[MURG] Uploading IP (intellectual property)

[Ed Minchau]

 --- Yan King Yin <y.k.y at lycos.com> wrote: > From:
cat13 at illrepute.org
> 
> The open source movement generally overlooks the
> reward
> issue and therefore has a serious problem.
> 

Are you certain of this?  In open-source software,
there are other rewards than money.  Take Linux as an
example: a myriad of programmers write useful code and
share it without monetary compensation, and the
rewards are that everyone who wants to can get a free
operating system.  The reward for them is not having
to use Windows.  Not only that, since Linux is
"peer-reviewed", it evolves in much the same way as
science, consistently improving.


> The open-source model is entirely inappropiate for >
MU because no one will ever be able to make 
> significant contributions to MU in their
> home-grown laboratories; whereas most people who own
> a PC can contribute code.
> 

Uploaded minds will reside on some kind of computing
system.

SETI at home is an example of average people making
significant contributions to a megaproject.  Much of
mind uploading research will in fact be software and
massively-parallel number crunching.

Consider the caenorhabditis elegans uploading project.
 This is the simplest nervous system, 303 or so
neurons, only 968 cells in the whole animal.  A lot of
processing power is needed to figure out exactly in
what way those neurons are connected and the way that
each synapse modifies the calculation.  The rest of
the cells of the body must be modelled as well,
showing how each interacts with (perhaps) all the
other cells.  Not only that, a primitive environment
must be created in cyberspace, sufficiently detailed
to provide stimuli to which the "uploaded" worm can
react. 

In all liklihood, the virtual worm would not be alone
in this environment; its environment would necessarily
include other worms.  Count on modelling more than one
worm, perhaps a lot more than one.

Through careful study of the interaction of the
virtual worm and its virtual environment, and the ways
in which those are reflected in its nervous system, we
can better determine the actual connections involved
in its nervous system.  On a more basic level, we
learn how neurons really work, and can work on ways to
improve our scanning techniques by focusing on
important features we learne from the worm upload.

That is a lot of number-crunching.  However, if the
SETI at home idea is followed, that can be distributed
across the internet.  

The AI industry runs parallel and intersects with mind
uploading; where AI is top-down, mind uploading is
bottom-up.  Their end goal is the same: human-level
intelligence on an artificial substrate.  Note that
the moment the first uploaded mind is created, a
backup copy can be made, and then activated.  This
backup copy is an AI.  So, as soon as mind uploading
is a reality, so is human-level AI.  

There are many software people working on AI in their
basements, who could just as easily fit in with the
mind uploading crowd.  And they could just as easily
contribute code, which would be needed for a project
like the c. elegans upload project.

> >Since you're not seriously suggesting that any
> governmental or private
> >backer will fund the entire uploading project,
> could you clarify what
> >you are suggesting?
> 
> The consortium will *complete* the whole brain
> emulation
> project, and then it will start to generate revenue
> from
> the product. Only then, it will begin to pay back
> the
> original contributors whom the consortium has owed
> tokens.
> 

Asking people to pay for something which will only pay
off, if at all, _after_ the Singularity: sounds like a
government program.

If you want individual investors to spend big dollars
on something, then they should have reason to expect
to be paid back (with a profit) within their
lifetimes.

> >If you're suggesting that a subset of the required
> research is done and
> >kept proprietary, who will pick it up and do more? 
> Since it's proprietary,
> >the only group that can further it is the group
> that started it, which
> >leads to a single group funding the entirety of the
> project, which is not
> >reasonable.
> 
> I'm actually suggesting a single consortium for the
> entire brain emulation project (separate from BCI).
> It
> is difficult but I think technically it is feasible.
> The reason why this is necessary is that a partial
> project cannot generate enough income to pay back
> the funding, which is a form of debt financing (ie
> we get funding now and pay back later).
> 

There is no way that consortium would attract
investment for such an all-or-nothing project.  The
above suggests that the brain emulation project would
have one product: one emulated human brain.  This
would be X number of years away.  This single product
would then receive profit by....?  being sold?  (i
don't think selling a sentient being is allowed under
slavery laws)  Or perhaps the technique is sold.

This misses out on any chance of making any money to
perpetuate the project for X number of years.  What
happens if the money runs out at X-4 years?  Are
investors expected to put more money into the
consortium, on a product which, if it works at all, is
still 4 years away?

Instead, go for the incremental, multifaceted
approach: use those guys in their home-grown
laboratories, for starters. 

Ed

______________________________________________________________________ 
Post your free ad now! http://personals.yahoo.ca