Discussion:
Approximate v. Exact
(too old to reply)
Acme Diagnostics
2005-02-21 18:03:03 UTC
Permalink
There are two types of thinking: Approximate and Exact.

Approximate always trumps Exact.

One reason is that Approximate includes Exact, and by
how much, approximately.

That's why people who can think in Approximate run
things and make the decisions.

Larry
Tron
2005-02-21 18:13:33 UTC
Permalink
Post by Acme Diagnostics
There are two types of thinking: Approximate and Exact.
Approximate always trumps Exact.
One reason is that Approximate includes Exact, and by
how much, approximately.
That's why people who can think in Approximate run
things and make the decisions.
Well, that's not quite accurate, but it holds in a general sense.

T
Acme Diagnostics
2005-02-21 18:51:05 UTC
Permalink
Post by Tron
Post by Acme Diagnostics
There are two types of thinking: Approximate and Exact.
Approximate always trumps Exact.
One reason is that Approximate includes Exact, and by
how much, approximately.
That's why people who can think in Approximate run
things and make the decisions.
Well, that's not quite accurate, but it holds in a general sense.
Agree. But I'm having trouble thinking of counter examples. The
first thing that came to mind was the Manhattan Project. But is
that really correct? It's physics and particle behavior based on
probability. The construction issues would be Approximate.
Was the final decision-maker involved in anything Exact?

Then I thought about the trajectory of a space-ship launch. But
as accurate as that needs to be, isn't it still only
approximate, i.e. to so many decimal places? Is the person who
actually runs the program, the final decision-maker, involved in
that?

Ok, I've got one: The ship captain says "full steam ahead."
That sounds like an exact instruction the must be followed
exactly. And the captain is surely running things and making
the decisions. But still, doesn't the captain know that .xxx% of
the time there will be a miscommunication or inability to
follow the order and must allow for that Approximately?

I'm sure there are plenty of counter-examples. I'd like to find
some to judge how Approximate my statement really is.

Larry

p.s. Thanks Tron.
Tron
2005-02-21 21:35:04 UTC
Permalink
Post by Acme Diagnostics
Post by Tron
Post by Acme Diagnostics
There are two types of thinking: Approximate and Exact.
Approximate always trumps Exact.
One reason is that Approximate includes Exact, and by
how much, approximately.
That's why people who can think in Approximate run
things and make the decisions.
Well, that's not quite accurate, but it holds in a general sense.
Agree.
I'm really sorry; I was trying to make a joke.

"Accurate ... exact .... not quite true, but in a general sense it is a good
approximation ..."
Sort of using your own words to formulate a counter argument.

Seriously:

0) Style and consistency
It would be more fitting if your thesis weren't an absolute ("
...always..."), but obeyed itself to state: "Approximate beats Exact in,
roughly, ninetynine out of hundred cases, give or take twenty" (the reverse
of "67,3 % of all statistics are bogus".)

1) Crux
"One reason is that Approximate includes Exact, and by
how much, approximately."
If there is any mistake here, it is in this analysis. As another poster
pointed out, the difference between exact and approximate can be seen as a
difference in degree, hence it is not a difference in kind.
In any case, it is a very important part of the analysis, because "included
in" implies that such knowledge may be called forth at times of need by
further investigation of available approximate knowledge. If it were true,
it would indeed confer great benefits. But that, I think you will find, is
not always so, and is really a point worth investigating.
While I agree with you wrt. your assertion, I suspect that you have not
identified the correct, at least not the chief reason that your statement is
true.

I don't have the Encyclopaedia answer for this at my fingertips, so indulge
me while I go through the philosophical analysis steps: What would make your
statement true? In what cases does App. include Ex., and in what cases, if
any, does it not? Are there cases in which Ex. includes App.?
First result: I suspect that there are several issues involved in your
statements, so that it is not entirely clear (to me) exactly (not kidding
now) what your statements cover.
Next step: listing possible areas where the terms could be employed.

Caution: I'll have to try to reach for exact, which is going to make this a
bit longer than if I had stuck with your fair approximation. So
approximation rules, also in part because exactitude is a resource glutton,
up to and including intellectual bulimia. 20 % effort for the first 80 %
yield, 80 % effort for the last 20 % yield, etc.
I haven't found any counterexamples, but I hope that my analysis will not be
in vain: in the best of cases, it will contribute to making looking for
counterexamples easier.


2) Synonymity
Several polarities come to mind, each with a similar, yet perhaps differing
flavour.
Praxis (case) - theory (rule)
Special - general (Details vs. The Big Picture)

First of all, I'd like to comment on the "Manhattan Project" example.
To me there is a difference between areas where exact knowledge is possible.
Like that old Sophist of yonderyear previously discussed, in some cases
there is no thing to be known; and then the chain stops. To call
probabilistic knowledge an example of approximate would in my view be wrong,
because the opposition A/E is only meaningful when E is possible, and A is
attained by stopping before the run of the intelligence mill is complete.
The polarity pair involved here is of another type than the others, and the
polarity itself takes place at a different level than the others. The others
pertain incomplete human knowledge, this one pertains "complete" human
knowledge (as exactly as possible calculated probability) of an unknowable
field. I don't have the english vocabulary for this, but it would be
something like "Actual"/"Calculable" - probabilistic/"stochastic"....?


a) Praxis - theory
Philosophers have noted the limitations of laws, rules and guidelines (here:
theory, for short). Science can be defined by it treating of classes of
objects, and classes of objects are only arrived at by more or less
abstracting, leaving out the specifics of a particular case.
So theory will almost never "reach down" to the world of actually occurring
cases, and for the Doer, there will be an intellectual bridge to gap.
This is readily seen in how much better the experienced, but unschooled
artisan (or artist) performs compared to the learned, but inexperienced
scholar; the difference between performers of every kind and critics of
every kind.
In this case, exact knowledge pertains the particular case, while
theoretical knowledge is approximate.
In this case, the Approximate doesn not contain the Exact; unless one is the
original Abstractor, the author of the theory, and the theory is arrived at
by induction from a range of cases; the one would have all the left out
detail at one's command (if one's memory serves).
Digression: Some of the best leaders often combine experience and education,
having, at least to some degree, worked their way up. First example that
springs to mind is Churchill, who, during WWII took an active interest not
only instatesmanship, but also in the minutest details of service and
equipment. Of course his entry level wasn't dishwashing, but he had seen
action as a young man, been a politician and a civil servant (or minister?
or both?), and so was supremely prepared.

b) special - general
Originally derived from Plato's dieresis technique of definition by sorting
under genus and distinguishing inside the genus by properties (properties
not being traits or qualities in general (common to all members of the
genus), but traits or qualities connected with what made an X an X, traits
that were "the property" of only X-es, that made any X a "proper" X ...
"properties"....). This hierarcy may of course have many levels (like
biological taxonomy).
Structurall this resembles the case-rule pyramid, but with all the details
left in.
Exact knowledge here is knowing exactly what genus something belongs to, and
by what criteria, and knowing exactly what subspecies something is
distinguished as, and by what criteria. Approximate knowledge is to know
sort of what a thing is (the original "kind'a" knowledge is, btw, knowing
the genus but not the species), and sort of why it is something of its own
kind.
The benefit of the technique is that species "inherit" traits from the
genus. Exact knowledge results in a list of traits. Inexact knowledge makes
for a shorter, and perhaps erroneous, list.
Then, of course, there is an intimate connection between the highest level
of investigation .... (there must be a better word for this? I mean levels
like "cosmological/galaxies and stars", "solar system", "global/geology,
meteorology etc.", "macro.-objects", "cells", "molecules", "atoms",
"particles", "quarks", "strings" ... "orders of magnification" ....) ...,
the "big picture", and long term strategy. The Big Picture involves minor
details as almost invisible building blocksl, while the elements in the
picture can be fairly large conglomerates of objects (like a galaxy ...).
There is a wonderful saying by Goethe, I can't really remember it, but it
goes something like "If you misbutton the first button, the coat will never
fit". Good strategy is always the result of correct interpretation of the
big picture; that is the first button, the first railway switch, determinng
the limits of the area of possible outcomes, which continues to shrink with
evey new choice (which is an elimination of possibilities), until, perhaps,
one can do something that "opens a new avenue" .... ah, language ...
So, the guy at the top does not need detailed knowledge, and that is why the
guy at the top in e.g. the army is a "general", not a "special".
"On War" by Clausewitz is big on this. Very good book, btw, and it is
online, AFAIK, at 6udotclausewitz.com.
Post by Acme Diagnostics
Then I thought about the trajectory of a space-ship launch. But
as accurate as that needs to be, isn't it still only
approximate, i.e. to so many decimal places? Is the person who
actually runs the program, the final decision-maker, involved in
that?
Back to epistemology, here you have the pragmatic criterion: if it works, it
is true.

I guess "it works" would be a sort of "window" (section of a continuum),
where the decision maker only needs to be informed by the specialist whether
or not something is inside or outside the window, or when it will be inside
or outside of the window. Other specialists will investigate, and answer,
whether it ever can be inside or outside the window, and how it can be made
to be or be prevented from being inside or outside the window.

"It works", in this case, is the definition of "exact", your
"(mathematically) exact" is beyond the scope of pragmatics, and rests in a
small sphere of ideality hovering just above the head of Plato.

In the pragmatic sense, a trajectory that works is exact and not
approximate, since "approximate is already built into pragmatics. Should go
well with your argument, btw, being perfectly suitable for pragmatists and
other doers.
Post by Acme Diagnostics
Ok, I've got one: The ship captain says "full steam ahead."
That sounds like an exact instruction the must be followed
exactly. And the captain is surely running things and making
the decisions. But still, doesn't the captain know that .xxx% of
the time there will be a miscommunication or inability to
follow the order and must allow for that Approximately?
Eeeeeeh ..... Imperfection. Plato again.
The Captain must allow for margins of error, and should institute double
checks during communication and during execution (correction of deviation).
To be specific, aboard ships (and e.g. between planes and flight control
towers) that is already instituted in the readback (?) of orders, where the
receiver repeats any order or information for the sender to collate what is
received with what is intended to send.
Also, ships and air traffic continuously, or at least periodically, check
the actual position and trajectory against the position and trajectory
plotted from the information last sent or order last given, to ensure proper
execution by monitoring for deviation.
I'd say this is a procedural, not an epistemological question.

My take is that you are on to something akin to the OODA loop of Boyd.
Information is a cost, both in acquisition and processing. Hence, he who can
reduce the spending of resources, primarily wrt. time, while yet maintaining
a precision level that is within the "works" window, will be superior to any
competitor, be it enemy or rival.
And if the unit one is responsible for leading, for performing OODA loops
for, is a complex unit, a ship, a business, an army, a country, the
OODAlooper needs a broad, general knowledge of its internal working, and of
the wnvironment and the competition.
And perhaps the level of generality simplifies the picture enough to
increase the cycle time of the the OODA.
But the details of the same are not necessarily included in this general
knowledge.

Well, enough rambling. HTH

"Pardon me for writing such a long letter, but I had so little time."
(Pascal).
T
Tron
2005-02-21 21:37:59 UTC
Permalink
"Tron" <***@frisurf.no> skrev i melding news:dusSd.9697

....
Post by Tron
First of all, I'd like to comment on the "Manhattan Project" example.
To me there is a difference between areas where exact knowledge is possible.
... and areas where it is not.

T
Acme Diagnostics
2005-02-22 20:52:05 UTC
Permalink
Post by Acme Diagnostics
Post by Acme Diagnostics
Post by Tron
Post by Acme Diagnostics
There are two types of thinking: Approximate and Exact.
Approximate always trumps Exact.
One reason is that Approximate includes Exact, and by
how much, approximately.
That's why people who can think in Approximate run
things and make the decisions.
Well, that's not quite accurate, but it holds in a general sense.
Agree.
I'm really sorry; I was trying to make a joke.
Appreciated! Sorry to be so slow. Hadn't had my coffee.
Post by Acme Diagnostics
0) Style and consistency
It would be more fitting if your thesis weren't an absolute ("
....always..."), but obeyed itself to state: "Approximate beats Exact in,
roughly, ninetynine out of hundred cases, give or take twenty" (the reverse
of "67,3 % of all statistics are bogus".)
Right. "Always" is a long, long time. 'Scuse my deviation from
the mean. <g>
Post by Acme Diagnostics
1) Crux
"One reason is that Approximate includes Exact, and by
how much, approximately."
If there is any mistake here, it is in this analysis. As another poster
pointed out, the difference between exact and approximate can be seen as a
difference in degree, hence it is not a difference in kind.
I now see that that is the obvious inference so the word is
misleading. But suppose, on reading your joke, I inferred: "70%
probability he's joking, 25% he's being literal, 5% he means
something I can't figure out. Those would be differences in kind,
and I was intending to include that in "Approximate," but now
realize I didn't say what I intended (Well I was trying to be
cute.)

You expose a big distinction then. I mostly meant a difference in
kind. When I meant a difference in degree, I meant a fuzzy
difference, not an exactly measurable one. Perhaps always an
element of judgment.
Post by Acme Diagnostics
In any case, it is a very important part of the analysis, because "included
in" implies that such knowledge may be called forth at times of need by
further investigation of available approximate knowledge.
I should have left off "and by how much." That was definitely
misleading. I didn't mean to be that specific. I was including
that, but also a range with definite end-points but no
probability distribution for Exact (or a "flat" one). Possibly
with the certain knowledge that Exact would be impossible to ever
ascertain. Most vague.
Post by Acme Diagnostics
it would indeed confer great benefits. But that, I think you will find, is
not always so, and is really a point worth investigating.
While I agree with you wrt. your assertion, I suspect that you have not
identified the correct, at least not the chief reason that your statement is
true.
Agreed. And if I expand out what I really meant by "Approximate"
to the long list of real-world reasoning issues, it would have to
include a result that was completely unanticipated. Which leads
into...
Post by Acme Diagnostics
I don't have the Encyclopaedia answer for this at my fingertips, so indulge
me while I go through the philosophical analysis steps: What would make your
statement true? In what cases does App. include Ex., and in what cases, if
any, does it not? Are there cases in which Ex. includes App.?
First result: I suspect that there are several issues involved in your
statements, so that it is not entirely clear (to me) exactly (not kidding
now) what your statements cover.
Yes, way too many to be covered by "Approximate." I can't think
of a way to explain without listing 40 terms, each of those
drilling down into huge dogmas. Let's try "objective methodology"
for Exact, and "best possible judgment" for Approximate. But I
don't intend the usual interpretation of "judgment." Rather a
developed skill heirarchy for best judgment. Not that this would
be formally taught, but just due to random experience it would be
layered more or less properly, and include more-or- less
Approximate thinking, and the end result might be a fundamental
difference in reasoning. Not better or worse, but more or less
suited to whatever goals. And Approximate would be better for
final decision-making wrt running things.

Then there is my brain programming dogma saying that we
are all very highly programmable. Even reading one book
programs you a little without you knowing it. So these skills
are being programmed all the time, and I'm wondering if
an unbalanced exposure to either Exact or Approximate
thinking helps or hinders related skills later on, and if we
could notice this in extreme cases. I think to noticing this. One
thing I do know - nobody seeks a logician to be CEO of their
company. But a logician might be the most *important* person for
profit, say in a computer design.

It gets infinitely complex. There are cases where I don't care
if I'm wrong - I will proceed no matter what. There are cases
where it doesn't matter if I'm wrong as long as in N cases I'm
right P percent of the time. But my point here isn't these cases
- I think the important point is the complexity, that thinking in
"Approximate" often isn't satisfying, often has no answer, can
never be completely "figured out."
Post by Acme Diagnostics
Next step: listing possible areas where the terms could be employed.
Caution: I'll have to try to reach for exact, which is going to make this a
bit longer than if I had stuck with your fair approximation. So
approximation rules, also in part because exactitude is a resource glutton,
up to and including intellectual bulimia. 20 % effort for the first 80 %
yield, 80 % effort for the last 20 % yield, etc.
Ok! Thanks for reminding me to include this important concept.
Sometimes Exact is just not worth the cost. We accept 2%
defective parts because it's cheaper to replace them. (Besides,
there's the marketing dept. <g>)
Post by Acme Diagnostics
I haven't found any counterexamples, but I hope that my analysis will not be
in vain: in the best of cases, it will contribute to making looking for
counterexamples easier.
It seems hard. We can think of trivial or naive examples, like
needing to know the exact count of widgets to ship on this order
or direction of an electron spin. A lot harder to think of
where Exact thinking is best in context of running things, final
decision making.
Post by Acme Diagnostics
2) Synonymity
Several polarities come to mind, each with a similar, yet perhaps differing
flavour. Praxis (case) - theory (rule), Special - general (Details vs. The Big Picture)
First of all, I'd like to comment on the "Manhattan Project" example.
To me there is a difference between areas where exact knowledge is possible.
Like that old Sophist of yonderyear previously discussed,
Gorgias. At least I learned to spell his name.
Post by Acme Diagnostics
in some cases
there is no thing to be known; and then the chain stops. To call
probabilistic knowledge an example of approximate would in my view be wrong,
You know I'm still working on that wording in my article "because
we live in a probabilistic universe." Thanks to you I know it
needs to be changed. Just haven't arrived at satisfactory
wording. Well, I'm working on 100 things at once. I'll get around
to it.
Post by Acme Diagnostics
because the opposition A/E is only meaningful when E is possible, and A is
attained by stopping before the run of the intelligence mill is complete.
The polarity pair involved here is of another type than the others, and the
polarity itself takes place at a different level than the others. The others
pertain incomplete human knowledge, this one pertains "complete" human
knowledge (as exactly as possible calculated probability) of an unknowable
field. I don't have the english vocabulary for this, but it would be
something like "Actual"/"Calculable" - probabilistic/"stochastic"....?
I'll chew on those distinctions on my own time. Maybe it will
help to revise my article.
Post by Acme Diagnostics
a) Praxis - theory
theory, for short). Science can be defined by it treating of classes of
objects, and classes of objects are only arrived at by more or less
abstracting, leaving out the specifics of a particular case.
So theory will almost never "reach down" to the world of actually occurring
cases, and for the Doer, there will be an intellectual bridge to gap.
That may be the main purpose of my original post - that gap. Or
perhaps a decision to develop a skill set to best serve goals on
one side or the other of that gap - at the expense of the other
side.
Post by Acme Diagnostics
This is readily seen in how much better the experienced, but unschooled
artisan (or artist) performs compared to the learned, but inexperienced
scholar; the difference between performers of every kind and critics of
every kind.
Two improved examples. Both obviously true from my POV and
experience. Now for the first time I'm thinking of how these
"scholars" do tend to perform mechanically. Well, art was far
removed from my mind when posting and generalizes my intended
case quite a bit. But I like the correspondence because I could
use that experience to find more examples. Switching to
argumentation, I put the Critic and Doer on equal footing (often
just as qualified in either role).
Post by Acme Diagnostics
In this case, exact knowledge pertains the particular case, while
theoretical knowledge is approximate.
The theory does not exactly match the particular (exact) case,
and is thus approximate.
Post by Acme Diagnostics
In this case, the Approximate doesn not contain the Exact; unless one is the
original Abstractor, the author of the theory, and the theory is arrived at
by induction from a range of cases; the one would have all the left out
detail at one's command (if one's memory serves).
The theory loses much information, i.e. the variation in samples.
Yes, this appears to reverse my case. But the final decision
maker is predicting too - maybe employing the theory and maybe
not, but using exact cases to inductively predict the next case
just bypassing the intermediate theory step, so that seems
to revert back to Approximate.
Post by Acme Diagnostics
Digression: Some of the best leaders often combine experience and education,
having, at least to some degree, worked their way up. First example that
springs to mind is Churchill, who, during WWII took an active interest not
only instatesmanship, but also in the minutest details of service and
equipment. Of course his entry level wasn't dishwashing, but he had seen
action as a young man, been a politician and a civil servant (or minister?
or both?), and so was supremely prepared.
I know someone who has this sort of mind, certainly not myself.
Geopolitics one minute, fix the boot buckle the next, put it all
together into a whole. So fast that one doesn't expend the other.
Perhaps Churchill was like that. I suspect that such capable
thinkers don't rise to that level very often due to political
considerations, unless they are Ghengis Khan. <g>
Post by Acme Diagnostics
b) special - general
Originally derived from Plato's dieresis technique of definition by sorting
under genus and distinguishing inside the genus by properties (properties
not being traits or qualities in general (common to all members of the
genus), but traits or qualities connected with what made an X an X, traits
that were "the property" of only X-es, that made any X a "proper" X ...
"properties"....). This hierarcy may of course have many levels (like
biological taxonomy).
Structurall this resembles the case-rule pyramid, but with all the details
left in.
Much like a skill hierarchy. I forgot that Plato did that. What a
guy.
Post by Acme Diagnostics
Exact knowledge here is knowing exactly what genus something belongs to, and
by what criteria, and knowing exactly what subspecies something is
distinguished as, and by what criteria. Approximate knowledge is to know
sort of what a thing is (the original "kind'a" knowledge is, btw, knowing
the genus but not the species), and sort of why it is something of its own
kind. The benefit of the technique is that species "inherit" traits from the
genus. Exact knowledge results in a list of traits. Inexact knowledge makes
for a shorter, and perhaps erroneous, list.
Now the skill hierarchy analogy is even better. The top level
"inherits traits" from the lower level. If the lower level is
deficient, the higher level is made deficient.

So now I ask, if this is true, then do we not form the bottom
rows of this hierarchy when very young? And wouldn't some be
lucky enough to get them organized just right for either Exact or
Approximate thinking, thus most benefiting from academic exposure
to things like math or decision analysis later on? Could this be
extreme enough in some people to observe?
Post by Acme Diagnostics
Then, of course, there is an intimate connection between the highest level
of investigation .... (there must be a better word for this?
I use "levels of description." Haven't found anything better yet.
Post by Acme Diagnostics
I mean levels
like "cosmological/galaxies and stars", "solar system", "global/geology,
meteorology etc.", "macro.-objects", "cells", "molecules", "atoms",
"particles", "quarks", "strings" ... "orders of magnification" ....) ...,
the "big picture", and long term strategy. The Big Picture involves minor
details as almost invisible building blocksl, while the elements in the
picture can be fairly large conglomerates of objects (like a galaxy ...).
"Field of view." Looking through a telescope at a busy airport,
never seeing a plane. Saturn becoming a pixel in the star field.
Post by Acme Diagnostics
There is a wonderful saying by Goethe, I can't really remember it, but it
goes something like "If you misbutton the first button, the coat will never
fit". Good strategy is always the result of correct interpretation of the
big picture;
This is so important. Relating to a discussion in another thread
about "standing outside the system." That's pretty pure
contextual reasoning, and I'd definitely label that as
Approximate. How can you ever know for sure that you have the
exact right context, except in theory.

Wrt complex decisions, I was taught to define the problem more
than go for a solution. In math, you famously define the problem,
but I think you also get used to going for a solution pretty
quick.
I think real-world decision makers spend a lot more time on
defining problems and a lot less time looking for solutions, or
should. I think defining the problem relates to seeing the big
picture.
Post by Acme Diagnostics
that is the first button, the first railway switch, determinng
the limits of the area of possible outcomes, which continues to shrink with
evey new choice (which is an elimination of possibilities), until, perhaps,
one can do something that "opens a new avenue" .... ah, language ...
You do amazingly well. I cannot detect that American English is
not your first language.
Post by Acme Diagnostics
So, the guy at the top does not need detailed knowledge, and that is why the
guy at the top in e.g. the army is a "general", not a "special".
"On War" by Clausewitz is big on this. Very good book, btw, and it is
online, AFAIK, at 6udotclausewitz.com.
I'd sort of disagree here, unless you just mean "sufficient" and
not "best." I guess I'm in the minority of those
who think a leader should be able to do every job. I've seen such
leaders and they always seem to lead the best.
Post by Acme Diagnostics
Post by Acme Diagnostics
Then I thought about the trajectory of a space-ship launch. But
Back to epistemology, here you have the pragmatic criterion: if it works, it
is true.
Well, it's not a perfect test, but it's the best one I've found.
Better than "empirically confirmed" even. And it includes both
science and logic equally and fully. The only thing it doesn't
include is faith, as far as I know. I think I might hear from
Gorgias on this. <g>
Post by Acme Diagnostics
I guess "it works" would be a sort of "window" (section of a continuum),
where the decision maker only needs to be informed by the specialist whether
or not something is inside or outside the window, or when it will be inside
or outside of the window. Other specialists will investigate, and answer,
whether it ever can be inside or outside the window, and how it can be made
to be or be prevented from being inside or outside the window.
Dubiously returning to my original terminology, I would call the
Exact thinker the "specialist" at most (Approximate thinkers
might need to be specialists too), but the decision maker needs
to be the Approximate thinker. Ok, just whether each element is
in the window. Specialists providing the window data. The
decision-maker able to make all the windows work together in the
highest appropriate level of description ("big picture").
Post by Acme Diagnostics
"It works", in this case, is the definition of "exact", your
"(mathematically) exact" is beyond the scope of pragmatics, and rests in a
small sphere of ideality hovering just above the head of Plato.
Hehe.
Post by Acme Diagnostics
In the pragmatic sense, a trajectory that works is exact and not
approximate, since "approximate is already built into pragmatics. Should go
well with your argument, btw, being perfectly suitable for pragmatists and
other doers.
Well, if someone sinks five ships, it doesn't matter
whether they bombed the deck or torpedoed it. They sunk
exactly five ships and that's all that matters. It "worked" five
times.

<snip my ship captain "full steam ahead" as exact command>

<Skip ATC feedback. As a lousy pilot, got that>
Post by Acme Diagnostics
My take is that you are on to something akin to the OODA loop of Boyd.
Boyd was a top fighter pilot. People seem surprised that he
should come up with OODA. I'm not the least surprised. Those
people develop everything we're talking about to the max. Hard to
find an activity that compares. Again, only talking about the
best.
Post by Acme Diagnostics
Information is a cost, both in acquisition and processing. Hence, he who can
reduce the spending of resources, primarily wrt. time, while yet maintaining
a precision level that is within the "works" window, will be superior to any
competitor, be it enemy or rival.
Basic decision analysis. I think contextual reasoning beats it.
You can't put the best decision-making on paper. But if I had an
army, I'd need a methodology, and Boyd's seems the one to beat.
I'd still promote the commanders who won most of their battles.
Make that "all." Cheney talked to Boyd about Iraq. But that's no
objective test of anything at all.
He needed to talk to MacArthur (post WWII Japan). Well, I haven't
walked in any of their shoes.

First year out of college and in business I walked around with a
slide rule in my pocket. Second year I left the slide rule home
and started making money. I imagine half of readers thinking,
"Slide rule? Is that like an abacus?"
Post by Acme Diagnostics
And if the unit one is responsible for leading, for performing OODA loops
for, is a complex unit, a ship, a business, an army, a country, the
OODAlooper needs a broad, general knowledge of its internal working, and of
the wnvironment and the competition.
And perhaps the level of generality simplifies the picture enough to
increase the cycle time of the the OODA.
But the details of the same are not necessarily included in this general
knowledge.
"Pardon me for writing such a long letter, but I had so little time."
(Pascal).
I thought that was Samuel Johnson (??) Well, probably a thousand
people. Now 1,001.

I am supposedly coauthoring a book containing several
"learn-while-doing" projects designed to promote contextual
reasoning (et al). That's neither here nor there, but I thought I
should mention it lest my motivation here be misinterpeted by
some others, not you. I really have nothing at all against
theorists or "exact" thinkers. OTOH, when you order a painter,
you don't expect a wallpaper hanger to show up.

Larry
Tron
2005-02-24 00:26:33 UTC
Permalink
Hi,

I think I'm beyond my usefulness on this topic. Still, some expiratory
flickers:


Let's try "objective methodology"
Post by Acme Diagnostics
for Exact, and "best possible judgment" for Approximate. But I
don't intend the usual interpretation of "judgment." Rather a
developed skill heirarchy for best judgment.
An educated guess?
As long as it is complete, correct, rapid in adapting to changed
circumstances and quicker than the next guy's, it'll work.
exactness, or science, would then be the effort to establish good patterns
to follow during training, like one traces clumsy letters when learning to
write. With more and more practice of the alphabet, the letters get more
even and graceful. When that is in place, one can start writing texts. Like
a clarinet player exercising by playing scales, which is not yet music.
The japanese have a word for it, called "haragei", literally "the art of the
belly". The curious name is due the fact that they believed the belly to be
the seat of emotions (evolution is catching up to them in the western
world). Diligent application to work and exercise will increase one's
haragei, and those with great haragei are able to instantly make correct
decisions.
Post by Acme Diagnostics
Then there is my brain programming dogma saying that we
are all very highly programmable. Even reading one book
programs you a little without you knowing it. So these skills
are being programmed all the time, and I'm wondering if
an unbalanced exposure to either Exact or Approximate
thinking helps or hinders related skills later on, and if we
could notice this in extreme cases.
I think we see it all the times, and nothing extreme about it.
_Everything_ has to be learned. There are in this or that culture skills
that are taken for granted, part of the basic apparatus of human beings,
which are nevertheless an expression of a high level of acquired skill. One
has to know several, widely differing cultures to see it, but once you are
moving among them, they are obvious.

One
Post by Acme Diagnostics
thing I do know - nobody seeks a logician to be CEO of their
company. But a logician might be the most *important* person for
profit, say in a computer design.
Well, the world might not be logical, and might be too complex for any
operable model.
However, a logician is an expert on talking about logic, not necessarily on
applying logic; you wouldn't want a cattle breeder to prepare your steak (on
the force of him being a cattle breeder; he might be an excellent cook
besides that).
And logic is per definition decoupled from empirical content. Just to excuse
the logicians.

The only example to the contrary is fictional, and that, of course, is Conan
Doyle's professor Moriarty. Probably of limited value .....


.....
Post by Acme Diagnostics
Ok! Thanks for reminding me to include this important concept.
Typing "80 20 rule" into Google give you near 8 million hits. So it is at
least well known.

....>>counterexamples ..
Post by Acme Diagnostics
It seems hard. We can think of trivial or naive examples, like
needing to know the exact count of widgets to ship on this order
or direction of an electron spin. A lot harder to think of
where Exact thinking is best in context of running things, final
decision making.
Perhaps the solution you present is somehow baked into the way you have
posed the problem. People who run things have to perfoem a lot of different
activities requiring
many different kinds of skill. In my (not entirely nonexistent) knowledge of
business, there is very little to be exact about, so trying to be exact is a
major error.

A leader's task is to aim, making the goal or target visible to the
"gunners". The leader's knowledge of "gunnery" needs only to encompass the
results (the window/it works), not the details of any subordinate process;
like in bookkeeping, the major column is a sum of sums, not the cost of
every every stamp and matchstick. Here approximate knowledge is unspecified
knowledge; it will "contain" the exact knowledge like a sum will contain its
addenda, but again the problem with exactitude is one of economy, of
resources, this time in "mental processing costs" for one's own sake, and in
signal economy if this is to be communicated.

As an aside:
In rhetorics, btw, an "approximate" approach to exactitude - the truth - can
be
very effective, provided one is not caught. This example could have been a
case of
true approximation, only the arrow of direction points the other way ...
divergence from exactitude.
Post by Acme Diagnostics
Post by Tron
a) Praxis - theory
......an intellectual bridge to gap.
Post by Acme Diagnostics
That may be the main purpose of my original post - that gap. Or
perhaps a decision to develop a skill set to best serve goals on
one side or the other of that gap - at the expense of the other
side.
Going from rule to case, isn't that casuistry? I know that that is a
derogatory term today, but in its time, it was an honoured method of
inference. I think it is still taught as a method in some law schools;
besides logic (which it resembles), it is almost the only method law can
use.
......using exact cases to inductively predict the next case
Post by Acme Diagnostics
just bypassing the intermediate theory step, so that seems
to revert back to Approximate.
You are very close to "intuiton" here. To avoid newagery, let's define
intuition as chains of reasoning performed so quickly that none of the steps
involved are performed consciously. Metaphorically, thinking without
self-reflection, the mental equivalent to riding a bike, or driving a car. A
lot of women are good at inferring like this (talking about the same things
from 13 to 23 results in an enormously fast cycle, while not applying any
theory undercuts the need for any conscious cognitive processes. The value
of training.).

Have you , btw, considered "heuristics" as a candidate? I was thinking of
fact finding missions in fields where there is no generally recognized
method for arriving at results, no gererally recognized layout of the field,
etc. Very advanced trial-and-error; equations with only unknowns.
...


The benefit of the technique is that species "inherit" traits from the
Post by Acme Diagnostics
Post by Tron
genus. Exact knowledge results in a list of traits. Inexact knowledge makes
for a shorter, and perhaps erroneous, list.
Now the skill hierarchy analogy is even better. The top level
"inherits traits" from the lower level. If the lower level is
deficient, the higher level is made deficient.
A bit confused here. Like in programming, the lower hierarchies (the
species, the offspring) inherit from above (the genus, the parent). Living
things have movement, metabolism, reproduction ... Animals are living things
that have a central nervous system and a wide range of sensory organs ....
man is a living animal with reason ....
I.e.: Man has movement, metabolism, reproduction, a central nervous system
and a wide range of sensory organs, and reason. (A bit ... approximate, but
I hope it is clear.) This is a purely logical thing (Concepts, properties,
sets ....).

A skill hierarchy is something like a process of increasing maturity and
experience?
Real world, biology (learning), history of ...?
Post by Acme Diagnostics
I use "levels of description." Haven't found anything better yet.
Thx. I had it on the tip of my keyboard.

...."standing outside the system." That's pretty pure
Post by Acme Diagnostics
contextual reasoning, and I'd definitely label that as
Approximate. How can you ever know for sure that you have the
exact right context, except in theory.
Here is a key point. What is contextual reasoning, anyway?
Post by Acme Diagnostics
Wrt complex decisions, I was taught to define the problem more
than go for a solution.
In math, you famously define the problem,
Post by Acme Diagnostics
but I think you also get used to going for a solution pretty
quick.
I think real-world decision makers spend a lot more time on
defining problems and a lot less time looking for solutions, or
should.
Nah ... to me these two functions surprisingly often coincide in the same
activity.
It is true, as you say, that ...." defining the problem relates to seeing
the big picture." But this is a different picture. It is also the big
picture, but
Post by Acme Diagnostics
it includes the "present position - goal definition - route mapping"
detail, where "defining the problem" is to sketch the start, the end and
the distance between, while "finding the solution" is to trace the steps
from start to goal. Obviously, all this happens in the same territory.
Possible example: people who come for advice, but solve the problem during -
and by - telling what is the matter.

....
Post by Acme Diagnostics
You do amazingly well. I cannot detect that American English is
not your first language.
Thank you, but you don't know my typing speed ....
...>>"On War" by Clausewitz is big on this. Very good book, btw, and it is
Post by Acme Diagnostics
Post by Tron
online, AFAIK, at 6udotclausewitz.com.
Nice quote by him, btw: "Strategy is not complex; but in war, that which is
simple isn't necessarily easy."
Post by Acme Diagnostics
I'd sort of disagree here, unless you just mean "sufficient" and
not "best." I guess I'm in the minority of those
who think a leader should be able to do every job. I've seen such
leaders and they always seem to lead the best.
There's always Thomas .. George? ... something ... Kennan? Kenney?
Something.
Management guru who held that managers didn't need knowledge of the business
they were running, only (preferably Thomas-George Kennan's own) management
techniques.
(He's so out of date he's not even googlable...)
Post by Acme Diagnostics
Dubiously returning to my original terminology, I would call the
Exact thinker the "specialist"
Yes.
Post by Acme Diagnostics
Post by Tron
My take is that you are on to something akin to the OODA loop of Boyd.
Boyd was a top fighter pilot. People seem surprised that he
should come up with OODA. I'm not the least surprised. Those
people develop everything we're talking about to the max. Hard to
find an activity that compares. Again, only talking about the
best.
Well, it is basically "anamnesis - diagnosis - therapy" (reading symptoms -
classifying disorder - applying remedy), which was formulated by
Hippocrates. Seems no one can beat the ancient Greeks. Aristotle's
"position - goal - route" formula for acting isn't far removed either. (And
people say philosophy needs to catch up to science, when science is still
struggling 2500 years behind ...).
...
Post by Acme Diagnostics
I thought that was Samuel Johnson (??) Well, probably a thousand
people. Now 1,001.
"I have made this //letter// longer, because I did not have the time to make
it shorter."
Pascal, Lettres provinciales, letter 16, 157.
Post by Acme Diagnostics
I am supposedly coauthoring a book containing several
"learn-while-doing" projects designed to promote contextual
reasoning (et al).
Interesting. I did my master's thesis in philosophy on Frege, the originator
of the context principle. But I guess you do not mean context in a
philosophy (of language), logic or programming setting, where it has rather
narrow (albeit slightly differing) definitions. The programmer's definition
is, AFAIK, a subset of all that one knows as prerequisite for reasoning,
instead of employing every fact that one knows; or, IOW, localization. It
also apporaches something akin to approximation, but only if one assumes
that all information is mutually connected, and that limits to investigation
are artificial. If there is truly independent and irrelevant
information,then a specific (and extremely well chosen) subset might contain
the exact amount of knowledge necessary.
If I can supply you with any background material from the Frege bench, even
if only to eliminate blind alleys, let me know.

T
Acme Diagnostics
2005-02-24 21:07:09 UTC
Permalink
Post by Tron
Hi,
I think I'm beyond my usefulness on this topic. Still, some expiratory
Just consider my posts templates for when you are bored and can't
find anything better to do! I always learn important things from
talking to you, so your replies are quite useful. I think you're
an excellent reasoner, so how you got there is interesting. I
lurk many of your posts, btw, as you are highlighted in my
unthreaded reader. Again, it takes me a long time to think some
things through, and that's the only reason for not being very
responsive sometimes. One reason I work on 100 things at a time
(working outlines, etc.).
Post by Tron
The japanese have a word for it, called "haragei", literally "the art of the
belly". The curious name is due the fact that they believed the belly to be
the seat of emotions (evolution is catching up to them in the western
world). Diligent application to work and exercise will increase one's
haragei, and those with great haragei are able to instantly make correct
decisions.
Great. I love that kind of thing, especially Japanese.

<excuse snippage. Interesting to read, agree, etc.>
Post by Tron
A leader's task is to aim, making the goal or target visible to the
"gunners". The leader's knowledge of "gunnery" needs only to encompass the
results (the window/it works), not the details of any subordinate process;
like in bookkeeping, the major column is a sum of sums, not the cost of
every every stamp and matchstick.
You arrived at the source of my starting point: leadership. But I
avoided the word because I have a definition too lengthy to type
here and which seems ever more contrary to the common definitions
which (academically) these days seem to have more to do with
psychology and psychology thinly veiled as organization.
- I don't mean that at all. One big difference from your
definition is the "works window." I do understand that and the
economy, studied that in college and high school as far back as
linear programming (geometry) in about the 9th grade (snip
dogma about remembering "useful" info, and I was already
well-practiced at that due to some prior learn-while-doing), but
I include in leadership the drill down into all detail made
possible by the skill heirarchy and other requirements of
leadership like intelligence.

Btw, I am the world's worst leader. Horrible. Lack the
intelligence, lack the discipline, lack much else. But I love to
study reasoning and "leadership" is where I find the best
examples, assuming my definition of drill-down ability. But
my examples come almost exclusively from business and military
which is the mother lode but still somewhat limiting. Well, I'm
embarking on Clinton's book to be read most analytically (i.e.
slowly) as I consider him a phenomenal reasoner.
Post by Tron
Here approximate knowledge is unspecified
knowledge; it will "contain" the exact knowledge like a sum will contain its
addenda, but again the problem with exactitude is one of economy, of
resources, this time in "mental processing costs" for one's own sake, and in
signal economy if this is to be communicated.
That's where the skill heirarchy comes in. If the CEO has not
only studied but *been* a bookkeeper (even for only a month
assuming he has a CEO learning curve) he has the deep
understanding to make decisions just on the balance sheet and a
crafty question or three. But if he hasn't, the subordinate will
outfox him/her. Better example would be bartending. I ran
a club one time and, unless you've been a bartender yourself
(for the month or whatever), you'll never figure out how they
are stuffing $60 a night, but they often do. My apologies to
honest bartenders everywhere. Here I think to see
a flaw in the "works window" philosophy including in the
military. Subordinates have their own agendas. Remember
Captain Binghamton? (McHale's Navy). One seasoned
officer told me, "Don't laugh - that's how it usually is!" In my
experience in business, subordinates are usually more clever than
they appear, perhaps purposely sometimes. But since the military
"works" I wonder how that problem is solved within the "works
window" methodology. I'm going to attempt to find something
out about this on my own too.
Post by Tron
Post by Acme Diagnostics
Post by Tron
a) Praxis - theory
.......an intellectual bridge to gap.
Post by Acme Diagnostics
That may be the main purpose of my original post - that gap. Or
Going from rule to case, isn't that casuistry?
Yes, thanks for that. And the legal analogy is right on target
for the distinction between "objective methodology" and "best
judgment" with a skill set.
Post by Tron
I know that that is a
derogatory term today, but in its time, it was an honoured method of
inference. I think it is still taught as a method in some law schools;
besides logic (which it resembles), it is almost the only method law can
use.
Wrt litigation here in the U.S. a flip of the coin would be more
fair than what we have now due to complexity. I really think the
code needs to be entirely re-written to be more general and more
reliance placed on judgment (casuistry), but accompanied by much
better selection and review of judges and proceedings (TV maybe).
</end rant>
Post by Tron
.......using exact cases to inductively predict the next case
Post by Acme Diagnostics
just bypassing the intermediate theory step, so that seems
to revert back to Approximate.
You are very close to "intuiton" here. To avoid newagery, let's define
intuition as chains of reasoning performed so quickly that none of the steps
involved are performed consciously.
That's my definition of subconscious reasoning! I add that,
with a properly constructed skill heirarchy (some military
training for example), it becomes a "difference in kind" from
intuition. But perhaps that's what you mean by intuition anyway.
I think of intuition as emotion, feelings, as in the misnomer (?)
"women's intuition" and it looks like you sort of agree...
Post by Tron
Metaphorically, thinking without
self-reflection, the mental equivalent to riding a bike, or driving a car. A
lot of women are good at inferring like this (talking about the same things
from 13 to 23 results in an enormously fast cycle, while not applying any
theory undercuts the need for any conscious cognitive processes. The value
of training.).
I highly respect that 13-23 intuition as you call it. I've seen
it to be superior in lots of cases.

I think there is a "change in kind" when things get very complex
so that the bike riding and driving examples no longer apply
(incl. metaphorically). I think those examples are plainly like
"intuition." But they don't require any skill hierarchy. More
like a few minutes or hours to get the knack. I would say that
racing in the Daytona 500 or flying a small plane on instruments
in bad weather a difference in kind. I hear the instructor
saying, "Forget what you think you ever learned about driving."
Post by Tron
Have you , btw, considered "heuristics" as a candidate? I was thinking of
fact finding missions in fields where there is no generally recognized
method for arriving at results, no gererally recognized layout of the field,
etc. Very advanced trial-and-error; equations with only unknowns.
Yes. It's big in the area of AI in which I am involved. Also one
of the cases of evidence I have for subconsious reasoning being
orders of magnitudes faster than conscious reasoning, e.g. the
"combinatorial explosion" that humans somehow miraculously
overcome. And how do I know it's reasoning? Well, I have direct
experience for it in playing organ music. But more generally,
because when there are errors they are always self-serving. If
the errors were random, they would be more or less divided
evenly between self-serving and not self-serving. This seems most
evident when reading political posts. I also have a list (link)
of "ways to promote objectivity" to overcome that a little
sometimes (two other posters contributing).
Post by Tron
Post by Acme Diagnostics
Post by Tron
The benefit of the technique is that species "inherit" traits from the
genus. Exact knowledge results in a list of traits. Inexact knowledge
makes for a shorter, and perhaps erroneous, list.
Now the skill hierarchy analogy is even better. The top level
"inherits traits" from the lower level. If the lower level is
deficient, the higher level is made deficient.
A bit confused here. Like in programming, the lower hierarchies (the
species, the offspring) inherit from above (the genus, the parent). Living
things have movement, metabolism, reproduction ... Animals are living things
that have a central nervous system and a wide range of sensory organs ....
man is a living animal with reason ....
Ok, now with the programming example I understand. I had the
direction reversed. Need more haragei. <bows respectfully>
Post by Tron
A skill hierarchy is something like a process of increasing maturity and
experience?
No. More methodological than that.

Like how on the piano you first learn scales, then build on those
to play chords, then build on those to add bass and melodies.
Then when the entire skill set is complete and you are playing a
song, you can just "think" an emotion, and that is converted into
an instruction in the top heirarchy and it filters down
throughout the hierarchy to the lowest levels and every element
is changed ("salted" as the OP wonderfully called it) to affect
the emotion you intend. Another player can distinguish the
"message" easily.

I could go into more detail about the previously mentioned
learn-while-doing programs to lay down a skill heirarchy. Think
of learning chess in a unique way to lay down about a dozen core
reasoning skills - not what most learn from playing chess. Chess
has overwhelming advantages I won't go into, but one is that it
is a good starter program for kids, no real prerequisites. Also
good for kids, independently add written humor for obvious
context switching.

Then think of those two forming a basis for laying down debating
skills which is quite similar to chess in a number of respects.
Next, researched analytical book. Inductive reasoning
(statistics). All expertly designed, implemented, and supervised.
Like with military training. And so forth. Not nearly that
simple, complete, or limited, but in a nutshell.
Post by Tron
Here is a key point. What is contextual reasoning, anyway?
Best examples are in humor and, within that, puns. I often read
alt.humor.puns (and rec.humor) for that very reason (and to laugh
sometimes). It's fascinating. They are continually taking both
words and the underlying concepts and, half-way through the
process you are purposely led to *expect*, they abruptly switch
to a new context but in which those words or concepts also fit.
Their posts are usually cascades and you can see context switches
with almost every line. Exactly the oppposite of here where we
try our best to remain in or at least find the same context,
often failing. This is such an obvious distinction.

Completely digressing, but so important: They understand each
other better than we do here! The reason: cooperation (like
we are almost uniquely doing now), which tells me volumes
about Usenet v. real-life. Some have suggested that it also had
to do with complexity and subject matter, but that really doesn't
hold up if you look at some examples. Maybe somewhat generally.

Then there are the more-or-less "real-world logic" distinctions
as per my article that you criticized: probability,
point-of-view, levels of description, language "reliability,"
inferencing language into logical propositions.

Then there's the detective or investigative reporter which we've
talked about before. Finding the context where "the story adds
up" (remove contradictions). Jack Anderson's "Peace, War,
and Politics" is interesting in this respect. Leaders,
i.e. administrators, are doing that a lot of the time, if not
most of the time. I do it constantly. I was heavily trained to do
it, more in line with cynical thinking as appropriate in business
or watching news/political broadcasts. Here, trying to figure out
kook.troll motivators, etc. One needs to stay in practice. <g>
Post by Tron
Post by Acme Diagnostics
I think real-world decision makers spend a lot more time on
defining problems and a lot less time looking for solutions, or
should.
Nah ... to me these two functions surprisingly often coincide in the same
activity.
It is true, as you say, that ...." defining the problem relates to seeing
the big picture." But this is a different picture. It is also the big
picture, but
Post by Acme Diagnostics
it includes the "present position - goal definition - route mapping"
detail, where "defining the problem" is to sketch the start, the end and
the distance between, while "finding the solution" is to trace the steps
from start to goal. Obviously, all this happens in the same territory.
Ok. I obviously need to give this some more thought. Thanks.
Post by Tron
Possible example: people who come for advice, but solve the problem during -
and by - telling what is the matter.
I describe it as "Defining the problem until the solution jumps
out and bites you on the leg." My favorite example here is the
Arab-Isreali conflict. I hear nothing but solutions. I have
never seen a popular documentary such that most voters would see
that really defines that problem, starting at the beginning of
history (I guess) and including such things as the Holocaust and
bad things Christians did to Arabs. (which just about exhausts
what I know about it <g>). Maybe too politically incorrect to do.
Post by Tron
....
Post by Acme Diagnostics
Boyd was a top fighter pilot. People seem surprised that he
should come up with OODA. I'm not the least surprised. Those
I shouldn't have implied that I know OODA, just know about it.
Readers Digest, probably. <g> It came after my time academically.

But I googled this:

The key is to obscure your intentions and make them
unpredictable to your opponent while you simultaneously clarify
his intentions. That is, operate at a faster tempo to generate
rapidly changing conditions that inhibit your opponent from
adapting or reacting to those changes and that suppress or
destroy his awareness. Thus, a "hodge-podge" of confusion and
disorder occur to cause him to over- or under-react to
conditions or activities that appear to be uncertain,
ambiguous, or incomprehensible. Put more succinctly, deny your
opponent the use of his maneuvering advantages against you
while you convert your strengths into an advantage over him and
cause him to make a wrong move, one that can be easily defeated.

That sounds a lot like some chess strategy. Even more, nasty
debating.
Post by Tron
Well, it is basically "anamnesis - diagnosis - therapy" (reading symptoms -
classifying disorder - applying remedy), which was formulated by
Hippocrates. Seems no one can beat the ancient Greeks. Aristotle's
"position - goal - route" formula for acting isn't far removed either. (And
people say philosophy needs to catch up to science, when science is still
struggling 2500 years behind ...).
They amaze the hell out of me, not that I know (or remember) much.
Aristotlian logic was the starting point for my debating program.
Had some philosophy in college. Read I.F. Stone's book on
Socrates as research for an AI book chapter on how news product
labels will solve his paradox of democracy. <g>
Post by Tron
....
Post by Acme Diagnostics
I thought that was Samuel Johnson (??) Well, probably a thousand
people. Now 1,001.
"I have made this //letter// longer, because I did not have the time to make
it shorter."
Pascal, Lettres provinciales, letter 16, 157.
Looks like Sam stole it. I shall revise my cite forthwith!
Post by Tron
Post by Acme Diagnostics
I am supposedly coauthoring a book containing several
"learn-while-doing" projects designed to promote contextual
reasoning (et al).
Interesting. I did my master's thesis in philosophy on Frege,
I sat out the Vietnam war in college and I distinctly remember
them mailing a degree now and then but it was so long ago I can't
remember such specific detail. But I do remember some
pretty damn flowery fonts for those days. <g>
Post by Tron
the originator
of the context principle. But I guess you do not mean context in a
philosophy (of language), logic or programming setting, where it has rather
narrow (albeit slightly differing) definitions.
Thanks for mentioning that! If this is a fair ridiculously short
approximation:

"Never to ask for the meaning of a word in isolation, but only
in the context of a proposition"

then I should bring this to the attention of a few critics who
now seem to have been arguing in bad faith. I had forgotten that
after googling it in writing the article you so worthily
criticized (I'm pretty sure it was listed as a google). After
all, given the givens about the approximation, it does seem
rather obvious to any 12-year-old class cutup like myself.

Anyway, it is extremely interesting and fundamental. Something
different from what I call "inferencing language into
propositions," based on a course that was pretty mechanical as in
rules for proposition re-arranging. As above described, the
Context Principle was most proximate in my thinking when starting
this thread, though the book I'm supposedly coauthoring is the
more general context with some work preceding said article.

But "contextual reasoning" is much more general than that.
It's most of what you'd need to do to win a moderated, timed,
judged debate. But not like a college or presidential debate
(i.e. speeches), but back-and-forth no-holes-barred sometimes
combative debate (designed to teach winning and decision-making
too). For example, there and quite noticeable here in the more
combative debates, you are trying to stay one loop outside of
your opponent, but not two.

Most generally, I mean what a leader needs to do ("people who run
things, final decision-maker"), but qualified in all jobs led,
and capable across the theory -> engineering -> building (or mfg
or operating) -> maintaining chain (at least).
Post by Tron
The programmer's definition
is, AFAIK, a subset of all that one knows as prerequisite for reasoning,
instead of employing every fact that one knows; or, IOW, localization.
Well I don't think that could be known before the fact, so we
would be employing every fact known and every inference that
could be made, and, in the case of the most capable, an artistic
talent component of creative reasoning (i.e. can't be taught),
which I describe as popping out of nowhere. Like intuition but
not to make the decision; instead to supply context for a
conscious, reasoned decision where those associations must
then pass a logic, science or "things that work" test. All
bearing upon a specific goal or decision. Not far from insanity,
I guess. But as Red Skelton said, as long as you're making money
they won't lock you up.
Post by Tron
It
also apporaches something akin to approximation, but only if one assumes
that all information is mutually connected, and that limits to investigation
are artificial. If there is truly independent and irrelevant
information,then a specific (and extremely well chosen) subset might contain
the exact amount of knowledge necessary.
What objective methodology could possibly decide that subset
before the fact? In some AI models there are naive relationships
among propositions and words to supply some context, but
hopelessly limited compared to human reasoning. I have my own
scheme, which doesn't differentiate me at all from any other AI
kook.
Post by Tron
If I can supply you with any background material from the Frege bench, even
if only to eliminate blind alleys, let me know.
I would love a paragraph or ten at your leisure, maybe including
a more mature understanding of the Context Principle than my
naive googling, or whatever you guess is pertinent in context of
my general meandering.

Use large garden snippers. No offense ever for non-response.
Did I ever tell you about the chess game with my uncle with
a 10-year delay in one of the moves? (But what a move it was!)

Thanks much,

Larry

Immortalist
2005-02-21 18:20:57 UTC
Permalink
Post by Acme Diagnostics
There are two types of thinking: Approximate and Exact.
Exact is just a "degree of truth" equal to 100%. Approximate is a degree of
truth between 1% and 99%. Therefore there is only one type of thinking:
Fuzzy Think!
Post by Acme Diagnostics
Approximate always trumps Exact.
Unlike traditional or classical logic, which attempts to categorize
information into binary patterns such as black / white, true / false, yes /
no, or all / nothing, Fuzzy Logic pays attention to the "excluded middle"
and tries to account for the "grays", the partially true and partially false
situations which make up 99.9% of human reasoning in everyday life. It
builds upon the assumption that everything consists of degrees on a sliding
scale-whether it be truth, age, beauty, wealth, color, race, or anything
else that is effected by the dynamic nature of human behavior and
perception. The question Zadeh always insists upon asking is, "To what
degree is something true or false?"

Zadeh looks around him in the real world which he finds pervaded by concepts
which do not have sharply defined boundaries, where information is often
incomplete or sometimes unreliable. In fact, he would classify most words as
having fuzzy meanings-virtually every adjective or adverb in ordinary
speech. These concepts become clear if seen in transition from membership to
non-membership in gradual, rather than abrupt, increments.

In quest for precision, scientists have generally attempted to manipulate
the real world into artificial mathematical models that make no provision
for gradation. They have tried to describe the laws governing the incredibly
complex behavior of humans, both singly and in groups, in mathematical terms
similar to those employed in the analysis of inanimate systems, which, in
Zadeh's view, has been, and will continue to be, a misdirected effort.

Because the human mind can't handle so many isolated separate ideas at one
time, it tends to bundle similarly-related objects into categories in such a
way as to reduce the complexity of the information processing task. It is
this incredible capacity of the human mind to manipulate these fuzzy or
unsharp categories that distinguishes human intelligence from the machine
intelligence of current generation computers.

Because Fuzzy Logic provides the tools to classify information into broad,
coarse categorizations or groupings, it has infinite possibilities for
application which have proven to be much cheaper, simpler and more effective
than other systems in handling complex information. Fuzzy Logic has
extremely broad implications for many fields not just electrical engineering
and computer technology which have been fairly quick to incorporate its
theoretical principles. Numerous consumer goods especially household
products and electronic equipment-microwaves, cameras, and camcorders
already incorporate Fuzzy Logic into their design. So have computer control
systems such as the famous subway of Sendai, Japan, or numerous complex
diagnostic and monitoring biomedical systems which are starting to be used
in hospitals.

But other fields such as the social sciences-economy, finance, psychology,
sociology, politics, religion, ethics, law, medicine, geography, folklore,
anthropology that deal with the complexity of human behavior-are just
beginning to explore the infinite possibilities of Fuzzy Logic.

Zadeh was not the first to think about "shades of gray". Philosophers such
as Plato indicated that there was a third region (beyond "true" and "false")
where opposites "tumbled about". Hegel, Marx, Engels and Lukasiewicz, among
others, also dealt with middle regions. But it was Zadeh who first developed
the general theory and laid the foundation for what Fuzzy Logic is today.

For a well-researched, very readable, popular description of Lotfi Zadeh and
the development of the field of Fuzzy Logic, refer to Daniel McNeill and
Paul Freiberger's award winning book, Fuzzy Logic: The Revolutionary
Computer Technology that is Changing our World, 1993. For a technical
introduction to the field, see Zadeh's Fuzzy Sets and Applications: Selected
Papers. Edited by Yager, Ovchnikov, Tong, and Nguyen. New York: Wiley, 1987.

http://www.google.com/search?hl=en&q=Zadeh+fuzzy+logic

#####################################
Post by Acme Diagnostics
One reason is that Approximate includes Exact, and by
how much, approximately.
That's why people who can think in Approximate run
things and make the decisions.
Larry
Loading...