John, Ideas from the following article may also be of interest in this context:
Yingxu Wang (2003) On Cognitive Informatics, Brain and Mind, Volume 4, Issue 2, pp 151-167 http://www2.enel.ucalgary.ca/IJCINI/ICfCI/JB&M-Vol4-No2-CI.pdf All the best, Gordana From: John Collier <colli...@ukzn.ac.za<mailto:colli...@ukzn.ac.za>> Date: Saturday, February 1, 2014 7:26 AM To: Bob Logan <lo...@physics.utoronto.ca<mailto:lo...@physics.utoronto.ca>> Cc: "fis@listas.unizar.es<mailto:fis@listas.unizar.es>" <fis@listas.unizar.es<mailto:fis@listas.unizar.es>> Subject: Re: [Fis] The Interaction Man Bob, Sometimes ignored in the mathematics of Shannon's approach are the coding and decoding steps, which he does not put in mathematical form, but appear in his diagrams. There has been some work in this area, the best of which I think to be Information Flow by Barwise and Seligman. It is a difficult book, and could have been a lot more clear. In any case there is a potential solution to the coding issue in the idea of infomorphisms being relations between two sets of classifications. The classifications don't have to be the same for information transfer, but they do have to satisfy certain conditions. The work is grounded in work in the 30s by engineers looking at distributed systems. My understadning is that there is a group at Stanford working on reconciling this approach with Shannon, but I haven't heard anything from them recently. Ted Gorenson, who was on fis some time ago, was giving reports. I have been making some progress of my own here, on the specific problem from my PhD thesis on commensurability across scientific theories with differing classifications. I have given a few talks on this, and will give a more advanced one at a meeting on New Approaches to Scientific Realism near Cape Town in August. Unfortunately, what initially looked promising is now leading me to some serious doubts about whether information transferred from one theoretical context to another can solve the problem, and I am going back to my thesis hypothesis that pragmatics are required to solve the problem, and that this cannot be formalized (the basis of a couple of papers I have on pragmatics -- the formal pragmatists really don't like it) I have done with a former student. Sorry for the vagueness, but this is not an easy problem, and to go into more detail would take far too much space right now. Incidentally, I had a massive hard drive problem, and lost much of my in box, hence the late reply. I hope it is still useful. John At 04:37 PM 2013-12-08, Bob Logan wrote: Dear John - I agree with your distinction between information and communication. What is essential for communication is the interpretation of the information. If I cannot interpret the information there is no communication. What Shannon leave out of his theory of signals (this is not a typo, I believe that the notion of Shannon's work as information theory is a category error) is the interpretation of the receiver. The notion that a random set of numbers is the maximum amount of information seems ludicrous to me as what interpretation can one make of a random set of numbers. John, one slight quibble. You refer to Shannon's "model of communication". How can he have a model of communication if he makes now allowance for interpretation. He was concerned with the accuracy of transmitting a set of signs from point A to point B. Krassimir wrote: "Communication is a process of exchanging of "signals, messages" with different degree of complexity (Shannon)." Without the ability to interpret the signals there is no communication. If someone speaks to me in Navaho or Mandarin they will exchange signals with me but their level of communication will be close to nil. All I will be able to infer is that they want to communicate with me. I hope that my exchange of signals is interpretable and that I have communicated with you, John and other members of FIS. The expression of this hope leads to the following thought. In an exchange between two intelligent agents who speak the same language but have made different assumptions about an issue they are discussing there is often a breakdown in communication because their interpretation of the assumptions upon which their exchange of signals are based are so different. This brings to mind I. A. Richards notion that in order for communication to occur one has to feedforward the context of what one wants to say. He once suggested that perfect communication only occurs if the two communicants have identical experiences and since this is not possible absolutely perfect communication is not possible. However one can improve one's communication by feedforwarding the context. So my feedforward to you and the FIS audience is that I worked with Marshall McLuhan from 1974 to his passing in 1980, he was a student of I. A. Richards and he (McLuhan) believed communication is effected by both the content of the message and the medium or channel by which the signals are exchanged so that "the medium is the message." all the best - Bob Logan On 2013-12-08, at 7:41 AM, John Collier wrote: At 12:38 AM 2013/12/05, Krassimir Markov wrote: Dear Pedro and FIS Colleagues, This discussion is full with interesting ideas. What I want to add is that I distinguish the concepts "communication" and "information interaction" which reflect similar phenomena but at different levels of live hierarchy. Communication is a process of exchanging of "signals, messages" with different degree of complexity (Shannon). Information interaction is exchanging of information models. It is specific only for intelligent agents but not for low levels of live mater (bio molecules, cells, organs). Main feature of intelligent agents is decision making based on information models. Information interaction is impossible without communication. Friendly regards Krassimir I would agree with distinguishing between communication and information interaction, but I infer exactly the opposite conclusion. Communication, it seems to me (and also according to the setup that is the basis for Shannon's approach) requires coding and decoding modules, but information transmission does not; it requires only a channel. Information needs to be decoded (given meaning) to be communication. At least that is what I read off of Shannon's model of communication theory. Maybe Krassimir is not talking about Shannon type information, and has a different model in mind. If this is the case, it would be nice if he were to make it explicit, since most people today at least start with Shannon's approach (or one of the two rough equivalents discussed by Kolmogorov) as basis that at least gives the syntax of information, with channel, coding and decoding required for a full communications channel, but left undefined by Shannon (though recent work has tried to fill these gaps). Cheers, John ---------- Professor John Collier colli...@ukzn.ac.za<mailto:colli...@ukzn.ac.za> Philosophy and Ethics, University of KwaZulu-Natal, Durban 4041 South Africa T: +27 (31) 260 3248 / 260 2292 F: +27 (31) 260 3031 Http://web.ncf.ca/collier _______________________________________________ fis mailing list fis@listas.unizar.es<mailto:fis@listas.unizar.es> https://webmail.unizar.es/cgi-bin/mailman/listinfo/fis ______________________ Robert K. Logan Chief Scientist - sLab at OCAD Prof. Emeritus - Physics - U. of Toronto http://utoronto.academia.edu/RobertKLogan www.physics.utoronto.ca/Members/logan<http://www.physics.utoronto.ca/Members/logan> ________________________________ Professor John Collier colli...@ukzn.ac.za<mailto:colli...@ukzn.ac.za> Philosophy and Ethics, University of KwaZulu-Natal, Durban 4041 South Africa T: +27 (31) 260 3248 / 260 2292 F: +27 (31) 260 3031 Http://web.ncf.ca/collier <http://web.ncf.ca/collier>
_______________________________________________ fis mailing list fis@listas.unizar.es https://webmail.unizar.es/cgi-bin/mailman/listinfo/fis