Hi all, My take is that measure theory and sigma algebra had to be introduced by folks like Kolmogorov, Lebesgue,etc to sort out issues of counting that involves discrete and continuous variables in event space. However, that is not what is required to address information as processed in physical and biological processes. It's more to do with the basic tenets of what can be distinguished by a process. The availability of colour is of little use to the colour blind. Transfer of momentum upon collision of particles is surely a casual process, but to determine the state, whether the substance is a gas or a liquid, that bears no information.
The example of wavelength limited information points to a short distance cutoff .... that's discrete. Causality is generic. Useful information is quite contextual and specific. Finding the relevant variables for description of the world (contextually defined) is always the nub. Regarding Bialek's book, there are a large number of examples, ranging from b visual systems responding to single photons, to fly embryos laying down precise spatial boundaries for protein expression that can be counted as a bit count even though the morphogen can be viewed as a continuous signal. Cheers, Sri -------- Original message -------- From: John Collier <colli...@ukzn.ac.za> Date:15/07/2014 06:19 (GMT+00:00) To: fis@listas.unizar.es,"Pedro C. Marijuan" <pcmarijuan.i...@aragon.es> Subject: Re: [Fis] FIS in Varna. Analogue Computation Dear fis members, I don't think that granularity per se is a necessary basis for the application of information theory to analog channels. In some cases it might be, and I agree that studying the relations between analog (continuous) and digital (discrete) processes is likely to be both interesting and productive. However the bandwidth of an analog channel typically can be defined even if there is no discreteness, for example if the information bearing process consists of waves so that the information bearing capacity is limited by the wavelength. Virtually all physical processes are cyclical in some way and thus have a limited bandwidth. A countercase would be a collision between particles that carries momentum from one to another. I can't think offhand right now (I just woke up), but I suspect that even in such cases there is a finite amount of information transferred. In any case, Shannon discussed the bandwidth of continuous process channels. It is worth looking at. John At 10:28 PM 2014-07-14, Srinandan Dasmahapatra wrote: >I think I agree with Joseph Brenner >here. Analogue computing is linked to real >processes, while living beings find ways of >transducing information out of dynamical states. >The graininess that information theories rely on >to define measures may be directly linked >to physical limits in the information carriers >(such as photons) or they might be limitations >of the processing organism, extracting the >sufficient "difference that makes a difference". >And yes, there's often a too hasty rush to view >analogue computing through pixellated perspectives. > >I'm not sure if this is well known to members of >this list, but Bill Bialek's biophysics text is >a profound reflection of the interplay between >the analogue and the digital, with selection >pressure forcing the sufficiency of the grainy >"difference that makes a difference" towards a >necessity for organisms, and hence pushing >sensory systems close to the physical limits of information transfer. >Cheers, >Sri > > >-------- Original message -------- >From: Joseph Brenner >Date:14/07/2014 18:12 (GMT+00:00) >To: Pridi Siregar ,"Pedro C. Marijuan" >Cc: fis@listas.unizar.es >Subject: Re: [Fis] FIS in Varna. Analogue Computation > >Dear Colleagues, > >My first reaction to this suggested project is that the logic and philosophy >of information (where I am more comfortable) would have little to >contribute. However, analogue computation is an area in which insights from >some complex theories of information might be useful. Analogue computation >has always appeared to me, perhaps incorrectly, as being closer to real >processes and therefore in principle better able to model their fuzzy, >qualitative aspects. But in some of the articles I've seen, the authors seem >almost apologetic at not being able to claim the 'power' of the digital >computer . . . > >Best wishes, > >Joseph > > >----- Original Message ----- >From: "Pridi Siregar" <pridi.sire...@ibiocomputing.com> >To: "Pedro C. Marijuan" <pcmarijuan.i...@aragon.es> >Cc: <fis@listas.unizar.es> >Sent: Monday, July 14, 2014 4:35 PM >Subject: Re: [Fis] FIS in Varna > > > > Thanks for the news Pedro. Sounds really exciting! As you might recall I'm > > interested in applications and I would be very keen on having a > > brainstorming session that would include pure researchers and > > application-oriented guys like me to explore technology transfer > > opportunities. I don't know if this could be part of some (possible) > > future agenda but I'm sure that business people may find it more than > > worthwile to attend such meetings! I'm sure Plamen would be interested > > too. > > > > best! > > > > Pridi > > > > > > > > ----- Mail original ----- > > De: "Pedro C. Marijuan" <pcmarijuan.i...@aragon.es> > > À: fis@listas.unizar.es > > Envoyé: Vendredi 11 Juillet 2014 14:41:42 > > Objet: [Fis] FIS in Varna > > > > Dear FISers, > > > > The fis summer conference in Varna just took place 5-6 July --our 20 > > years of activities were celebrated too, FIS 20th. Rather unfortunately > > not many people attended: half dozen from Spain related to Juan > > Castellanos and me (from Madrid and Zaragoza); and a few parties around > > Krassimir from Bulgaria and Ukraine. But we had a great time > > (discussions and exchanges, banquets, beach) and the place is really > > beautifull & prices quite affordable. The idea, quite possible to > > realize, is that every year that we do not have a plenary fis or isis > > conference, we arrange a small summer school in Varna. > > > > Among the exchanges this year, the retinue of "basic concepts" around > > information generated the most intense debate--is there any concept > > prior to information? Joseph's contribution was also discussed by > > Krassimir addressed to the Russian colleagues (in Russian). Computer > > related ontologies, new schemes to handle Big Data, and brain > > exploration through AI and EEG by a very advanced Egyptian team were > > quite exciting discussion topics too. For the future, we think that > > spinoff companies could be enticed to participate, developing new > > products and taking profit from some of those initiatives. In any case, > > the interaction with brilliant ITHEA colleagues from Bulgaria, Russia, > > Ukraine, Armenia, Belarus, Egypt... is a valuable experience itself. > > > > And that's all! > > > > best wishes---Pedro > > > > -- > > ------------------------------------------------- > > Pedro C. Marijuán > > Grupo de Bioinformación / Bioinformation Group > > Instituto Aragonés de Ciencias de la Salud > > Centro de Investigación Biomédica de Aragón (CIBA) > > Avda. San Juan Bosco, 13, planta X > > 50009 Zaragoza, Spain > > Tfno. +34 976 71 3526 (& 6818) > > pcmarijuan.i...@aragon.es > > http://sites.google.com/site/pedrocmarijuan/ > > ------------------------------------------------- > > > > _______________________________________________ > > Fis mailing list > > Fis@listas.unizar.es > > http://listas.unizar.es/cgi-bin/mailman/listinfo/fis > > > > _______________________________________________ > > Fis mailing list > > Fis@listas.unizar.es > > http://listas.unizar.es/cgi-bin/mailman/listinfo/fis > > > >_______________________________________________ >Fis mailing list >Fis@listas.unizar.es >http://listas.unizar.es/cgi-bin/mailman/listinfo/fis ---------- Professor John Collier 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 http://listas.unizar.es/cgi-bin/mailman/listinfo/fis
_______________________________________________ Fis mailing list Fis@listas.unizar.es http://listas.unizar.es/cgi-bin/mailman/listinfo/fis
