Mark F. Cook · 12 July 1991 · reply to #2588
James Perkins ([email removed]) writes:
> Mark F. Cook [email removed] writes:
> > advance (which is exponential), the rate at which the human race advances
> > technologically should 'go vertical' in less than 300 years.
>
> One possible problem with Vinge's supposition is that it fails to take
> into account that people must apply the technology that has been
> discovered, and this takes time. The biggest limiting factor to the
> growth of technology, as I see it, are the factors of societal
> communication and understanding. It takes time for engineers to
> incorporate new ideas, it takes time to develop proficiency in their
> use, it takes time to bring people up to speed, and large projects where
> many new technological advancements are put together have a high risk of
> failure.
This is only true if the underlying technology needs to be widely
communicated, which may not be the case. Joe Citizen does not need
to re-learn how to drive a car, just because fuel injection, power
steering, and catalytic convertors get introduced. A closer example
would be the introduction of RISC architectures in UN*X workstation
SPUs. While these provide an order-of-magnitude (in some cases)
performance improvement in the computer, only a small sub-set of the
total user base (the system administrators) need any supplemental
training to take advantage of the new technology. To all the other
users, the change is transparent; their programs just run faster.
If the advances in technology occur in the information and communication
industry, then the rate of increase may go even faster.
> ... Look at the failure of the first n Mariner space probes, the
> incredible cost of the Space Shuttle, the Sergeant York program, Hubble,
> Galileo, the Soviet Mars moon probes, and the NeXT computer.
With the possible exception of Galileo and the NeXT computer, these failures
can be directly attributed to mismanagement and political infighting, not
the inability to disseminate new technology. (This is *not*, however, true
for the Sgt. York. It was a piece of sh*t from the start. But I digress...)
> I feel that, since there is a limited capacity and speed to which humans
> can (or WANT to) absorb information and knowledge, and there is a
> limited speed with which Systems Technology can be pushed forward, that:
I don't think you can make a blanket statement about the entire human race
like that and get it to hold water. While a fundementalist Christian (or
some other individual with strong political, cultural, or religious views)
may say, "I want nothing to do with TV, even though it will make me better
informed, since it can be a corrupting influence and is therefore a tool
of the devil.", a starving Ethiopian would almost certainly embrace radical,
new farming techniques, if the alternative was death from hunger. Closer
to home, what technophile TML member would not snatch up a new technology
offered to them, if it would make them wealthy or make life easier for
them. I maintain the once a society becomes comfortable with 'high
technology', it becomes easier to continue to introduce 'higher technology'.
> [[[ James' Maxim of Technology Advancement ]]]
>
> You can't push technology faster than it can be
> communicated, understood, incorporated, and perfected.
No argument here. I just think that the new technology is being
communicated, understood, incorporated, and perfected faster all
the time. I cannot believe that mankind has some inherent 'upper
limit' for accepting these changes. The limit is cultural, and
since the culture is modified by the new technology, and vice-versa,
I don't see any immediate obstacle to accelerated assimilation.
> To bring 10,000 worlds, or even 100 along with you, vastly slows the
> pace of technology growth, and results in much duplication of effort.
> In today's world, even bringing the 50 United States, Europe, and Japan
> along with you on a world with instant communication, takes time. How
> many wonderful new ideas have been sitting on desks for 20 years waiting
> for the rest of technology to catch up?
OK, now this is a different story. I had intended the Vinge example to
apply to a single world. When you talk about the Imperium, you are suddenly
faced with the light (or jump-6) barrier as an absolute propagation speed
limiter. Also, one cannot assume that a new technology will be disseminated
planet wide the instant it arrives from space, *unless* it is the latest
in a smooth series of innovations. However, this is an ideal situation,
and hardly likely to occur.
> Yes, we are in the age of technological growth. I see advancements in
> complexity going exponential, but I see systems technology growing
> linearly. I see improvements in computer speed growing exponentially,
> but the software technology bogging down in complexity management.
If the hardware speed can advance, why can't the software's ability to
maximize the performance do likewise? Just because it hasn't yet doesn't
mean it won't. As for linear growth, processor speed is currently
increasing geometrically (currently about 2x per year), not linearly.
> What about ceramic high-temperature turbine-powered drive-by-wire
> automobiles? One innovation at a time, please! Serial, not parallel
> improvement. New ideas on a proven base. It's they way human
> engineering and technology transfer works. The human mind has a limited
> capacity to absorb and use new ideas.
Not true! all three concepts (ceramic engines, turbines, and drive-by-wire)
were developed independently of one another. None of those ideas had to
wait for any of the others to be perfected before they could work.
Synthesis is the fundemental nature of innovation. Right now, the only
thing preventing all three concepts from being incorporated in next year's
Chevrolet is one person with a sufficiently large checkbook. :-)
As for the human mind having a limited capacity to absorb and use new
ideas, that may be, but we're nowhere near the limit. Using a work-
station running X-Windows, I can absorb an order of magnitude more info
today than I could have on an ASCII terminal 5 years ago. I maintain
that the *true* limiting factor is the efficiency of the data interface.
Vinge agreed with this and postulated a direct mind-machine neural link.
the most efficient way to inform a person is to pump the date directly
into the head, in a fashion that doesn't overwhelm the recipient. How
one does that is still in the realm of SF, but I believe it *will* take
place, eventually.
Later,
Mark F. Cook
USMail: User Interface Technical Support
Hewlett-Packard - Interface Technology Operation
[address removed]
INTERNET: [email removed]
[email removed]