Fw: [FIS] Following our exchange with Loet LeydersdorffFw: [FIS] Following our exchange with Loet Leydersdorff
From: Igor Rojdestvenski <igor.rojdestvenski@plantphys.umu.se>
Date: Mon 30 Oct 2006 - 18:00:49 CET
----- Original Message -----
Agree to one extent. Physics went a very long way to create operational concepts -- physical
variables. Their necessity is beyond doubt. But we should be sure we do not mix them with
philosophical concepts, of which these are but particular properties.
We do this mixing here very often. That is why our debates sometimes become like "is a brick
heavy, or is it red?". And the third opinion is "no, it is solid and rectangular".
This is the danger of operational concepts.
In my view we definitely should go operational, but each time we use a certain operational concept,
we should:
Otherwise ......You know....
Yours, Igor
----- Original Message -----
Dear Igor,
In the social sciences, concepts need to be operationalized. Conceptually, one can
attribute all kinds of properties and have philosophical debates about these constructs. The value
of empirical research is precisely to take the next step.
with best wishes,
Loet
PS. I am also out of quotum.
------------------------------------------------------------------------------
Loet Leydesdorff
----------------------------------------------------------------------------
Dear Loet, sending directly as it seems that I am over the quota.
I would like to reiterate. Your operational definitions are, in fact, what I
call variables. They are ontologically helpful, but only that.
What would be the operational definition of matter, in your opinion?
Again, my point in terminology.
Information is a concept. As a concept, it has particular properties.
Matter is a concept. It has particular properties of mass, density,
structural params, etc., which are measurable. May we call mass an operational definition of matter?
I do not think so.
Same with information. Shannons information refers to information as a
concept in the same way as mass refers to matter. Brillouin's information refers to information as a
concept in the same way as density refers to matter. These properties are measurable and, to a
limited extent, describe certain aspects of the concept of information. But this concept may never
be reduced to either of them.
Igor
------------------------------------------------------------------------
We cannot consider a concept as a variable. Simply
because a concept is not measured in bits, grams, joules, etc. And a variable always is.
We may, instead, associate variables with a concept,
these variables describing certain measurable aspects of a concept. Similarly to the concept of
matter, the variables for which represent mass, density, structural parameters, etc, etc.
This is the key point, in my opinion. Information is
a concept. And what we call information in Shannon's definition is, in fact, a variable associated
with the concept of information. One of many possible variables.
Igor
Yes, the same concept can differently be operationalized.
However, in the case of information we should not confuse two concepts: Shannon-type information and
meaningful information. The Chinese language has two different expressions for these two concepts:
Both words contain two char�acters . The above one, 'sjin sji',
corresponds to the mathe�matical definition of informa�tion as uncertainty. The sec�ond, 'tsjin
bao,' means infor�mation but also intelligence. In other words, it means infor�mation which informs
us, and which is thus considered meaningful.
The first concept can be operationalized as Shannon-type
information. The second perhaps as Brillouin's "negentropy". "Meaningful
information" assumes a system for which the information can have meaning. One can also call
this "observed information", that is, the information is "observed" by the
receiving system. Shannon-type information remains expected information content (of a message).
It seems to me that operational definitions thoroughly solve the
conceptual confusion.
With best wishes,
Loet
--------------------------------------------------------------------------
Loet Leydesdorff
Now available: The Knowledge-Based Economy: Modeled, Measured,
Simulated. 385 pp.; US$ 18.95
--------------------------------------------------------------------------
_______________________________________________
--------------------------------------------------------------------------
No virus found in this incoming message.
------------------------------------------------------------------------------
No virus found in this incoming message.
_______________________________________________
|
This archive was generated by hypermail 2.1.8 on Mon 30 Oct 2006 - 18:30:28 CET