Replies

But when asking the underlying question “how much data are needed to ascertain information?” there are no non-physical aspects. There may be gaps that require inference, but no model contains “this gap bridged by God.” A representation of an object (e.g. a relational variable versus a relation) is far from a non-physical entity. It has what we call “informational heft” — meaning from a physical Universe in a physical Universe.

Sorry not to have gotten back sooner, Freedumb2003; I spent yesterday with my folks.

I have some questions regarding the passages in the above italics. It's good to know you're in the information processing business, since that makes you a good "go-to guy" for answers.

Do you consider data as physical quantities? It seems to me data are outputs of inputs that must be in "computable" form in the first place. Where you seem to see the "physical" here, I see only the formalism of input, output, and computability. Plus the thought occurs to me that not all problems within the human sphere are reducible to "computable" form.

How does data get translated into information, absent a subjective mind to do it? Is this the "God of the gaps" to which you were referring? But we're not even speaking of God here. We are speaking of subjective human intelligence, and whether or to what extent it has any right to be present in "science" today.

If that sounds silly, just consider: Science has relentlessly been purging the "subjective" ever since Francis Bacon. No science that is not completely "objective" can pass muster as science under this regime.

And yet the transition from data to information would seem to require a mind to effect it, the outcome of which Shannon referred to as "the reduction of uncertainty in the receiver." How can one have a model of information without taking the receiver into account? Or the sender, for that matter?

It seems to me that to say that something has "informational heft" is to speak analogically. It does not confer actual physicality on anything; it is little more than a figure of speech.

Just sending my questions/observations along to you, dear Freedumb2003, in this fit of "amorphous musing" I seem to be having....

Thank you ever so much for writing!

And yet the transition from data to information would seem to require a mind to effect it, the outcome of which Shannon referred to as "the reduction of uncertainty in the receiver." How can one have a model of information without taking the receiver into account? Or the sender, for that matter?

I think you need to re-read Shannon ...

Thank you so very much for sharing your insights, dearest sister in Christ!

I suspect the stumbling block on the receiver being informed is that the receiver in Shannon's mathematical model of communications can be a mind or spirit just as surely as a molecular machine or digital receiver or component in computers, robotics or software including artificial intelligence and so on.

It seems to me that to say that something has "informational heft" is to speak analogically. It does not confer actual physicality on anything; it is little more than a figure of speech.

Precisely so!

Shannon's is a mathematical model and for that very reason it is transportable to analyze communications of all kinds.

Some argue that Shannon's theory while obvious applicable in the discrete case can result in negative Shannon entropy (increase of uncertainty) in the continuous case and therefore is somehow inadequate.

I do not see this as a fault but the logical conclusion of its being applied beyond telecommunications or computing, e.g. Rosen's relational model for biology which is circular.

And in a very real sense, when I (receiver) read (decode) a letter (message) sent via USPS (channel) written to me (encoded) by a sender (author) my uncertainty (Shannon entropy) may be increased rather than reduced! LOLOL!