantum systems 
    without disturbing them. See for instance Braginsky and Khalili, or 
    Kwiat, Weinfurter and Zeilinger, Tombesi and Mancini,... &c.