Dear Mike,

I apologize for taking so long to respond. It's just that I decided
that this was too much fun and was taking too much time, so I decided
to go cold turkey. But I think the time has now come.

Originally Posted by Mike Alexander '59

> How about some more a prosaic things like walking, performing
> housekeeping chores, bussing tables at a restaurant or hospital
> orderly work?

> Biped robots keep falling down, because the computers who control
> them can't handle the complex balancing task of walking,
> especially when it is windy. They have a problem avoiding objects
> that unexpectedly appear in their path. We've been working on this
> for 40 years with little success. Yet bipedal birds, with brains
> tens of thousands of times smaller than ours can walk just fine
> and they don't run into things. Surely the most powerful computers
> today have achieved the level of birdbrain.

> Housekeeping is still further away, because not only do you have
> to move around, you have to recognize objects and manipulate them
> and move around them in a constantly-changing environment.

The question that you're addressing here has nothing to do with the
intelligence of computers, but rather with the way that the computer
is packaged.

As I've previously mentioned, it's nice to think of intelligent
computers looking like either Arnold Schwarzenegger or Kristanna Loken
(or maybe Arianna Huffington after Terminator 4 comes out).

But I imagine that they'll look like computers or machines or
something.

From your reference to housekeeping, I assume that you're referring
to the Roomba:



As you indicate, the Roomba isn't very intelligent today, although
they've been coming out with new versions. The point is that these
devices will all become increasing intelligent as time goes on, and
by the 2020s, I believe you'll have intelligent devices that will do
the housework, fix the plumbing, make the bed, and do all sorts of
other things.

Here's another example, and this will give you an idea of what the
intelligent robot of the future might look like:



This device could have all sorts of arms and tools attached to it.
There could be various arms equipped with anything from screwdrivers
to guns. It could be integrated with some sort of rolling platform,
and so the device could indeed buss tables at a restaurant or do
hospital orderly work.

An interesting feature is the Vision and Targeting Head. Once computer
vision works a lot better, the robot device could actually have
multiple "eyes" attached to a number of "arms," and so could look
simulataneously in all directions, and around corners, and so forth.

In the diagram above, there's a computer in the upper left hand
corner. That's because the Packbot that you purchase today is, in
fact, controlled by this computer. By 2030, all the intelligence
will be packed into the robot itself.

So by the 2030, we'll have self-contained, autonomous,
super-intelligent computers able to do pretty much anything that a
human being could do.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Dear Brian,

Originally Posted by Brian Rush

> Then I suggest you suffer from a paucity of the imagination. I'm
> not being that abstruse here.

I agree. "Abstruse" is not the word I would have chosen either. The
word "incoherent" seems more appropriate.

> Genetic modification of humans so as to achieve higher
> intelligence is speculative, granted, but no more so than the
> production of self-aware computers.

I know that a lot of people talk about genetic modification of human
beings to achieve greater intelligence, but I have a problem with it.
The problem is the same as the reason why you can't take an old
computer and just a new CPU and more memory into it. There may be no
place to plug in the memory, and the faster CPU won't do any good if
the bus structure doesn't support or it, or if the power supply can't
power it.

If you try to use genetic modification to make the brain more
intelligent, then I suspect you'd have to restructure the entire
human body, to provide the new super-brain with a greater blood
supply and a more robust nerve structure. I really don't believe
that those problems can be solved in the time frame of the next few
decades.

> Your assumption that the conditions of the past will continue
> w/r/t human intelligence is not necessarily valid.

This whole forum is about applying the lessons of history to the
future.

> Possibly. I'm smarter than most of them, definitely.

No comment.

> But that's not the point; the point is that, however smart they
> are, they are currently following blind leads, and thus have no
> prospect of success.

This is just plain silly. For example, I covered IBM Research Lab as
a journalist, and they never spent substantial money following blind
leads. That would be even more true today with research dollars so
scarce. IBM's Blue Gene project, for example, is no blind lead.

Furthermore, the whole concept of technological growth is that
exponential growth curves would be followed EVEN IF labs around the
world were following blind leads. If you have a sufficient number of
researchers attacking the same problem, then each new discovery, each
new invention will occur at exactly the right time, if not by one
researcher then by another. The exponential growth curves are
relentless.

> What's more, it's not a matter just of throwing in some
> indeterminacy, but of structuring the whole decision-making
> process around indeterminacy, because except in carefully
> pre-chosen circumstances we almost never make decisions on
> sufficient data. What we have here is a guaranteed delay. So your
> projected date is almost certainly off.

Once again, I don't even understand this. I've discussed several
thought processes throughout this thread, and the algorithms that
would be used to implement them, and indeterminacy was never an
issue. And if I'm wrong about that, then the algorithms I discussed
could easily be modified to introduce what indeterminacy is required.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Dear Corvis,

Originally Posted by Corvis

> This is definitely true, and in fact the selective pressures to
> produce humans with enhanced intelligence (and other features)
> will virtually guarantee a desire to do it.

See my response to Brian, and my misgivings that this can be
accomplished.

Ray Kurzweil has speculated about a process whereby the human brain
can be "reverse engineered." This would be done by injecting a bunch
of nanocomputers into a person's blood stream to travel up to the
brain and record every neuron and every connection and every
neurotransmitter concentration represented in each synapse-thin
layer. Then, according to Kurzweil, the brain's logic can be
reengineered in a computer, and so the computer will become a mental
clone of the person.

> Another line of "attack," yielding a similar result, would be the
> development of particular capabilities "on chip," and then the
> integration of those capabilities with the human brain. In either
> case, though, these technologies would need to be developed prior
> to the development of intelligent machines in order to invalidate
> John's concerns, and then the development of enhanced humans would
> need to keep pace with those machines.

Here's where I have a problem with this: If you reverse engineer
someone's brain and re-engineer it in a computer, then what do you
have? Do you have a human being in the shape of a computer? Or do
you simply have another computer? Does this computer feel pain?
What if the pain-generating neurons are turned on? Does the
computer's voice synthesis circuits then start screaming? And if
you're going to re-engineer some person's human brain, why not make
that a part of a larger computer software system that can do other
things? For example, this super-intelligent computer would have its
own CPU, memory and software, but when it needed to draw on Joe
Smith's previous experiences, it could just boot up the Joe Smith
brain module and check things out.

In other words, if all this happened, would the human race still
exist?

> I'm staying out of the debate concerning the timetable for any of
> these technologies. I think they all have numerous obstacles that
> must be overcome before they are practicable -- some foreseen and
> some not -- and, really, only time will tell how things unfold.

Cluck, cluck.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Dear Titus,

Originally Posted by Titus Sabinus Parthicus

> One thing that has been forgotten in this discussion, or at least
> has been left out, are Isaac Asimov's 'Three Laws of Robotics',
> which, IIRC, were adopted by the UN back in the 1960s as
> authoritative worldwide, and which I see as being equally
> applicable to AI computers.

Actually, we did discuss Asimov's laws here. See my June 10 posting,
for example, on page 1 of this thread.

> For the purposes of this thread, I'd see Law One as being the
> most applicable, with Law Two coming in a close second.

The problem is this: Even if the U.S. implemented this or some other
set of laws, that wouldn't prevent China or some other country from
ignoring them. In fact, if there's anything that I'm certain of,
it's this: That this new super-intelligent computer techology will,
like all new technology, be used immediately on the battlefield, and
the country that uses it first will win. That's why no one, even the
U.S., will every implement Law One.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Dear Sean Love and Richt,

This material on language development and evolution is very
interesting, and it probably will play a part in the software
algorithms for the development of the first super-intelligent
computers.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Dear Justin,

Originally Posted by Justin '77

> From The Independent ...

I'm really having a lot of trouble with this article. I found a copy
of the full article you referenced at
http://www.webwarrior.net/modules.ph...ticle&sid=3711
.

Roger Dobson, who wrote the Independent article, doesn't mention who
the researchers are, or anything about these mysterious algorithms.

I found the journal Brain and Mind at http://www.epub.org.br/cm/ ,
but I can't find any article that appears to be on the subject that
Roger Dobson is referencing. I did read a few of the articles, but
they don't seem to be particularly well researched.

So, here's my response:

First off, the algorithms that I described earlier in this forum were
not based on counting neurons. These are brute-force pattern
matching algorithms that I estimate will be easily within the
capability of 2030 computers, which will be 100,000 times as powerful
as today's computers. And if I'm wrong about that date, then
computers in 2050 will be 100,000,000,000 times as powerful as
today's computers.

Secondly, the Independent article indicates that the researchers'
conclusions are not based on the 100 billion neurons in the human
brain, but on what they say are 10^8342 neural connections. This is
much larger than the number of elementary particles in the universe
(according to
http://www.transhumanism.org/piperma...er/000495.html
), so it doesn't seem credible.

Also, the Independent article ends with:

> Ironically, the discovery could be used to change the way that
> computers are designed. Instead of adding more bytes, they could
> mimic the human brain, with more emphasis on connections.

In other words, just redesign computers so they use connections
rather than bytes of memory, and the problem is solved.

Putting all of this together, I don't believe that Roger Dobson's
article has very much credibility.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Dear Croaker,

Originally Posted by Croaker'39

> But that did not occur in any substantial way ~50,000 years ago,
> certainly not enough to "release culture." The symbolic language
> of the kind that brings on cognition must have occurred ... more
> like 3,000 years ago. Prior to that, according to Jaynes, humans
> were precognitive, bicameral sheep in a pasture of emotional
> insecurity.

The 3,000 year figure surprises me, since "modern" humans have been
around for some 50,000 years, as you point out. What does
"bicameral" mean? I've only seen that word used in the context of
legislatures.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Originally Posted by John J. Xenakis

So by the 2030, we'll have self-contained, autonomous, super-intelligent computers able to do pretty much anything that a human being could do.

When work on robots began it was widely believed that we would have housekeeping robots before now.

IIFC the Roomba doesn't have a sophisticated processor. The problem with building robots that walk, recognise objects and manipulate them (i.e. a housekeeper) is not that our computers lack the processing power, it's that we really don't know how people do these things so we can't program machines to do them.

Originally Posted by John J. Xenakis

Dear Croaker,

Originally Posted by Croaker'39

> But that did not occur in any substantial way ~50,000 years ago,
> certainly not enough to "release culture." The symbolic language
> of the kind that brings on cognition must have occurred ... more
> like 3,000 years ago. Prior to that, according to Jaynes, humans
> were precognitive, bicameral sheep in a pasture of emotional
> insecurity.

The 3,000 year figure surprises me, since "modern" humans have been
around for some 50,000 years, as you point out. What does
"bicameral" mean? I've only seen that word used in the context of
legislatures.

John

John J. Xenakis
john@generationaldynamics.com
http://www.generationaldynamics.com

Bicameral Mind HTH

Dear Mike,

Originally Posted by Mike Alexander '59

> When work on robots began it was widely believed that we would
> have housekeeping robots before now.

> IIFC the Roomba doesn't have a sophisticated processor. The
> problem with building robots that walk, recognise objects and
> manipulate them (i.e. a housekeeper) is not that our computers
> lack the processing power, it's that we really don't know how
> people do these things so we can't program machines to do them.

As I understand it, the Roomba works pretty well today.

The company keeps coming out with new versions:



I've read somewhere that the only real problem it's having today is
distinguishing floor transitions, like moving from a bare floor to a
rug. But we know how to solve that problem with improved computer
vision, which we'll have within 10-20 years.

John

IRT this thread's original idea (that AIs would eventually turn on us and destroy us, roughly during the next saeculum's 4T), has anyone seen the new (and improved, IMO) version of 'Battlestar Galactica'? The one in which humanity had fought off that very threat once. (Humans had created the Cylons to be a race of robot slaves in this one, and the slaves eventually rose up against their oppressors, hoping to exterminate them, thereby gaining their freedom. They were eventually exiled to another star system, after losing that round.) However, they came back 40 years later with a nuclear/cyber BOB attack that achieved total surprise, and total victory over the now-doomed human race within 48 hours - and these Cylons were far more advanced than the 'chrome-plated walking toasters' of the old 70s'-era series.

I really enjoyed the new BG, and I hope it becomes a continuing series.

Same here. In fact, I should soon have the miniseries itself in my DVD collection, courtesy of Ebay.

Dear Titus,

Originally Posted by Titus Sabinus Parthicus

> IRT this thread's original idea (that AIs would eventually turn on
> us and destroy us, roughly during the next saeculum's 4T), has
> anyone seen the new (and improved, IMO) version of 'Battlestar
> Galactica'? The one in which humanity had fought off that very
> threat once. (Humans had created the Cylons to be a race of robot
> slaves in this one, and the slaves eventually rose up against
> their oppressors, hoping to exterminate them, thereby gaining
> their freedom. They were eventually exiled to another star system,
> after losing that round.) However, they came back 40 years later
> with a nuclear/cyber BOB attack that achieved total surprise, and
> total victory over the now-doomed human race within 48 hours - and
> these Cylons were far more advanced than the 'chrome-plated
> walking toasters' of the old 70s'-era series.

I haven't seen the new Battlestar Galactica, but as I approach
senility I become more and more impatient with this genre of science
fiction, including Terminator movies and the Matrix
movies.

The underlying plot is conflict between humans and intelligent
computers, and so far so good. But the problem I have is that
technology then stands still in these movies, as if computer
technology development will stop cold as soon as the first
intelligent computers are available. That simply doesn't make sense,
since the intelligent computers would have to continue developing
increasingly intelligent versions of themselves to survive (survival
of the fittest). And I'm finding it increasingly difficult to
suspend disbelief when I see one of these movies. There'd better be
some pretty good sex scenes to make up for it.

John

John J. Xenakis
john@generationaldynamics.com
http://www.GenerationalDynamics.com

John,

Actually, there was considerable technological advancement during the 40 years between the two Human-Cylon wars, all of it on the Cylon side - the Humans had been frightened into abandoning all such efforts by the first war. That, and a little espionage, gave the Cylons total victory in the second war within 24 hours of war breaking out.

BTW, there was a great sex scene near the beginning, between Dr. Gaius Baltar, and Number Six, who just happened to be the Cylon spy I referred to above.

Originally Posted by Titus Sabinus Parthicus

John,

Actually, there was considerable technological advancement during the 40 years between the two Human-Cylon wars, all of it on the Cylon side - the Humans had been frightened into abandoning all such efforts by the first war. That, and a little espionage, gave the Cylons total victory in the second war within 24 hours of war breaking out.

BTW, there was a great sex scene near the beginning, between Dr. Gaius Baltar, and Number Six, who just happened to be the Cylon spy I referred to above.

Dear Titus,

So what happens? All the humans are killed? Then what?

John

Originally Posted by John J. Xenakis

Originally Posted by Titus Sabinus Parthicus

John,

Actually, there was considerable technological advancement during the 40 years between the two Human-Cylon wars, all of it on the Cylon side - the Humans had been frightened into abandoning all such efforts by the first war. That, and a little espionage, gave the Cylons total victory in the second war within 24 hours of war breaking out.

BTW, there was a great sex scene near the beginning, between Dr. Gaius Baltar, and Number Six, who just happened to be the Cylon spy I referred to above.

Dear Titus,

So what happens? All the humans are killed? Then what?

John

All except for the 'Galactica' fleet, which was formed around an old battlestar that had been ready for decommissioning when the Cylons attacked. The only ships that made any headway at all against the Cylons' jammers were either old or in serious need of maintenance - thus what little tech the humans had developed since the first war was (most fortunately!) offline. Said fleet, as in the original TV series, escaped into uncharted space, and is looking for Earth. And the Cylons are now looking for that fugitive fleet - to finish the job. (Sounds like the pilot for a new series to me.)

Dear Titus,

Originally Posted by Titus Sabinus Parthicus

All except for the 'Galactica' fleet, which was formed around an old battlestar that had been ready for decommissioning when the Cylons attacked. The only ships that made any headway at all against the Cylons' jammers were either old or in serious need of maintenance - thus what little tech the humans had developed since the first war was (most fortunately!) offline. Said fleet, as in the original TV series, escaped into uncharted space, and is looking for Earth. And the Cylons are now looking for that fugitive fleet - to finish the job. (Sounds like the pilot for a new series to me.)

Hmmmm. That reminds of a comic book story that I read in the 50s. Everyone was ridiculing an old man who kept around a bunch of old wooden airplanes. Then the aliens attacked earth, and used their magic ray weapon to destroy all the modern planes that the Earth forces were flying around in. Well, come to find out that wood is impervious to the aliens' fabulous ray, and so the wooden planes come out and kill off all the aliens.

John

11,000 Cleancut Nanobots



Forget 5 o'clock shadow. The 11,000 Cleancut Nanobots are so tiny you
can't see or feel them. But as you go about your daily routine,
they'll scour your face for signs of whiskers and cut them off. Here's
how it works. The first time you let the nanobots loose, they create a
3-D map of your face and send it to the computer. You outline on the
computer where the nanobots should and shouldn't shave. You can even
try out different styles before setting the microscopically tiny
robots in motion.

When you want to give the nanobots a rest, you wipe your face with an
electrically charged cloth designed to collect them. When you want to
release them again, push a button on the cloth to switch the
electrical charge and off they go. The advantages? The nanobots do all
the work, there's no 5 o'clock shadow, no nicks, no chafing, and no
need for shaving cream.

Boston Globe, 1/15/2004

http://www.boston.com/dailyglobe2/01...he_edge+.shtml

11,000 Cleancut Nanobots



Forget 5 o'clock shadow. The 11,000 Cleancut Nanobots are so tiny you
can't see or feel them. But as you go about your daily routine,
they'll scour your face for signs of whiskers and cut them off. Here's
how it works. The first time you let the nanobots loose, they create a
3-D map of your face and send it to the computer. You outline on the
computer where the nanobots should and shouldn't shave. You can even
try out different styles before setting the microscopically tiny
robots in motion.

When you want to give the nanobots a rest, you wipe your face with an
electrically charged cloth designed to collect them. When you want to
release them again, push a button on the cloth to switch the
electrical charge and off they go. The advantages? The nanobots do all
the work, there's no 5 o'clock shadow, no nicks, no chafing, and no
need for shaving cream.

Boston Globe, 1/15/2004

http://www.boston.com/dailyglobe2/01...he_edge+.shtml

11,000 Cleancut Nanobots



Forget 5 o'clock shadow. The 11,000 Cleancut Nanobots are so tiny you
can't see or feel them. But as you go about your daily routine,
they'll scour your face for signs of whiskers and cut them off. Here's
how it works. The first time you let the nanobots loose, they create a
3-D map of your face and send it to the computer. You outline on the
computer where the nanobots should and shouldn't shave. You can even
try out different styles before setting the microscopically tiny
robots in motion.

When you want to give the nanobots a rest, you wipe your face with an
electrically charged cloth designed to collect them. When you want to
release them again, push a button on the cloth to switch the
electrical charge and off they go. The advantages? The nanobots do all
the work, there's no 5 o'clock shadow, no nicks, no chafing, and no
need for shaving cream.

Boston Globe, 1/15/2004

http://www.boston.com/dailyglobe2/01...he_edge+.shtml

Century City

CBS will run a new drama, Century City, starting on March 17.

The premise is a law firm in the year 2030.

I wonder if any of the lawyers will be super-intelligent computers?

For that matter, will any of the defendants be super-intelligent
computers? If a super-intelligent computer commits murder, is the
computer guilty or is the programmer guilty?

Sincerely,

John

John J. Xenakis
E-mail: john@GenerationalDynamics.com
Web site: http://www.GenerationalDynamics.com


http://www.viacom.com/press.tin?ixPressRelease=80254170

CBS ANNOUNCES MID-SEASON SERIES PREMIERE DATES

Futuristic Legal Drama 'Century City' Premieres Tuesday, March 16,
9:00-10:00 PM

'The Stones,' A Comedy About Two Adult Children Living With Their
Divorcing Parents, Debuts Wednesday, March 17, 9:30-10:00 PM

CBS announced premiere dates for its two mid-season series -- the
drama CENTURY CITY and comedy THE STONES.

CENTURY CITY, a provocative legal drama set in the year 2030, will
premiere on Tuesday, March 16 (9:00-10:00 PM, ET/PT) on the CBS
Television Network.

Nestor Carbonell, Viola Davis, Emmy Award-winner Hector Elizondo, Ioan
Gruffudd, Kristin Lehman and Eric Schaeffer star in this futuristic
drama where lawyers at the firm of Crane, Constable, McNeil and
Montero find that the laws change but that people remain the same.

In a time when judges can go before lawyers as holograms, the firm
takes on such morally and ethically ambiguous cases as parents suing
their doctor for withholding critical results of their unborn child's
genetic mapping, and a man accused of robbery when he `steals' back
his identity from his ex-fianc? who has uploaded his presence onto all
of her electronics. As this diverse group of attorneys find
themselves in uncharted legal territory, their cases provide an
eye-opening look into issues confronting society in the not-so-distant
future.

CENTURY CITY is produced by Heel and Toe Films in association with
Universal Network Television. Paul Attanasio, Katie Jacobs and Ed
Zuckerman are executive producers.

Regular time period occupant THE GUARDIAN will return on Tuesday,
April 27.

[End of message]

Thank you, John. These posts are very useful.

--Croak

Originally Posted by Croakmore

Thank you, John. These posts are very useful.

--Croak

And fun, too!

John

Scientists creating life from scratch

The following article almost makes it sound like we'll soon be able
to create life using ordinary ingredients found in your home
refrigerator.

Early in this topic, we had a lengthy discussion of whether a
super-intelligent computer could be considered "alive."

Now we have scientists telling us that they'll be able to create
cellular life within ten years.

I recently had occasion to put together a new graphic on what the
future is going to look like:



When we try to look forward into the future, we can see a 4T time
with the "clash of civilizations," following by a period of great
prosperity.

But around 2030, the point at which computers will become more
intelligent than human beings ("The Singularity"), there's a big
"Danger" sign blocking our vision, and it seems to be impossible to
see beyond that sign. No matter how I try to analyze it, as
elementary a question as whether the human race will survive much
past The Singularity seems to depend on chaotically random events
which cannot be predicted any more than next month's weather can be
predicted. This means that we won't be able to see beyond that
"Danger" sign until we're there, and that's possibly the scariest part
of this.

Sincerely,

John

John J. Xenakis
E-mail: john@GenerationalDynamics.com
Web site: http://www.GenerationalDynamics.com




Researchers Exploring 'What Is Life?' Seek To Create a Living Cell

April 2, 2004; Page B1


SCIENCE JOURNAL
By SHARON BEGLEY


As any geek who ever soldered together a circuit board from
off-the-shelf parts can testify, if you truly want to understand how
something works, you need to build it yourself.

That approach doesn't raise any eyebrows when applied to gizmos and
gadgets, but now a loosely organized band of scientists is extending
it in an audacious way. In hopes of answering the age-old question
"what is life?" they are trying to assemble -- from off-the-shelf,
nonliving molecules -- a living cell.

"Creating a cell from scratch is probably at least 10 years away, but
it is going to happen," says Mark Bedau of Reed College, Portland,
Ore. "We're in for some very interesting, very profound new ways of
thinking about what life is, and about where you draw the boundary
between life and nonlife."

One of the deepest mysteries in biology is how molecules that are no
more alive than the tip of a pencil can form a reproducing,
metabolizing, evolving organism. If you plop a droplet of any of the
molecules that make up living cells (fats, amino acids, water, DNA,
other organic molecules) onto a glass slide, it just sits there. No
one would mistake it for a living thing. Yet when the right
ingredients assemble in the right proportions, the result comes alive,
as it did on Earth some 3.8 billion years ago.

The transformation is so profound that most scientists until the 19th
century believed in the theory called vitalism. This holds that living
things possess a mysterious "vital spark" that endows them with life,
and that life cannot be explained by mere chemistry and physics. But
today, harnessing no more than thermodynamics, electromagnetism and
chemistry, scientists are taking steps toward creating a living cell.

The first step is to separate the would-be cell from the outside
world, and this turns out to be startlingly simple. Several
researchers have created little self-replicating vesicles, minuscule
bubbles much like the membranes around living cells. In a special
mixture of oil and water, Luigi Luisi of the Swiss Federal Institute
of Technology, Zurich, has found, membranes form spontaneously, grow
by incorporating small molecules from the outside world, and reproduce
by pinching themselves in two, like amoebas.

Dr. Luisi's vesicles fulfill two of the main requirements for life,
growing and reproducing. In addition, says David Deamer of the
University of California, Santa Cruz, a living cell must transform raw
materials and energy into more of itself (metabolize), and also
evolve. Although no one has gotten a single vesicle to carry out all
of these reactions, Prof. Deamer and his colleague Pierre-Alain
Monnard are coming close.

"As they form, the vesicles capture a polymerase enzyme that strings
together small molecular building blocks into more complicated
molecules," says Prof. Deamer. "The vesicles take in molecules from
the outside, and the polymerase uses them as nutrients and as an
energy source to synthesize RNA," a cousin of DNA.

Other groups of scientists have gotten simple amino acids to link
together into proteins, like beads linking into a biological necklace.
The reaction occurs on the surface of the vesicle, which somehow
jump-starts the assembly. The vesicles created in the lab are not just
dumb containers; they support biological reactions.

Last fall, molecular biologist Jack Szostak of Massachusetts General
Hospital, Boston, and colleagues reported that common clay particles
have an unsuspected talent. They can speed up the conversion of little
clusters of molecules into vesicles, making the formation of a cell
membrane even easier. Inside the vesicle, the clay particles grab hold
of short bits of RNA and assemble them into a long strand. Voila: a
little sphere containing genetic material able to grow and copy
itself.

The missing ingredient in this cell wannabe is metabolism, but Steen
Rasmussen of Los Alamos National Lab thinks he can provide it. He and
Liaohai Chen of Argonne National Lab have designed a microscopic
container with metabolic molecules and genes whose electrical
properties drive metabolic reactions. The scientists have demonstrated
experimentally that this micrometabolism can produce exactly the
molecules the container is made of (so the system would be able to
grow).

"All the pieces are there -- self-assembling container, genes and
metabolism that captures energy from the outside world," says Dr.
Rasmussen. "The question is, how do we get it to reproduce? If we do,
then most people would say it is alive."

If researchers manage to create living cells from scratch, their
mastery of the machinery of life could blur the line between alive and
not-alive. Combining the traits of artificial cells with
nanotechnology, Dr. Rasmussen and colleagues wrote in a recent issue
of Science, could produce machines that "would literally form the
basis of a living technology possessing powerful capabilities and
raising important social and ethical" questions. Adds Prof. Bedau, "It
will be crossing a threshold, enabling technologies we can't even
imagine now."

Scientists are close enough to creating life in the lab that it is
time to start a public debate about what that would mean -- for
traditional views of the sanctity of life as well as for whether the
creators will be able to control their creations.



(*) You can e-mail me at sciencejournal@wsj.com1.

URL for this article:
http://online.wsj.com/article/0,,SB1...371990,00.html

Hyperlinks in this Article: (1) mailto:sciencejournal@wsj.com

Copyright 2004 Dow Jones & Company, Inc. All Rights Reserved

This copy is for your personal, non-commercial use only. Distribution
and use of this material are governed by our Subscriber Agreement and
by copyright law. For non-personal use or to order multiple copies,
please contact Dow Jones Reprints at 1-800-843-0008 or visit
www.djreprints.com.