Re: [Gnu-arch-users] OT: trained dependency

[Andrew Suffield]

On Thu, Nov 11, 2004 at 03:39:12PM +0900, Stephen J. Turnbull wrote:
>     Andrew> Probably deservedly; most people have no more than a
>     Andrew> handful of original thoughts in their entire lives.
> 
> I doubt that.  I would agree that most people have no more than a
> handful of _useful_ original thoughts, and that as they get older
> their patterns of thought become more and more habitual.  But any six-
> year-old can produce a double handful of original thoughts, suitable
> for driving parents to hard drink, in a matter of an hour or less.

Bah, word games, that's what I meant :P

[I tend to assume people will take the interpretation that makes
sense, and skip defining every detail]

>     Andrew> As best I can tell, it's normally left to natural
>     Andrew> aptitude. I've never seen any evidence that there is a
>     Andrew> practical way to teach this sort of thing.
> 
> The teacher's side is called "mentoring", the student's side is
> "apprenticeship".  If you mean "can I [== Steve] lead a horse to
> calculus and then make him think?", you're right: I can't.  But if I
> watch a student and find out what interests her and what she's good
> at, I certainly can foster originality.

I am reasonably confident that this only works with a subset of people
(there's probably a high concentration of them to be found in a
university). That's approximately what I was referring to.

> You are correct
> that it is not possible to simply employ credentialed individuals and
> get the desired effect systematically.

And that's the relevant conclusion anyway.

>     Andrew> Summarising information is a learned skill, and a fairly
>     Andrew> mechanical one. It's not a particularly good measure of
>     Andrew> anything else (despite a number of popular testing systems
>     Andrew> which are based on it).
> 
> Nonsense.  The mechanical aspect is what I mean by "sed".  However, to
> effectively summarize information, ie compress it beyond what can be
> done with bzip2, one must start by choosing an audience and
> abstracting with that audience in mind.

Mmph... from my perspective this is as simple as a plane projection in
geometry; the space of information and the target plane are features
of the problem, and there's no real creative effort involved in
collapsing one onto the other. On the other hand I suspect you're
starting from a position where this problem is incompletely defined
and the task is first to finish specifying it.

> This is a highly social,
> creative activity---unless you live in a society that confuses
> "conventional wisdom" with "common (aka 'uncommon') sense"

Well, I do, but *I* am Discordian. "Common sense is what tells you
that the world is flat". [As a general rule I decline to comment on
British society; I'm sitting too close to see it clearly].

> Keynes's "beauty contest" as a
> metaphor for all of society, and not just financial bubbles.

I prefer to think of it as a sheep farm (very little difference, only
with less cute girls and more shit).

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'                          |
   `-             -><-          |

signature.asc
Description: Digital signature