Pages

Monday, May 25, 2009

Networks

Networks are everywhere

The study of networks and the related branch of mathematics, graph theory, have received an increasing amount of attention during the past few years. It's a highly interdisciplinary area, connecting the natural sciences with the social sciences, maths and the computer sciences. Did I leave out anybody who is interested in anything? Not so surprising, since networks can be found everywhere throughout our closely connected world: from computer networks and electrical power grids, to social networks, neural networks, to food webs - even the bible has a network:
[Click to enlarge. Via.]


Just consider the amount of different networks woven through your life: there's the route maps of airlines, the internet and its virtual link-structure, the world wide web (for example the blogosphere), the production and distribution processes of consumption goods, electricity networks and, of course, the sexual networks. And though they are different in many aspects, they share a similar underlying fundamental structure that can be mathematically captured and analyzed. Needless to say, with our rapidly improving capacities for storing and handling large amounts of data, the study of networks has flourished tremendously within the last decade.

What is a network?

Pictorially speaking a network is a collection of dots connected by lines. Physicists tend to call the dots nodes and the lines links, mathematicians call it a graph with vertices and edges, but it's the same thing really. The number of connections a dot has is also called the 'degree' of the node or, if you prefer, the 'valency' of the vertex. The great beauty of networks is their generality. From such a simple mathematical description one arrives at a great variety of structures and phenomena. 

There are different properties a network can have:
  • The links between the nodes can just be connectors that are on or off, or they can be arrows indicating a preferred direction, called a 'directed graph'. Links on websites for example have a direction, friendship networks on Facebook don't.
  • Both the nodes and the links can carry information, eg about what is produced in the nodes or transmitted in the links.

There are two important special types of networks:

  • Random: A random network is one in which from a collection of nodes pairs are picked randomly and connected. 
  • Scale free: A scale-free network is one in which there is a specific relation between the number of nodes and their degree, called a 'power law'. In these networks, there are a lot of nodes with only a few links, and a few nodes with a lot of links - these are also called 'hubs'. Real world networks, eg the www or airline networks, turn out to be often scale-free (to some approximation) with consequences for their robustness and vulnerability e.g. the spread of viruses or the dissemination of information.
Other important information in a network's structure are highly connected clusters with sparse connections (gaps) between them (there are different algorithms for their identification), and flows around loops. Particularly interesting are growth mechanisms and phase transitions that can occur with that growth.
Network Science - what is it good for

So why am I telling you that? Because the analysis of networks can help us to understand many aspects of the inanimate and animate world whose interdependence obscures local analysis. In particular, the growth of specific structures and the dynamical properties of networks play an increasingly important role for managing large scale effects. Understanding the conditions necessary for resilience - may it be of a social, economical or ecological network - is essential to ensure stability of these networks that are vital parts of our lives.

While my enthusiasm about "complex systems" is limited due to the vagueness and ambiguity of a lot of of this research, network science is its backbone. Needless to say, there is some overenthusiasm also in this area. One thing I would like to know better for example is what the limits are of modeling systems as networks. With enough abstraction, I can probably describe everything as some sort of graph. But under which circumstances is that insightful?

This post was inspired by last week's colloquium by Raissa D'Souza from UC Davis on "Growing, Jamming and Changing Phase" that you find on PIRSA 09050004.



Further Reading

53 comments:

  1. Great post! It would be interesting to know whether using the results from this field could improve the professional (e.g. academic) networking skills which in the today's economic turmoil are to play an ever more important role in one's career.

    ReplyDelete
  2. Hi Researcher,

    I don't know about the networking 'skills,' but I find it certainly possible that the available tools become more useful and more widely used. Presently it is probably mostly those who work in fields only few people work on who profit the most, but the vast majority still has to be convinced that improving the network structure can be advantageous. Best,

    B.

    ReplyDelete
  3. It is most certainly nice to be helped out by the younger generation, to see in ways not thought of before. A whole new layer to thinking? Where did the kernel begin? It then spreads by equatorial design.

    Although the perceptron initially seemed promising, it was eventually proved that perceptrons could not be trained to recognise many classes of patterns. This led to the field of neural network research stagnating for many years, before it was recognised that a feedforward neural network with two or more layers (also called a multilayer perceptron) had far greater processing power than perceptrons with one layer (also called a single layer perceptron). Single layer perceptrons are only capable of learning linearly separable patterns; in 1969 a famous book entitled Perceptrons by Marvin Minsky and Seymour Papert showed that it was impossible for these classes of network to learn an XOR function. They conjectured (incorrectly) that a similar result would hold for a perceptron with three or more layers. Three years later Stephen Grossberg published a series of papers introducing networks capable of modelling differential, contrast-enhancing and XOR functions. (The papers were published in 1972 and 1973, see e.g.: Grossberg, Contour enhancement, short-term memory, and constancies in reverberating neural networks. Studies in Applied Mathematics, 52 (1973), 213-257, online [1]). Nevertheless the often-cited Minsky/Papert text caused a significant decline in interest and funding of neural network research. It took ten more years until neural network research experienced a resurgence in the 1980s. This text was reprinted in 1987 as "Perceptrons - Expanded Edition" where some errors in the original text are shown and corrected.

    More recently, interest in the perceptron learning algorithm has increased again after Freund and Schapire (1998) presented a voted formulation of the original algorithm (attaining large margin) and suggested that one can apply the kernel trick to it.Perceptron

    ReplyDelete
  4. I find this subject really fascinating. Do you take reader requests? More posts on networks!

    ReplyDelete
  5. Yes, sometimes I take reader requests. I'll consider that an encouragement to dump networky stuff here that crosses my way :-)

    ReplyDelete
  6. Here is how I understand the notion of a network in a broader sense. Everyone and everything is part of some short of a network because nobody and nothing is self contained or self defined and thus is forced to interact to acquire a substance. Only God is self contained because he contains everything. God does not need to interact to be. Similarly only God is self defined. That's why when Moses asked God for his identity, he replied "I am what I am"; exactly because God cannot be defined with the help of any other notion. Brilliant! I wonder how people 3000 years ago were able to grasp that beautiful idea.

    ReplyDelete
  7. Hi Giotis,

    I think you are confusing a network with an open system. To speak about a network, you need a clear definition (or at least interpretation) of nodes and links to begin with, and it's far from clear you can always achieve that. It is true however that there is in reality no such thing as a closed system, so any such notion is only approximate. We are always part of a larger system unless you want to speak of the whole universe (or multiverse, or god, depending on taste). Best,

    B.

    ReplyDelete
  8. missing:

    Lilith,
    Satan,
    God,
    ...

    best,
    A.

    ReplyDelete
  9. Bee, may I suggest 2 essential things you did not mentioned. (You mentioned dots and interconnections.)

    1) Transformation. How one type of network transforms into another. While mathematically interesting, this is far more interesting when placed in a physical setting where transformations consume energy and hindered by various factors.

    2) Pathways, i.e. routing of information. This happens for both static networks and networks undergoing physical transformations. The latter presents the ultimate challenge theoretically and engineering-wise. The Internet is of this type - routing pathways change dynamically and quite fast, while the physical 'graph' also changes but less frequently. The working dynamics of the Internet core (i.e. excluding the web), which is controlled by many dozens of complex protocols within its backbones, represent one of man's greatest interdisciplinary achievements in the same order as inventions such as the aircrafts and spacecrafts.

    ReplyDelete
  10. Hi Tkk,

    Thanks. What do you mean with 'transformation' of 'network types'? I mentioned phase transitions? Best,

    B.

    ReplyDelete
  11. "What do you mean with 'transformation' of 'network types'?"

    Using networks we now use so commonly, let me list different types of physical networks:

    - the telephone network (a point to point type topology of simple wires carrying analog voice signals)

    - the TV cable network which also carry Internet connections (a tree type topology of coaxial wires carrying analog HF modulated signals)

    - the Internet core (a meshed topology of optical wires carrying modulated photon waves)

    - the cellphone (or wi-fi) network (a point to point topology of ultra UHF RF transmissions carrying digital signals)

    Clearly, all of the above couldn't be more different from each other. Now imagine you sit in Toronto airport and performs an update of Backreaction web server from your PC wirlessly. The information you sent is packaged in some protocols in your PC; these protocols assume one type of network. Yet it get routed, across all of the above physical network types and topologies, successfully. Along the way, there is vast amount of transformations, both in the controlling protocols and physical adjustments (especially at the interfaces between different networks).

    In fact the actual information you sent is broken apart into 'fragments', sent in different paths through all of the network types, and reassembled upon arrival, even when the bits may arrived lost, damaged, or out of order from traveling through so much physical and logical transformations. It also possible that some of the fragments, while traveling to the web server, will encounter a lost connection, a frequent event for wi-fi connections. This is dynamical change of pathway while carrying a signal. Yet the entire signal will arrive with zero error.

    So I am not talking about phase transition.

    ReplyDelete
  12. "
    - the telephone network (a point to point type topology of simple wires carrying analog voice signals)"

    Not quite right, the simple wires meet in a hub where switching takes place. The simple wires carry both the payload (voice) and signaling; the signaling may be broken out into its own network once it reaches the hub. Alternately, the signalling network can begin at phone (e.g., ISDN PRI).

    Second: protocol translations are not the same as network transformations. An example of what might qualify as a network transformation is e.g., the transformation of a frame relay point-to-point mesh network into a IP-enabled Frame Relay any-to-any VPN, with a hub-and-spoke type of structure, with the VPN being the hub.

    ReplyDelete
  13. Hello, it's probably false to say "the www [... turns] out to be often scale-free". See: W. Willinger, D. Alderson, J. C. Doyle, Mathematics and the Internet: A Source of Enormous Confusion and Great Potential. Notices of the AMS, Volume 56, Number 5 for an extended and up to date discussion.

    ReplyDelete
  14. Hi Bee,

    As you have demonstrated in this article the study of networks is gaining momentum as of late, yet really what’s the difference between referring to something as a system or a network other than to emphasizes a little more strongly the connectivity aspect of what one is trying to understand. I must say I was also impressed by the talk you pointed to given by Raissa D’Souza, for it certainly was an excellent synopsis of the subject’s history and where it is headed.

    I wouldn’t pretend to claim I understand all or in depth what she brought forth, yet the overall gist is they are still struggling to define and qualify the fundamentals of network science as it’s called. I was particularly intrigued about the differences in opinion that exists as to what are the core aspects involved in the ordering of networks as exemplified between Simon and Mandelbrot.

    Unfortunately this is where she abbreviated the lecture for lack of time, which was unfortunate as it appears her group has made some headway in this area as of late in forming some sort of synthesis between what appeared before as opposing views. I must admit that when it comes to the nature of things that Mandelbrot’s position that much is driven by economy I feel as something that can’t be ignored. If she ever is to give another recorded lecture that expands upon all this I would be grateful if made aware of it. I found her to be a excellent speaker and her enthusiasm for the subject is nothing short of infectious.

    Best,

    Phil

    ReplyDelete
  15. Hi Anonymous,

    Thanks for the reference, I will look at this. Real world networks aren't truly scalefree anyways. Best,

    B.

    ReplyDelete
  16. Hi Phil,

    Well, I think describing something as a 'system' is always possible. It is almost tautological to call something a system. It only becomes scientifically useful if you add qualifiers (closed system, open system, complex system, self-aware or reproducing system etc) and system properties (feedback, circles, etc) that allows you to classify behaviors.

    As I said above, it is a point I don't quite understand, but it isn't clear to me under which circumstances a system can be described as a network in a meaningful way. Take eg a self-aware system (like, Phil Warnell). There are certain properties you might be able to describe as a network (organs? chemical reactions?), but I'm not sure it is always possible to disentangle them and still have a complete picture. Or take some typical physical experiment like the interferometry we were discussing recently. In what sense is that a network? Does that provide any useful picture or interpretation?

    Raissa mentioned in her talk a similar question as one she is presently working on, how to disentangle the different layers of real world networks (biological, transport, social) etc.

    Best,

    B.

    ReplyDelete
  17. I used the term in its broader sense following the reasoning that any entity that is not standalone is part of a network and interacts.

    Nobody and nothing is a standalone entity except the abstract theological notion of God.

    In theological terms the Universe is a derived concept (not a standalone entity) because it needs God to acquire a substance or to have a meaning.

    Now that I'm thinking about it, I wonder whether theology is a science. It has a theroretical structure after all. Something like mathematics.

    ReplyDelete
  18. Hi Bee! Thanks for answering my comment. Your remark about the people in scarcely populated subfields benefitting the most sounds interesting, but I'm afraid not too many people in such subfields use social networks. Should think this over :) Best, S.R.

    ReplyDelete
  19. Hi Researcher,

    Why do you think so? I have noticed repeatedly that people who work on very specialized fields tend to know very well the other few people on the globe who work on similar things. That isn't so surprising, is it? It's the people who have sufficient others in their vicinity who don't see a reason to look farther than that. I would agree that they don't (yet!) use social networks in any organized way, but it is becoming more and more common to set up a 'virtual' meeting point for people with common interests. Best,

    B.

    ReplyDelete
  20. Hi Giotis,

    "any entity that is not standalone is part of a network and interacts."

    What I was trying to say is that I would agree on 'any entity that is not standalone is part of a SYSTEM and interacts,' but why a network in particular?

    Best,

    B.

    ReplyDelete
  21. Because the entities interact in a certain way. You could call it a protocol.

    ReplyDelete
  22. What I'm trying to say is that a network or a protocol etc presupposes you can decompose your system into discrete entities that are connected via other discrete entities through which they do exchange something, and that the behavior of the system is not only entirely contained in that decomposition but useful in addition (let's not even mention quantization). If you want to go so far to say every differential equation is a protocol and can be written as a network, I think then I would agree with you, just that in this case graph theory seems somewhat an overkill.

    ReplyDelete
  23. Hi Bee,

    of course the people in small subfields mostly know each other, what I wanted to stress was indeed the insufficient use of social networks. Some people are incredibly conservative :)

    Speaking of the virtual meetings, do you know any reasonably easy-to-use solution for writing formulas if some of the parties have no access to the computer writing pads? I mean, something for talking (e.g. like Skype) combined with the possibility of writing formulas (say, in LaTeX at least at the level available at certain (e.g. WordPress) blogs)?

    By the way, do you think we can expect larger-scale virtual conferences (say, in the high energy physics or quantum gravity) any time soon?

    This would be great because if one wants to branch out or change (sub)fields, it can be quite difficult to get funds for attending the conferences in the new (sub)field.

    Thanks in advance for your input.

    Best,
    S.R.

    ReplyDelete
  24. Hi Researcher,

    These are interesting points you are raising. As to the equations: Needless to say this problem has crossed my way from time to time. Meanwhile however the resolution of webcams is good enough and the bandwidth is high enough to just write on a piece of paper and hold it in the camera. If you write large enough that is and if it's not too long or too many equations. It works better for diagrams.

    The problem with tools, say, for drawing etc, is that they are pretty useless without a magnet pen. If you've ever tried to write with a mouse you know what I mean.

    As to conferences. I think that we will see increasinly mixed conferences where people attend or speak by video-audio-link. The technology is by now good and stable enough to allow that. I don't think however that it will become common to have conferences specifically designed this way. One of the best reasons to organize and attend conferences is still to just meet people face to face. There are however some folks holding conferences in SecondLife which is kinda funny. I think there is some niche for that. Still again I think it will be the exception rather than the rule.

    Best,

    B.

    ReplyDelete
  25. @Arun:

    " Not quite right, ..."

    Of course not quite right. How can one correctly give a technical description of the vast complicated phone network in one sentence? My purpose is to simply list there are many types of communications networks.

    "Second: protocol translations are not the same as network transformations."

    Protocol translation is not transformation. A good example of transformation is at the Internet exchange points (such as various MAE facilities). Study MAE architectures around the world and you know what I mean.

    "the transformation of a frame relay point-to-point mesh network into a IP-enabled Frame Relay any-to-any VPN, ..."

    This is more like protocol translations. Frame relay SVCs are used to map TCP VCs to make VPNs. All these are protocol manipulations. The underlying physical layers have not fundamentally changed.

    I simply want to point out that today's Internet is a prime example of managing network pathways and transformations successfully. The Internet exchange networks around the world do consume a great deal of energy/resource and exhibit unique constraints not found anywhere else. Building and managing these most essential and complex exchange networks do represent a high challenge that millions of Internet users do not even know they exist!

    However, a blog like this is no place to discuss any technicality in-depth.

    ReplyDelete
  26. Hi again Bee,

    and thanks for commenting. You're certainly right about the conferences mostly surviving in the classical form, my speculation was probably a bit too wild :)

    By the way, do you think that linking to one's publications in one's social network profile can significantly increase academic visibility (say, in terms of citations or getting invitations to speak at the conferences)? Or are there some smarter ways of enhancing academic visibility using social networks?

    Best, S.R.

    ReplyDelete
  27. Hi Researcher,

    "do you think that linking to one's publications in one's social network profile can significantly increase academic visibility (say, in terms of citations or getting invitations to speak at the conferences)? "

    While I think it does increase academic visibility (in that the effect is at worst none) I think that whether or not it has a significant effect depends on, well, the quality of your work. That is to say while scientific networking can help you to spread the word, it doesn't spread if it's not worth spreading. Presently, if you're a scientist into Web2.0 and good with software development and/or have lots of experience using such, it gets you invitations to conferences, that being those conferences dedicated to the topic. What scores much higher than adding publications to one's profile is still getting in contact with and having an interesting exchange with others. Best,

    B.

    ReplyDelete
  28. Hi Bee,

    Thanks for the reply. I guess one of the distinctions between networks and systems is this notion of information as to what underlies everything. In the system picture information transfer is more of a consequence, rather than an objective or purpose for lack of a better way to put it. There are of course metaphysical underpinnings to this network picture, which on the face of things are not so evident which places limits on structure and dynamics which are in a sense predetermined or at least determinable. The question then to ask is this how nature works or rather is it simply a reflection of how we would like it to work?

    Best,

    Phil

    ReplyDelete
  29. AcademicLurker8:52 AM, May 28, 2009

    Bee,

    Hi. I've been occasionally reading your blog for a while now, but haven't commented before.

    I've also been interested in network theory, particularly in ways to coarse grain networks based on preserving various dynamical properties.

    On the subject of over-enthusiasm, there was in interesting and rather caustic article by Evelyn Fox Keller a while back:

    Revisiting "scale free" networks
    BioEssays vol 27 pp 1060-1068 (2005)

    ReplyDelete
  30. Hi Lurker,

    Thanks for the reference, will give it a look. It was also entertaining what Raissa said in her talk about folks who see power-laws everywhere :-) Best,

    B.

    ReplyDelete
  31. When a living, real system dies (rigorous meaning available), it becomes that kind of “network”, close to what's implied in their “network science”. This is what has happened, by the way, to (official) science today: it used to be a living, creative system once (I still remember those happy days) and now it's only a network of corrupt, selfish “relations” determining everything and leaving no place for creation by real problem solution (“life”). It's not surprising that this dead entity gives preference to equally dead, simplified, unreal abstractions, such as that standard network theory (a post-modern version of statistical physics?). It is easily disproved as a whole by an elementary comparison with reality, where an arbitrary new “network element” can appear “unpredictably” and through the well-known property of “amplification/propagation of perturbations” (omnipresent instability) completely destroy the “averaged” statistical imitation. It just happens like that all the time in real life, which is opposed by its very essence favouring everything “individual” and “special” to statistics and statistical (artificially simplifying) averaging. Those dead, fixed-structure network-systems may also represent an interest of course, but a very limited one and which in reality can always be “interrupted” by “living” interventions and influences that instantaneously “kill” the disproportionally inflated simplification of network theory. But take it easy, of course, Raissa and other girls from the network :) : the talk looks amusing, the really advanced post-modern science, with exciting pictures, charming presenter and even some “tricky formulas” on the background (of which nobody really cares!) to show that in addition to and despite the above, it's serious science! The only disappointment is the standard statement at the beginning that (consistent) definition of complexity is absent being replaced by arts and letters, as usually. If you were visiting this blog more often, Raissa (check my links anyway), you would know that absolutely universal and consistent concept of (dynamic) complexity in arbitrary systems exists for more than a decade already and is multiply confirmed by successful applications to various systems, including real (rather than abstract), living and naturally creative “networks” (in that case they are not really different from “systems”). Happy networking to everybody, but don't forget to stay alive, despite everything...

    ReplyDelete
  32. @tkk, needless to say, I disagree that the change in topology via moving from frame to IP is merely a protocol change.

    @giotis: ""any entity that is not standalone is part of a network and interacts." - how do I understand the Sun and the planets as a network?

    ReplyDelete
  33. Hi Bee,

    The one network that certainly can’t be denied is the one we are currently communicating across. It’s interesting to note that as science struggles to better understand the application and utility of networks, their existence has already become a battle ground for some. It leaves one to wonder if there is nothing that the human network conceives that can be left untouched by its flaws. I think if any focus should be given to this emerging aspect of science, it should be to address the network which is in the greatest need to be understand so it might be improved as not to have as many flaws.

    Best,

    Phil

    ReplyDelete
  34. Phil Harmsworth9:08 AM, May 29, 2009

    This TV documentary may also be of interest:

    http://www.abc.net.au/tv/documentaries/interactive/futuremakers/ep4/

    ReplyDelete
  35. Yes why not, you can think of them as a network. The defining property of a network is communication. I'm telling you something, you understand it (by using certain rules i.e. a protocol) and you respond accordingly. So you can think of the sun and the planets as a set of entities communicating via the metric by using as a protocol the GR field equations. You could say that they communicate with the language (protocol) of GR.

    ReplyDelete
  36. The spacetime is not merely a protocol. It is a dynamical entity itself. Thing graviational waves. If you follow up on that, you'll come to the conclusion that the only way to make sense of it is to reformulate the field equations as mediating (protocol) between the sources (eg planets) and the spacetime (metric, curvature, etc). As I said above, I think you can probably generically do that by taking all the input into a differential equation and declare it 'nodes' and taking the differential operator itself and declaring it 'protocol,' yet I doubt doing this is of any practical use.

    ReplyDelete
  37. Bee, I didn't say that the metric is the protocol, I said that the field equations are the protocol. Read again.

    I said that the entities communicate and interact via (and with of course) the metric.

    I never claimed that this has a practical use. I was speaking theoretically.

    ReplyDelete
  38. Sorry, I mistakenly interpreted 'communicating via' as 'that's the links'. I think we mean the same then.

    ReplyDelete
  39. Communication is not the defining property of networks - I don't see any communication in the network Bee has in the original post. There are relationships there, but no communication.

    Communication involves the transfer of information. What information are the celestial bodies exchanging? Surely you should be able to quantify it a la Shannon? I wait with bated breath.

    Spacetime - the bearer of the metric - can hardly be represented as a link connecting two nodes. What two nodes?

    I guess silliness is the name of the game.

    ReplyDelete
  40. Giotis:
    " sun-planet ... network"

    I am on Bee side. GR field equations, nor space-time, are protocols. I think all agree that space-time is matter-energy, a real thing that interacts with other matters. (F Wilczek calls it the Grid.) GR equations is human's limited and partially-correct summarization of this physical reality. Other advanced civilization will understand space-time with their 'equations', which likely to be quite different depending on how much or less their 'equations' cover the space-time reality compared to GR. We're tying to do more with QG are we not? GR, or any such equations, are not protocols. And therefore the universe is not a network.

    ReplyDelete
  41. Arun, There is no need to resort to characterizations and irony to make a point.

    You asked me something and I tried to give you an honest answer. You should have respected that.

    As you understand I can't continue talking to someone who doesn't respect some basic rules of social behaviour.


    Tkk, Yes you have a point there.

    ReplyDelete
  42. Hi Arun,

    “Communication is not the defining property of networks - I don't see any communication in the network Bee has in the original post. There are relationships there, but no communication.”

    I would agree with Giotis that the whole metaphysical underpinning of the network concept ia communication or rather information transfer if you prefer. In Bee’s example it was the concept and proliferation of a certain conception of a specific deity that was being propagated and shared in common.. In the planetary network example the information being transferred is the distance, location and mass/energy of all the nodes in relation to each other that are being communicated and acted upon in accordance.

    Towards the end of his life Einstein himself proposed that matter/energy and space-time where not separate entities, yet rather extensions of one another. I’m not contending this picture be true, yet rather conceptually consistent with what we know. You could say for instance this is one of Smolin’s central problems with string theory being background dependant, while loop quantum gravity is not and thus more in line with the underpinnings of GR. Also, as I eluded to earlier QM today is looked at more from an information standpoint then had been in the past. Personally I think this information perspective is over extended yet it certainly would be a stretch to contend that it can be so easily summarily dismissed,

    Best,

    Phil

    ReplyDelete
  43. Giotis: you wrote something that doesn't make sense and can't defend it, so you invoke social behavior.

    Phil: "the whole metaphysical underpinning of the network concept ia communication or rather information transfer if you prefer." The underpinning of networks is pair-wise relationships of any kind - with or without communication.

    It doesn't even hold in Bee's example. Where is the propagation of a certain notion of deity between Jesus and Jesus' brother?

    Wherever physics may be going with information, until you can give me a information-theoretic description of gravitational motions, saying it is communications is absurd.

    In the case of the solar system, think of the sun and planets as nodes, and do what Giotis wants to do, draw lines from each of them to spacetime - that is his network. Problem is that spacetime doesn't behave like any other node. It is itself expandable into nodes and links and so expanded till it turns into a continuum. It is easier to visualize the solar system as a network with Newton's action at a distance gravity.

    To give you another analogy, broadcast television is called network television not because we are all immersed in the continuum of radio waves, but because various television stations are networked together to broadcast the same programs. Nobody sees a network between the broadcast stations and the innumerable receiving television sets. Likewise, this continuum of spacetime is not a network; and heroic efforts to model it as such have so far failed.

    ReplyDelete
  44. Likewise consider the "cellular network". Here is the wikipedia definition "A cellular network is a radio network made up of a number of radio cells (or just cells) each served by at least one fixed-location transceiver known as a cell site or base station. These cells cover different land areas to provide radio coverage over a wider area than the area of one cell, so that a variable number of portable transceivers can be used in any one cell and moved through more than one cell during transmission."

    The communications happen between the portable transceivers - but the network consists of the fixed transceivers and the communications facilities between these fixed transceivers. The network, in common usage, **does not** include the portable transceivers (e.g., "Verizon's cellular network is down"; We telecom guys debate putting intelligence in the network or in the handset (the portable transceiver). But the network is not the initiator of communications. Nor is the cellular network plus the handsets in operation conceived of as a network in its own right.

    ReplyDelete
  45. Your attempting to connect neurons?:)LOL

    For some people, it just doesn't work beyond the obvious.

    Who was Jesus's brother?

    The points mute if you recognize that the human brain is indeed diverse with it's characterizations and final summations. Glaucon or maybe a Phaedrus perhaps, to write a counter view to move forward in progression.

    How widely can you link those "same neurons" while offering a network that is listed to so many characters by comment then what can be thought of, by writing your own story?

    You want to prove a point right? Lead us to think of, and see perspective in new ways? A perception you found?

    Best,

    ReplyDelete
  46. Hi Bee,

    if you don't mind, I'd like to come back to your sentence "What scores much higher than adding publications to one's profile is still getting in contact with and having an interesting exchange with others."

    Do you mean here using social networks or "live" socializing at the conferences etc., or both? And in the first case, I wonder whether using social networks brings about something radically different from, say, contacting the people you don't know in person via e-mail (by the way, are there any ways to increase the chances of getting response to such e-mails, in addition to being polite and up to the point?).

    Thanks, S.R.

    ReplyDelete
  47. Hi Arun,

    As I said from the beginning I’m not myself convinced that the network picture as you envision it without having a dynamic component is equivalent to referring to things as systems. Yet when the dynamics is added the difference between the two begin to blur. I would concede that the network outlook tends to capture mainly the structure (order) yet the action is implied for without it the structure itself cannot manifest to form. This is where one could contend the information resides, that is between the order and the form. There has always been this struggle concerning the kinetic and dynamic perspectives, where one or the other thought to be the better context for describing as to understand what is real. All I’m saying it’s sometimes difficult to distinguish what constitutes being the difference between a node and a link without considering what happens between them .

    Best,

    Phil

    ReplyDelete
  48. Hi Phil,
    Not sure I follow. Anyway, maybe time will bring clarity.

    -Arun

    ReplyDelete
  49. Hi Researcher,

    "Do you mean here using social networks or "live" socializing at the conferences etc., or both?"

    Both.

    "And in the first case, I wonder whether using social networks brings about something radically different from, say, contacting the people you don't know in person via e-mail (by the way, are there any ways to increase the chances of getting response to such e-mails, in addition to being polite and up to the point?)."

    I haven't noticed something radically different, except maybe that the level of formality of the exchange differs drastically from one person to the next. For some the written word is what you publish in, thus has to be flawless and accurate. For others writing is considerably less formal and not much different from a casual verbal exchange.

    As I said above, what you can do to increase your chances of getting replies is to write about something interesting for the person you are contacting.

    This is not addressed specifically at you, but there seems to be a very large fraction of people who completely misjudge how 'interesting' it is what they might have to say. This is a very extreme example.

    Best,

    B.

    ReplyDelete
  50. One of the biggest networks of all is the set of web-pages accessed by URLs.

    The network is -
    1. Each web-page has a URL.
    2. Each web-page refers to the URLs of other web-pages.
    A simplified dynamics is -
    1. A person creates a web-page, and thus a URL.
    2. A person links to other web-pages, perhaps more likely to those found by search engines.
    3. Search engines catalog web-pages.
    4. Search engines rank a web-page, by among other things, the links to the web-page.

    None of the dynamics resides in the network, in that web-pages and URLs are static things, with no activity of their own. If this network embodies communication, the communication is at the level of the users of the network; the network by itself does not transmit or receive information.

    ReplyDelete
  51. This comment has been removed by the author.

    ReplyDelete
  52. Hi Bee,

    While poking around the newly inspired Whitehouse website this weekend, I came across a piece on their blog that reminded me of this post of yours . It seems that Obama and his people find some utility in looking and dealing with problems regarding organization using networks as the methodology of choice. This one all has to do with the promotion and implementation of the transparency in government.

    I must say I'm impressed that Obama seems certainly determined to utilize all means at his disposal in accomplishing this goal. However, despite all the bells and whistles one still gets the feeling that it lacks the input that is the most important and that being direct feedback and the transparency to be found only in open dialogue. It’s one thing to know what they are thinking, yet another it be assured that people know they are being heard. That is despite all the carefully thought out nodes and links, it’s not transparent to me how this will serve to accommodate this most vital component or can I find a node or link that stands to directly address this.

    Best,

    Phil

    ReplyDelete

COMMENTS ON THIS BLOG ARE PERMANENTLY CLOSED. You can join the discussion on Patreon.

Note: Only a member of this blog may post a comment.