f



Is concept creation a consequence of the language being a serial representation of a parallel process, reality?

IMO language is a serial representation of a parallel process, reality
and the creation of abstract concepts is just the absolutely necessary
tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points.
OTOH a concept can become any of these nodes or any combination of
connections between nodes or points, contiguous or not.
IOW in order to describe all the signs from A to Z we need n (26) steps
and operators, but we can replace all these steps with only one
concept, specifically alphabet.
This replacement process or concept creation works as long as the steps
are clearly defined but it becomes blurry if we fail to do it properly.
So is concept creation just the tool needed to overcome the limitations
of the serial representation (2 or 3 dimensions?) of a parallel
process, reality (4 or more dimensions?). 
JP

0
gms2004 (18)
1/4/2006 2:37:47 PM
comp.ai.philosophy 6718 articles. 1 followers. Post Follow

82 Replies
907 Views

Similar Articles

[PageSpeed] 38

Just Playing wrote:
> IMO language is a serial representation of a parallel process, reality
> and the creation of abstract concepts is just the absolutely necessary
> tool to achieve this approximation.
> In a serial representation the points are connected sequentially using
> and/or operators to describe nodes and complex configurations of
> points.
> OTOH a concept can become any of these nodes or any combination of
> connections between nodes or points, contiguous or not.
> IOW in order to describe all the signs from A to Z we need n (26) steps
> and operators, but we can replace all these steps with only one
> concept, specifically alphabet.
> This replacement process or concept creation works as long as the steps
> are clearly defined but it becomes blurry if we fail to do it properly.
> So is concept creation just the tool needed to overcome the limitations
> of the serial representation (2 or 3 dimensions?) of a parallel
> process, reality (4 or more dimensions?).
> JP

Let's try again.
Sound is the only type of information that we can perceive externally
(hear) and also create, control and perceive internally (speak).
OTOH the amount of information that can be processed and transmitted by
sound is very small in comparison with all the information we perceive
and process (see, hear, smell, feel, taste, etc.).
As we need to process and transmit much more information that can be
processed thru sound we create "the verbal language" which is a
form of packing the information thru creation of abstract concepts.
IMO we need to start looking at concepts as methods of increasing the
amount of information to be processed and transmitted.
Maybe a starting point would be the calculation of the differences
between the amount of information that is necessary to process and
transmit in comparison to the capacity of sound to do it in order to
understand the creation of abstract concepts.
It could be just that simple.
JP

0
Just
1/5/2006 3:08:00 PM
Just Playing wrote:
> Just Playing wrote:
> > IMO language is a serial representation of a parallel process, reality
> > and the creation of abstract concepts is just the absolutely necessary
> > tool to achieve this approximation.
> > In a serial representation the points are connected sequentially using
> > and/or operators to describe nodes and complex configurations of
> > points.
> > OTOH a concept can become any of these nodes or any combination of
> > connections between nodes or points, contiguous or not.
> > IOW in order to describe all the signs from A to Z we need n (26) steps
> > and operators, but we can replace all these steps with only one
> > concept, specifically alphabet.
> > This replacement process or concept creation works as long as the steps
> > are clearly defined but it becomes blurry if we fail to do it properly.
> > So is concept creation just the tool needed to overcome the limitations
> > of the serial representation (2 or 3 dimensions?) of a parallel
> > process, reality (4 or more dimensions?).
> > JP
>
> Let's try again.
> Sound is the only type of information that we can perceive externally
> (hear) and also create, control and perceive internally (speak).
> OTOH the amount of information that can be processed and transmitted by
> sound is very small in comparison with all the information we perceive
> and process (see, hear, smell, feel, taste, etc.).
> As we need to process and transmit much more information that can be
> processed thru sound we create "the verbal language" which is a
> form of packing the information thru creation of abstract concepts.
> IMO we need to start looking at concepts as methods of increasing the
> amount of information to be processed and transmitted.
> Maybe a starting point would be the calculation of the differences
> between the amount of information that is necessary to process and
> transmit in comparison to the capacity of sound to do it in order to
> understand the creation of abstract concepts.
> It could be just that simple.
> JP

In terms of subjects and relations, one must first understand the
difference between a related subject and an extrinsic subject.
Subjects that are combined together are of course related.  The subject
given to a relationship is an extrinsic subject.  Also, the subjects in
a relationship are intrinsic to the extrinsic subject.  Intrinsic
subjects can never be related to an extrinsic subject.  The instant
this happens a new extrinsic subject is created.

Now, you say we need to "start looking at concepts as methods of
increasing the amount of information to be processed and transmitted."
 This is what we do already.  Concepts are extrinsic subjects.  They
encapsulate a relationship.  Encapsulating extrinsic subjects creates a
new concept.  This is how language grows.

If you're looking for a formula or something other than this system,
forget it.  This is the system we are in and no shortcuts are possible.

0
jhn_hbr
1/6/2006 6:39:22 AM
jhn_hbr@yahoo.com wrote:
> Just Playing wrote:
> > Just Playing wrote:
> > > IMO language is a serial representation of a parallel process, reality
> > > and the creation of abstract concepts is just the absolutely necessary
> > > tool to achieve this approximation.
> > > In a serial representation the points are connected sequentially using
> > > and/or operators to describe nodes and complex configurations of
> > > points.
> > > OTOH a concept can become any of these nodes or any combination of
> > > connections between nodes or points, contiguous or not.
> > > IOW in order to describe all the signs from A to Z we need n (26) steps
> > > and operators, but we can replace all these steps with only one
> > > concept, specifically alphabet.
> > > This replacement process or concept creation works as long as the steps
> > > are clearly defined but it becomes blurry if we fail to do it properly.
> > > So is concept creation just the tool needed to overcome the limitations
> > > of the serial representation (2 or 3 dimensions?) of a parallel
> > > process, reality (4 or more dimensions?).
> > > JP
> >
> > Let's try again.
> > Sound is the only type of information that we can perceive externally
> > (hear) and also create, control and perceive internally (speak).
> > OTOH the amount of information that can be processed and transmitted by
> > sound is very small in comparison with all the information we perceive
> > and process (see, hear, smell, feel, taste, etc.).
> > As we need to process and transmit much more information that can be
> > processed thru sound we create "the verbal language" which is a
> > form of packing the information thru creation of abstract concepts.
> > IMO we need to start looking at concepts as methods of increasing the
> > amount of information to be processed and transmitted.
> > Maybe a starting point would be the calculation of the differences
> > between the amount of information that is necessary to process and
> > transmit in comparison to the capacity of sound to do it in order to
> > understand the creation of abstract concepts.
> > It could be just that simple.
> > JP
>
> In terms of subjects and relations, one must first understand the
> difference between a related subject and an extrinsic subject.
> Subjects that are combined together are of course related.  The subject
> given to a relationship is an extrinsic subject.  Also, the subjects in
> a relationship are intrinsic to the extrinsic subject.  Intrinsic
> subjects can never be related to an extrinsic subject.  The instant
> this happens a new extrinsic subject is created.
>
> Now, you say we need to "start looking at concepts as methods of
> increasing the amount of information to be processed and transmitted."
>  This is what we do already.  Concepts are extrinsic subjects.  They
> encapsulate a relationship.  Encapsulating extrinsic subjects creates a
> new concept.  This is how language grows.

Actually I was trying to explain it as simple as possible.
We create as many concepts as the physical limitations of the channel
of communications allow us.
We can replace everything with only 1 concept or with the  largest
number possible. We create "everything" and "nothing" and in between a
lot of other concepts.
JP

> If you're looking for a formula or something other than this system,
> forget it.  This is the system we are in and no shortcuts are possible.

0
Just
1/6/2006 3:04:04 PM
This is the 2nd time that I am responding. It seems that there is some
problem with my server.
If I understand you correctly you are talking about a representation,
specifically the concepts encapsulating a relationship.
This is exactly what I am talking about but I am focusing on the way
this representation is done.
I consider that the complexity of what is represented is much higher
than the one of one that is doing the representation.
IOW the verbal language has this limitation, which IMO can be
quantified, and the only way to overcome it is to create "meaning"
and abstract concepts.
The amount of information or complexity that has to be encapsulated in
the concepts can then be calculated.
We can have a concept that encapsulates "everything" or
"nothing" as well as concepts that fill in between these two.
JP

0
Just
1/7/2006 12:07:17 AM
Just Playing wrote:
> This is the 2nd time that I am responding. It seems that there is some
> problem with my server.
> If I understand you correctly you are talking about a representation,
> specifically the concepts encapsulating a relationship.
> This is exactly what I am talking about but I am focusing on the way
> this representation is done.
> I consider that the complexity of what is represented is much higher
> than the one of one that is doing the representation.
> IOW the verbal language has this limitation, which IMO can be
> quantified, and the only way to overcome it is to create "meaning"
> and abstract concepts.
> The amount of information or complexity that has to be encapsulated in
> the concepts can then be calculated.
> We can have a concept that encapsulates "everything" or
> "nothing" as well as concepts that fill in between these two.
> JP

I tend to disagree with your proposition, but then I'm open for trying.
 The reservations that I have are that I believe verbal language is a
higher system than calculation.  IOW the subject system is extrinsic to
the unit system.  Therefore, verbal langage can't be calculated.
Nevertheless, let's give it a go.  Let's use a simple example such as
the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
Sat'  is more complex than 'week'.  Certainly this is so since there
are seven subjects we are using as opposed to the extrinsic subject
'week'.  Maybe what your proposing is that 'week' should be also known
as week(7).  But then what about a work week?  Should we call this
week(5)?  Is this what your getting at?

0
jhn_hbr
1/7/2006 7:04:25 AM
jhn_...@yahoo.com wrote:
> Just Playing wrote:
> > This is the 2nd time that I am responding. It seems that there is some
> > problem with my server.
> > If I understand you correctly you are talking about a representation,
> > specifically the concepts encapsulating a relationship.
> > This is exactly what I am talking about but I am focusing on the way
> > this representation is done.
> > I consider that the complexity of what is represented is much higher
> > than the one of one that is doing the representation.
> > IOW the verbal language has this limitation, which IMO can be
> > quantified, and the only way to overcome it is to create "meaning"
> > and abstract concepts.
> > The amount of information or complexity that has to be encapsulated in
> > the concepts can then be calculated.
> > We can have a concept that encapsulates "everything" or
> > "nothing" as well as concepts that fill in between these two.
> > JP
>
> I tend to disagree with your proposition, but then I'm open for trying.
>  The reservations that I have are that I believe verbal language is a
> higher system than calculation.  IOW the subject system is extrinsic to
> the unit system.  Therefore, verbal langage can't be calculated.
> Nevertheless, let's give it a go.  Let's use a simple example such as
> the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
> Sat'  is more complex than 'week'.  Certainly this is so since there
> are seven subjects we are using as opposed to the extrinsic subject
> 'week'.  Maybe what your proposing is that 'week' should be also known
> as week(7).  But then what about a work week?  Should we call this
> week(5)?  Is this what your getting at?

This is not exactly what I am talking about.
I start with what is to be represented which I will call R.
The second is the representation system, verbal language in this case,
and I will call it V.
The problem is that the amount of information in R is of such a
magnitude that V cannot represent it in real time or in a 1 to 1 ratio.
To overcome this limitation V starts to pack information in concepts
and this is what can be calculated.
If we know how much information is to be transmitted we can transmit in
as many concepts as V allows us to do it, from one to the maximum
number possible.
In your example with days and week you are working inside the V or the
representation system and it seems that there is no limitation between
R and V anymore.
Your example shows that you can replace 7 subjects with one, but as you
correctly point out you lose information and create confusion this way.
You need a limitation in order to replace 7 days with a week in absence
of this limitation there is no reason to do it.
Thank you,
JP

0
Just
1/7/2006 2:40:44 PM
Just Playing wrote:
> jhn_...@yahoo.com wrote:
> 
>>Just Playing wrote:
>>
>>>This is the 2nd time that I am responding. It seems that there is some
>>>problem with my server.
>>>If I understand you correctly you are talking about a representation,
>>>specifically the concepts encapsulating a relationship.
>>>This is exactly what I am talking about but I am focusing on the way
>>>this representation is done.
>>>I consider that the complexity of what is represented is much higher
>>>than the one of one that is doing the representation.
>>>IOW the verbal language has this limitation, which IMO can be
>>>quantified, and the only way to overcome it is to create "meaning"
>>>and abstract concepts.
>>>The amount of information or complexity that has to be encapsulated in
>>>the concepts can then be calculated.
>>>We can have a concept that encapsulates "everything" or
>>>"nothing" as well as concepts that fill in between these two.
>>>JP
>>
>>I tend to disagree with your proposition, but then I'm open for trying.
>> The reservations that I have are that I believe verbal language is a
>>higher system than calculation.  IOW the subject system is extrinsic to
>>the unit system.  Therefore, verbal langage can't be calculated.
>>Nevertheless, let's give it a go.  Let's use a simple example such as
>>the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
>>Sat'  is more complex than 'week'.  Certainly this is so since there
>>are seven subjects we are using as opposed to the extrinsic subject
>>'week'.  Maybe what your proposing is that 'week' should be also known
>>as week(7).  But then what about a work week?  Should we call this
>>week(5)?  Is this what your getting at?
> 
> 
> This is not exactly what I am talking about.
> I start with what is to be represented which I will call R.
> The second is the representation system, verbal language in this case,
> and I will call it V.
> The problem is that the amount of information in R is of such a
> magnitude that V cannot represent it in real time or in a 1 to 1 ratio.
> To overcome this limitation V starts to pack information in concepts
> and this is what can be calculated.
> If we know how much information is to be transmitted we can transmit in
> as many concepts as V allows us to do it, from one to the maximum
> number possible.
> In your example with days and week you are working inside the V or the
> representation system and it seems that there is no limitation between
> R and V anymore.
> Your example shows that you can replace 7 subjects with one, but as you
> correctly point out you lose information and create confusion this way.
> You need a limitation in order to replace 7 days with a week in absence
> of this limitation there is no reason to do it.
> Thank you,
> JP
> 

"A picture is worth a thousand words."

Yup.

Which 1,000 words?

..............

NB that R --> V is different for different people. Hence you cannot 
"calculate" how much information is "packed into" a concept. Whose 
version of the concept? And what algorithm will you use? Consider a 
(relatively) simple example: What does "classical greek sculpture" mean 
to you? Suppose you start talking, explaining the concept as you 
understand it. At what point does one say "That's enough", and start 
"calculating" the information "packed into" the concept?

The example of "7 days is a week, a week is 7 days" is little more than 
synonymy. It doesn't explain much about "packing information into a 
concept." Was that your point?
0
Wolf
1/7/2006 10:20:43 PM
Wolf Kirchmeir wrote:
> Hence you cannot 
> "calculate" how much information is "packed into" a concept.

Information Theory? Kolmogorov Complexity? (And no, people do not 
naturally do anything of the kind.) It is more theoretically useful than 
pratical though.

-- Risujin
0
Risujin
1/8/2006 12:35:21 AM
Risujin wrote:
> Wolf Kirchmeir wrote:
> 
>> Hence you cannot "calculate" how much information is "packed into" a 
>> concept.
> 
> 
> Information Theory? Kolmogorov Complexity? (And no, people do not 
> naturally do anything of the kind.) It is more theoretically useful than 
> pratical though.
> 
> -- Risujin

If you mean Shannon's information theory and its developments, I don't 
think it's relevant to OP's notions, which is about meaning, not 
information in Shannon's limited sense.

Kolgomorov complexity: I've read an account/demonstration of it, but 
don't understand it well enough to comment.

HTH
0
Wolf
1/8/2006 3:01:14 AM
"Just Playing" <gms2004@lycos.com> wrote in message 
news:1136385467.431070.35790@o13g2000cwo.googlegroups.com...
> IMO language is a serial representation of a parallel process, reality
> and the creation of abstract concepts is just the absolutely necessary
> tool to achieve this approximation.
> In a serial representation the points are connected sequentially using
> and/or operators to describe nodes and complex configurations of
> points.

   language is a serial compression of a parallel process.
   Which is why it distorts reality more than it represents it.



> OTOH a concept can become any of these nodes or any combination of
> connections between nodes or points, contiguous or not.
> IOW in order to describe all the signs from A to Z we need n (26) steps
> and operators, but we can replace all these steps with only one
> concept, specifically alphabet.
> This replacement process or concept creation works as long as the steps
> are clearly defined but it becomes blurry if we fail to do it properly.
> So is concept creation just the tool needed to overcome the limitations
> of the serial representation (2 or 3 dimensions?) of a parallel
> process, reality (4 or more dimensions?).
> JP
> 


0
zzbunker
1/8/2006 1:39:54 PM
Wolf Kirchmeir wrote:
> Just Playing wrote:
> > jhn_...@yahoo.com wrote:
> >
> >>Just Playing wrote:
> >>
> >>>This is the 2nd time that I am responding. It seems that there is some
> >>>problem with my server.
> >>>If I understand you correctly you are talking about a representation,
> >>>specifically the concepts encapsulating a relationship.
> >>>This is exactly what I am talking about but I am focusing on the way
> >>>this representation is done.
> >>>I consider that the complexity of what is represented is much higher
> >>>than the one of one that is doing the representation.
> >>>IOW the verbal language has this limitation, which IMO can be
> >>>quantified, and the only way to overcome it is to create "meaning"
> >>>and abstract concepts.
> >>>The amount of information or complexity that has to be encapsulated in
> >>>the concepts can then be calculated.
> >>>We can have a concept that encapsulates "everything" or
> >>>"nothing" as well as concepts that fill in between these two.
> >>>JP
> >>
> >>I tend to disagree with your proposition, but then I'm open for trying.
> >> The reservations that I have are that I believe verbal language is a
> >>higher system than calculation.  IOW the subject system is extrinsic to
> >>the unit system.  Therefore, verbal langage can't be calculated.
> >>Nevertheless, let's give it a go.  Let's use a simple example such as
> >>the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
> >>Sat'  is more complex than 'week'.  Certainly this is so since there
> >>are seven subjects we are using as opposed to the extrinsic subject
> >>'week'.  Maybe what your proposing is that 'week' should be also known
> >>as week(7).  But then what about a work week?  Should we call this
> >>week(5)?  Is this what your getting at?
> >
> >
> > This is not exactly what I am talking about.
> > I start with what is to be represented which I will call R.
> > The second is the representation system, verbal language in this case,
> > and I will call it V.
> > The problem is that the amount of information in R is of such a
> > magnitude that V cannot represent it in real time or in a 1 to 1 ratio.
> > To overcome this limitation V starts to pack information in concepts
> > and this is what can be calculated.
> > If we know how much information is to be transmitted we can transmit in
> > as many concepts as V allows us to do it, from one to the maximum
> > number possible.
> > In your example with days and week you are working inside the V or the
> > representation system and it seems that there is no limitation between
> > R and V anymore.
> > Your example shows that you can replace 7 subjects with one, but as you
> > correctly point out you lose information and create confusion this way.
> > You need a limitation in order to replace 7 days with a week in absence
> > of this limitation there is no reason to do it.
> > Thank you,
> > JP
> >
>
> "A picture is worth a thousand words."
>
> Yup.
>
> Which 1,000 words?
>
> .............
>
> NB that R --> V is different for different people.

This is my second attempt to reply. I got a copy of my first reply in
my email last night but it does not show up here. Wonder what is going
on?

Yes but it there may be enough information in the model to move a
little further than we are doing it now.
JP


> Hence you cannot "calculate" how much information is "packed into" a concept. Whose
> version of the concept? And what algorithm will you use?

That is an interesting point but there are at least 2 ways to overcome
it.
Firstly we do not start from scratch but we learn the concepts in
society and thus we can approximate how much information they contain.
Secondly we can create an experiment where we present the subjects the
same information R and a time limit for the answer or verbal
representation V of R.
While we do not know the amount of information in R we will have
information in the V answers to find what you call an algorithm.
JP

> Consider a (relatively) simple example: What does "classical greek sculpture" mean
> to you? Suppose you start talking, explaining the concept as you
> understand it. At what point does one say "That's enough", and start
> "calculating" the information "packed into" the concept?

IMO your example is much too complex at this stage.
JP

>
> The example of "7 days is a week, a week is 7 days" is little more than
> synonymy. It doesn't explain much about "packing information into a
> concept." Was that your point?

No. And this is not my example.
I cross-posted in alt.philosophy too and some replies got posted only
there.
I will try to copy a part of one of my responses to one of these
replies:

According to CNRS (in France): "Mathematics operates in two
complementary ways. In the *visual* one the meaning of a theorem is
perceived instantly on a geometric figure. The *written* one leans on
language, on algebra; it operates in time."
Have a nice day, Thisbe :)


"Written" language is just the visual form of the "spoken" one
and as you point out it "operates in time" while the "visual"
language operates in space.
While I do not agree that the visual language does not operate in time
the main point is that there is a difference in the number of
dimensions used by the 2 languages.
Let's say that the "visual" one uses N dimensions while the
"spoken" one uses N-1. This makes the visual amount of information
at least twice that much as the one that can be expressed in the same
unit of time verbally.
IMO this is an extremely important limitation and the only way we as a
species have overcome it is by the creation of "meaning" or
abstract concepts.
JP

0
Just
1/8/2006 4:11:37 PM
You are right and I am just wondering if there is a way to corect the
distortions.
JP

0
Just
1/8/2006 4:17:32 PM
Wolf Kirchmeir wrote:
> Risujin wrote:
>> Wolf Kirchmeir wrote:
>>> Hence you cannot "calculate" how much information is "packed into" a 
>>> concept.
>>
>> Information Theory? Kolmogorov Complexity? (And no, people do not 
>> naturally do anything of the kind.) It is more theoretically useful 
>> than pratical though.
>>
>> -- Risujin
> 
> If you mean Shannon's information theory and its developments, I don't 
> think it's relevant to OP's notions, which is about meaning, not 
> information in Shannon's limited sense.

Well if we're talking about meaning then any symbol can stand for any 
amount of associations.

> Kolgomorov complexity: I've read an account/demonstration of it, but 
> don't understand it well enough to comment.

It defines complexity in terms of "the smallest possible program" which 
can describe something. So while it is possible to get an upper bound on 
the *most* information an algorithm could contain there could always be 
a smaller program (unless proven otherwise). Its really only useful 
theoretically AFAIK.

-- Risujin
0
Risujin
1/8/2006 4:43:29 PM
"zzbunker" <jimhunter1@comcast.net> wrote:
> "Just Playing" <gms2004@lycos.com> wrote in message
> news:1136385467.431070.35790@o13g2000cwo.googlegroups.com...
> > IMO language is a serial representation of a parallel process, reality
> > and the creation of abstract concepts is just the absolutely necessary
> > tool to achieve this approximation.
> > In a serial representation the points are connected sequentially using
> > and/or operators to describe nodes and complex configurations of
> > points.
>
>    language is a serial compression of a parallel process.
>    Which is why it distorts reality more than it represents it.

The hidden complexity of langauge is not the simple serial stream of
tokens, but the meaning.  It's not just the parallel nature that causes the
distrotion.  The universe is both parallel and analog.  How do you
represent the complex behavior of even a single water molecule in a glass
of water with a serial stream of tokens taken from a finite set of symbols?
Saying that that there is some distortion here is a gross understatemet.
There's nearly an infinite amount of information left out of the meaning
when you try to use langauge to represent even a trivially small part of
the universe.

And that's what meaning is all about.  It's the mapping from from the
langauge, to some aspect of the universe which is being represented by the
language.

> > OTOH a concept can become any of these nodes or any combination of
> > connections between nodes or points, contiguous or not.

A "concept" is a combination of a language symbol, and it's meaning.

You have attempted to define "meaning" as some finite combination of other
langauge symbols.  That won't do it.  Meaning is more complex than that.
You have to create an assocation between the langauge symbol, and the
aspect of the universe you are trying to describe with the language symbol.

If the universe was full of nothing but langauge symbols, then mapping from
symbol to symbol is all you would need.  But it's not.  It's full of
complex behavior which must be assocated with langauge symbols.  You can't
describe the meaning of natural langauge only using natural language.
Somewhere, the meaning of the words must be tied back to the aspects of the
physical world they are representing.  It's got to all be grounded to
physcial reality.

This is done for us by our sensors.  The meaning of "light" is defined by
the stream of pulses produces by our eyes.  Each pulse, coming from the
eye, represents an aspect of the universe we call light.  Each pulse, is a
symbol, in a langauge, who's meaning is defined by the sensor that "spoke"
the word.

Notice also that this low level language on which we base all our high
level meaning, is temporal in nature.  The meaning isn't timeless like the
words we write down on paper are.  A pulse coming from the eye means
something to the effect of "I see light NOW".

All the meaning, of all the words, and all concepts, used by humans, are
defined in terms of these temporal pulses, which all have meaning, defined
by the hardware which "spoke" the symbol.

> > IOW in order to describe all the signs from A to Z we need n (26) steps
> > and operators, but we can replace all these steps with only one
> > concept, specifically alphabet.
> > This replacement process or concept creation works as long as the steps
> > are clearly defined but it becomes blurry if we fail to do it properly.
> > So is concept creation just the tool needed to overcome the limitations
> > of the serial representation (2 or 3 dimensions?) of a parallel
> > process, reality (4 or more dimensions?).
> > JP

This "replacement" processes you talk about is simply how the meaning of
the symbols are defined.  But few people grasp the complexity of meaning
even though we all use it every day.  One of the largest road blocks to
understanding meaning correctly is our extensive use of written langauge.
When we communiate via written words, we are forced to work with symbols
that have little to know temporal meaning.  We are forced to map all our
temporal meaning into the spatial relationship of the symbols on the page.
Unlike our "eye" we can't say, "I see light NOW" by putting an X on a piece
of paper.  When someone reads the message, they won't know when the "Now"
was that you wrote the message.  So you map the temporal information into
the spatial domain by writing, "I saw the sun come up at 6:10 AM".

We have been forced to translate all our temporal meaning into this spatial
language so that we can communicate with others using written langauge.
And we have learned so much about the nature of the universe reading these
types of words, that we have even been trained to think in these prurely
spatial terms.  We spend so much time thinking about everything in these
spatial terms, that we end up believing that all meaning can be defined in
spatial terms when it can't.

The meaning behind all the words are not spatial, they contain a huge an
very important temporal component.  In the langauge that the brain hardware
uses, there is only only one word spoken.  It's the pulse.  All meaning in
the brain, is represented with on word.  Not even two symbols like our
computers - only one word is used.

All meaning for these words are defined in terms of the hardware which
produced the word, and the time that the word was spoken.  Even though the
sign is the same in all caues "the pulse", the meaning is different for
every device int eh brain that can "talk" this langauge.  THe eye is saying
"I see light NOW", "I see light NOW", the ear is saying, "I hear sound
NOW", "I hear sound NOW".  some nuron is saying, "I hear the ear speak
NOW", "or the ear is really talking alot NOW".

All meaning has to be tied back to the hardware which spoke the word, and
to the the exact point in time the word was generated.

You can't define a concept correctly, unless you ground it's meaning all
the way back to the hardware that gave the concept meaning in the first
place.  It's the hardware that defines the assocation between the symbol,
and the aspect of the universe it represents.

You can't even use words to define the hardware, because those words are
meaningless as well, until their meaning is tied back the universe with the
hardware that created them.

Humans can share meaning easilly, because we all share the same hardware
for defining meaning (our sensors and our brains).  We take this fact for
granted and mostly forget about it the crutial role it plays in defining
meaning.

But, if you try to take the human out of the loop, and replace it with a
machine that understands meaning without the help of humans, then you have
to understand that it's internal representation of "meaning" must also be
grounded to the universe with sensors of some type.  And then its internal
understanding of meaning, will all be in terms of its own sensors.

So, if you want to talk about concepts and the storage of concepts, you
have to understand that a concept is a langauge symbol combined with
meaning.  And meaning can be in terms of other language symbols, but they
must all eventually be grounded back to the hardware that defines their
true relationship to different aspects of the universe.  And even when you
talk about defining symbols in terms of other symboos, you are making
reference to some type of hardware which does that combining.

So, to understand concepts, machines must have their own hardware for
defining their own langauge (sensors and langauge processors like neurons
or and gates or whatever you choose), and must define for themselves, the
meaning of any other langauge, in terms of it's own langauge as defined by
it's own hardware.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/8/2006 7:15:21 PM
Curt Welch wrote:
> "zzbunker" <jimhunter1@comcast.net> wrote:
> 
>>"Just Playing" <gms2004@lycos.com> wrote in message
>>news:1136385467.431070.35790@o13g2000cwo.googlegroups.com...
>>
>>>IMO language is a serial representation of a parallel process, reality
>>>and the creation of abstract concepts is just the absolutely necessary
>>>tool to achieve this approximation.
>>>In a serial representation the points are connected sequentially using
>>>and/or operators to describe nodes and complex configurations of
>>>points.
>>
>>   language is a serial compression of a parallel process.
>>   Which is why it distorts reality more than it represents it.
> 
> 
> The hidden complexity of langauge is not the simple serial stream of
> tokens, but the meaning. [...]

Which means that language, when considered as an expression or encoding 
of meanings, is not serial. Although the signs or tokens are delivered 
serially, they are not interpreted serially: the meaning of some sign S 
is conditional on other signs, {S',S",...} which may occur before or 
after S or both. Or even outside the language stream entirely. That's 
why a parsing diagram of a sentence is not a string but a tree (at 
least: I've seen parsing schemes  that attempt to go beyond syntax, and 
all of them result in networks, some of them multi-dimensional.) NB that 
"sign" here means any behaviour, linguistic or otherwise, that 
conditions the interpretation of the utterance.

Simple example: what is the meaning of "minute"? Both sound and meaning 
of this token depend on other tokens in the sentence. (The fact that in 
English we have a large number of homographs explains why phonics alone 
is insufficient to teach the reading of English, BTW. It's a good thing 
that operant conditioning includes delayed response, or we couldn't 
parse such terms at all.)

Other tokens' meanings are even more complexly determined by the context 
of the utterance in which they occur. Consider the many meanings of 
"fix", for example (and if you have access to an Oxford English 
Dictionary, look it up. Study the examples of usage. You'll develop a 
glimpse of a clue of a notion of how context determines meaning.) Let 
alone the fact that we must write an extended discussion of any new 
meaning for an old token, or merely to explain how we interpret a token...

I suggest that several of the participants in this dialogue take a few 
courses in literary and/or linguistic analysis. A lot of what you say 
points to a vague apprehension of fairly well-understood concepts of how 
language means. Unfortunately, these concepts are at a level of analysis 
that doesn't (as yet?) have much practical bearing on AI, IMO.

HTH
0
Wolf
1/8/2006 7:54:51 PM
Come to think of it, sound waves and light waves are exactly the same
thing, except at different frequency levels.

0
RyanT
1/8/2006 10:05:33 PM
Wolf Kirchmeir <wolfekir@sympatico.ca> wrote:

> I suggest that several of the participants in this dialogue take a few
> courses in literary and/or linguistic analysis. A lot of what you say
> points to a vague apprehension of fairly well-understood concepts of how
> language means. Unfortunately, these concepts are at a level of analysis
> that doesn't (as yet?) have much practical bearing on AI, IMO.

Yeah, the solution to AI defines the foundation on which all meaning is
built.  Solving AI requires lowing the level of analysis of langauge all
the way down to physics.  You have to fill that large gap between what we
know about how atoms works, and what we know about language with some
precise hardware made out of atoms that is able to learn and to use natural
language like humans do.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/8/2006 10:41:02 PM
Wolf Kirchmeir wrote:
> Just Playing wrote:
> > jhn_...@yahoo.com wrote:
> >
> >>Just Playing wrote:
> >>
> >>>This is the 2nd time that I am responding. It seems that there is some
> >>>problem with my server.
> >>>If I understand you correctly you are talking about a representation,
> >>>specifically the concepts encapsulating a relationship.
> >>>This is exactly what I am talking about but I am focusing on the way
> >>>this representation is done.
> >>>I consider that the complexity of what is represented is much higher
> >>>than the one of one that is doing the representation.
> >>>IOW the verbal language has this limitation, which IMO can be
> >>>quantified, and the only way to overcome it is to create "meaning"
> >>>and abstract concepts.
> >>>The amount of information or complexity that has to be encapsulated in
> >>>the concepts can then be calculated.
> >>>We can have a concept that encapsulates "everything" or
> >>>"nothing" as well as concepts that fill in between these two.
> >>>JP
> >>
> >>I tend to disagree with your proposition, but then I'm open for trying.
> >> The reservations that I have are that I believe verbal language is a
> >>higher system than calculation.  IOW the subject system is extrinsic to
> >>the unit system.  Therefore, verbal langage can't be calculated.
> >>Nevertheless, let's give it a go.  Let's use a simple example such as
> >>the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
> >>Sat'  is more complex than 'week'.  Certainly this is so since there
> >>are seven subjects we are using as opposed to the extrinsic subject
> >>'week'.  Maybe what your proposing is that 'week' should be also known
> >>as week(7).  But then what about a work week?  Should we call this
> >>week(5)?  Is this what your getting at?
> >
> >
> > This is not exactly what I am talking about.
> > I start with what is to be represented which I will call R.
> > The second is the representation system, verbal language in this case,
> > and I will call it V.
> > The problem is that the amount of information in R is of such a
> > magnitude that V cannot represent it in real time or in a 1 to 1 ratio.
> > To overcome this limitation V starts to pack information in concepts
> > and this is what can be calculated.
> > If we know how much information is to be transmitted we can transmit in
> > as many concepts as V allows us to do it, from one to the maximum
> > number possible.
> > In your example with days and week you are working inside the V or the
> > representation system and it seems that there is no limitation between
> > R and V anymore.
> > Your example shows that you can replace 7 subjects with one, but as you
> > correctly point out you lose information and create confusion this way.
> > You need a limitation in order to replace 7 days with a week in absence
> > of this limitation there is no reason to do it.
> > Thank you,
> > JP
> >
>
> "A picture is worth a thousand words."
>
> Yup.
>
> Which 1,000 words?
>
> .............
>
> NB that R --> V is different for different people.

Yes but it there may be enough information in the model to move a
little further than we are doing it now.
JP


> Hence you cannot  "calculate" how much information is "packed into" a concept. Whose
> version of the concept? And what algorithm will you use?

That is an interesting point but there are at least 2 ways to overcome
it.
Firstly we do not start from scratch but we learn the concepts in
society and thus we can approximate how much information they contain.
Secondly we can create an experiment where we present the subjects the
same information R and a time limit for the answer or verbal
representation V of R.
While we do not know the amount of information in R we will may have
have enough data in the V answers to find what you call an algorithm.
JP

 Consider a (relatively) simple example: What does "classical greek
sculpture" mean
> to you? Suppose you start talking, explaining the concept as you
> understand it. At what point does one say "That's enough", and start
> "calculating" the information "packed into" the concept?


IMO your example is much too complex at this stage.
JP

>
> The example of "7 days is a week, a week is 7 days" is little more than
> synonymy. It doesn't explain much about "packing information into a
> concept." Was that your point?

No. And this is not my example.
I cross-posted in alt.philosophy too and some replies got posted only
there.
I will try to copy a part of one of my responses to one of these
replies:

Thisbe:
According to CNRS (in France): "Mathematics operates in two
complementary ways. In the *visual* one the meaning of a theorem is
perceived instantly on a geometric figure. The *written* one leans on
language, on algebra; it operates in time."
Have a nice day, Thisbe :)

JP:
"Written" language is just the visual form of the "spoken" one
and as you point out it "operates in time" while the "visual"
language operates in space.
While I do not agree that the visual language does not operate in time
the main point is that there is a difference in the number of
dimensions used by the 2 languages.
Let's say that the "visual" one uses N dimensions while the
"spoken" one uses N-1. This makes the visual amount of information
at least twice that much as the one that can be expressed in the same
unit of time verbally.
IMO this is an extremely important limitation and the only way we as a
species have overcome it is by the creation of "meaning" or
abstract concepts.
JP

0
Just
1/9/2006 12:50:58 AM
Just Playing: IMO language is a serial representation of a parallel
process, reality and the creation of abstract concepts is just the
absolutely necessary
tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points.
---------
Reality, a parallel process is represented not by language, but by
subliminal parallel processes, for lack of a better word by intuition.
Language is the culturally developed tool of serial communication of
selected concepts.

Serial, rational thought is highly specialized. If we had to think
serially to meet the demands of reality, we would die of hunger on the
first day of life.

0
YadaYada
1/9/2006 1:51:25 AM
Wolf Kirchmeir wrote:
> Just Playing wrote:
> > jhn_...@yahoo.com wrote:
> >
> >>Just Playing wrote:
> >>
> >>>This is the 2nd time that I am responding. It seems that there is some
> >>>problem with my server.
> >>>If I understand you correctly you are talking about a representation,
> >>>specifically the concepts encapsulating a relationship.
> >>>This is exactly what I am talking about but I am focusing on the way
> >>>this representation is done.
> >>>I consider that the complexity of what is represented is much higher
> >>>than the one of one that is doing the representation.
> >>>IOW the verbal language has this limitation, which IMO can be
> >>>quantified, and the only way to overcome it is to create "meaning"
> >>>and abstract concepts.
> >>>The amount of information or complexity that has to be encapsulated in
> >>>the concepts can then be calculated.
> >>>We can have a concept that encapsulates "everything" or
> >>>"nothing" as well as concepts that fill in between these two.
> >>>JP
> >>
> >>I tend to disagree with your proposition, but then I'm open for trying.
> >> The reservations that I have are that I believe verbal language is a
> >>higher system than calculation.  IOW the subject system is extrinsic to
> >>the unit system.  Therefore, verbal langage can't be calculated.
> >>Nevertheless, let's give it a go.  Let's use a simple example such as
> >>the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
> >>Sat'  is more complex than 'week'.  Certainly this is so since there
> >>are seven subjects we are using as opposed to the extrinsic subject
> >>'week'.  Maybe what your proposing is that 'week' should be also known
> >>as week(7).  But then what about a work week?  Should we call this
> >>week(5)?  Is this what your getting at?
> >
> >
> > This is not exactly what I am talking about.
> > I start with what is to be represented which I will call R.
> > The second is the representation system, verbal language in this case,
> > and I will call it V.
> > The problem is that the amount of information in R is of such a
> > magnitude that V cannot represent it in real time or in a 1 to 1 ratio.
> > To overcome this limitation V starts to pack information in concepts
> > and this is what can be calculated.
> > If we know how much information is to be transmitted we can transmit in
> > as many concepts as V allows us to do it, from one to the maximum
> > number possible.
> > In your example with days and week you are working inside the V or the
> > representation system and it seems that there is no limitation between
> > R and V anymore.
> > Your example shows that you can replace 7 subjects with one, but as you
> > correctly point out you lose information and create confusion this way.
> > You need a limitation in order to replace 7 days with a week in absence
> > of this limitation there is no reason to do it.
> > Thank you,
> > JP
> >
>
> "A picture is worth a thousand words."
>
> Yup.
>
> Which 1,000 words?
>
> .............
>
> NB that R --> V is different for different people.
Yes but it there may be enough information in the model to move a
little further than we are doing it now.
JP


>Hence you cannot "calculate" how much information is "packed into" a concept. Whose
> version of the concept? And what algorithm will you use?
 That is an interesting point but there are at least 2 ways to overcome
it.
Firstly we do not start from scratch but we learn the concepts in
society and thus we can approximate how much information they contain.
Secondly we can create an experiment where we present the subjects the
same information R and a time limit for the answer or verbal
representation V of R.
While we do not know the amount of information in R we will have
information in the V answers to find what you call an algorithm.
JP



>Consider a (relatively) simple example: What does "classical greek sculpture" mean
> to you? Suppose you start talking, explaining the concept as you
> understand it. At what point does one say "That's enough", and start
> "calculating" the information "packed into" the concept?

IMO your example is much too complex at this stage.
JP
>
> The example of "7 days is a week, a week is 7 days" is little more than
> synonymy. It doesn't explain much about "packing information into a
> concept." Was that your point?

No. And this is not my example.
I cross-posted in alt.philosophy too and some replies got posted only
there.
I will try to copy a part of one of my responses to one of these
replies:
Thisbe:
According to CNRS (in France): "Mathematics operates in two
complementary ways. In the *visual* one the meaning of a theorem is
perceived instantly on a geometric figure. The *written* one leans on
language, on algebra; it operates in time."
Have a nice day, Thisbe :)
JP:
"Written" language is just the visual form of the "spoken" one
and as you point out it "operates in time" while the "visual"
language operates in space.
While I do not agree that the visual language does not operate in time
the main point is that there is a difference in the number of
dimensions used by the 2 languages.
Let's say that the "visual" one uses N dimensions while the
"spoken" one uses N-1. This makes the visual amount of information
at least twice that much as the one that can be expressed in the same
unit of time verbally.
IMO this is an extremely important limitation and the only way we as a
species have overcome it is by the creation of "meaning" or
abstract concepts.
JP

0
Just
1/9/2006 1:52:13 AM
What ARE you looking for uh? - are you trying to fit a language
arithmetic algorithm to a thesis? are you trying to prove that
something U learned a priori fits, or must fit a natural human
growth pattern - cos it won't work

0
mimo_545
1/9/2006 1:55:16 AM
Just Playing: IMO language is a serial representation of a parallel
process, reality and the creation of abstract concepts is just the
absolutely necessary
tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points.
---------
Reality, a parallel process is represented not by language, but by
subliminal parallel processes, for lack of a better word by intuition.
Language is the culturally developed tool of serial communication of
selected concepts.

Serial, rational thought is highly specialized. If we had to think
serially to meet the demands of reality, we would die of hunger on the
first day of life.

0
YadaYada
1/9/2006 2:00:35 AM
Just Playing wrote:
IMO language is a serial representation of a parallel process, reality
and the creation of abstract concepts is just the absolutely necessary
tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points.
---------
Reality, a parallel process is represented not by language, but by
subliminal parallel processes, for lack of a better word by intuition.
If we had to think serially to meet the demands of reality, we would
die of hunger on the first day of life.

Serial, rational thought is highly specialized. Its expression,
language, is the culturally developed tool of serial communication of
selected concepts.

0
YadaYada
1/9/2006 2:15:18 AM
Just Playing: IMO language is a serial representation of a parallel
process, reality and the creation of abstract concepts is just the
absolutely necessary
tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points.
---------
Reality, a parallel process is represented not by language, but by
subliminal parallel processes, for lack of a better word by intuition.
Language is the culturally developed tool of serial communication of
selected concepts.

Serial, rational thought is highly specialized. If we had to think
serially to meet the demands of reality, we would die of hunger on the
first day of life.

0
YadaYada
1/9/2006 2:31:24 AM
Just Playing wrote:
IMO language is a serial representation of a parallel process, reality
and the creation of abstract concepts is just the absolutely necessary
tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points.
---------
Reality, a parallel process is represented not by language, but by
subliminal parallel processes, for lack of a better word by intuition.
If we had to think serially to meet the demands of reality, we would
die of hunger on the first day of life.

Serial, rational thought is highly specialized. Its expression,
language, is the culturally developed tool of serial communication of
selected concepts.

0
YadaYada
1/9/2006 2:45:22 AM
Curt Welch wrote:
> You have to create an assocation between the langauge symbol, and the
> aspect of the universe you are trying to describe with the language symbol.
>
> If the universe was full of nothing but langauge symbols, then mapping from
> symbol to symbol is all you would need.  But it's not.  It's full of
> complex behavior which must be assocated with langauge symbols.  You can't
> describe the meaning of natural langauge only using natural language.
> Somewhere, the meaning of the words must be tied back to the aspects of the
> physical world they are representing.  It's got to all be grounded to
> physcial reality.

This analysis has a crucial step missing. Not one of us is capable of
making such connections between our internal a priori concepts and the
unknowable empirical reality that is out there. Given our limited
sensory and rational capabilities it is impossible, even in theory.

0
YadaYada
1/9/2006 4:33:02 AM
Just Playing wrote:
>> IMO language is a serial representation of a parallel process, reality and the creation of abstract concepts is just the absolutely necessary tool to achieve this approximation.
In a serial representation the points are connected sequentially using
and/or operators to describe nodes and complex configurations of
points. <<
---------
Reality, a parallel process is represented not by language, but by
subliminal parallel processes, for lack of a better word by intuition.
Language is the culturally developed tool of serial communication of
selected concepts.

Serial, rational thought is highly specialized. If we had to think
serially to meet the demands of reality, we would die of hunger on the
first day of life.

0
YadaYada
1/9/2006 4:42:15 AM
Wolf Kirchmeir wrote:
> Just Playing wrote:
> > jhn_...@yahoo.com wrote:
> >
> >>Just Playing wrote:
> >>
> >>>This is the 2nd time that I am responding. It seems that there is some
> >>>problem with my server.
> >>>If I understand you correctly you are talking about a representation,
> >>>specifically the concepts encapsulating a relationship.
> >>>This is exactly what I am talking about but I am focusing on the way
> >>>this representation is done.
> >>>I consider that the complexity of what is represented is much higher
> >>>than the one of one that is doing the representation.
> >>>IOW the verbal language has this limitation, which IMO can be
> >>>quantified, and the only way to overcome it is to create "meaning"
> >>>and abstract concepts.
> >>>The amount of information or complexity that has to be encapsulated in
> >>>the concepts can then be calculated.
> >>>We can have a concept that encapsulates "everything" or
> >>>"nothing" as well as concepts that fill in between these two.
> >>>JP
> >>
> >>I tend to disagree with your proposition, but then I'm open for trying.
> >> The reservations that I have are that I believe verbal language is a
> >>higher system than calculation.  IOW the subject system is extrinsic to
> >>the unit system.  Therefore, verbal langage can't be calculated.
> >>Nevertheless, let's give it a go.  Let's use a simple example such as
> >>the days of the week.  You propose that 'Sun, Mon, Tues, Wed, Thu, Fri,
> >>Sat'  is more complex than 'week'.  Certainly this is so since there
> >>are seven subjects we are using as opposed to the extrinsic subject
> >>'week'.  Maybe what your proposing is that 'week' should be also known
> >>as week(7).  But then what about a work week?  Should we call this
> >>week(5)?  Is this what your getting at?
> >
> >
> > This is not exactly what I am talking about.
> > I start with what is to be represented which I will call R.
> > The second is the representation system, verbal language in this case,
> > and I will call it V.
> > The problem is that the amount of information in R is of such a
> > magnitude that V cannot represent it in real time or in a 1 to 1 ratio.
> > To overcome this limitation V starts to pack information in concepts
> > and this is what can be calculated.
> > If we know how much information is to be transmitted we can transmit in
> > as many concepts as V allows us to do it, from one to the maximum
> > number possible.
> > In your example with days and week you are working inside the V or the
> > representation system and it seems that there is no limitation between
> > R and V anymore.
> > Your example shows that you can replace 7 subjects with one, but as you
> > correctly point out you lose information and create confusion this way.
> > You need a limitation in order to replace 7 days with a week in absence
> > of this limitation there is no reason to do it.
> > Thank you,
> > JP
> >
>
> "A picture is worth a thousand words."
>
> Yup.
>
> Which 1,000 words?
>
> .............
>
> NB that R --> V is different for different people. Hence you cannot
> "calculate" how much information is "packed into" a concept. Whose
> version of the concept? And what algorithm will you use? Consider a
> (relatively) simple example: What does "classical greek sculpture" mean
> to you? Suppose you start talking, explaining the concept as you
> understand it. At what point does one say "That's enough", and start
> "calculating" the information "packed into" the concept?

I had a very bad experience with my replies being posted in the last
few days. I still don't see all of them and some of the them got
posted with a delay of 12 to 24 hours.
In regards to your example I had some time to read and think it over.
My objection is that we are talking about completely different things.
I am talking about reality as is perceived thru all the senses while
your example deals only with concepts inside the verbal language.
IOW you do not have a R and V but only a V.
Greek sculpture is a concept and not reality. In order to calculate you
have to have something perceived not exclusively thru verbal language
and your example lacks this characteristic.
Jan 8, 11.55pm, EST
JP


> The example of "7 days is a week, a week is 7 days" is little more than
> synonymy. It doesn't explain much about "packing information into a
> concept." Was that your point?

0
Just
1/9/2006 4:56:06 AM
Curt Welch wrote:
> "zzbunker" <jimhunter1@comcast.net> wrote:
> > "Just Playing" <gms2004@lycos.com> wrote in message
> > news:1136385467.431070.35790@o13g2000cwo.googlegroups.com...
> > > IMO language is a serial representation of a parallel process, reality
> > > and the creation of abstract concepts is just the absolutely necessary
> > > tool to achieve this approximation.
> > > In a serial representation the points are connected sequentially using
> > > and/or operators to describe nodes and complex configurations of
> > > points.
> >
> >    language is a serial compression of a parallel process.
> >    Which is why it distorts reality more than it represents it.
>
> The hidden complexity of langauge is not the simple serial stream of
> tokens, but the meaning.  It's not just the parallel nature that causes the
> distrotion.  The universe is both parallel and analog.  How do you
> represent the complex behavior of even a single water molecule in a glass
> of water with a serial stream of tokens taken from a finite set of symbols?
> Saying that that there is some distortion here is a gross understatemet.
> There's nearly an infinite amount of information left out of the meaning
> when you try to use langauge to represent even a trivially small part of
> the universe.
>
> And that's what meaning is all about.  It's the mapping from from the
> langauge, to some aspect of the universe which is being represented by the
> language.
>
> > > OTOH a concept can become any of these nodes or any combination of
> > > connections between nodes or points, contiguous or not.
>
> A "concept" is a combination of a language symbol, and it's meaning.

Why do you need a concept in the first place?
Jan 9, 12.25pm
JP

>
> You have attempted to define "meaning" as some finite combination of other
> langauge symbols.  That won't do it.  Meaning is more complex than that.
> You have to create an assocation between the langauge symbol, and the
> aspect of the universe you are trying to describe with the language symbol.
>
> If the universe was full of nothing but langauge symbols, then mapping from
> symbol to symbol is all you would need.  But it's not.  It's full of
> complex behavior which must be assocated with langauge symbols.  You can't
> describe the meaning of natural langauge only using natural language.
> Somewhere, the meaning of the words must be tied back to the aspects of the
> physical world they are representing.  It's got to all be grounded to
> physcial reality.
>
> This is done for us by our sensors.  The meaning of "light" is defined by
> the stream of pulses produces by our eyes.  Each pulse, coming from the
> eye, represents an aspect of the universe we call light.  Each pulse, is a
> symbol, in a langauge, who's meaning is defined by the sensor that "spoke"
> the word.
>
> Notice also that this low level language on which we base all our high
> level meaning, is temporal in nature.  The meaning isn't timeless like the
> words we write down on paper are.  A pulse coming from the eye means
> something to the effect of "I see light NOW".
>
> All the meaning, of all the words, and all concepts, used by humans, are
> defined in terms of these temporal pulses, which all have meaning, defined
> by the hardware which "spoke" the symbol.
>
> > > IOW in order to describe all the signs from A to Z we need n (26) steps
> > > and operators, but we can replace all these steps with only one
> > > concept, specifically alphabet.
> > > This replacement process or concept creation works as long as the steps
> > > are clearly defined but it becomes blurry if we fail to do it properly.
> > > So is concept creation just the tool needed to overcome the limitations
> > > of the serial representation (2 or 3 dimensions?) of a parallel
> > > process, reality (4 or more dimensions?).
> > > JP
>
> This "replacement" processes you talk about is simply how the meaning of
> the symbols are defined.  But few people grasp the complexity of meaning
> even though we all use it every day.  One of the largest road blocks to
> understanding meaning correctly is our extensive use of written langauge.
> When we communiate via written words, we are forced to work with symbols
> that have little to know temporal meaning.  We are forced to map all our
> temporal meaning into the spatial relationship of the symbols on the page.
> Unlike our "eye" we can't say, "I see light NOW" by putting an X on a piece
> of paper.  When someone reads the message, they won't know when the "Now"
> was that you wrote the message.  So you map the temporal information into
> the spatial domain by writing, "I saw the sun come up at 6:10 AM".
>
> We have been forced to translate all our temporal meaning into this spatial
> language so that we can communicate with others using written langauge.
> And we have learned so much about the nature of the universe reading these
> types of words, that we have even been trained to think in these prurely
> spatial terms.  We spend so much time thinking about everything in these
> spatial terms, that we end up believing that all meaning can be defined in
> spatial terms when it can't.
>
> The meaning behind all the words are not spatial, they contain a huge an
> very important temporal component.  In the langauge that the brain hardware
> uses, there is only only one word spoken.  It's the pulse.  All meaning in
> the brain, is represented with on word.  Not even two symbols like our
> computers - only one word is used.
>
> All meaning for these words are defined in terms of the hardware which
> produced the word, and the time that the word was spoken.  Even though the
> sign is the same in all caues "the pulse", the meaning is different for
> every device int eh brain that can "talk" this langauge.  THe eye is saying
> "I see light NOW", "I see light NOW", the ear is saying, "I hear sound
> NOW", "I hear sound NOW".  some nuron is saying, "I hear the ear speak
> NOW", "or the ear is really talking alot NOW".
>
> All meaning has to be tied back to the hardware which spoke the word, and
> to the the exact point in time the word was generated.
>
> You can't define a concept correctly, unless you ground it's meaning all
> the way back to the hardware that gave the concept meaning in the first
> place.  It's the hardware that defines the assocation between the symbol,
> and the aspect of the universe it represents.
>
> You can't even use words to define the hardware, because those words are
> meaningless as well, until their meaning is tied back the universe with the
> hardware that created them.
>
> Humans can share meaning easilly, because we all share the same hardware
> for defining meaning (our sensors and our brains).  We take this fact for
> granted and mostly forget about it the crutial role it plays in defining
> meaning.
>
> But, if you try to take the human out of the loop, and replace it with a
> machine that understands meaning without the help of humans, then you have
> to understand that it's internal representation of "meaning" must also be
> grounded to the universe with sensors of some type.  And then its internal
> understanding of meaning, will all be in terms of its own sensors.
>
> So, if you want to talk about concepts and the storage of concepts, you
> have to understand that a concept is a langauge symbol combined with
> meaning.  And meaning can be in terms of other language symbols, but they
> must all eventually be grounded back to the hardware that defines their
> true relationship to different aspects of the universe.  And even when you
> talk about defining symbols in terms of other symboos, you are making
> reference to some type of hardware which does that combining.
>
> So, to understand concepts, machines must have their own hardware for
> defining their own langauge (sensors and langauge processors like neurons
> or and gates or whatever you choose), and must define for themselves, the
> meaning of any other langauge, in terms of it's own langauge as defined by
> it's own hardware.
>
> --
> Curt Welch                                            http://CurtWelch.Com/
> curt@kcwc.com                                        http://NewsReader.Com/

0
Just
1/9/2006 5:23:57 AM
"YadaYada" <yadayada___@excite.com> wrote:
> Curt Welch wrote:
> > You have to create an assocation between the langauge symbol, and the
> > aspect of the universe you are trying to describe with the language
> > symbol.
> >
> > If the universe was full of nothing but langauge symbols, then mapping
> > from symbol to symbol is all you would need.  But it's not.  It's full
> > of complex behavior which must be assocated with langauge symbols.  You
> > can't describe the meaning of natural langauge only using natural
> > language. Somewhere, the meaning of the words must be tied back to the
> > aspects of the physical world they are representing.  It's got to all
> > be grounded to physcial reality.
>
> This analysis has a crucial step missing. Not one of us is capable of
> making such connections between our internal a priori concepts and the
> unknowable empirical reality that is out there. Given our limited
> sensory and rational capabilities it is impossible, even in theory.

I don't undestand your point.  What is this idea of your's that there is
something "unknownable" that is somehow important for us to know?

If something is unknownable, then it will never affect us and we will never
know about and we will never sense it and we will never talk about and we
will never define any of our words using it.  So why are you trying to talk
about something that doesn't exist to us?  And how is that going to prevent
us from grounding the meaning of our words?

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/9/2006 9:48:57 AM
"Just Playing" <gms2004@lycos.com> wrote:
> Curt Welch wrote:
> > "zzbunker" <jimhunter1@comcast.net> wrote:
> > > "Just Playing" <gms2004@lycos.com> wrote in message
> > > news:1136385467.431070.35790@o13g2000cwo.googlegroups.com...
> > > > IMO language is a serial representation of a parallel process,
> > > > reality and the creation of abstract concepts is just the
> > > > absolutely necessary tool to achieve this approximation.
> > > > In a serial representation the points are connected sequentially
> > > > using and/or operators to describe nodes and complex configurations
> > > > of points.
> > >
> > >    language is a serial compression of a parallel process.
> > >    Which is why it distorts reality more than it represents it.
> >
> > The hidden complexity of langauge is not the simple serial stream of
> > tokens, but the meaning.  It's not just the parallel nature that causes
> > the distrotion.  The universe is both parallel and analog.  How do you
> > represent the complex behavior of even a single water molecule in a
> > glass of water with a serial stream of tokens taken from a finite set
> > of symbols? Saying that that there is some distortion here is a gross
> > understatemet. There's nearly an infinite amount of information left
> > out of the meaning when you try to use langauge to represent even a
> > trivially small part of the universe.
> >
> > And that's what meaning is all about.  It's the mapping from from the
> > langauge, to some aspect of the universe which is being represented by
> > the language.
> >
> > > > OTOH a concept can become any of these nodes or any combination of
> > > > connections between nodes or points, contiguous or not.
> >
> > A "concept" is a combination of a language symbol, and it's meaning.
>
> Why do you need a concept in the first place?
> Jan 9, 12.25pm
> JP

Why do I need it?  Or why do humans need it?  Or why do they exist?  What
are you asking?

I think we have them because they control the motion of our arms and legs.
Evolution built us a "concept creation machine" for the purpose of creating
arm and leg motions that were useful for our survival.  The brain is just a
complex context sensitive arm and leg controller.  The hardware in our
brain which we use to create and understand langauge is just more of the
same hardware which evolution built for us to move our arms and legs.  The
problem of speaking and understanding words is no different at the lowest
level from the problem of moving our arms and legs and understing the world
around us.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/9/2006 9:57:45 AM
Curt Welch wrote:
> "Just Playing" <gms2004@lycos.com> wrote:
> > Curt Welch wrote:
> > > "zzbunker" <jimhunter1@comcast.net> wrote:
> > > > "Just Playing" <gms2004@lycos.com> wrote in message
> > > > news:1136385467.431070.35790@o13g2000cwo.googlegroups.com...
> > > > > IMO language is a serial representation of a parallel process,
> > > > > reality and the creation of abstract concepts is just the
> > > > > absolutely necessary tool to achieve this approximation.
> > > > > In a serial representation the points are connected sequentially
> > > > > using and/or operators to describe nodes and complex configurations
> > > > > of points.
> > > >
> > > >    language is a serial compression of a parallel process.
> > > >    Which is why it distorts reality more than it represents it.
> > >
> > > The hidden complexity of langauge is not the simple serial stream of
> > > tokens, but the meaning.  It's not just the parallel nature that causes
> > > the distrotion.  The universe is both parallel and analog.  How do you
> > > represent the complex behavior of even a single water molecule in a
> > > glass of water with a serial stream of tokens taken from a finite set
> > > of symbols? Saying that that there is some distortion here is a gross
> > > understatemet. There's nearly an infinite amount of information left
> > > out of the meaning when you try to use langauge to represent even a
> > > trivially small part of the universe.
> > >
> > > And that's what meaning is all about.  It's the mapping from from the
> > > langauge, to some aspect of the universe which is being represented by
> > > the language.
> > >
> > > > > OTOH a concept can become any of these nodes or any combination of
> > > > > connections between nodes or points, contiguous or not.
> > >
> > > A "concept" is a combination of a language symbol, and it's meaning.
> >
> > Why do you need a concept in the first place?
> > Jan 9, 12.25pm
> > JP
>
> Why do I need it?  Or why do humans need it?  Or why do they exist?  What
> are you asking?
>
> I think we have them because they control the motion of our arms and legs.
> Evolution built us a "concept creation machine" for the purpose of creating
> arm and leg motions that were useful for our survival.  The brain is just a
> complex context sensitive arm and leg controller.  The hardware in our
> brain which we use to create and understand langauge is just more of the
> same hardware which evolution built for us to move our arms and legs.  The
> problem of speaking and understanding words is no different at the lowest
> level from the problem of moving our arms and legs and understing the world
> around us.
>
> --
> Curt Welch                                            http://CurtWelch.Com/
> curt@kcwc.com                                        http://NewsReader.Com/

My point is very simple.
A concept is a method of packing data as a method to overcome a
limitation and IMO this is the only reason we create concepts.
As an example of limitation is the fact that we cannot process all the
information in "real time".
I tried to use a simple model where what is to be represented (R) is
much more complex than the representation system, which in this case is
verbal language (V).
I started first by stating that all the senses are perceiving the data
(R) and then the verbal language (V) has to pack into the data into
concepts to process and transmit.
The main point is that this is a very "limitation overcome"
oriented process of concept creation or data packaging and after a
while the new concepts created increase the amount of data to be
processed and new concepts have to be created.
Now I can look at the verbal language only and consider it as R and
then we start the process of packing the data inside the language
whenever we have a limitation, as having to answer in such a manner
that we do not have the option to use more or descriptive concepts.
This is the case with the 26 signs of the being replaced by the concept
of alphabet, or the 7 days being replaced by the concept of week, or
all the statues found in some areas of the world, created at some
period of time, by a common segment of population as being Greek
sculpture.
OTOH I agree with almost everything you said about concepts in your
previous reply and I look at human body as the hardware that defines
our language.
Jan 9, 9.45 am
JP

0
Just
1/9/2006 2:41:06 PM
Curt Welch wrote:
> > You have to create an assocation between the langauge symbol, and the
> > aspect of the universe you are trying to describe with the language
> > symbol. . . . It's got to all be grounded to physcial reality.

This analysis has a crucial step missing. Not one of us is capable of
making such connections between our internal a priori concepts and the
unknowable empirical reality that is out there. Given our limited
sensory and rational capabilities it is impossible, even in theory.

> > I don't undestand your point.  What is this idea of your's that there is
> > something "unknownable" that is somehow important for us to know?
> > If something is unknownable, then it will never affect us and we will never
> > know about and we will never sense it and we will never talk about and we
> > will never define any of our words using it.  So why are you trying to talk
> > about something that doesn't exist to us?  And how is that going to prevent
> > us from grounding the meaning of our words?

By "we" do you mean an individual or culture? This distinction is
crucial in the connection between language, which is a public, a priori
set of words and definitions, and concepts which are private instances
of public language.

The reality connection is from public (a priori) knowledge to public (a
priori) language. The innate ability to form concepts is stoked by
cultural transmission, from adult to child. Personal life experience is
a thin overlay on top of what we acquire culturally.

0
YadaYada
1/9/2006 2:51:11 PM
Just Playing wrote:

<snip>
> My point is very simple.
> A concept is a method of packing data as a method to overcome a
> limitation and IMO this is the only reason we create concepts.

i thought it was to communicate ? in the sense of if i didn't want to
communicate then language would be unecessary and so we wouldn't need
to create concepts, or ?

sammi
0
sam
1/9/2006 2:53:52 PM
> Curt Welch wrote:
> > > You have to create an assocation between the langauge symbol, and the
> > > aspect of the universe you are trying to describe with the language
> > > symbol. . . . It's got to all be grounded to physcial reality.
I agree.  What we think and communicate must closely correspond to the
universe if it is to have any survival value.
> YadaYada wrote:
> This analysis has a crucial step missing. Not one of us is capable of
> making such connections between our internal a priori concepts and the
> unknowable empirical reality that is out there. Given our limited
> sensory and rational capabilities it is impossible, even in theory.
I agree.  We do not deal with true reality, only qualia.

Earlier, Curt Welch noted that language must be grounded in physics.

It is.

The 3 fundamental laws of the empirical universe are:
#1 Differences Exist.  Because of this law, language has nouns,
mathematics has variables (a, b, c, x, y, z, etc.), and physics has
objects (galaxies, planets, molecules, atoms, quarks, etc.)
#2 Differences are Dynamic.  Because of this law, language has verbs,
mathematics has operators (+, -, *, /, etc.), and physics has laws of
motion.
#3 Differences are Relative.  Because of this law, language has
modifiers (adjectives and adverbs), mathematics has relative symbols
(>, <, =, etc.), and physics has scalar references (magnitude, gravity,
etc.)

In the end, it is the dynamic (the verb, if you will) which garners our
interest.  There is a classic cartoon about two germans listening to a
countryman who is speaking.  One listener turns to the other and says,
"What's he talking about?"  The neighbor replies, "I don't know.  He
hasn't gotten to the verb."

And it is the verb which is central to our every day speech:
"Where are you going?"
"I'm going to the store to get eggs and milk."
This dialectic exchange is short and efficient in communicating the
essence of the requested information.  However, in reality, "going to
the store" is a grand routine made up of numerous subroutines:
- get into the car
- drive to the store and park
- walk through the aisles and get eggs and milk
- pay for the items
- drive home
- put the items in the refrigerator
In turn, each of these subroutines are composed of even smaller
modules.  "Pay for the items" includes multiple communication routines
with the clerk as well as many coordinated eye-hand movements.  Each of
these, in turn, are control system feedback loops.

Thus, in the end, all of our behaviors are tree-like hierarchies of
process modules.  To communicate with a computer requires that an AI
system learn the human hierarchies and be able to retrieve the level
being communicated by a human and orient its behaviors to that level,
knowing the implications of the specific tree structure being acted
upon.

If you tell a robot to go to the store and get eggs and milk, the robot
needs to have the whole tree of routines, subroutines, sub-subroutines,
etc., in order to comply.

Perhaps the future of AI may not be in programming a robot to drive a
car, but in teaching it how to learn tree structures of process
modules,  In other words, "monkey see, monkey do."

0
JAK
1/9/2006 4:04:38 PM
"YadaYada" <yadayada___@excite.com> wrote:
> Curt Welch wrote:
> > > You have to create an assocation between the langauge symbol, and the
> > > aspect of the universe you are trying to describe with the language
> > > symbol. . . . It's got to all be grounded to physcial reality.
>
> This analysis has a crucial step missing. Not one of us is capable of
> making such connections between our internal a priori concepts and the
> unknowable empirical reality that is out there. Given our limited
> sensory and rational capabilities it is impossible, even in theory.
>
> > > I don't undestand your point.  What is this idea of your's that there
> > > is something "unknownable" that is somehow important for us to know?
> > > If something is unknownable, then it will never affect us and we will
> > > never know about and we will never sense it and we will never talk
> > > about and we will never define any of our words using it.  So why are
> > > you trying to talk about something that doesn't exist to us?  And how
> > > is that going to prevent us from grounding the meaning of our words?
>
> By "we" do you mean an individual or culture?

I mean you, me, and anyone else that we want to include.

> This distinction is
> crucial in the connection between language, which is a public, a priori
> set of words and definitions, and concepts which are private instances
> of public language.

That's just nonsense to me.  There is no public languge.  There's nothing
to use it.  All we have is a few billion private languages.  You can't find
one single word written in a public language.  It was all created by the
act of some individual using their own idea private langauge.

The culture is nothing more than the sum of it's parts.  If you and I were
all that existed in the world, you and I together would be the "culture"
and all we would be is two people, each with our own private language.

By communicating and trying to solve common problems, our private langauges
would tend to converge, but still, there would never be a "public" langauge
between us.  All there would be, is your languge and my language.

At best, you could describe public langauge as the total combination of our
two private languages.

> The reality connection is from public (a priori) knowledge to public (a
> priori) language. The innate ability to form concepts is stoked by
> cultural transmission, from adult to child. Personal life experience is
> a thin overlay on top of what we acquire culturally.

Yeah, but you talk as if there's some odd magic to this concept of culture
when there's nothing magically here at all.

If you and I were the culture, then my personal expernece doesn't include
your personal experinece.  Big deal.  That's obvious.

So, when you wrote:

> This analysis has a crucial step missing. Not one of us is capable of
> making such connections between our internal a priori concepts and the
> unknowable empirical reality that is out there.

All you seem to be saying is that I can't know what you know.  Well yeah,
that's obvious.  But what does that have to do with what I wrote?  Why is
that a "crucial missing step"?

The meaning of our natural langauge is defined by the hardware which maps
aspects of the universe into the langauge (our body).  It's the physical
properties of our body that grounds the meaning of each of our own private
langauges to the physical world.

My body doesn't define your langauge any more than your body defines mine.

The langauge (/knowledge) of the culture you make reference to doesn't
exist - expect as a collection of private langauges.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/9/2006 6:59:57 PM
On 9 Jan 2006 08:04:38 -0800, "JAK" <jak@theoryofmind.org> in
comp.ai.philosophy wrote:

>> Curt Welch wrote:
>> > > You have to create an assocation between the langauge symbol, and the
>> > > aspect of the universe you are trying to describe with the language
>> > > symbol. . . . It's got to all be grounded to physcial reality.
>I agree.  What we think and communicate must closely correspond to the
>universe if it is to have any survival value.
>> YadaYada wrote:
>> This analysis has a crucial step missing. Not one of us is capable of
>> making such connections between our internal a priori concepts and the
>> unknowable empirical reality that is out there. Given our limited
>> sensory and rational capabilities it is impossible, even in theory.
>I agree.  We do not deal with true reality, only qualia.
>
>Earlier, Curt Welch noted that language must be grounded in physics.
>
>It is.
>
>The 3 fundamental laws of the empirical universe are:
>#1 Differences Exist.  Because of this law, language has nouns,
>mathematics has variables (a, b, c, x, y, z, etc.), and physics has
>objects (galaxies, planets, molecules, atoms, quarks, etc.)
>#2 Differences are Dynamic.  Because of this law, language has verbs,
>mathematics has operators (+, -, *, /, etc.), and physics has laws of
>motion.
>#3 Differences are Relative.  Because of this law, language has
>modifiers (adjectives and adverbs), mathematics has relative symbols
>(>, <, =, etc.), and physics has scalar references (magnitude, gravity,
>etc.)
>
>In the end, it is the dynamic (the verb, if you will) which garners our
>interest.  There is a classic cartoon about two germans listening to a
>countryman who is speaking.  One listener turns to the other and says,
>"What's he talking about?"  The neighbor replies, "I don't know.  He
>hasn't gotten to the verb."
>
>And it is the verb which is central to our every day speech:
>"Where are you going?"
>"I'm going to the store to get eggs and milk."
>This dialectic exchange is short and efficient in communicating the
>essence of the requested information.  However, in reality, "going to
>the store" is a grand routine made up of numerous subroutines:
>- get into the car
>- drive to the store and park
>- walk through the aisles and get eggs and milk
>- pay for the items
>- drive home
>- put the items in the refrigerator
>In turn, each of these subroutines are composed of even smaller
>modules.  "Pay for the items" includes multiple communication routines
>with the clerk as well as many coordinated eye-hand movements.  Each of
>these, in turn, are control system feedback loops.
>
>Thus, in the end, all of our behaviors are tree-like hierarchies of
>process modules.  To communicate with a computer requires that an AI
>system learn the human hierarchies and be able to retrieve the level
>being communicated by a human and orient its behaviors to that level,
>knowing the implications of the specific tree structure being acted
>upon.
>
>If you tell a robot to go to the store and get eggs and milk, the robot
>needs to have the whole tree of routines, subroutines, sub-subroutines,
>etc., in order to comply.
>
>Perhaps the future of AI may not be in programming a robot to drive a
>car, but in teaching it how to learn tree structures of process
>modules,  In other words, "monkey see, monkey do."

All you've really done is state a series of problematic conclusions.
What is your proof and demonstration that such things are true?

For that matter if language is grounded in physics as you claim are
you talking about equal and opposite ideas or exactly what kind of
physics did you have in mind?

What kind of physics supports propositions and propositional truth?

What kind of physics supports self contradictions and lying?

And finally what kind of physics supports the relations between
predicates? I don't see any predicates physically walzing around
through space hand in hand. For that matter I don't see predicates
of any kind in space at all.

~v~~

0
lesterDELzick
1/9/2006 9:12:51 PM
On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
comp.ai.philosophy wrote:

>"YadaYada" <yadayada___@excite.com> wrote:
>> Curt Welch wrote:
>> > > You have to create an assocation between the langauge symbol, and the
>> > > aspect of the universe you are trying to describe with the language
>> > > symbol. . . . It's got to all be grounded to physcial reality.
>>
>> This analysis has a crucial step missing. Not one of us is capable of
>> making such connections between our internal a priori concepts and the
>> unknowable empirical reality that is out there. Given our limited
>> sensory and rational capabilities it is impossible, even in theory.
>>
>> > > I don't undestand your point.  What is this idea of your's that there
>> > > is something "unknownable" that is somehow important for us to know?
>> > > If something is unknownable, then it will never affect us and we will
>> > > never know about and we will never sense it and we will never talk
>> > > about and we will never define any of our words using it.  So why are
>> > > you trying to talk about something that doesn't exist to us?  And how
>> > > is that going to prevent us from grounding the meaning of our words?
>>
>> By "we" do you mean an individual or culture?
>
>I mean you, me, and anyone else that we want to include.
>
>> This distinction is
>> crucial in the connection between language, which is a public, a priori
>> set of words and definitions, and concepts which are private instances
>> of public language.
>
>That's just nonsense to me.  There is no public languge.  There's nothing
>to use it.  All we have is a few billion private languages.  You can't find
>one single word written in a public language.  It was all created by the
>act of some individual using their own idea private langauge.

Then why are you having this conversation?

~v~~

0
lesterDELzick
1/9/2006 9:14:18 PM
lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
> comp.ai.philosophy wrote:
>
> >"YadaYada" <yadayada___@excite.com> wrote:
> >> Curt Welch wrote:
> >> > > You have to create an assocation between the langauge symbol, and
> >> > > the aspect of the universe you are trying to describe with the
> >> > > language symbol. . . . It's got to all be grounded to physcial
> >> > > reality.
> >>
> >> This analysis has a crucial step missing. Not one of us is capable of
> >> making such connections between our internal a priori concepts and the
> >> unknowable empirical reality that is out there. Given our limited
> >> sensory and rational capabilities it is impossible, even in theory.
> >>
> >> > > I don't undestand your point.  What is this idea of your's that
> >> > > there is something "unknownable" that is somehow important for us
> >> > > to know? If something is unknownable, then it will never affect us
> >> > > and we will never know about and we will never sense it and we
> >> > > will never talk about and we will never define any of our words
> >> > > using it.  So why are you trying to talk about something that
> >> > > doesn't exist to us?  And how is that going to prevent us from
> >> > > grounding the meaning of our words?
> >>
> >> By "we" do you mean an individual or culture?
> >
> >I mean you, me, and anyone else that we want to include.
> >
> >> This distinction is
> >> crucial in the connection between language, which is a public, a
> >> priori set of words and definitions, and concepts which are private
> >> instances of public language.
> >
> >That's just nonsense to me.  There is no public languge.  There's
> >nothing to use it.  All we have is a few billion private languages.  You
> >can't find one single word written in a public language.  It was all
> >created by the act of some individual using their own idea private
> >langauge.
>
> Then why are you having this conversation?

We do it for many reasons.  But for the most part, we are just attempting
to reduce the differences in our private languages.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/9/2006 9:58:57 PM
The proof is definitional: there is that which is "P" and that which is
"NOT P" (sentential logic, as I recall).  The existence of "not" proves
the existence of difference.  A ball is not a sign post, a sign post is
not the ocean, and the ocean is not a turtle.  "Not" is the proof that
whatever "is" has limits.  It is limited by that which it is not.
Similarly, processes (verbs) are defined by "not."  Empirical proof is
endless.  Where would you like to start?

For physics, again, where would you like to start.  If anything in
physics can be defined by nouns, verbs, and modifiers, it adheres to
the 3 laws.  If physics is soley described by mathematics - vectors,
for example - then onces again the 3 laws apply.  It is impossible to
communicate outside of the 3 laws.

Yet, they are human inventions created to deal with the "real" world,
whatever that may be.  The 3 laws are merely the basis of our
perceptions of the universe in factors which we can recognize, record,
and communicate.

Propositions, propositional truth, self contradictions, and lying are
all constructs created by humans.  Each includes processes which convey
meaning to all of us (given standard definitions).  Self contradictions
and lying involve "not" in some degree and are a by-product of
conscious thought.  Anything which can be thought exists within the
brain.  "Predicates waltz around" in the brain - a physical and
empirical facet of the universe.

Obviously, you perceive a conflict.  Is your concern in regards to
Zermelo-Fraenkel set theory or what?  I can address your concern more
adroitly if you give your personal definition of "predicate" along with
an example.

Let me forwarn you that attempts to justify thoughts being "outside of
the physical universe" (a "dual universe" scenario) fail when
communication between the supposed "other universe" and our brains
comes into play.  At that juncture, some mechanism to cross the
threshold between the two universes and activate neurons of the brain
requires a physical action which would fall under the perview of
physics.

0
JAK
1/10/2006 12:46:29 AM
I agree.

0
JAK
1/10/2006 12:51:24 AM
Curt Welch wrote:
> > Then why are you having this conversation?
> We do it for many reasons.  But for the most part, we are just attempting
> to reduce the differences in our private languages.

How can you and I have such wide disparity in understanding the world
if I have a reasonably satisfactory understanding of science?

Without cultural training, what means did I use to come to an
understanding of 
chemical bonds that I have never experienced?

0
YadaYada
1/10/2006 12:58:46 AM
On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
comp.ai.philosophy wrote:

>lesterDELzick@worldnet.att.net (Lester Zick) wrote:
>> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
>> comp.ai.philosophy wrote:
>>
>> >"YadaYada" <yadayada___@excite.com> wrote:
>> >> Curt Welch wrote:
>> >> > > You have to create an assocation between the langauge symbol, and
>> >> > > the aspect of the universe you are trying to describe with the
>> >> > > language symbol. . . . It's got to all be grounded to physcial
>> >> > > reality.
>> >>
>> >> This analysis has a crucial step missing. Not one of us is capable of
>> >> making such connections between our internal a priori concepts and the
>> >> unknowable empirical reality that is out there. Given our limited
>> >> sensory and rational capabilities it is impossible, even in theory.
>> >>
>> >> > > I don't undestand your point.  What is this idea of your's that
>> >> > > there is something "unknownable" that is somehow important for us
>> >> > > to know? If something is unknownable, then it will never affect us
>> >> > > and we will never know about and we will never sense it and we
>> >> > > will never talk about and we will never define any of our words
>> >> > > using it.  So why are you trying to talk about something that
>> >> > > doesn't exist to us?  And how is that going to prevent us from
>> >> > > grounding the meaning of our words?
>> >>
>> >> By "we" do you mean an individual or culture?
>> >
>> >I mean you, me, and anyone else that we want to include.
>> >
>> >> This distinction is
>> >> crucial in the connection between language, which is a public, a
>> >> priori set of words and definitions, and concepts which are private
>> >> instances of public language.
>> >
>> >That's just nonsense to me.  There is no public languge.  There's
>> >nothing to use it.  All we have is a few billion private languages.  You
>> >can't find one single word written in a public language.  It was all
>> >created by the act of some individual using their own idea private
>> >langauge.
>>
>> Then why are you having this conversation?
>
>We do it for many reasons.  But for the most part, we are just attempting
>to reduce the differences in our private languages.

Then any reduction in differences in your private languages makes it a
public language to that extent.

~v~~

0
lesterDELzick
1/10/2006 1:02:42 AM
On 9 Jan 2006 16:46:29 -0800, "JAK" <jak@theoryofmind.org> in
comp.ai.philosophy wrote:

It's not clear to whom you're replying. If you're replying to me:

>The proof is definitional: there is that which is "P" and that which is
>"NOT P" (sentential logic, as I recall).  The existence of "not" proves
>the existence of difference.  A ball is not a sign post, a sign post is
>not the ocean, and the ocean is not a turtle.  "Not" is the proof that
>whatever "is" has limits.  It is limited by that which it is not.
>Similarly, processes (verbs) are defined by "not."  Empirical proof is
>endless.  Where would you like to start?

Well I don't necessarily disagree that "not" is true of everything. In
fact my signature logo says exactly that. And I've been proving so for
the last twenty plus years. Where I disagree is saying that you've
proved anything. Definitional proof? Proof by definition? So now
you're defining what's true? That's just the empirical assumption of
truth in action. I supposed I should be encouraged because it seems
we've reached the fourth law of ideation: "I thought of it first".

>For physics, again, where would you like to start.  If anything in
>physics can be defined by nouns, verbs, and modifiers, it adheres to
>the 3 laws.  If physics is soley described by mathematics - vectors,
>for example - then onces again the 3 laws apply.  It is impossible to
>communicate outside of the 3 laws.

Great. The problem was never whether physics requires predication but
whether predication follows the laws of physics.Either you can produce
equal and opposite and ideas or predication doesn't follow the laws of
physics.

>Yet, they are human inventions created to deal with the "real" world,
>whatever that may be.  The 3 laws are merely the basis of our
>perceptions of the universe in factors which we can recognize, record,
>and communicate.
>
>Propositions, propositional truth, self contradictions, and lying are
>all constructs created by humans.  

So what?

>                                                    Each includes processes which convey
>meaning to all of us (given standard definitions).  Self contradictions
>and lying involve "not" in some degree and are a by-product of
>conscious thought.  Anything which can be thought exists within the
>brain.  "Predicates waltz around" in the brain - a physical and
>empirical facet of the universe.

Which you have yet to establish as true.

>Obviously, you perceive a conflict.  Is your concern in regards to
>Zermelo-Fraenkel set theory or what?  I can address your concern more
>adroitly if you give your personal definition of "predicate" along with
>an example.

I'm just asking for your proof because I've already given mine. Don't
expect to change the subject.I'm not interested in anyones set theory.

>Let me forwarn you that attempts to justify thoughts being "outside of
>the physical universe" (a "dual universe" scenario) fail when
>communication between the supposed "other universe" and our brains
>comes into play.  At that juncture, some mechanism to cross the
>threshold between the two universes and activate neurons of the brain
>requires a physical action which would fall under the perview of
>physics.

Hell I'd just consider your claim proven if you could show me some
equal and opposite ideas out there in space. If not perhaps you could
get back to work on the physics of ideas before proclaiming a solution
to the problem in terms of "not" which you have no idea how to prove.

~v~~

0
lesterDELzick
1/10/2006 1:23:27 AM
"YadaYada" <yadayada___@excite.com> wrote:
> Curt Welch wrote:
> > > Then why are you having this conversation?
> > We do it for many reasons.  But for the most part, we are just
> > attempting to reduce the differences in our private languages.
>
> How can you and I have such wide disparity in understanding the world
> if I have a reasonably satisfactory understanding of science?

Because we have been exposed to different things in our lives, and
because we were born different (aka, nature and nuture).

> Without cultural training, what means did I use to come to an
> understanding of
> chemical bonds that I have never experienced?

Of course we all learn a huge amount of information from our culture.

But the culture doesn't exist as some holy deity separate from the people
that make it up.  I learned some things from my Dad, some from my Mom, some
from my sister, some from various teachers in school, some from the people
that wrote the books I've read even thought they died 200 years ago.  Some
from people I talk to on Usenet.  It's all stuff the culture taught me, but
in all cases, it was exposure to other people each with their own personal
opinions which I learned from.

The culture we talk about is just a large group of individual people
trying to find a common and useful set of behaviors to share to help us all
meet our needs.  But in the end, there is no common set of behaviors, there
are only a few billion people each with their own unique behaviors defined
by their own unique bodies.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/10/2006 2:29:30 AM
lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
> comp.ai.philosophy wrote:
>
> >lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> >> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
> >> comp.ai.philosophy wrote:
> >>
> >> >"YadaYada" <yadayada___@excite.com> wrote:
> >> >> Curt Welch wrote:
> >> >> > > You have to create an assocation between the langauge symbol,
> >> >> > > and the aspect of the universe you are trying to describe with
> >> >> > > the language symbol. . . . It's got to all be grounded to
> >> >> > > physcial reality.
> >> >>
> >> >> This analysis has a crucial step missing. Not one of us is capable
> >> >> of making such connections between our internal a priori concepts
> >> >> and the unknowable empirical reality that is out there. Given our
> >> >> limited sensory and rational capabilities it is impossible, even in
> >> >> theory.
> >> >>
> >> >> > > I don't undestand your point.  What is this idea of your's that
> >> >> > > there is something "unknownable" that is somehow important for
> >> >> > > us to know? If something is unknownable, then it will never
> >> >> > > affect us and we will never know about and we will never sense
> >> >> > > it and we will never talk about and we will never define any of
> >> >> > > our words using it.  So why are you trying to talk about
> >> >> > > something that doesn't exist to us?  And how is that going to
> >> >> > > prevent us from grounding the meaning of our words?
> >> >>
> >> >> By "we" do you mean an individual or culture?
> >> >
> >> >I mean you, me, and anyone else that we want to include.
> >> >
> >> >> This distinction is
> >> >> crucial in the connection between language, which is a public, a
> >> >> priori set of words and definitions, and concepts which are private
> >> >> instances of public language.
> >> >
> >> >That's just nonsense to me.  There is no public languge.  There's
> >> >nothing to use it.  All we have is a few billion private languages.
> >> >You can't find one single word written in a public language.  It was
> >> >all created by the act of some individual using their own idea
> >> >private langauge.
> >>
> >> Then why are you having this conversation?
> >
> >We do it for many reasons.  But for the most part, we are just
> >attempting to reduce the differences in our private languages.
>
> Then any reduction in differences in your private languages makes it a
> public language to that extent.

Sure.  The only way to define public language is in terms of shared
behaviors.  The larger the set of people which share a similar behavior the
more public the behavior is said to be - whether it's langauge behavior or
any other type of social behavior.  That's what we mean when we talk about
a public language.

But the meaning of this public language itself doesn't exist.  Only the
concept of the language exists.  This is because the true and complete
meaning of my langauge is defined by how my body and brain works.  This
can't be fully documented in words of any langauge.  It can only be
documented by the existence of my body.

We can document the total sum of all our private languages with the
existence of all our bodies as a large group.  But there's no way to
accurately define or document the common sub-set of behavior all those
bodies produce which we try to call our shared public langauge.  How close
do the behaviors have to be before we call them part of the the shared
public langauge?  When we have two different, but very similar behaviors,
how do we define what should be the exact behavior of our public langauge?
It's all very arbitrary.

The concept of a shared public langauge, or the concept of a langauge
defined by the culture is nothing but a pink elephant.  It doesn't and
can't exist.  Only the idea and language used to define the idea exists.
Yet, it's still very useful to all of us to talk as if it did exist - so we
talk that way.

If anything, it's an unrechable goal that is useful for us to try and get
as close to as we can.  But to pretend it actually exists as anything but a
dream, is to mis-understand how the world works.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/10/2006 2:52:38 AM
Curt Welch wrote:
> lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> > On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
> > comp.ai.philosophy wrote:
> >
> > >lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> > >> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
> > >> comp.ai.philosophy wrote:
> > >>
> > >> >"YadaYada" <yadayada___@excite.com> wrote:
> > >> >> Curt Welch wrote:
> > >> >> > > You have to create an assocation between the langauge symbol,
> > >> >> > > and the aspect of the universe you are trying to describe with
> > >> >> > > the language symbol. . . . It's got to all be grounded to
> > >> >> > > physcial reality.
> > >> >>
> > >> >> This analysis has a crucial step missing. Not one of us is capable
> > >> >> of making such connections between our internal a priori concepts
> > >> >> and the unknowable empirical reality that is out there. Given our
> > >> >> limited sensory and rational capabilities it is impossible, even in
> > >> >> theory.
> > >> >>
> > >> >> > > I don't undestand your point.  What is this idea of your's that
> > >> >> > > there is something "unknownable" that is somehow important for
> > >> >> > > us to know? If something is unknownable, then it will never
> > >> >> > > affect us and we will never know about and we will never sense
> > >> >> > > it and we will never talk about and we will never define any of
> > >> >> > > our words using it.  So why are you trying to talk about
> > >> >> > > something that doesn't exist to us?  And how is that going to
> > >> >> > > prevent us from grounding the meaning of our words?
> > >> >>
> > >> >> By "we" do you mean an individual or culture?
> > >> >
> > >> >I mean you, me, and anyone else that we want to include.
> > >> >
> > >> >> This distinction is
> > >> >> crucial in the connection between language, which is a public, a
> > >> >> priori set of words and definitions, and concepts which are private
> > >> >> instances of public language.
> > >> >
> > >> >That's just nonsense to me.  There is no public languge.  There's
> > >> >nothing to use it.  All we have is a few billion private languages.
> > >> >You can't find one single word written in a public language.  It was
> > >> >all created by the act of some individual using their own idea
> > >> >private langauge.
> > >>
> > >> Then why are you having this conversation?
> > >
> > >We do it for many reasons.  But for the most part, we are just
> > >attempting to reduce the differences in our private languages.
> >
> > Then any reduction in differences in your private languages makes it a
> > public language to that extent.
>
> Sure.  The only way to define public language is in terms of shared
> behaviors.  The larger the set of people which share a similar behavior the
> more public the behavior is said to be - whether it's langauge behavior or
> any other type of social behavior.  That's what we mean when we talk about
> a public language.
>
> But the meaning of this public language itself doesn't exist.  Only the
> concept of the language exists.  This is because the true and complete
> meaning of my langauge is defined by how my body and brain works.  This
> can't be fully documented in words of any langauge.  It can only be
> documented by the existence of my body.
>
> We can document the total sum of all our private languages with the
> existence of all our bodies as a large group.  But there's no way to
> accurately define or document the common sub-set of behavior all those
> bodies produce which we try to call our shared public langauge.  How close
> do the behaviors have to be before we call them part of the the shared
> public langauge?  When we have two different, but very similar behaviors,
> how do we define what should be the exact behavior of our public langauge?
> It's all very arbitrary.
>
> The concept of a shared public langauge, or the concept of a langauge
> defined by the culture is nothing but a pink elephant.  It doesn't and
> can't exist.  Only the idea and language used to define the idea exists.
> Yet, it's still very useful to all of us to talk as if it did exist - so we
> talk that way.
>
> If anything, it's an unrechable goal that is useful for us to try and get
> as close to as we can.  But to pretend it actually exists as anything but a
> dream, is to mis-understand how the world works.
>
> --
> Curt Welch                                            http://CurtWelch.Com/
> curt@kcwc.com                                        http://NewsReader.Com/

I disagree with you in regards to private and public languages.
While I do not use these terms I can understand the differentiation
that these terms reflect.
IMO the private language is a nonverbal one and it is the one you are
mostly  talking about.
OTOH this private language becomes public when it starts using verbal
language as it has to employ concepts created by society and  culture.
The private language has all the characteristics you describe but the
public one is very clear and logical.
The main reason we have arguments is not because of the linguistics,
which  despite Wolf's suggestion has nothing to do with this post,
but because the concepts we use do not have the same meaning.
I guess I could say that my point in all this thread is the translation
from private to public language, thru the creation of concepts.
Jan 10, 10.55am
BTW I still have problems having my replies posted. Anybody else has
this problem?  
JP

0
Just
1/10/2006 3:54:07 PM
On 10 Jan 2006 02:52:38 GMT, curt@kcwc.com (Curt Welch) in
comp.ai.philosophy wrote:

>lesterDELzick@worldnet.att.net (Lester Zick) wrote:
>> On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
>> comp.ai.philosophy wrote:
>>
>> >lesterDELzick@worldnet.att.net (Lester Zick) wrote:
>> >> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
>> >> comp.ai.philosophy wrote:
>> >>
>> >> >"YadaYada" <yadayada___@excite.com> wrote:
>> >> >> Curt Welch wrote:
>> >> >> > > You have to create an assocation between the langauge symbol,
>> >> >> > > and the aspect of the universe you are trying to describe with
>> >> >> > > the language symbol. . . . It's got to all be grounded to
>> >> >> > > physcial reality.
>> >> >>
>> >> >> This analysis has a crucial step missing. Not one of us is capable
>> >> >> of making such connections between our internal a priori concepts
>> >> >> and the unknowable empirical reality that is out there. Given our
>> >> >> limited sensory and rational capabilities it is impossible, even in
>> >> >> theory.
>> >> >>
>> >> >> > > I don't undestand your point.  What is this idea of your's that
>> >> >> > > there is something "unknownable" that is somehow important for
>> >> >> > > us to know? If something is unknownable, then it will never
>> >> >> > > affect us and we will never know about and we will never sense
>> >> >> > > it and we will never talk about and we will never define any of
>> >> >> > > our words using it.  So why are you trying to talk about
>> >> >> > > something that doesn't exist to us?  And how is that going to
>> >> >> > > prevent us from grounding the meaning of our words?
>> >> >>
>> >> >> By "we" do you mean an individual or culture?
>> >> >
>> >> >I mean you, me, and anyone else that we want to include.
>> >> >
>> >> >> This distinction is
>> >> >> crucial in the connection between language, which is a public, a
>> >> >> priori set of words and definitions, and concepts which are private
>> >> >> instances of public language.
>> >> >
>> >> >That's just nonsense to me.  There is no public languge.  There's
>> >> >nothing to use it.  All we have is a few billion private languages.
>> >> >You can't find one single word written in a public language.  It was
>> >> >all created by the act of some individual using their own idea
>> >> >private langauge.
>> >>
>> >> Then why are you having this conversation?
>> >
>> >We do it for many reasons.  But for the most part, we are just
>> >attempting to reduce the differences in our private languages.
>>
>> Then any reduction in differences in your private languages makes it a
>> public language to that extent.
>
>Sure.  The only way to define public language is in terms of shared
>behaviors.  The larger the set of people which share a similar behavior the
>more public the behavior is said to be - whether it's langauge behavior or
>any other type of social behavior.  That's what we mean when we talk about
>a public language.

>But the meaning of this public language itself doesn't exist. 

Yeah, look, Curt, I really don't think you have acumen to be a decent
behaviorist. Glen would have long since seen this self contradiction
coming and retreated into double talk.First you say there is no public
language then you agree there is public language, now you say the
meaning of public language doesn't exist.

>                                                                                                Only the
>concept of the language exists.  This is because the true and complete
>meaning of my langauge is defined by how my body and brain works.  This
>can't be fully documented in words of any langauge.  It can only be
>documented by the existence of my body.

So now you're the definition of public language? You're an idiot,
Curt. You're not the definition of anything except your own private
language which seems to be pretty much nonsense at this point.

It isn't the words themselves taken in isolation which constitutes
public language but the significance of words in relation to other
words. Words considered in isolation obviously have considerable
private significance but such considerations are easily factored from
word combinations precisely to reduce and eliminate subjective
components.

>We can document the total sum of all our private languages with the
>existence of all our bodies as a large group.  But there's no way to
>accurately define or document the common sub-set of behavior all those
>bodies produce which we try to call our shared public langauge.  How close
>do the behaviors have to be before we call them part of the the shared
>public langauge?  When we have two different, but very similar behaviors,
>how do we define what should be the exact behavior of our public langauge?
>It's all very arbitrary.

What's very arbitrary is your view of what part of this nonsense
anyone else will accept. I think the only thing left to analyze is
whether and to what extent double talk can be a public language.

>The concept of a shared public langauge, or the concept of a langauge
>defined by the culture is nothing but a pink elephant.  It doesn't and
>can't exist.  Only the idea and language used to define the idea exists.
>Yet, it's still very useful to all of us to talk as if it did exist - so we
>talk that way.

So now you're saying there is no public language but we behave as if
there were? Jesus, just how stupid do you think we are and how much
more of this shit do you seriously expect anyone to swallow?

>If anything, it's an unrechable goal that is useful for us to try and get
>as close to as we can.  But to pretend it actually exists as anything but a
>dream, is to mis-understand how the world works.

Yeah well one of us certainly misunderstands how the world works.

~v~~

0
lesterDELzick
1/10/2006 4:06:48 PM
Lester Zick wrote:

> Hell I'd just consider your claim proven if you could show me some
> equal and opposite ideas out there in space.

I guess I'm grappling to understand your complaint.  Let's start with
your statement above.  Your challenge is to "show ... some equal and
opposite ideas out there in space."  I can interpret that multiple
ways: ideas that are equivalent, ideas that are opposite, and ideas
that are contradictions (simultaneously equal and opposite).  I will
need further clarification of your challenge from you, but let me try
to address the three which I have discerned from your statement:

a.) equivalent ideas - the idea of "planet" is a categorical idea.  In
terms of categories, Earth and Mars would both fall into the category
of planets.  Thus, Earth and Mars can be equated based upon category -
they are both planets.  In specifics, they differ, but in generalities,
they are the same.  (Knowledge is scalar with Parmenides at one end and
Heraclitus at the other - both can be shown as true given the
"general-specific" scale.)

b.) opposing ideas - a planet is typically defined by the edge of its
atmosphere.  Thus, the edge of Earth is defined by the edge of outer
space.  Space is not Earth.  A true opposition, however, would require
a negative.  The belief in opposing charges (such as positrons and
electrons which annihilate each other to create 2 photons) provides an
instance of true opposition.  Many other oppositions exist including
opposite spin on orbits or opposite planetary spins.  Time appears to
have a direction.  Though it proceeds ever forward, in our minds, we
can seemingly "go backwards".  Yet, this activity is usually a jump to
a preceding era and forward action (instant replay).  Playing a tape
backwards is closer to opposition, but as we play it, the tape device
is going forward again.  (I need your assistance in fine-tuning my
response.  Are you looking for opposition, negation, or something
else?)

c.) contradictions - some well known contradictions are "lens
galaxies".  Some distant galaxies appear to have other galaxies
surrounding them in halos.  The belief is that the central galaxy acts
like a lens with light from more distant galaxies bending around it to
produce the halo effect.  Thus, the most distant galaxies appear (in
the halo) where they actually do not exist.  The growth of astronomy
has been greatly due to contradictions found between beliefs about the
universe and actual observations (a la Galileo).  In the latter case,
the idea in the mind is a "reference signal" (Perceptual Control Theory
- William T. Powers) which does not match feedback from observations.
Given the contradiction, the original idea is thrown out (geocentricity
in the case of Galileo).  Nevertheless, prior to the observations, the
idea contradicts empirical facts yet to be seen.  However, the
contradictory idea is rooted in neural structures of the brain which
have a real and physical nature.  Thus, a physical representation of
the universe in the brain contradicts the true universe.

Creating unreal ideas is a central ability of the imagination - to
conjure ideas which do not need to be consistent with the "real world".
 This allows for forethought, planning, and creativity.  As Freud said,
"thinking is rehearsal work."  The brain cannot be restrained by past
reality as it cobbles together new ideas which may or may not prove
useful or viable.

I suspect I am still not addressing your issue of "equal and opposite"
sufficiently.  However, I hope my attempts show that I am truly trying
to understand your position and that my viewpoints are in earnest with
interest and respect.

Could you provide a specific "idea" which emulates your position?

0
JAK
1/10/2006 5:54:26 PM
>Sure.  The only way to define public language is in terms of shared
>behaviors.

Agreed.  Definitions and the processing of language must have common
procedures (behaviors) shared by all using the public language.

>The larger the set of people which share a similar behavior the
>more public the behavior is said to be - whether it's langauge behavior or
>any other type of social behavior.  That's what we mean when we talk about
>a public language.

Agreed.  A family refering to discarded corn cobs as "bones", is not
public enough to be considered part of the public language.  If others
incorporate the new term and it becomes commonplace, then it becomes
part of the public language.

>But the meaning of this public language itself doesn't exist.  Only the
>concept of the language exists.  This is because the true and complete
>meaning of my langauge is defined by how my body and brain works.  This
>can't be fully documented in words of any langauge.  It can only be
>documented by the existence of my body.

Very True, but not well conveyed.  I may not do much better, but I'll
try.

As we write cursively, our handwriting has a personal and identifiable
style.  This is well known and easily shown.  Yet, it also shows that
the public language does not have a cookie-cutter portrayal.  And some
handwriting is almost unintelligible.  Thus, handwriting is defined by
how the body and brain works and is documented by the body of the
writer.

Another good example is color.  Some folks are color-blind.  For you to
talk about red and green, a color-blind person may only know the
difference based upon stoplights - red is on top and green is below.
Worse for this person, it is an area of personal weakness and may bring
increased stress.  For someone else with normal vision, red and green
conjures up Christmas and holly as well as happy times.  Both people
will use the same terms, red and green, but the two individuals will
have completely different perspectives toward it.  And the internal
representation of red and green within each of these two people will be
significantly different.

This thread is a further example.  Curt's thoughts are close enough to
mine to make me accept them.  On the other hand they differ
significantly with Lester's beliefs.  As a result, we can infer that
Curt's and my internal (brain) representation of these ideas are closer
than Curt's and Lester's.  Yet, Curt, Lester, and I read the exact same
words (I hope).  Thus, the external representation is identical, but
the internal representations within the three of us are unique to each
individual.

0
JAK
1/10/2006 6:39:16 PM
On 10 Jan 2006 10:39:16 -0800, "JAK" <jak@theoryofmind.org> in
comp.ai.philosophy wrote:

>>Sure.  The only way to define public language is in terms of shared
>>behaviors.
>
>Agreed.  Definitions and the processing of language must have common
>procedures (behaviors) shared by all using the public language.
>
>>The larger the set of people which share a similar behavior the
>>more public the behavior is said to be - whether it's langauge behavior or
>>any other type of social behavior.  That's what we mean when we talk about
>>a public language.
>
>Agreed.  A family refering to discarded corn cobs as "bones", is not
>public enough to be considered part of the public language.  If others
>incorporate the new term and it becomes commonplace, then it becomes
>part of the public language.
>
>>But the meaning of this public language itself doesn't exist.  Only the
>>concept of the language exists.  This is because the true and complete
>>meaning of my langauge is defined by how my body and brain works.  This
>>can't be fully documented in words of any langauge.  It can only be
>>documented by the existence of my body.
>
>Very True, but not well conveyed.  I may not do much better, but I'll
>try.
>
>As we write cursively, our handwriting has a personal and identifiable
>style.  This is well known and easily shown.  Yet, it also shows that
>the public language does not have a cookie-cutter portrayal.  And some
>handwriting is almost unintelligible.  Thus, handwriting is defined by
>how the body and brain works and is documented by the body of the
>writer.
>
>Another good example is color.  Some folks are color-blind.  For you to
>talk about red and green, a color-blind person may only know the
>difference based upon stoplights - red is on top and green is below.
>Worse for this person, it is an area of personal weakness and may bring
>increased stress.  For someone else with normal vision, red and green
>conjures up Christmas and holly as well as happy times.  Both people
>will use the same terms, red and green, but the two individuals will
>have completely different perspectives toward it.  And the internal
>representation of red and green within each of these two people will be
>significantly different.
>
>This thread is a further example.  Curt's thoughts are close enough to
>mine to make me accept them.  On the other hand they differ
>significantly with Lester's beliefs. 

Yeah I have a problem with your describing my comments as beliefs.
They are nothing of the kind. Public objectivity in language is well
documented and easily explained. Curt is just engaging in revisionist
doubletalk in support of totally nonsensical naturalized epistemology
favored by materialists and behaviorists. If you want to agree with
Curt perhaps you could get down to the epistemological nitty-gritty.

>                                                      As a result, we can infer that
>Curt's and my internal (brain) representation of these ideas are closer
>than Curt's and Lester's.  Yet, Curt, Lester, and I read the exact same
>words (I hope).  Thus, the external representation is identical, but
>the internal representations within the three of us are unique to each
>individual.

I can't imagine why you think Curt's nonsense represents a serious
issue for science. Curt is not a critical thinker. I can't tell if you
are or not but I'm beginning to wonder. Curt just pulls whatever he
wants out of his ass to justify his extraordinarily idiotic claims in
support of naturalized epistemology and behaviorism. I can't comment
on his programming skills but his ideas on science are beyond stupid.

~v~~

0
lesterDELzick
1/10/2006 7:20:44 PM
In reading all of the replies, I am troubled by the general direction.
In our search for expanded communication - passing more complex
information - we may be missing a key.  As Curt Welch noted, the amount
of information available from a tiny water droplet is immense.
Extracting and communicating that information would be intensive.  Even
more, the information may be useless.  Presently, the information we do
get from our 5 senses is partially thrown out by the mind as not
useful.  The feeling of your shoes on your feet or the cuff of your
shirt is ignored.  In psychology, this is know as habituation.  Per
Gestalt psychology, we visually pick out "figure" and ignore "ground"
information.  If we throw out half or more of what we perceive, what is
essential?  How do we transfer this crucial bit of editing to an AI
system?

Perhaps we should take a hint from evolution.  With rats, they are
known to explore their environment - avoiding a direct route to food.
Novelty is known to enrich rats brains and help us remember things.
"Variety is the spice of life", as they say.  If we are to capture some
information and pass it as communication, perhaps we should focus on
passing that which is different, that which is new.  If someone already
knows about an earthquake, why tell them again?  If information is to
be useful, it must be new to the receiving system.  Pass a "headline".
If it is new to the receiving system, it will request more detail.

In addition to novelty, efficiency seems to be another important
factor.  Natural selection, with its cycles of feast and famine as well
as predation, demands efficiency.  It is quickly seen in the sleekness
of fish and birds.  The sluggards survive, but only through efficient
systems.  (For instance, bears hibernate.)  We can see our own demand
for efficiency by virtue of every budget in place at corporations,
universities, governments, and our personal bank accounts.  If we pass
infomation within an AI system, not only should it be useful
(different) to the receiving system, but only the essential information
should be passed, and passed quickly.  Humans perform this efficient
transfer by using verbal terms which condense complex information
(jargon) in order to convey it quickly and efficiently.

I suggest that rather than focusing on the quantity of infomation
conveyed, the quality (primarily novelty and efficiency) should be of
greater concern.

0
JAK
1/10/2006 7:40:58 PM
On 10 Jan 2006 09:54:26 -0800, "JAK" <jak@theoryofmind.org> in
comp.ai.philosophy wrote:

>Lester Zick wrote:
>
>> Hell I'd just consider your claim proven if you could show me some
>> equal and opposite ideas out there in space.
>
>I guess I'm grappling to understand your complaint.  Let's start with
>your statement above. 

Let's not. My comment was facetious and not intended to be taken
seriously unless you really intend to maintain ideas are physical.

>                                   Your challenge is to "show ... some equal and
>opposite ideas out there in space."  I can interpret that multiple
>ways: ideas that are equivalent, ideas that are opposite, and ideas
>that are contradictions (simultaneously equal and opposite).  I will
>need further clarification of your challenge from you, but let me try
>to address the three which I have discerned from your statement:

No interpretation required. If ideas and other cognitive effects obey
the laws of physics as your comment to Curt suggested I would like to
see you show experimentally definitive evidence that ideas obey f=ma
and equal and opposite reactions which material interactions obey. If
not I suggest we move on to other issues and stop pretending ideas are
material artifacts.

>a.) equivalent ideas - the idea of "planet" is a categorical idea.  In
>terms of categories, Earth and Mars would both fall into the category
>of planets.  Thus, Earth and Mars can be equated based upon category -
>they are both planets.  In specifics, they differ, but in generalities,
>they are the same.  (Knowledge is scalar with Parmenides at one end and
>Heraclitus at the other - both can be shown as true given the
>"general-specific" scale.)
>
>b.) opposing ideas - a planet is typically defined by the edge of its
>atmosphere.  Thus, the edge of Earth is defined by the edge of outer
>space.  Space is not Earth.  A true opposition, however, would require
>a negative.  The belief in opposing charges (such as positrons and
>electrons which annihilate each other to create 2 photons) provides an
>instance of true opposition.  Many other oppositions exist including
>opposite spin on orbits or opposite planetary spins.  Time appears to
>have a direction.  Though it proceeds ever forward, in our minds, we
>can seemingly "go backwards".  Yet, this activity is usually a jump to
>a preceding era and forward action (instant replay).  Playing a tape
>backwards is closer to opposition, but as we play it, the tape device
>is going forward again.  (I need your assistance in fine-tuning my
>response.  Are you looking for opposition, negation, or something
>else?)
>
>c.) contradictions - some well known contradictions are "lens
>galaxies".  Some distant galaxies appear to have other galaxies
>surrounding them in halos.  The belief is that the central galaxy acts
>like a lens with light from more distant galaxies bending around it to
>produce the halo effect.  Thus, the most distant galaxies appear (in
>the halo) where they actually do not exist.  The growth of astronomy
>has been greatly due to contradictions found between beliefs about the
>universe and actual observations (a la Galileo).  In the latter case,
>the idea in the mind is a "reference signal" (Perceptual Control Theory
>- William T. Powers) which does not match feedback from observations.
>Given the contradiction, the original idea is thrown out (geocentricity
>in the case of Galileo).  Nevertheless, prior to the observations, the
>idea contradicts empirical facts yet to be seen.  However, the
>contradictory idea is rooted in neural structures of the brain which
>have a real and physical nature.  Thus, a physical representation of
>the universe in the brain contradicts the true universe.
>
>Creating unreal ideas is a central ability of the imagination - to
>conjure ideas which do not need to be consistent with the "real world".
> This allows for forethought, planning, and creativity.  As Freud said,
>"thinking is rehearsal work."  The brain cannot be restrained by past
>reality as it cobbles together new ideas which may or may not prove
>useful or viable.
>
>I suspect I am still not addressing your issue of "equal and opposite"
>sufficiently.  However, I hope my attempts show that I am truly trying
>to understand your position and that my viewpoints are in earnest with
>interest and respect.

Forget it. If you have to engage in all this circumlocution just to
evade the implications of your reply to Curt my point has been made.

>Could you provide a specific "idea" which emulates your position?

Look. My issue with what you had to say regarding differences is not
so much that it's wrong but that it's not proven true by anything you
say nor is it orginal since I have been saying and demonstrating what
you are trying to say all over the usenet for the last several years
and in various books and essays for two decades prior to that. Maybe
imitation is the sincerest form of flattery but I'd at least like you
get the story straight if you're going to promote the idea. If you'd
like documentation of what I've had to say and proof for my claims on
the usenet I'll be happy to cite thread, chapter, and verse over the
last year. But I deal in mechanics and science not philosophy.

~v~~

0
lesterDELzick
1/10/2006 7:42:46 PM
"Just Playing" <gms2004@lycos.com> wrote:
> Curt Welch wrote:
> > lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> > > On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
> > > comp.ai.philosophy wrote:
> > >
> > > >lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> > > >> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
> > > >> comp.ai.philosophy wrote:
> > > >>
> > > >> >"YadaYada" <yadayada___@excite.com> wrote:
> > > >> >> Curt Welch wrote:
> > > >> >> > > You have to create an assocation between the langauge
> > > >> >> > > symbol, and the aspect of the universe you are trying to
> > > >> >> > > describe with the language symbol. . . . It's got to all be
> > > >> >> > > grounded to physcial reality.
> > > >> >>
> > > >> >> This analysis has a crucial step missing. Not one of us is
> > > >> >> capable of making such connections between our internal a
> > > >> >> priori concepts and the unknowable empirical reality that is
> > > >> >> out there. Given our limited sensory and rational capabilities
> > > >> >> it is impossible, even in theory.
> > > >> >>
> > > >> >> > > I don't undestand your point.  What is this idea of your's
> > > >> >> > > that there is something "unknownable" that is somehow
> > > >> >> > > important for us to know? If something is unknownable, then
> > > >> >> > > it will never affect us and we will never know about and we
> > > >> >> > > will never sense it and we will never talk about and we
> > > >> >> > > will never define any of our words using it.  So why are
> > > >> >> > > you trying to talk about something that doesn't exist to
> > > >> >> > > us?  And how is that going to prevent us from grounding the
> > > >> >> > > meaning of our words?
> > > >> >>
> > > >> >> By "we" do you mean an individual or culture?
> > > >> >
> > > >> >I mean you, me, and anyone else that we want to include.
> > > >> >
> > > >> >> This distinction is
> > > >> >> crucial in the connection between language, which is a public,
> > > >> >> a priori set of words and definitions, and concepts which are
> > > >> >> private instances of public language.
> > > >> >
> > > >> >That's just nonsense to me.  There is no public languge.  There's
> > > >> >nothing to use it.  All we have is a few billion private
> > > >> >languages. You can't find one single word written in a public
> > > >> >language.  It was all created by the act of some individual using
> > > >> >their own idea private langauge.
> > > >>
> > > >> Then why are you having this conversation?
> > > >
> > > >We do it for many reasons.  But for the most part, we are just
> > > >attempting to reduce the differences in our private languages.
> > >
> > > Then any reduction in differences in your private languages makes it
> > > a public language to that extent.
> >
> > Sure.  The only way to define public language is in terms of shared
> > behaviors.  The larger the set of people which share a similar behavior
> > the more public the behavior is said to be - whether it's langauge
> > behavior or any other type of social behavior.  That's what we mean
> > when we talk about a public language.
> >
> > But the meaning of this public language itself doesn't exist.  Only the
> > concept of the language exists.  This is because the true and complete
> > meaning of my langauge is defined by how my body and brain works.  This
> > can't be fully documented in words of any langauge.  It can only be
> > documented by the existence of my body.
> >
> > We can document the total sum of all our private languages with the
> > existence of all our bodies as a large group.  But there's no way to
> > accurately define or document the common sub-set of behavior all those
> > bodies produce which we try to call our shared public langauge.  How
> > close do the behaviors have to be before we call them part of the the
> > shared public langauge?  When we have two different, but very similar
> > behaviors, how do we define what should be the exact behavior of our
> > public langauge? It's all very arbitrary.
> >
> > The concept of a shared public langauge, or the concept of a langauge
> > defined by the culture is nothing but a pink elephant.  It doesn't and
> > can't exist.  Only the idea and language used to define the idea
> > exists. Yet, it's still very useful to all of us to talk as if it did
> > exist - so we talk that way.
> >
> > If anything, it's an unrechable goal that is useful for us to try and
> > get as close to as we can.  But to pretend it actually exists as
> > anything but a dream, is to mis-understand how the world works.
> >
> > --
> > Curt Welch
> > http://CurtWelch.Com/ curt@kcwc.com
> > http://NewsReader.Com/
>
> I disagree with you in regards to private and public languages.
> While I do not use these terms I can understand the differentiation
> that these terms reflect.
> IMO the private language is a nonverbal one and it is the one you are
> mostly  talking about.
> OTOH this private language becomes public when it starts using verbal
> language as it has to employ concepts created by society and  culture.
> The private language has all the characteristics you describe but the
> public one is very clear and logical.

Have you seen the way I write?  What's clear and logical about that! :)

We all act differently in public than we do in private.  Our behavior in
all regards is a function of our context created by the environment we are
in.  We learn from experience how we need to act in each context to get
what we want.

Now, it's natural to talk about priviate behavior and public behavior as
being different and it's natrual to limit that idea to private langauge and
public langauge.  And that seems to be how you want to use the terms.  And
that's a fine way to use them.

However, it has nothing to do with what I was trying to talk about.  I was
talking about a "public langauge" as being the langauge of our culture.  A
language created and used by the culture.  That's the langauge which
doesn't actually exist even though we talk as if it does.

I was using "priviate langauge" to talk about the full set of language
behaviors we use as an individual, whether we are using it in our head as
private thoughts, or whether we are using it on a TV broadcast to the
world.

> The main reason we have arguments is not because of the linguistics,
> which  despite Wolf's suggestion has nothing to do with this post,
> but because the concepts we use do not have the same meaning.

Sure, of course.  We also have arguments just because we want different
things.

> I guess I could say that my point in all this thread is the translation
> from private to public language, thru the creation of concepts.
> Jan 10, 10.55am
> BTW I still have problems having my replies posted. Anybody else has
> this problem?

No, but I don't waste my time trying to post through google groups.

I've seen a lot of duplicate posts from people that post from google
lately.

> JP

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/10/2006 8:23:33 PM
lesterDELzick@worldnet.att.net (Lester Zick) wrote:

> So now you're saying there is no public language but we behave as if
> there were? Jesus, just how stupid do you think we are ...

I expect you to be smart enough to read past any confusion I create by my
lack of skills in communicating.  That's why I write so much at times - to
make sure I have supplied you with enough material to understand the
concept behind the words even if my choice of words does a poor job of
communicating it.  If you are not smart enough to extract the clear and
simple idea I was trying to communicate in the post I wrote, you should
just give up trying.  It's beyond your skill level.  You are just wasting
your time.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/10/2006 8:58:56 PM
On 10 Jan 2006 20:58:56 GMT, curt@kcwc.com (Curt Welch) in
comp.ai.philosophy wrote:

>lesterDELzick@worldnet.att.net (Lester Zick) wrote:
>
>> So now you're saying there is no public language but we behave as if
>> there were? Jesus, just how stupid do you think we are ...
>
>I expect you to be smart enough to read past any confusion I create by my
>lack of skills in communicating.  That's why I write so much at times - to
>make sure I have supplied you with enough material to understand the
>concept behind the words even if my choice of words does a poor job of
>communicating it.  If you are not smart enough to extract the clear and
>simple idea I was trying to communicate in the post I wrote, you should
>just give up trying.  It's beyond your skill level.  You are just wasting
>your time.

Curt, you're not even smart enough to do a decent job of pretending
to be stupid. You're the one wasting my time not me. You're just smart
enough to make believe double talk is a matter of "poor communication"
because you're too stupid to realize people aren't quite the fools you
take them to be.You write so much to conceal the fact you haven't said
anything substantial. Just remember Lincoln's dictum. Glen is slightly
cleverer.You are a first class idiot and a true credit to behaviorism.

~v~~

0
lesterDELzick
1/10/2006 10:14:01 PM
"JAK" <jak@theoryofmind.org> wrote:
> In reading all of the replies, I am troubled by the general direction.
> In our search for expanded communication - passing more complex
> information - we may be missing a key.

The missing key is that you can't understand the meaning of infomration
until you solve AI.  So much our our understanding is based on a foundation
that is unknown to most of us. Words like "data" and "information" are all
a part of this great unknown.

> As Curt Welch noted, the amount
> of information available from a tiny water droplet is immense.
> Extracting and communicating that information would be intensive.  Even
> more, the information may be useless.  Presently, the information we do
> get from our 5 senses is partially thrown out by the mind as not
> useful.  The feeling of your shoes on your feet or the cuff of your
> shirt is ignored.  In psychology, this is know as habituation.  Per
> Gestalt psychology, we visually pick out "figure" and ignore "ground"
> information.  If we throw out half or more of what we perceive, what is
> essential?  How do we transfer this crucial bit of editing to an AI
> system?
>
> Perhaps we should take a hint from evolution.  With rats, they are
> known to explore their environment - avoiding a direct route to food.
> Novelty is known to enrich rats brains and help us remember things.
> "Variety is the spice of life", as they say.  If we are to capture some
> information and pass it as communication, perhaps we should focus on
> passing that which is different, that which is new.  If someone already
> knows about an earthquake, why tell them again?  If information is to
> be useful, it must be new to the receiving system.  Pass a "headline".
> If it is new to the receiving system, it will request more detail.

Yes, information which is already known is not very useful to us.  But the
newness of the information has nothing to do with.  I can write 1000 words
about the dirt spots on my floor and all of it would be totaly new to you -
and totaly useless.

> In addition to novelty, efficiency seems to be another important
> factor.  Natural selection, with its cycles of feast and famine as well
> as predation, demands efficiency.  It is quickly seen in the sleekness
> of fish and birds.  The sluggards survive, but only through efficient
> systems.  (For instance, bears hibernate.)  We can see our own demand
> for efficiency by virtue of every budget in place at corporations,
> universities, governments, and our personal bank accounts.  If we pass
> infomation within an AI system, not only should it be useful
> (different) to the receiving system, but only the essential information
> should be passed, and passed quickly.  Humans perform this efficient
> transfer by using verbal terms which condense complex information
> (jargon) in order to convey it quickly and efficiently.
>
> I suggest that rather than focusing on the quantity of infomation
> conveyed, the quality (primarily novelty and efficiency) should be of
> greater concern.

The key here is to understand how data is selected.  Why is some data seen
as useful information and other data as useless information - or even noise
(the term we use to describe data which is not only useless, but seems to
have no content at all).  The answer is that the data must have a purpose.
The collection or the creation of the data must have a purpose or else you
have no way to define how information is separated from noise.

Our purpose was given to us by evolution.  We are machines built to
survive.  And as such, we also have data processing hardware designed for
the purpose of the purpose of extracting useful information from the
enviroment.  And the ultimate purpose of that data is to guide us in the
flapping of our arms and legs and lips so we can survive.  The brain is a
very complex device which finds the data needed to drive the arms and legs
and ignores everything else.  If the data is useful in producing arm
motions that help us survive, we call it "information".  If it's not, we
call it noise.

You don't need to know about the dirt spots on my floor because that data
will never (most likely) help you survive.  It will never be needed to help
you make the correct arm and leg motions.  However, if I turned as crazy
one day as Lester believes I am, and I hunt down all the people in this
group and kill the ones that can't accuratly tell me about the dirt spots
on my floor, then that noise would instantly become vital information to
you.

So, if you want to solve AI, you have to understand how the brain finds
information in the noise.  There's nothing about the data itself that will
tell you if it's good or bad or useless.  That only happens when the data
is correlated to a purpose.  And this is what the reinforcement problem I
always talk about is all about.  You can't create intelligent behavior, and
you can't build a machine to extract data from noise, unless you first give
it a purpose.  And the most general way to give it it's own purpose,
instead of just building our own view of purpose into it, is by building a
reinforcement learning machine - one which has it's own dumb critic which
is signalling purpose through reward and punishment signals.  The job of
the learning machine then becomes one of using those signals as a guide to
locating the useful data in the noise, which the sensors are collecting,
and sending the useful data, to the arms and legs, while throwing the rest
out.

We spend to much time dealing with the data in the middle of the
transformation from noise to arm and leg control, that we loose track of
the fact that everything we learn, is there only to support the purpose of
moving our body parts.  That data that the brain has determined to have a
role in that, is the data it "keeps" as "useful information".

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/10/2006 11:31:01 PM
JAK wrote:

> In reading all of the replies, I am troubled by the general direction.
> In our search for expanded communication - passing more complex
> information - we may be missing a key.  As Curt Welch noted, the amount
> of information available from a tiny water droplet is immense.
> Extracting and communicating that information would be intensive.  Even
> more, the information may be useless.  Presently, the information we do
> get from our 5 senses is partially thrown out by the mind as not
> useful.  The feeling of your shoes on your feet or the cuff of your
> shirt is ignored.  In psychology, this is know as habituation.  Per
> Gestalt psychology, we visually pick out "figure" and ignore "ground"
> information.  If we throw out half or more of what we perceive, what is
> essential?  How do we transfer this crucial bit of editing to an AI
> system?
> 
> Perhaps we should take a hint from evolution.  With rats, they are
> known to explore their environment - avoiding a direct route to food.
> Novelty is known to enrich rats brains and help us remember things.
> "Variety is the spice of life", as they say.  If we are to capture some
> information and pass it as communication, perhaps we should focus on
> passing that which is different, that which is new.  If someone already
> knows about an earthquake, why tell them again?  If information is to
> be useful, it must be new to the receiving system.  Pass a "headline".
> If it is new to the receiving system, it will request more detail.
> 
> In addition to novelty, efficiency seems to be another important
> factor.  Natural selection, with its cycles of feast and famine as well
> as predation, demands efficiency.  It is quickly seen in the sleekness
> of fish and birds.  The sluggards survive, but only through efficient
> systems.  (For instance, bears hibernate.)  We can see our own demand
> for efficiency by virtue of every budget in place at corporations,
> universities, governments, and our personal bank accounts.  If we pass
> infomation within an AI system, not only should it be useful
> (different) to the receiving system, but only the essential information
> should be passed, and passed quickly.  Humans perform this efficient
> transfer by using verbal terms which condense complex information
> (jargon) in order to convey it quickly and efficiently.
> 
> I suggest that rather than focusing on the quantity of infomation
> conveyed, the quality (primarily novelty and efficiency) should be of
> greater concern.

What you are looking for is "predictive information". There are a variety of
equivalent ways to formulate it. One way that is fairly intuitive is to
consider a time series {x_i} (where x can be a vector) and take the mutual
information between x_past and x_future, or in probabilistic terms
P(x_future|x_past) -- the probability of the future of x given its past.
The most information x_past can give about x_future is just the entropy of
x_past. This is an upper limit, but it would only hold if all the
information in x_past were predictive of x_future. This cannot be the case
because the entropy of x_past is extensive (proportional to its length)
while the predictive fraction is subextensive (grows at a rate that is less
than linear). The predictive information can be finite (after some amount
of finite observation no new predictive information is gained by further
observation), logarithmic, or exponential with a fractional exponent less
than one (so greater than logarithmic, but still sublinear). These are the
bounds of "phase transitions" in complexity space, and also mark the bounds
of optimal learning algorithms (different algorithms are optimal in each
complexity regime). This gives a precise sense in which "most information
is useless". Only predictive information is useful. As to what information
is "relevant" it depends on what features of x_future you want to predict.
That too can be made precise in terms of "fitness". See, for example,
Geisler and Diehl's paper "Bayesian natural selection and the evolution of
perceptual systems", which is available on-line (I don't have a URL handy,
but the paper is easy to find).

For a motivation, derivation, and some applications of "predictive
information" see the papers:

"Predictability, Complexity, and Learning", by Bialek, Nemenman, and Tishby

"Complexity through nonextensivity", also by Bialek, Nemenman, and Tishby
(This is a much shorter version of the former, but it is more explicit in
bringing out the relationship between predictive information and Kolmogorov
complexity.)

"Fluctuation-Dissipation Theorem and Models of Learning", Ilya Nemenman
(looks at ideal learners for different regimes, and examines the question
"what is the learning algorithm of a rat?")

"Information Theory and Learning: A Physical Approach", Ilya Nemenman
(this is Ilya's PhD dissertation.)

"The Information Bottleneck Method", by Tishby, Pereira, and Bialek
(a short paper and a fundamental breakthrough in formalizing and quantifying
"meaningful" and "relevant" information.)

That should be enough to begin putting water droplets and dirt spots on the
floor in perspective. All these papers are avialable here:

http://www.princeton.edu/~wbialek/wbialek.html

and/or here:

http://www.menem.com/ilya/digital_library/full.html

-- Michael


0
Michael
1/11/2006 2:05:58 AM
Curt Welch wrote:
> "Just Playing" <gms2004@lycos.com> wrote:
> > Curt Welch wrote:
> > > lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> > > > On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
> > > > comp.ai.philosophy wrote:
> > > >
> > > > >lesterDELzick@worldnet.att.net (Lester Zick) wrote:
> > > > >> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
> > > > >> comp.ai.philosophy wrote:
> > > > >>
> > > > >> >"YadaYada" <yadayada___@excite.com> wrote:
> > > > >> >> Curt Welch wrote:
> > > > >> >> > > You have to create an assocation between the langauge
> > > > >> >> > > symbol, and the aspect of the universe you are trying to
> > > > >> >> > > describe with the language symbol. . . . It's got to all be
> > > > >> >> > > grounded to physcial reality.
> > > > >> >>
> > > > >> >> This analysis has a crucial step missing. Not one of us is
> > > > >> >> capable of making such connections between our internal a
> > > > >> >> priori concepts and the unknowable empirical reality that is
> > > > >> >> out there. Given our limited sensory and rational capabilities
> > > > >> >> it is impossible, even in theory.
> > > > >> >>
> > > > >> >> > > I don't undestand your point.  What is this idea of your's
> > > > >> >> > > that there is something "unknownable" that is somehow
> > > > >> >> > > important for us to know? If something is unknownable, then
> > > > >> >> > > it will never affect us and we will never know about and we
> > > > >> >> > > will never sense it and we will never talk about and we
> > > > >> >> > > will never define any of our words using it.  So why are
> > > > >> >> > > you trying to talk about something that doesn't exist to
> > > > >> >> > > us?  And how is that going to prevent us from grounding the
> > > > >> >> > > meaning of our words?
> > > > >> >>
> > > > >> >> By "we" do you mean an individual or culture?
> > > > >> >
> > > > >> >I mean you, me, and anyone else that we want to include.
> > > > >> >
> > > > >> >> This distinction is
> > > > >> >> crucial in the connection between language, which is a public,
> > > > >> >> a priori set of words and definitions, and concepts which are
> > > > >> >> private instances of public language.
> > > > >> >
> > > > >> >That's just nonsense to me.  There is no public languge.  There's
> > > > >> >nothing to use it.  All we have is a few billion private
> > > > >> >languages. You can't find one single word written in a public
> > > > >> >language.  It was all created by the act of some individual using
> > > > >> >their own idea private langauge.
> > > > >>
> > > > >> Then why are you having this conversation?
> > > > >
> > > > >We do it for many reasons.  But for the most part, we are just
> > > > >attempting to reduce the differences in our private languages.
> > > >
> > > > Then any reduction in differences in your private languages makes it
> > > > a public language to that extent.
> > >
> > > Sure.  The only way to define public language is in terms of shared
> > > behaviors.  The larger the set of people which share a similar behavior
> > > the more public the behavior is said to be - whether it's langauge
> > > behavior or any other type of social behavior.  That's what we mean
> > > when we talk about a public language.
> > >
> > > But the meaning of this public language itself doesn't exist.  Only the
> > > concept of the language exists.  This is because the true and complete
> > > meaning of my langauge is defined by how my body and brain works.  This
> > > can't be fully documented in words of any langauge.  It can only be
> > > documented by the existence of my body.
> > >
> > > We can document the total sum of all our private languages with the
> > > existence of all our bodies as a large group.  But there's no way to
> > > accurately define or document the common sub-set of behavior all those
> > > bodies produce which we try to call our shared public langauge.  How
> > > close do the behaviors have to be before we call them part of the the
> > > shared public langauge?  When we have two different, but very similar
> > > behaviors, how do we define what should be the exact behavior of our
> > > public langauge? It's all very arbitrary.
> > > The concept of a shared public langauge, or the concept of a langauge
> > > defined by the culture is nothing but a pink elephant.  It doesn't and
> > > can't exist.  Only the idea and language used to define the idea
> > > exists. Yet, it's still very useful to all of us to talk as if it did
> > > exist - so we talk that way.

You are talking about theoretical concepts and not reality, as there is
no greek sculpture, week, alphabet or language, public or private.
These are just shortcuts, replacements for descriptive terms.
In respect to these concepts I said that the rules we use them are very
clear and logical. The real  issue is the meaning of these concepts.
JP


> > >
> > > If anything, it's an unrechable goal that is useful for us to try and
> > > get as close to as we can.  But to pretend it actually exists as
> > > anything but a dream, is to mis-understand how the world works.
> > >
> > > --
> > > Curt Welch
> > > http://CurtWelch.Com/ curt@kcwc.com
> > > http://NewsReader.Com/
> >
> > I disagree with you in regards to private and public languages.
> > While I do not use these terms I can understand the differentiation
> > that these terms reflect.
> > IMO the private language is a nonverbal one and it is the one you are
> > mostly  talking about.
> > OTOH this private language becomes public when it starts using verbal
> > language as it has to employ concepts created by society and  culture.
> > The private language has all the characteristics you describe but the
> > public one is very clear and logical.
>
> Have you seen the way I write?  What's clear and logical about that! :)
>
> We all act differently in public than we do in private.  Our behavior in
> all regards is a function of our context created by the environment we are
> in.  We learn from experience how we need to act in each context to get
> what we want.
>
> Now, it's natural to talk about priviate behavior and public behavior as
> being different and it's natrual to limit that idea to private langauge and
> public langauge.  And that seems to be how you want to use the terms.  And
> that's a fine way to use them.
>
> However, it has nothing to do with what I was trying to talk about.  I was
> talking about a "public langauge" as being the langauge of our culture.  A
> language created and used by the culture.  That's the langauge which
> doesn't actually exist even though we talk as if it does.
>
> I was using "priviate langauge" to talk about the full set of language
> behaviors we use as an individual, whether we are using it in our head as
> private thoughts, or whether we are using it on a TV broadcast to the
> world.
>
> > The main reason we have arguments is not because of the linguistics,
> > which  despite Wolf's suggestion has nothing to do with this post,
> > but because the concepts we use do not have the same meaning.
>
> Sure, of course.  We also have arguments just because we want different
> things.
>
> > I guess I could say that my point in all this thread is the translation
> > from private to public language, thru the creation of concepts.
> > Jan 10, 10.55am
> > BTW I still have problems having my replies posted. Anybody else has
> > this problem?
>
> No, but I don't waste my time trying to post through google groups.
>
> I've seen a lot of duplicate posts from people that post from google
> lately.
>
> > JP
>
> --
> Curt Welch                                            http://CurtWelch.Com/
> curt@kcwc.com                                        http://NewsReader.Com/

0
Just
1/11/2006 12:24:52 PM
IMO quality is just another theoretical concept, useful to have a
discussion but with no correspondence in reality.
Looking for quality is just another way to say looking for meaning and
the philosophers have been doing it for a few thousand years.
Maybe we should give the quantity a chance and something meaningful,
"quality" wise, will come out of this.
To me personally the most important issue is clearing our language, our
communication.
JP

0
Just
1/11/2006 12:39:15 PM
"Just Playing" <gms2004@lycos.com> wrote in message 
news:1136982292.636173.48290@g49g2000cwa.googlegroups.com...
>
> Curt Welch wrote:
>> "Just Playing" <gms2004@lycos.com> wrote:
>> > Curt Welch wrote:
>> > > lesterDELzick@worldnet.att.net (Lester Zick) wrote:
>> > > > On 09 Jan 2006 21:58:57 GMT, curt@kcwc.com (Curt Welch) in
>> > > > comp.ai.philosophy wrote:
>> > > >
>> > > > >lesterDELzick@worldnet.att.net (Lester Zick) wrote:
>> > > > >> On 09 Jan 2006 18:59:57 GMT, curt@kcwc.com (Curt Welch) in
>> > > > >> comp.ai.philosophy wrote:
>> > > > >>
>> > > > >> >"YadaYada" <yadayada___@excite.com> wrote:
>> > > > >> >> Curt Welch wrote:
>> > > > >> >> > > You have to create an assocation between the langauge
>> > > > >> >> > > symbol, and the aspect of the universe you are trying to
>> > > > >> >> > > describe with the language symbol. . . . It's got to all 
>> > > > >> >> > > be
>> > > > >> >> > > grounded to physcial reality.
>> > > > >> >>
>> > > > >> >> This analysis has a crucial step missing. Not one of us is
>> > > > >> >> capable of making such connections between our internal a
>> > > > >> >> priori concepts and the unknowable empirical reality that is
>> > > > >> >> out there. Given our limited sensory and rational 
>> > > > >> >> capabilities
>> > > > >> >> it is impossible, even in theory.
>> > > > >> >>
>> > > > >> >> > > I don't undestand your point.  What is this idea of 
>> > > > >> >> > > your's
>> > > > >> >> > > that there is something "unknownable" that is somehow
>> > > > >> >> > > important for us to know? If something is unknownable, 
>> > > > >> >> > > then
>> > > > >> >> > > it will never affect us and we will never know about and 
>> > > > >> >> > > we
>> > > > >> >> > > will never sense it and we will never talk about and we
>> > > > >> >> > > will never define any of our words using it.  So why are
>> > > > >> >> > > you trying to talk about something that doesn't exist to
>> > > > >> >> > > us?  And how is that going to prevent us from grounding 
>> > > > >> >> > > the
>> > > > >> >> > > meaning of our words?
>> > > > >> >>
>> > > > >> >> By "we" do you mean an individual or culture?
>> > > > >> >
>> > > > >> >I mean you, me, and anyone else that we want to include.
>> > > > >> >
>> > > > >> >> This distinction is
>> > > > >> >> crucial in the connection between language, which is a 
>> > > > >> >> public,
>> > > > >> >> a priori set of words and definitions, and concepts which are
>> > > > >> >> private instances of public language.
>> > > > >> >
>> > > > >> >That's just nonsense to me.  There is no public languge. 
>> > > > >> >There's
>> > > > >> >nothing to use it.  All we have is a few billion private
>> > > > >> >languages. You can't find one single word written in a public
>> > > > >> >language.  It was all created by the act of some individual 
>> > > > >> >using
>> > > > >> >their own idea private langauge.
>> > > > >>
>> > > > >> Then why are you having this conversation?
>> > > > >
>> > > > >We do it for many reasons.  But for the most part, we are just
>> > > > >attempting to reduce the differences in our private languages.
>> > > >
>> > > > Then any reduction in differences in your private languages makes 
>> > > > it
>> > > > a public language to that extent.
>> > >
>> > > Sure.  The only way to define public language is in terms of shared
>> > > behaviors.  The larger the set of people which share a similar 
>> > > behavior
>> > > the more public the behavior is said to be - whether it's langauge
>> > > behavior or any other type of social behavior.  That's what we mean
>> > > when we talk about a public language.
>> > >
>> > > But the meaning of this public language itself doesn't exist.  Only 
>> > > the
>> > > concept of the language exists.  This is because the true and 
>> > > complete
>> > > meaning of my langauge is defined by how my body and brain works. 
>> > > This
>> > > can't be fully documented in words of any langauge.  It can only be
>> > > documented by the existence of my body.
>> > >
>> > > We can document the total sum of all our private languages with the
>> > > existence of all our bodies as a large group.  But there's no way to
>> > > accurately define or document the common sub-set of behavior all 
>> > > those
>> > > bodies produce which we try to call our shared public langauge.  How
>> > > close do the behaviors have to be before we call them part of the the
>> > > shared public langauge?  When we have two different, but very similar
>> > > behaviors, how do we define what should be the exact behavior of our
>> > > public langauge? It's all very arbitrary.
>> > > The concept of a shared public langauge, or the concept of a langauge
>> > > defined by the culture is nothing but a pink elephant.  It doesn't 
>> > > and
>> > > can't exist.  Only the idea and language used to define the idea
>> > > exists. Yet, it's still very useful to all of us to talk as if it did
>> > > exist - so we talk that way.
>
> You are talking about theoretical concepts and not reality, as there is
> no greek sculpture, week, alphabet or language, public or private.

What exists is the behavior of individual speakers and listeners, and 
"language" may be treated as such. This leads to a very different sort of 
science than has been suggested by mentalistic, mainstream psychological, 
theories. "Meanings" in the behavioral sense, then, are to be found among 
the variables that, put simply, cause utterances.




> These are just shortcuts, replacements for descriptive terms.
> In respect to these concepts I said that the rules we use them are very
> clear and logical. The real  issue is the meaning of these concepts.
> JP
>
>
>> > >
>> > > If anything, it's an unrechable goal that is useful for us to try and
>> > > get as close to as we can.  But to pretend it actually exists as
>> > > anything but a dream, is to mis-understand how the world works.
>> > >
>> > > --
>> > > Curt Welch
>> > > http://CurtWelch.Com/ curt@kcwc.com
>> > > http://NewsReader.Com/
>> >
>> > I disagree with you in regards to private and public languages.
>> > While I do not use these terms I can understand the differentiation
>> > that these terms reflect.
>> > IMO the private language is a nonverbal one and it is the one you are
>> > mostly  talking about.
>> > OTOH this private language becomes public when it starts using verbal
>> > language as it has to employ concepts created by society and  culture.
>> > The private language has all the characteristics you describe but the
>> > public one is very clear and logical.
>>
>> Have you seen the way I write?  What's clear and logical about that! :)
>>
>> We all act differently in public than we do in private.  Our behavior in
>> all regards is a function of our context created by the environment we 
>> are
>> in.  We learn from experience how we need to act in each context to get
>> what we want.
>>
>> Now, it's natural to talk about priviate behavior and public behavior as
>> being different and it's natrual to limit that idea to private langauge 
>> and
>> public langauge.  And that seems to be how you want to use the terms. 
>> And
>> that's a fine way to use them.
>>
>> However, it has nothing to do with what I was trying to talk about.  I 
>> was
>> talking about a "public langauge" as being the langauge of our culture. 
>> A
>> language created and used by the culture.  That's the langauge which
>> doesn't actually exist even though we talk as if it does.
>>
>> I was using "priviate langauge" to talk about the full set of language
>> behaviors we use as an individual, whether we are using it in our head as
>> private thoughts, or whether we are using it on a TV broadcast to the
>> world.
>>
>> > The main reason we have arguments is not because of the linguistics,
>> > which  despite Wolf's suggestion has nothing to do with this post,
>> > but because the concepts we use do not have the same meaning.
>>
>> Sure, of course.  We also have arguments just because we want different
>> things.
>>
>> > I guess I could say that my point in all this thread is the translation
>> > from private to public language, thru the creation of concepts.
>> > Jan 10, 10.55am
>> > BTW I still have problems having my replies posted. Anybody else has
>> > this problem?
>>
>> No, but I don't waste my time trying to post through google groups.
>>
>> I've seen a lot of duplicate posts from people that post from google
>> lately.
>>
>> > JP
>>
>> --
>> Curt Welch 
>> http://CurtWelch.Com/
>> curt@kcwc.com 
>> http://NewsReader.Com/
> 


0
Glen
1/11/2006 12:39:54 PM
On Wed, 11 Jan 2006 07:39:54 -0500, "Glen M. Sizemore"
<gmsizemore2@yahoo.com> in comp.ai.philosophy wrote:

>
>"Just Playing" <gms2004@lycos.com> wrote in message 
>news:1136982292.636173.48290@g49g2000cwa.googlegroups.com...
>>
>> Curt Welch wrote:
>>> "Just Playing" <gms2004@lycos.com> wrote:
>>> > Curt Welch wrote:
>>> > > lesterDELzick@worldnet.att.net (Lester Zick) wrote:

[. . .]

>> You are talking about theoretical concepts and not reality, as there is
>> no greek sculpture, week, alphabet or language, public or private.
>
>What exists is the behavior of individual speakers and listeners, and 
>"language" may be treated as such. This leads to a very different sort of 
>science than has been suggested by mentalistic, mainstream psychological, 
>theories. "Meanings" in the behavioral sense, then, are to be found among 
>the variables that, put simply, cause utterances.

Which are what, pray tell? Which variables cause you to utter what you
utter? Which variables cause Curt to utter his utter nonsense? If you
could at least identify them perhaps we could get beyond utterances
and onto meanings. Or if you can't identify them perhaps you would be
so utterly kind as to stop uttering them.

~v~~

0
lesterDELzick
1/11/2006 3:55:06 PM
JAK wrote:
> In reading all of the replies, I am troubled by the general direction.
> In our search for expanded communication - passing more complex
> information - we may be missing a key.  As Curt Welch noted, the amount
> of information available from a tiny water droplet is immense.
> Extracting and communicating that information would be intensive.  Even
> more, the information may be useless.  Presently, the information we do
> get from our 5 senses is partially thrown out by the mind as not
> useful.  The feeling of your shoes on your feet or the cuff of your
> shirt is ignored.  In psychology, this is know as habituation.  Per
> Gestalt psychology, we visually pick out "figure" and ignore "ground"
> information.

Why don't we replace Gestalt with a finite capacity to process
information?
We throw out data not as much because it is not essential but because
we can not process it in "real time".
Maybe this limitation forces us to find  the "essential" in the data
perceived.
Calculate this capacity to process data and maybe we can define the
"essential".
JP

 If we throw out half or more of what we perceive, what is
> essential?  How do we transfer this crucial bit of editing to an AI
> system?
>
> Perhaps we should take a hint from evolution.  With rats, they are
> known to explore their environment - avoiding a direct route to food.
> Novelty is known to enrich rats brains and help us remember things.
> "Variety is the spice of life", as they say.  If we are to capture some
> information and pass it as communication, perhaps we should focus on
> passing that which is different, that which is new.  If someone already
> knows about an earthquake, why tell them again?  If information is to
> be useful, it must be new to the receiving system.  Pass a "headline".
> If it is new to the receiving system, it will request more detail.
>
> In addition to novelty, efficiency seems to be another important
> factor.  Natural selection, with its cycles of feast and famine as well
> as predation, demands efficiency.  It is quickly seen in the sleekness
> of fish and birds.  The sluggards survive, but only through efficient
> systems.  (For instance, bears hibernate.)  We can see our own demand
> for efficiency by virtue of every budget in place at corporations,
> universities, governments, and our personal bank accounts.  If we pass
> infomation within an AI system, not only should it be useful
> (different) to the receiving system, but only the essential information
> should be passed, and passed quickly.  Humans perform this efficient
> transfer by using verbal terms which condense complex information
> (jargon) in order to convey it quickly and efficiently.
>
> I suggest that rather than focusing on the quantity of infomation
> conveyed, the quality (primarily novelty and efficiency) should be of
> greater concern.

0
Just
1/11/2006 8:34:45 PM
"Just Playing" <gms2004@lycos.com> wrote:
> Curt Welch wrote:
> > "Just Playing" <gms2004@lycos.com> wrote:
> > > Curt Welch wrote:

> > > > We can document the total sum of all our private languages with the
> > > > existence of all our bodies as a large group.  But there's no way
> > > > to accurately define or document the common sub-set of behavior all
> > > > those bodies produce which we try to call our shared public
> > > > langauge.  How close do the behaviors have to be before we call
> > > > them part of the the shared public langauge?  When we have two
> > > > different, but very similar behaviors, how do we define what should
> > > > be the exact behavior of our public langauge? It's all very
> > > > arbitrary. The concept of a shared public langauge, or the concept
> > > > of a langauge defined by the culture is nothing but a pink
> > > > elephant.  It doesn't and can't exist.  Only the idea and language
> > > > used to define the idea exists. Yet, it's still very useful to all
> > > > of us to talk as if it did exist - so we talk that way.
>
> You are talking about theoretical concepts and not reality,

Well, these clearly are my ideas, but to me this is no theory. It's fact.

I suspect you however are not completely following the point I'm trying to
make.  Or maybe it's just me not following your point.

> as there is
> no greek sculpture, week, alphabet or language, public or private.

Huh?  There are greek sculptures, that is fact.  And there is langauge to
talk about greek sculptures, that too is fact.  And there are physical
brains that create and respond to that lanauge, that too is fact and it's
all part of what I call physical reality.  So what are you trying to say by
"there are no..."?

> These are just shortcuts, replacements for descriptive terms.

"week" is no replacement for "7 days".  Week is just as real as second or
day, or year.  "7 days" is as much a replacement for week as week is a
replacement for 7 days.

Are you trying to say that some words are descriptive terms and the rest of
the words in a lanauge are just replacments for these base descriptive
terms?

I don't think that idea is going to get you very far in understanding
languge.  It's just not that simple.

> In respect to these concepts I said that the rules we use them are very
> clear and logical.

When you write "clear and logical", I think of ideas related to absolute
truth.  Math is is clear, logical and precise.  There are no grey areas.

But normal day to day natural langauge is nothing but grey areas. It's full
of fuzz.  The common meaning of most words are defined roughly by their
relative location with other words.  When does water turn from cold to warm
or warm to hot?  When does attraction turn to love?  When does red become
pink? Is it still a keyboard if it's just a drawing of the keys on a touch
screen? How many impurities do you have to add to a glass of water before
it stops being a glass of water and turns into a glass of dirty water?

Nothing about our natural langauge is precise.  It's all fuzzy as hell and
each of us put the location of the borders of the meaning of our our words
at different places.

I for example called a glass of water with a lot of extra stuff added
"dirty water", what might someone else have called it?

Generally speaking, I can use a term like "dirty water" and anyone who
speaks english will have a rough idea of what I'm talking about.  Yet,
everyone that reads my words will have a different image of just what I'm
talking about.  Some might think of water with dirt added.  Some my think
of the soapy and dirty water that might be left in a sink after washing
dirty hands.  Some might think of pond water.

The point I'm making is that the words we use only give people a rough idea
of what we are talking about.  If the space of all possible ideas was
thought of as a large 2D map, then each word would be assocated with some
rough location in the map.  Some words are very vague in meaning such as
"in the north" would be for a map.  Some are more precise, like "at the
park" would be for the map.  Or something very preciseas to the boundies
such as, "inside the room".

But there are always grey areas when it comes to any words that attempt to
describe the universe.  The only way to remove all the grey areas and make
it clear and logical, is when you define words only in terms of other
words.  When you do that, you get a language like math.  And only those
langauges are clear and logical.  Everything they are based on, and all our
normal langauge, I would never call clear and precise.  If it were, we
wouldn't need to waste so much time trying to teach people how to be clear
and precise with their langauge - it would be automatic.

> The real  issue is the meaning of these concepts.
> JP

Yeah, that's true. :)

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/11/2006 10:11:36 PM
Curt Welch wrote:
> "Just Playing" <gms2004@lycos.com> wrote:
> > Curt Welch wrote:
> > > "Just Playing" <gms2004@lycos.com> wrote:
> > > > Curt Welch wrote:
>
> > > > > We can document the total sum of all our private languages with the
> > > > > existence of all our bodies as a large group.  But there's no way
> > > > > to accurately define or document the common sub-set of behavior all
> > > > > those bodies produce which we try to call our shared public
> > > > > langauge.  How close do the behaviors have to be before we call
> > > > > them part of the the shared public langauge?  When we have two
> > > > > different, but very similar behaviors, how do we define what should
> > > > > be the exact behavior of our public langauge? It's all very
> > > > > arbitrary. The concept of a shared public langauge, or the concept
> > > > > of a langauge defined by the culture is nothing but a pink
> > > > > elephant.  It doesn't and can't exist.  Only the idea and language
> > > > > used to define the idea exists. Yet, it's still very useful to all
> > > > > of us to talk as if it did exist - so we talk that way.
> >
> > You are talking about theoretical concepts and not reality,
>
> Well, these clearly are my ideas, but to me this is no theory. It's fact.
>
> I suspect you however are not completely following the point I'm trying to
> make.  Or maybe it's just me not following your point.
>
> > as there is
> > no greek sculpture, week, alphabet or language, public or private.
>
> Huh?  There are greek sculptures, that is fact.  And there is langauge to
> talk about greek sculptures, that too is fact.  And there are physical
> brains that create and respond to that lanauge, that too is fact and it's
> all part of what I call physical reality.  So what are you trying to say by
> "there are no..."?

These are such obvious replacements that I should not even have to
respond.
What is a greek sculpture? a sculpture created by my greek neighbor?
No.
They are statues created by people belonging to a population that lived
in a specific part of the world during a specific period of time, or
for short greek sculptures.
JP


>
> > These are just shortcuts, replacements for descriptive terms.
>
> "week" is no replacement for "7 days".  Week is just as real as second or
> day, or year.

I doubt that you have a week outside the human civilization and
culture. It is correct that you have a second, day or year but there is
nothing in the nature that will differentiate 7 days from 6, 8  or any
other number of days maybe except for a month.
JP


  "7 days" is as much a replacement for week as week is a
> replacement for 7 days.

Correct.
JP

>
> Are you trying to say that some words are descriptive terms and the rest of
> the words in a lanauge are just replacments for these base descriptive
> terms?

Not exactly. I use 2 types of terms: descriptive and theoretical.
IMO it is very helpful to differentiate between descriptive and
theoretical concepts.
First ones are subject to the laws of the physical world, while the
second ones are subject only to logic.
JP


>
> I don't think that idea is going to get you very far in understanding
> languge.  It's just not that simple.
>
> > In respect to these concepts I said that the rules we use them are very
> > clear and logical.
>
> When you write "clear and logical", I think of ideas related to absolute
> truth.  Math is is clear, logical and precise.  There are no grey areas.
>
> But normal day to day natural langauge is nothing but grey areas. It's full
> of fuzz.  The common meaning of most words are defined roughly by their
> relative location with other words.  When does water turn from cold to warm
> or warm to hot?

Ok, this is the best example to show why I differentiate between
descriptive and theoretical concepts.
Cold and warm are descriptive terms, they describe what our sensors are
perceiving and these perceptions can not be described in opposite,
logical  terms.
Trying to apply logic to descriptive terms will result in this type of
arguments.
JP


  When does attraction turn to love?  When does red become
> pink?


Descriptive terms, same response.
JP

 Is it still a keyboard if it's just a drawing of the keys on a touch
> screen?

Shortcut for a descriptive term.
JP


 How many impurities do you have to add to a glass of water before
> it stops being a glass of water and turns into a glass of dirty water?

Shortcut for a descriptive term.
JP
>
> Nothing about our natural langauge is precise.  It's all fuzzy as hell and
> each of us put the location of the borders of the meaning of our our words
> at different places.
>
> I for example called a glass of water with a lot of extra stuff added
> "dirty water", what might someone else have called it?
>
> Generally speaking, I can use a term like "dirty water" and anyone who
> speaks english will have a rough idea of what I'm talking about.  Yet,
> everyone that reads my words will have a different image of just what I'm
> talking about.  Some might think of water with dirt added.  Some my think
> of the soapy and dirty water that might be left in a sink after washing
> dirty hands.  Some might think of pond water.
>
> The point I'm making is that the words we use only give people a rough idea
> of what we are talking about.  If the space of all possible ideas was
> thought of as a large 2D map, then each word would be assocated with some
> rough location in the map.

Actually you might try to look at reality as  3D and language as 2D.
JP


 Some words are very vague in meaning such as
> "in the north" would be for a map.  Some are more precise, like "at the
> park" would be for the map.  Or something very preciseas to the boundies
> such as, "inside the room".

You are talking about a frame of reference but you have a 2D model of a
3D reality.
JP


> But there are always grey areas when it comes to any words that attempt to
> describe the universe.  The only way to remove all the grey areas and make
> it clear and logical, is when you define words only in terms of other
> words.

You seem to have an issue with trying to make the physical world work
according to the logic.
Descriptive terms as the culture has created them are subject not only
to logic as are the theoretical concepts, but to the laws of the
physical world too.
There is a lot to be done but first is necessary to separate apples
from oranges, or in this case descriptive from theoretical terms.
JP
Jan 11, 10.53PM

When you do that, you get a language like math.  And only those
> langauges are clear and logical.  Everything they are based on, and all our
> normal langauge, I would never call clear and precise.  If it were, we
> wouldn't need to waste so much time trying to teach people how to be clear
> and precise with their langauge - it would be automatic.
>
> > The real  issue is the meaning of these concepts.
> > JP
>
> Yeah, that's true. :)
>
> --
> Curt Welch                                            http://CurtWelch.Com/
> curt@kcwc.com                                        http://NewsReader.Com/

0
Just
1/12/2006 3:53:29 AM
Curt Welch wrote:
[...]>
> When you write "clear and logical", I think of ideas related to absolute
> truth.  Math is is clear, logical and precise.  There are no grey areas.
> 
[...]

Oh, but there are. Vagueness and grey areas abound in math, as in every 
other human endeavour. All advances in math have come from recognition 
and clarification of a grey area, from recognition and sharpening of a 
vagueness. Some of those clarifications really, rally upset people, eg, 
what is meant by "infinite?" Cantor's clarification of that vagueness 
still discombulates people (recall the endless thread around it last 
year and the year before.)

More recently, we have the problem of randomness, still not resolved, 
but which affects not only  statistical computation issues etc (and 
hence impinges on AI), but also on "deep" questions in philosophy and 
theology. (A "deep" question is one that causes people to shout at each 
other, and sometimes leads to murder.)
0
Wolf
1/12/2006 4:04:06 PM
Curt Welch wrote:
[...]
> I for example called a glass of water with a lot of extra stuff added
> "dirty water", what might someone else have called it?
> 
> Generally speaking, I can use a term like "dirty water" and anyone who
> speaks english will have a rough idea of what I'm talking about.  Yet,
> everyone that reads my words will have a different image of just what I'm
> talking about.  Some might think of water with dirt added.  Some my think
> of the soapy and dirty water that might be left in a sink after washing
> dirty hands.  Some might think of pond water.[...]

I regularly did an exercise with my English classes in which I asked 
them to make stick-figure sketches of what came to mind when they heard 
words like tree, bird, couch (sofa), table, cup, house, and so on. I 
then asked a student to put his or her sketches on the board, then asked 
for different versions of each, and then determined the numbers of 
students who had made each type of sketch. Eg, deciduous trees and 
conifers were about equally represented for "tree", but there were 
always a few students who sketched bare branched tree.

I usually introduced the exercise in the study of poetry, as a means of 
helping students realise that a poet relies on the reader's ability to 
visualise trees, say, and tries to control that visualisation by use of 
other words. ER, "Gentle trees" fits well with leafy deciduous trees, 
but less well with conifers. So perhaps people who habitually see 
conifers when reading "trees" may have a slight difficulty reading a 
phrase like "gentle trees,", as they would have to shift to another 
visualisation to make the phrase "intelligible." Thus, reading poetry 
"trains the imagination," as the old claim has it. Etc.

It also seemed reasonable to infer that similar differences existed for 
more complex concepts like love, justice, freedom, health, wealth, etc. 
But these differences would be more difficult to discover, since one 
couldn't compare sketches in a notebook. So how does one discover the 
differences? By argument and discussion... Of course. What else?

HTH
0
Wolf
1/12/2006 4:22:04 PM
Please do provide references to your material.  I am greatly interested
in learning more.  (I, too, have been promoting similar ideas - since
1985.)

0
JAK
1/12/2006 5:45:05 PM
http://faculty.smu.edu/bthompso/spatialcontent.html

0
JAK
1/12/2006 5:48:19 PM
I think Curt's reference to "spots on the floor" is why I am not
whole-heartedly enthusiastic about processing quantitatively.  Even
though, I must admit that Chaos Theory would note that the spots should
not be totally ignored.

0
JAK
1/12/2006 6:09:45 PM
I believe that Curt's position agree's with those supporting "qualia":
http://plato.stanford.edu/entries/qualia/

The internal issue of phenomenological experience is also addressed
here: http://faculty.smu.edu/bthompso/spatialcontent.html

0
JAK
1/12/2006 6:17:46 PM
I agree.  In my creativity seminars, I ask people to imagine a cat.  I
then ask everyone to write down what kind of cat they imagined (house
cat, lion, tiger, etc.), what color their cat is, what position is it
in (sitting, lying down, running, etc.), and what perspective the cat
is seen (cat in a frontal, facing left, facing right, etc.).  To me
this line of thinking on both of our parts further cements Curt's
contention that the internal representation of words and language are
unique to each individual despite the public standards.  As an English
teacher, you can also attest to the importance of connotations as well
as metaphors and similes (the basis of analogies).  To evoke any of
these is a result of the individual's internal makeup and not part of
the standardized language.  Looking at a plowed field, should this
evoke television?  I suggest not.  But it did.  It led to the invention
of television:
http://www.time.com/time/time100/scientist/profile/farnsworth.html

0
JAK
1/12/2006 6:35:26 PM
"Just Playing" <gms2004@lycos.com> wrote:
> IMO quality is just another theoretical concept, useful to have a
> discussion but with no correspondence in reality.
> Looking for quality is just another way to say looking for meaning and
> the philosophers have been doing it for a few thousand years.
> Maybe we should give the quantity a chance and something meaningful,
> "quality" wise, will come out of this.
> To me personally the most important issue is clearing our language, our
> communication.
> JP

Well, we can't understand the full meaning of quality until you solve AI.
In other words, all these ideas come from humans, and unless you can fully
understand what a human is and how we get ideas and what ideas are, you
can't answer these questions.  And as you point out, they have not been
easy questions to answer because lots of very smart people have debated it
for thosands of years.

But, they didn't have access to all the scientific disoveries made over the
past thousands of years which we have access to today.

The word quality is easy to understand at one leave beacuse it's a
reference to the value of something.  And if that value is easily specified
and measured, then the quality of the item is easy to understand.

However, most use of the idea of quaility is in reference to an items value
to a human.  And when you don't have an easiy and obvious way to measure
it's value to a human, then you are left trying to understand how humans
assign value to everything.  Why do we love on thing and hate the next?
Why do we seek out some things, and avoid others?

But all this can easily be brought back to the same idea of why we select
one behavior over another.  The fact that we are forced to do that for
everyhing we do, shows that we have some type of system for selecting
behavior based on their quality.  And this same system, whatever it is, is
the source of all human evaluation of quality.  We see things as having
quality if our behavior selection system detects greater value in them. So,
to fully understand how humans detect quaility in all things big and
little, tangible and intangible, you have to uncover how the brain selects
behaviors.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/12/2006 8:28:29 PM
Curt Welch wrote:
> "Just Playing" <gms2004@lycos.com> wrote:
> > IMO quality is just another theoretical concept, useful to have a
> > discussion but with no correspondence in reality.
> > Looking for quality is just another way to say looking for meaning and
> > the philosophers have been doing it for a few thousand years.
> > Maybe we should give the quantity a chance and something meaningful,
> > "quality" wise, will come out of this.
> > To me personally the most important issue is clearing our language, our
> > communication.
> > JP
>
> Well, we can't understand the full meaning of quality until you solve AI.

Quality is just another concept created to replace N unknown variables
with only one in order to overcome the restrictions placed by logic on
our verbal communication.
JP

> In other words, all these ideas come from humans, and unless you can fully
> understand what a human is and how we get ideas and what ideas are, you
> can't answer these questions.  And as you point out, they have not been
> easy questions to answer because lots of very smart people have debated it
> for thosands of years.
>
> But, they didn't have access to all the scientific disoveries made over the
> past thousands of years which we have access to today.
>
> The word quality is easy to understand at one leave beacuse it's a
> reference to the value of something.  And if that value is easily specified
> and measured, then the quality of the item is easy to understand.
>
> However, most use of the idea of quaility is in reference to an items value
> to a human.  And when you don't have an easiy and obvious way to measure
> it's value to a human, then you are left trying to understand how humans
> assign value to everything.  Why do we love on thing and hate the next?
> Why do we seek out some things, and avoid others?
>
> But all this can easily be brought back to the same idea of why we select
> one behavior over another.  The fact that we are forced to do that for
> everyhing we do, shows that we have some type of system for selecting
> behavior based on their quality.  And this same system, whatever it is, is
> the source of all human evaluation of quality.  We see things as having
> quality if our behavior selection system detects greater value in them. So,
> to fully understand how humans detect quaility in all things big and
> little, tangible and intangible, you have to uncover how the brain selects
> behaviors.
>
> --
> Curt Welch                                            http://CurtWelch.Com/
> curt@kcwc.com                                        http://NewsReader.Com/

0
Just
1/13/2006 5:40:14 PM
On 13 Jan 2006 09:40:14 -0800, "Just Playing" <gms2004@lycos.com> in
comp.ai.philosophy wrote:

>
>Curt Welch wrote:
>> "Just Playing" <gms2004@lycos.com> wrote:
>> > IMO quality is just another theoretical concept, useful to have a
>> > discussion but with no correspondence in reality.
>> > Looking for quality is just another way to say looking for meaning and
>> > the philosophers have been doing it for a few thousand years.
>> > Maybe we should give the quantity a chance and something meaningful,
>> > "quality" wise, will come out of this.
>> > To me personally the most important issue is clearing our language, our
>> > communication.
>> > JP
>>
>> Well, we can't understand the full meaning of quality until you solve AI.
>
>Quality is just another concept created to replace N unknown variables
>with only one in order to overcome the restrictions placed by logic on
>our verbal communication.
>JP

I think you'll find that "quality" is just another concept used
to refer to non material dualism. I doubt matter and material
interactions use or refer to "quality".

>> In other words, all these ideas come from humans, and unless you can fully
>> understand what a human is and how we get ideas and what ideas are, you
>> can't answer these questions.  And as you point out, they have not been
>> easy questions to answer because lots of very smart people have debated it
>> for thosands of years.
>>
>> But, they didn't have access to all the scientific disoveries made over the
>> past thousands of years which we have access to today.
>>
>> The word quality is easy to understand at one leave beacuse it's a
>> reference to the value of something.  And if that value is easily specified
>> and measured, then the quality of the item is easy to understand.
>>
>> However, most use of the idea of quaility is in reference to an items value
>> to a human.  And when you don't have an easiy and obvious way to measure
>> it's value to a human, then you are left trying to understand how humans
>> assign value to everything.  Why do we love on thing and hate the next?
>> Why do we seek out some things, and avoid others?
>>
>> But all this can easily be brought back to the same idea of why we select
>> one behavior over another.  The fact that we are forced to do that for
>> everyhing we do, shows that we have some type of system for selecting
>> behavior based on their quality.  And this same system, whatever it is, is
>> the source of all human evaluation of quality.  We see things as having
>> quality if our behavior selection system detects greater value in them. So,
>> to fully understand how humans detect quaility in all things big and
>> little, tangible and intangible, you have to uncover how the brain selects
>> behaviors.
>>
>> --
>> Curt Welch                                            http://CurtWelch.Com/
>> curt@kcwc.com                                        http://NewsReader.Com/
>


~v~~

0
lesterDELzick
1/13/2006 6:23:57 PM
Wolf Kirchmeir wrote:
> Curt Welch wrote:
> [...]
> > I for example called a glass of water with a lot of extra stuff added
> > "dirty water", what might someone else have called it?
> >
> > Generally speaking, I can use a term like "dirty water" and anyone who
> > speaks english will have a rough idea of what I'm talking about.  Yet,
> > everyone that reads my words will have a different image of just what I'm
> > talking about.  Some might think of water with dirt added.  Some my think
> > of the soapy and dirty water that might be left in a sink after washing
> > dirty hands.  Some might think of pond water.[...]
>
> I regularly did an exercise with my English classes in which I asked
> them to make stick-figure sketches of what came to mind when they heard
> words like tree, bird, couch (sofa), table, cup, house, and so on. I
> then asked a student to put his or her sketches on the board, then asked
> for different versions of each, and then determined the numbers of
> students who had made each type of sketch. Eg, deciduous trees and
> conifers were about equally represented for "tree", but there were
> always a few students who sketched bare branched tree.
>
> I usually introduced the exercise in the study of poetry, as a means of
> helping students realise that a poet relies on the reader's ability to
> visualise trees, say, and tries to control that visualisation by use of
> other words. ER, "Gentle trees" fits well with leafy deciduous trees,
> but less well with conifers. So perhaps people who habitually see
> conifers when reading "trees" may have a slight difficulty reading a
> phrase like "gentle trees,", as they would have to shift to another
> visualisation to make the phrase "intelligible." Thus, reading poetry
> "trains the imagination," as the old claim has it. Etc.
>
> It also seemed reasonable to infer that similar differences existed for
> more complex concepts like love, justice, freedom, health, wealth, etc.
> But these differences would be more difficult to discover, since one
> couldn't compare sketches in a notebook. So how does one discover the
> differences? By argument and discussion... Of course. What else?
>
> HTH

I sort my models out primarily thus:-

Reality   = 5 sensory imputs, local field of perception.
            Vision.
Socially  = linguistics A; verbatim, spoken words, NVC
            linguistics B; written communications, grammar,
            logic.
Conceptual= Secondary,tertiary mechanical structures
            devised
            from reality with social convention.

Of course there may be 'reality based conceptions' they
might very well be the basic algorithms that include
fundamental motivational drives.

...........

Linguistics A & B are particulary interesting. 'A' refers
to social codes, 'B' refers to artificial languages,
(these are one place removed from 'reality') I try to find
out if the debate is regards A & B,  or 'A' and reality
or 'B' and reality. Reality here becomes the ID the 'I' the
personal privately sense of identity. Concept suggests the
rationalizing by this identity of the artificial and
secondary means to communicate. In other words 'I' only
exists in circumstances where secondary and artificial
means of communication prevail. The rest of the time
during 'reality' e.g. everyday living experiences we're
just not aware of it. (tough for the Bhuddist)

...........

You tend not to get these problems with pictographic,
ideographic scripts,(the alphabet of course refers to the
musicality of language) If you are using a picture
script you'll get;
Reality    = 5 sensory imputs.
socially A  = Verbatim, NVC. (a bit like a strip cartoon)
Conceptual = either a priori (private language) or structures
                    contrived  from picture based social, verbatin
                    referentials.
There is no 'imaginary' referential as there is in the alphabet.

What basic difference is there between a picture 'referential'
based comms system and an alphabetical one? Theres more social
and specific information in pictures.It is silent. Alphabetical
systems are similar to mathematical systems (by their calculus).
Verbatim real time alphabetical systems are also sound based so
we can infer meaning by guessing intonation.
..........


N.

0
mimo_545
1/13/2006 11:58:10 PM
mimo_545@hotmail.com wrote:
[...]
> 
> I sort my models out primarily thus:-
> 
> Reality   = 5 sensory imputs, local field of perception.
>             Vision.
[...]

There are at least 9 external sensory inputs, and at least that many 
internal ones. "Touch" is at least four different senses, for example. 
I've not kept track of research into sensation for about 20 years, yet 
by the early 80s it was known that pressure sensing via touch is both 
"surface" and "deep", etc. (My most recent information is a driblet I 
encountered by accident: pressure sensors in the feet work with 
proprioceptors in the legs and lower torso to prevent us from falling 
down; some of the neural circuits involved apparently bypass spinal 
cord. The fiance of a former student had just returned from a post-doc 
project in Sweden that dealt with this problem. I met him at their 
engagement party.}

The rest of your comments are mostly speculation, with little factual 
data support. Eg, you seem to assume that ideographic scripts are less 
abstract than alphabetic ones. There is a good deal of research that 
contradicts that assumption. And "social convention" is a label for a 
very large suitcase.

IMO, you also place too much weight on vision. This seems to be a common 
trait of AI workers, and may be connected with an equal overemphasis on 
abstract reasoning, knowledge structures, etc. Vision is the most 
abstract sense. There is also a curious tendency to identify the 
capacity to abstract as a sine qua non of intelligence.

HTH
0
Wolf
1/16/2006 2:09:29 PM
Curt Welch wrote:
> Our purpose was given to us by evolution.  We are machines built to
> survive.  And as such, we also have data processing hardware designed for
> the purpose of the purpose of extracting useful information from the
> enviroment.  And the ultimate purpose of that data is to guide us in the
> flapping of our arms and legs and lips so we can survive.  The brain is a
> very complex device which finds the data needed to drive the arms and legs
> and ignores everything else.  If the data is useful in producing arm
> motions that help us survive, we call it "information".  If it's not, we
> call it noise.
....
> ... That only happens when the data
> is correlated to a purpose.  And this is what the reinforcement problem I
> always talk about is all about.  You can't create intelligent behavior, and
> you can't build a machine to extract data from noise, unless you first give
> it a purpose.
....
> We spend to much time dealing with the data in the middle of the
> transformation from noise to arm and leg control, that we loose track of
> the fact that everything we learn, is there only to support the purpose of
> moving our body parts.  That data that the brain has determined to have a
> role in that, is the data it "keeps" as "useful information".

Curt, let's dig into your declared purpose of "only to support the
purpose of moving our body parts."  I know you were likely being a bit
glib for effect and speed, but if you are strategically correct, then
"purpose" is the most important issue before you.

Please expand upon your ideas.  What "purpose" do you believe has
evolved in humans?  How does the brain support that purpose?

0
JAK
1/17/2006 4:55:16 PM
On 17 Jan 2006 08:55:16 -0800, "JAK" <jak@theoryofmind.org> in
comp.ai.philosophy wrote:

>Curt Welch wrote:
>> Our purpose was given to us by evolution.  We are machines built to
>> survive.  And as such, we also have data processing hardware designed for
>> the purpose of the purpose of extracting useful information from the
>> enviroment.  And the ultimate purpose of that data is to guide us in the
>> flapping of our arms and legs and lips so we can survive.  The brain is a
>> very complex device which finds the data needed to drive the arms and legs
>> and ignores everything else.  If the data is useful in producing arm
>> motions that help us survive, we call it "information".  If it's not, we
>> call it noise.
>...
>> ... That only happens when the data
>> is correlated to a purpose.  And this is what the reinforcement problem I
>> always talk about is all about.  You can't create intelligent behavior, and
>> you can't build a machine to extract data from noise, unless you first give
>> it a purpose.
>...
>> We spend to much time dealing with the data in the middle of the
>> transformation from noise to arm and leg control, that we loose track of
>> the fact that everything we learn, is there only to support the purpose of
>> moving our body parts.  That data that the brain has determined to have a
>> role in that, is the data it "keeps" as "useful information".
>
>Curt, let's dig into your declared purpose of "only to support the
>purpose of moving our body parts."  I know you were likely being a bit
>glib for effect and speed, but if you are strategically correct, then
>"purpose" is the most important issue before you.
>
>Please expand upon your ideas.  What "purpose" do you believe has
>evolved in humans?  How does the brain support that purpose?

You know, JAK, all you're going to get out of Curt is doubletalk. I
don't say this to preclude dialog but Curt, Wolf, and Glen are all
about behaviorism first and intelligence second. Apparently you see
the brain as a functional device supporting purpose. I do they don't.

~v~~

0
lesterDELzick
1/17/2006 6:34:39 PM
"JAK" <jak@theoryofmind.org> wrote:
> Curt Welch wrote:
> > Our purpose was given to us by evolution.  We are machines built to
> > survive.  And as such, we also have data processing hardware designed
> > for the purpose of the purpose of extracting useful information from
> > the enviroment.  And the ultimate purpose of that data is to guide us
> > in the flapping of our arms and legs and lips so we can survive.  The
> > brain is a very complex device which finds the data needed to drive the
> > arms and legs and ignores everything else.  If the data is useful in
> > producing arm motions that help us survive, we call it "information".
> > If it's not, we call it noise.
> ...
> > ... That only happens when the data
> > is correlated to a purpose.  And this is what the reinforcement problem
> > I always talk about is all about.  You can't create intelligent
> > behavior, and you can't build a machine to extract data from noise,
> > unless you first give it a purpose.
> ...
> > We spend to much time dealing with the data in the middle of the
> > transformation from noise to arm and leg control, that we loose track
> > of the fact that everything we learn, is there only to support the
> > purpose of moving our body parts.  That data that the brain has
> > determined to have a role in that, is the data it "keeps" as "useful
> > information".
>
> Curt, let's dig into your declared purpose of "only to support the
> purpose of moving our body parts."  I know you were likely being a bit
> glib for effect and speed, but if you are strategically correct, then
> "purpose" is the most important issue before you.
>
> Please expand upon your ideas.  What "purpose" do you believe has
> evolved in humans?  How does the brain support that purpose?

It's all answerd by the framework of reinforcement learning.  We have a
learning brain which is a self adapting arm and leg controller which has
the simple purpose of maximizing all future rewards.  It attempts to find
which arm and leg motions produce the maximum rewards from the environment
(make the best life possible for us).

The body also has hardware in it separate from the learning part of the
brain, which generates these reward signals for the purpose of defining our
values - the hardware which generates the rewards is what defines good and
bad for us.  That hardware is what makes burning our fingers give us pain
and eating when hungry give us pleasure.

That reward hardware was created by evolution as a guide to control and
give the learning brain purpose.  Evolution picks hardware based on
survivial, so the ultimate purpose is just survial.  But that purpose is
passed to us indirectly by our reward hardware.  And the learning brain
then creates yet more indirection by attempting to predict future rewards,
and using those predictions to shape our behavior as well.  So experience
tells us that money is good so money is recognized by the brain has having
high reward values for us and behaviors which give us money gets
reinforced.

So, our prime purpose is survival.  But evolution gave us a learning brain,
and couldn't directly give it "survial" as a purpose.  You can't build
hardware to directly evaluate the survival advantage of every behavior.  So
it did it a simple way by creating a reward system to monitor and reward
the metrics that were shown by natrual selection to closely correlate with
survival.  Tearing our arms off was shown to be bad for survival so we have
sensors to test for damage to the body and send punishment signals to the
arm/leg controller to allow it adapt it's behavior to reduce the odds of
having an arm torn off.

The whole good vs evil battle we constantly talk about is nothing more than
an extension of what our brain was built to do - identify which arm and leg
motions were "good" and use those, while not using the ones identified as
"bad".  We are slaves to our pain and pleasure signals and to the large and
complex secondary values which the brain has calculated for us from a
lifetime of experience.

All our complex thoughts are just behaviors learned for their value in
maximizing all future rewards from our internal reward system.

If we had access to our internal reward system and had the power to turn on
a pure pleasure signal and block the pain signals, we would be forced in
the end to do it.  We would be unable to stop ourselves from doing it
because to do that would be to reach heaven.  It's the ultimate best
behavior for us because it would mazimize our reward function.  This is
exactly why drugs and alcohol can be so dangerous to us. It allows us some
small control over our pain and pleasure system and once you get that
control, you can't stop using it.

Of course, if we defeat our pleasure system and turn on a pure pleasure
signal, we would do nothing but sit there and die.  And then natrual
selection would have the ultimate control and stop building humans that
were smart enough to modify their pleasure hardware.

So, evolution built us as a survival machine.   But evolution found it
justified to use learning hardware to allow the system to adapt far quicker
to a changing environment than evolution could do with the DNA based
learning system alone, and that gave "us" (our learning brain which is the
center of what most of us know as as conscious selves), the sub purpose of
simply maximizing rewardws (maximize personal "good").  But the reward
system selected by evolution has a lot of different things it will reward
of punish us for, and many of them are based on being good members of
society (having and raising children for example) and how knows what else
evolution has build in and hidden down in our motivational hardware.

-- 
Curt Welch                                            http://CurtWelch.Com/
curt@kcwc.com                                        http://NewsReader.Com/
0
curt
1/18/2006 7:16:57 AM
Wolf Kirchmeir wrote:
Apologies for the delay..I had not forgotten (defrag + new
chips installation, Oh what a day! -I'm also sober! :)

>There are at least 9 external sensory inputs, and at least that many
>internal ones. "Touch" is at least four different senses, for example.
>I've not kept track of research into sensation for about 20 years, yet
>by the early 80s it was known that pressure sensing via touch is both
>"surface" and "deep", etc. (My most recent information is a driblet I
>encountered by accident: pressure sensors in the feet work with
>proprioceptors in the legs and lower torso to prevent us from falling
>down; some of the neural circuits involved apparently bypass spinal
>cord. The fiance of a former student had just returned from a post-doc
>project in Sweden that dealt with this problem. I met him at their
>engagement party.}

There was an article in last years New Scientist regards senses..
(Jan-05)the list just seems to keep growing! I keep it to 5 senses
in general although I do flirt with the notion that descriptive
analysis is subject to the strictures demanded by grammatical
/logical/cultural conventions and will alter the interpreted data.
If I am correct that language in thought and also via literacy,
e.g. during reading, generates a consciousness that is
misinterpreted as being--what?--hmmm, The foundation of all
intelligence ? its going to be wrong isn't it?. What you will
end up with is something like the tree JAK wrote posted in
:-Message-ID: <1136822678.319669.228280@z14g2000cwz.googlegroups.com>

 >The 3 fundamental laws of the empirical universe are:
 >#1 Differences Exist.  Because of this law, language has nouns,
 >mathematics has variables (a, b, c, x, y, z, etc.), and physics has
 >objects (galaxies, planets, molecules, atoms, quarks, etc.)
 >#2 Differences are Dynamic.  Because of this law, language has verbs,

 >mathematics has operators (+, -, *, /, etc.), and physics has laws of

 >motion.
 >#3 Differences are Relative.  Because of this law, language has
 >modifiers (adjectives and adverbs), mathematics has relative symbols
 >(>, <, =, etc.), and physics has scalar references (magnitude,
gravity,
 >etc.)

(for example its not hard finding a link between musical score and
mathemetics, because of the geometry of Pythagoras, but, it is
incorrect to assume that all human reasoning and thought processes
are based on his formulas and observations. I saw a program on
synaesthesia today with Ramachandran. When I was doing my foundation
we were encouraged to associate sounds with shapes and colours - it
was well known by our tutors and used in Graphics and media. Its not
new to us. We assumed it was part of memetics. I'm convinced that
many a battle waged in physics is caused by the use and interpretation
of they symbol, and its hard not to get bogged down in the descriptive
even when you do know your time is better spent elsewhere working
on in self initiated projects.)

An example of the influence of language and use of senses could be;
'You' & 'thou', which were once used in both English and French, now
confined to Latin based langs. It would seem that in todays middle
English 'thou'  is not often expressed unless it is non-verbally.
It requires senses such as sight for NVC & perception of intonality
to detect the intimation. 'Thou' (which my grandparents used,) as
I recall also had a quality in that it deliniated the class
boundries, for example, family & extended from employers & unknowns.
It exists, but not for us it seems, and anyone unfamiliar with the
concept of 'thou' and how it functions in a social environment (its
social coding) will feel they're missing something, might even feel
out of place.

>The rest of your comments are mostly speculation, with little factual
>data support. Eg, you seem to assume that ideographic scripts are less
>abstract than alphabetic ones. There is a good deal of research that
>contradicts that assumption. And "social convention" is a label for a
>very large suitcase.
On the contrary, so much emphasis is placed upon language and textual
communications (to the extent that people might assume that it
comprises the totality of human consciousness) that visual perception
and communications are secondarily requisite to human comprehensions.

>IMO, you also place too much weight on vision. This seems to be a common
>trait of AI workers, and may be connected with an equal overemphasis on
>abstract reasoning, knowledge structures, etc. Vision is the most
>abstract sense. There is also a curious tendency to identify the
>capacity to abstract as a sine qua non of intelligence.
I'm slightly disappointed that with your expertise you hav'nt
offered a possible solution in uniting the visual and aural ?

>HTH 

N.

0
mimo_545
1/24/2006 11:31:34 PM
mimo_545@hotmail.com wrote:
[...]
> An example of the influence of language and use of senses could be;
> 'You' & 'thou', which were once used in both English and French, now
> confined to Latin based langs. [...]

A nit-pick: the vast majority of languages worldwide use 'thou' 
(singular) and 'you' (plural). A majority also use other pronouns or 
pronoun forms to differentiate polite/infomal relationships, and/or 
class, and/or gender, and/or age, etc. Many languages extend these 
distinctions to the other pronouns as well: in one language (I can't 
recall which) when addressing a person, your 'I' has to express your 
respective social classes, for example.

However, as people's notions of such distinctions change, so does the 
language change. In German, in my lifetime the language has shifted from 
'Sie' (polite sing and plur 'you') to 'Du' and 'Ihr' (informal sing and 
plur 'you'). People my age and older have not all adapted to this usage 
change, however, so there's occasional perception of rudeness where none 
was intended. A mere 150 years or so ago, it was still customary to 
address servants as 'he' and 'she' in German. The master could not 
address a person of such inferior rank directly, you see: it would imply 
at least near-equality. Etc.

Relation to AI?  Being able to discern/learn such usages (among many, 
many others) is a sine-qua-non of natural language processing.

If ref. to subject of this thread. Sound is serial, hence the spoken 
language is serially presnted. But what it encodes is not parsed 
serially, and when context is included (as it must), is not parsed in 
parallel, but in a complex network, with a great deal of reiteration and 
recursion, over a multi-dimensional domain or "field of meaning." Some 
experiments in writing down poems attempt to demonstrate this, poorly IMO.

The amazing thing is that we communicate as well as we do.

HTH
0
Wolf
1/25/2006 3:51:31 PM
Reply: