Loading ...
Sorry, an error occurred while loading the content.
 

Re: Nano in SF Weekly/Wow

Expand Messages
  • steve wishnevsky
    ... i won t argue the point, just suggest that we are seeing the difference between hindsight and foresight...few people at the time thought that cars were
    Message 1 of 10 , Dec 30, 1999
      Clements, Robert wrote:

      > From: "Clements, Robert" <Robert.Clements@...>
      >
      > Wrong analogy. The motor car was a relatively simple fusion of two
      > preexisting systems (the carriage & the internal combustion engine) built
      > during a period where initial designs could literally be built in backyard
      > workshops & minimarketed before (even after: qv the infamous Morgan motor
      > co.) it became a mass produced product. It also had a series of obvious
      > broadly marketable applications (mostly in the mass transportation area) to
      > inspire construction.
      >
      > NT is a hi cost/hi tech concept which could conceivably do a lot of things
      > but which is currently _necessary_ for nothing... which is why the GM
      > analogy is uncomfortably important in this case.
      >
      > All the best,
      > Robert Clements <Robert.Clements@...>
      >

      i won't argue the point, just suggest that we are seeing the difference between
      hindsight and foresight...few people at the time thought that cars were more
      than toys for the rich and the eccentric, at the time. Horse had worked fine for
      two or three thousand years atthat point...steve w.
    • Christopher J. Phoenix
      ... I read that claim in a book on the Soviet space program. I can t find the book now, so I won t try to defend the claim. IIRC it was based on zero-G
      Message 2 of 10 , Jan 4, 2000
        At 10:19 AM 12/31/99 +1100, Clements, Robert wrote:
        >> As for who's paying, let's not forget that Mir has just about paid for
        >> itself over a time span of several decades.
        >>
        >MIR has been a capital drain on the Soviet/CIS/Russian economies since day
        >one. When this drain could be justified as a testbed for the secret manned
        >Mars program, the mission was politically ok; but when the politicoeconomic
        >will for such joyrides vanished, so did MIR's reason for being. In terms of
        >the money spent putting - & keeping - it in orbit, MIR could never have paid
        >its way; even if they ran commercial flights to it from here 'til
        >eternity....

        I read that claim in a book on the Soviet space program. I can't find the
        book now, so I won't try to defend the claim. IIRC it was based on zero-G
        crystal growth and other industrial products.

        >Whether the technology works or not isn't my point: it's whether it becomes
        >something people will pay good money to buy. Electronics has done splendidly
        >without NT: apart from limited specialist technologies (robotic space
        >exploration, say), why do you think people will continue to want to pay for
        >new technolgies every two years just because the capacity to create the
        >technologies exists?; & if they don't play ball, how do you expect to
        >finance the massive infrastructure costs involved in generating these
        >products?

        Why do people buy SUVs? It's not because they need them!

        Every few years, someone asks whether we really need all that compute power.
        And for five decades we have really needed it all to run the applications we
        want. Granted, the apps are inefficient--look at Windows vs. Linux memory
        requirements--but there are also things that really need that much crunch
        power, such as handwriting and speech recognition. Soon, video CODEC
        (especially driven by new display technologies). I don't know what the next
        killer app will be, but I'm pretty sure of two things: that there will be
        one, and that it will need at least a Pentium II-300 to run. The one after
        that will need a Pentium III-500.

        >The explosion in computer technologies - often used as a justification in
        >this context - is solely due to the fact that it started from a zero
        >comparative base (ie, not only were there no computers, there was nothing
        >seriously like computers in the amrket): even now, the nearexponential
        >growth is slowing (the main push for computing technologies is in gaming);
        >so what makes you think that NT can generate novelties which will induce a
        >new rush to buy in the future?

        NT "can" generate novelties in almost any field. Medicine, neural
        interface, infrastructure, construction, aerospace... Which industries would
        a replicating assembler make a difference in? I can't think of one where it
        wouldn't. Given that, which industries will want it badly enough to pay for
        it? Well, I'd guess that biotech, computers, and aerospace will want a lot
        of the preliminary steps, and then defense will push it over the edge
        (unless something else does first). Consider the military advantage of
        unlimited production capacity and nano-weapons. There's no way we could let
        someone else get there first.

        >Moore's law is already slowing; as it grinds against twin barriers of
        >technical limitation & market saturation. It's been a hell of a ride; but
        >the ridiculous stage of information technology profitability is ending,
        >kids. What possible mass market technology could refire it once computing
        >technology is inserted into all significant consumer items & video interface
        >technology reaches HDTV standard? None that i can see.

        Moore's law is slowing? That's news to me! What are your references? I've
        always heard that it was speeding up (the doubling time was decreasing).

        How about communications? There are several orders of magnitude to go
        before we can efficiently supply point-to-point HDTV video analogous to the
        phone system.

        Computing technology in all significant consumer items? Well... computers
        have been in cars for years, but GPS mapping is still in its early/luxury
        stages. We don't have anything close to a Pentium in most cars--or even in
        most houses.

        >(The rest of your analogies are at best speciality niches; although each in
        >themselves might be quite profitable... the military cash cow in particular.
        >I note, however, that you haven't provided a single unambiguous product in
        >your justification)

        A product? You want products? OK... understand that this will be a lower
        bound, not a realistic forecast. First, let's look at the gizmos that can
        be combined to make the products. Electric motors with 10,000 times the
        power-to-weight ratio of today's, at size scales down to 50 nm. "Smart"
        materials with embedded sensors and effectors. Fuel cells at all scales,
        99% efficient. Materials 10-100 times as strong as today's best. Computing
        too cheap to meter. Unlimited ability to specify complexity in the
        product--1,000 sensors are as cheap as one.

        An aircar, 100% reliable, with dial-and-forget navigation.

        A suit that adjusts thermal conductivity and keeps you clean and comfortable
        in environments from the South Pole to the Sahara, but when in a comfortable
        environment becomes almost undetectable to the wearer.

        A one-pound flying bomb (with no metal parts) that can reconfigure from
        quiet/stealthy to supersonic; has a range of up to hundreds of miles; can be
        programmed with complex maneuvers, reactions, and tropisms, and precise
        targeting. Hundreds of thousands can be configured, built, launched, and
        controlled each hour.

        A suspension of a few grams of cellular-sized air tankers; when injected,
        provides oxygen for several hours, and can recharge themselves from the
        lungs as appropriate. Useful for emergencies and recreation (SCUBA). This
        is R. Freitas' respirocytes.

        A surgical instrument that can create a painless, bloodless incision, and
        then heal it in seconds without a scar.

        An eye implant that taps into the optic nerve, and can capture what the user
        sees and replay it (to the same person). I think it's likely that we'll be
        able to decode and transfer the recording, but I'm not 100% sure of that.
        Also, this device might cause dizziness or other problems.

        A home power generator that'll run on almost any fuel; is 99% efficient;
        requires less space than an electric meter; is silent; has all the necessary
        electronics to sell power back to the grid; can regenerate fuel (store
        energy) from inputs such as solar or cheap nighttime power.

        >In a mature economic structure, the real money is made in servicing a
        >technology rather than building it (this statement is true of information,
        >too, by the way). Design products which will sell & then use the technology
        >necessary to build 'em is the way to succeeed... not invent a technology; &
        >hope like hell there's a sustainable market out there for it....

        A mature economic structure... does that mean stable, adjusted to the
        technology? I'm not sure we'll ever have that again. I'm not sure we
        should want it.

        Products that will sell... how much would you pay for a car that was safer
        than today's, more fuel efficient, got you places at 200 mph as the crow
        flies, and you didn't have to drive? Personally, I'd pay about three times
        what my current car is worth. If it took 100 kg of fairly pure materials to
        build... that would mean a manufacturing cost of less than $500. Prototypes
        would also cost $500 rather than $100,000, so the engineering cost would be
        quite low compared to today's process. Of course, it's possible that by the
        time nanotech gets cheap, there will be other technologies that can compete
        more-or-less effectively.

        But here's the paradigm I don't think you're getting. Nanotech is not for
        building individual products. It's a new, and much cheaper, way of building
        almost anything that can be designed. How many industries would find it
        worthwhile to own a technology that would allow them to cut their product
        engineering costs in half, and their manufacturing costs by 95%?

        How big is the cosmetics/fashion industry? Nanotech would completely
        revolutionize those industries, with products that can't be duplicated by
        other means.

        Imagine a new form of mass transit: it costs only $100 per mile to install,
        including dealing with the old infrastructure. You can install a mile in a
        day. It costs less than a penny per passenger-mile (including maintenance).
        Each passenger has their own capsule, which can be routed directly from
        source to destination at a top speed of 150 MPH (obviously less if there are
        sharp curves in the track). How long would it take to catch on? How much
        would it be worth?

        I could keep writing for hours, describing products that people would want
        if only they were cheap enough. And I wouldn't even have touched on the
        products that we can't imagine now--the real killer apps that will become
        possible when we can build 10^18 times as much complexity into our products
        for the same price. There's absolutely no doubt that one organization which
        controlled this technology (owned key patents, or security classifications)
        would become incredibly rich and/or powerful. It is an incredibly good
        long-term investment. Will it happen? I believe it will, because everyone
        will be afraid of other people getting to it first, and because within 20
        years it will be easy to do.

        Chris
        --
        Chris Phoenix cphoenix@... http://www.best.com/~cphoenix
        Work (Reading Research Council): http://www.dyslexia.com
        Is your paradigm shift automatic or stick?
      Your message has been successfully submitted and would be delivered to recipients shortly.