Re: Hypothetical scenario


On Tue, 26 Aug 2008 21:17:51 -0500
John Novak <john.novak@xxxxxxxxx> wrote:

There is some of that, true, but there are also cross-disciplinary
journals, and the cost of publication drops every year. Also,
industry works on a very different model than the one you describe
above, so that certainly can't be the end of the story.

Indeed, industry works with a model that looks like there is a little
more coming out of it than what is going in because of increased
efficiency in using external resources. It is also not a guild like
hierarchical social system like academia. But as industrial companies
become larger and larger they become more more and more bureaucratic
and soon they are just as inefficient as governments or large
universities.

And don't forget there is also competition and thus closed source
material, hampering progress. But anyway, a limited extra gain is better
than a system where one mans gain is another mans loss, a zero sum
system, like a system of academic titles with no interdisciplinary
collaboration. But both academia and industry fail to realize larger
exponential growth for their followers, unlike systems like the
Internet or the hypothetical molecular assemblers that profit form the
fact that every newly created element can itself create new elements.

The problem with that scenario has nothing to do with academia and
nothing to do with industry, per se. The problem is that the scenario
you outline is a hell of a lot harder to put into practice than it is
to describe. No one actually knows how to build a completely self-
replicating device at *any* scale whatsoever, except pre-existing
living systems, or slight modifications on the same.

I claimed that academia was bad at cross fertilization between
different disciplines. Also, I didn't make the claim that reprap was
already hundred percent able to copy itself. I believe at the moment it
can make sixty percent of its parts and these parts still need to be
assembled by hand as far as I know.

What I was trying to say is that one can use small insights one gains
from similar projects even if they operate on a different scale or with
incomparable substrates. In practice, the whole of scientific progress
is determined for a large part by just such adaptations of insights
that appeared more or less randomly in other fields. But academia is not
as good at this as it could be.

As an example, someone in the reprap project found a notation system
(from a book chapter notation method?) to number the individual
machines that were produced. In contrast to sexual reproduction,
reprap machines have mostly only one parent so they can be numbered like
3.2.4.5.6 meaning the sixth descendant of the fifth descendant of the
fourth descendant etc. How long would it take this little meme to
spread if it had to wait for a publication?

Seriously.

It's very easy to say, "Oh, just build a machine tool that can make a
copy of itself," and very difficult to do that, because automation and
programmability are at the heart of this concept. Machine tools,
though, don't build circuit cards, and they don't build integrated
circuits to populate those boards, and they don't build chemical
batteries to power themselves or lay extra power cables from the
substation, etc. Also, each drop in scale takes effort and research
and a completely new program on your automated tool. So your just-so
inductive approach has a base-case we can't build and an inductive
iteration we can't generalize.

I'm not trying to prove my idea works, in fact I was claiming it fails.
However, we would learn a lot by establishing the actual point were it
fails. Just like transistor technology will ultimately fail in placing
more elements onto smaller and smaller substrates. But it has come a
lot further than most people thought.

Yes, and the idea of traveling to Alpha Centauri is pretty much the
same as the idea of traveling to the Moon.
Except for the details.

But one can learn a lot from finding out where the actual point of
failure is.

The different jargons and notational systems are like trade
protection barriers. If one looks at mathematical notation for
example, it is a totally insufficient system if one compares it
with modern programming languages.

I don't even understand what this means, since modern programming
languages are described in more abstract real-world mathematics, not
(in practice) the other way around.

Here's a starting point:

http://sandersn.com/blog/index.php?title=math_notation_is_terrible&more=1&c=1&tb=1&pb=1

P.

.

Relevant Pages

  • How industry is undermining academia, Jennifer Washburn, author of "University, Inc.", 200
    ... Murray aspartame research reviews 2005.01.18 ... while serving on the Food and Drug Administration's ... schools have been more infiltrated by industry than any other sector of the ... But when it comes to academia, ...
    (sci.med.nutrition)
  • Re: lyle mcdougal--diet plagiarist?
    ... Although I loathe Big Pharm and a what a lot of industry engenders, ... not great things about academia either. ... Talented scientists are able to choose to work in institutions ... The segregation of institutions by quality is inevitable, ...
    (misc.fitness.weights)
  • Re: lyle mcdougal--diet plagiarist?
    ... Most industry researchers would function *very* well in academia, ... One of the best was the NYTimes magazine section story on EJ Corey's Lab ... The segregation of institutions by quality is inevitable, ...
    (misc.fitness.weights)
  • Re: Wheres the YSI action??
    ... >>>US immigration restrictions are probably just right. ... >>>industry or academia) needs you...you'll get in without a sweat. ... > jump through hoops and pay through the nose for the pleasure. ...
    (rec.music.hip-hop)