What's Wrong with Bandwidth?
Information is not Meaning or Understanding
Limits to Bandwidth
The theme of Limits to Mobility (read the summary volume Here) is about the parameters of interaction as they relate to the stability of the form of our interactions. That form is our ecology, whether we take that as just the network of what we know personally or a network of arbitrarily large extent. The particular issues of information, bandwidth, risk and prediction are emphasized in LTM Part I and in an essay about prediction called Prediction Needs Regulation.
My experience as a systems engineer combines work in transport systems with human/computer/communications systems. That includes transit, air traffic control and the Intelligent Transportation System (ITS). Information is inseparable from our actions and our geography, but it is more complete to call that our ecology.
The two parameters of our interactions are mobility in space and bandwidth for our logical or message interactions. The respective forms are the urban ecology and what may be called the ideology. The literal meaning of ecology is the concept of what we live in and ideology is a more direct statement that what we are in and how we behave is “all in our head”. We locate in different places and it is increasingly evident that we are polarized in our ideology. To me the most important dimension of this polarity is whether we believe our government (the “State”) is to preserve our ecology from risk and inequity or to consume it for inequitable power. Guess where we are now.
Bandwidth and mobility are parameters of the efficiency of consumption, for information and space respectively. But mobility is not places to be in and information is not understanding. The efficiency of these consumptions however affects the stability of their form. We have seen that abundantly in the urban form as transit-oriented development (TOD) ended about 1920 and was replaced by auto-oriented sprawl (AOS). We are only now recognizing the change in our social form with the bandwidth provided by the technology of individual channels we generally call the “Internet” or “social media”. The equivalent of drive-by shooting is the fire and forget flame on the social medium of your attention.
Are we unaware that mobility disrupts place and social media displace social interactions in places?
Too Much to Understand
Complexity may be defined as what we cannot understand fully enough to control. That happens among any number of people (agents) who all have their own wants and so their own ecology and ideology. We can only participate in negotiation, unless we have power to bully and dominate. The role of the State with hierarchical eminent-domain power is its own case, previously discussed regarding its imposed projects, exemplified by urban highways and urban removal.
We all have a bandwidth limit, localized around our head. We have by biology and learning a filter on all the information that can be in the channels to us. That selection is always local in space, but with communications technologies what we understand, our ideology, becomes very selective in what we know and believe.
“Communications” may be taken in its historical meaning of any exchange among people, by travel or message. That already integrates mobility and bandwidth. But after the posts were supplanted by telegraph and all the channel layers that we now use as the social media, we tend to mean the space-spanning information channels when we say “communications”. This narrowed sense has implications for our ideology and the physical ecology.
The Library of Babel
Jorge Luis Borges [The Library of Babel, 1941. Collected in Ficciones, 1956, Grove Press trans., 1962] wrote a parable about information and meaning. Scientists, mathematicians and engineers meanwhile defined information in ways that begged a relation to understanding. I have emphasized Norbert Wiener’s definition [in Cybernetics, 1948] by way of “control and communications in the animal and the machine”. Wiener was concerned with information in behavior and preservation of ecological form (homeostasis).
Better known is Claude Shannon’s codification of prior definitions for the “bit” [The Mathematical Theory of Communication, 1948]. Shannon provoked the relation between information and thermodynamic (Boltzmann) entropy. Thermodynamic entropy is also a measure of how many possible physical states (small-s state of physical attributes) are within one formal constraint of total energy-mass. Maximum entropy is least specification of a definite state (what particle or location has what attribute). Shannon made clear on his first page that:
“The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem. The significant aspect is that the actual message is one selected from a set of possible messages.”
The “selection” however is associated with intention by the message source for meaning at its destination. That agency and mutual-semiotic understanding is outside any engineering-channel model that ends at an “application” interface somewhere close to your head. What gets into your head and ideology is the mysterious process of how the perception of the bits affects your internal-network representation of the message, and cumulatively the ecology. But that depends on your selection of channels (where you are, what you perceive) that in turn makes you dependent on the message originators.
The Internet and social media include the interpersonal channels that are equivalent to a letter or telephone call. But now the form of the bandwidth is dominated by the small-worlds distribution of a few sources with many destinations. We are back to broadcast except without the regulation and public scruples of, for instance, Walter Cronkite. The effect is a the polarization of our ideology.
The physical state of the ecology is “true” as being objectively measurable. Any message from another source bears an arbitrary relation to that truth. It is “artificial data”, a term used for what AI produces but more generally applicable for anything removed from local reality. As people are separated in space by channels their relation becomes arbitrarily untrue, and so unaccountable for effect at a destination. Lies and cyber-bullying occur and for the unselective may become cyber-suicide. At least it pulls our ideology into inconsistent poles about the same ecology.
Read Nonsense
Borges’ parable supposes a very large library containing all sequences of bits. We rather imagine that as all sequences of the alphanumeric ASCII symbols. The sequences are partitioned into finite lengths that we imagine as books. The books could encode every state the universe could be in, and even states inconsistent with physics. Any formal constraint, like physics, vastly reduces the possibilities.
Given just this image by Borges, anything about entropy, self-organization, statistics, prediction and meaning can be considered. The image is after all a form in our minds that we immediately try to fit with what we already understand, our ideology.
The pure mathematics of information in the Shannon sense, and of Boltzmann’s entropy follows the same form, the intriguing -log p for information and log W for entropy. Information is about the probability p of some message element, entropy about the many ways W the state of a closed physical system can be. The mathematical form is equivalent to number size needed to enumerate all the possibilities. It is a count of the binary digits (bits) in a binary number large enough to encode all the possibilities. There is a finite information measure only for finite possibilities, as constrained by some form.
Borges does the neat trick of implying our semiotic constraint from our social evolution of language. That is from our genomic history of interaction in the ecology and so highly constrained even within physics. Then there are a finite number of books that are “meaningful”. Compared to all the random-sequence books in the Library of Babel, the meaningful texts are so few they have “measure 0” probability (in terms of Wiener’s application of measure theory to statistics and prediction. That is a key to ergodic theory that arises from Boltzmann’s entropy and that leads to the difference between randomly mixed energy-particles and the self-organized complexity of the ecology.
And yet there is so much variety because the ecology also attains the specificity that is as rare as a Shakespeare play in the Library of Babel. You can find many volumes that appear to be Shakespeare plays, some with only one symbol variant from the standard Folio edition, and so on to the plays entitled “Hamlet” but otherwise gobbledygook. This about what infinite monkeys with typewriters could produce. It is what happens from pure diffusion of ecological specificity. And there is a deep connection through ergodic theory to definitions of complexity [e.g., Greg Chaitin’s Algorithmic Complexity], deductive truth [Gödel] and self-organization of the ecology that has measure-0 probability of occurring through random permutations of particles. Then it also is a contemplation on Simon’s “Architecture of Complexity”.
If we take the ecology as the Library of Babel we will almost always read gobbledygook. We could not function in any “meaningful” way. We need the specificity—that requires immense information to describe—but it emerges from real and accountable interaction under a common form.
Too Much Bandwidth
Like mobility, more and more bandwidth is an aspiration. In server barns more bandwidth—measured as bits or some logical operations per time—means more channels and more “proof of work” (entropy production) for cyber-wealth. More, more, More! What could possibly go wrong.
For mobility, what goes wrong is the loss of accountability to space and its form of access. For bandwidth we lose accountability to each other. Mobility is not place and bandwidth is not meaning.
We do have a strong formal constraint in semiotics. We have a shared dictionary and grammar within local language modules. I have commented on the significance of modularity in complexity. This essay could not be random symbols. Few enough read it as is. It is on a channel among very many and has correspondingly few (but not by random probability) destinations. Its little network self-organized as did the message and the ecology it is in.
All that is special and specific about any meaning is in defiance of the Library of Babel and the infinite bandwidth sought to read it all, or bits needed to store it as the e-Library of Babel. We need a post-Internet Borges. What is special and specific needs its form. In self-organization that is a matter of a particular recursion between a potential field and the parts that field emerges from. That is how the original fields and part of physics become the ecology. The urban form is access and whether we have a shared ideological form is the difference between having a community or polarized factions.
I started with mobility and its effect on access form. It is also a parameter in the cooperation necessary for community. Bandwidth has an equal role. The effects of too much of either are evident in formal instability. The urban effect was obvious and historical fact. The ideological effect is becoming obvious.
The Bandwidth of the Ecology
The ecology has a huge bandwidth. It produces all the information on the ecological state that changes within all the possibilities of physics. But it is scaled. It marks every sparrow’s fall (or virus replication) and aggregate measures like the solar insolation rate and the global-average temperature.
The scaling of ecological bandwidth is in us. Remember that the ecology is our concept of what we are in. It is what we choose to resolve and measure according to our bandwidth-limited filters. The whole ecology—the eco-system that operates beyond us—produces a scale continuum of change by the densities of physical power it applies in any physical volume. As Wiener identified (concurrently with Kolmogorov, ca. 1942) in his prediction theory, the power spectral density (PSD) is the key measure of any system. Self-organizing systems generally are associated with a PSD that increases with scale: There is more power of change at larger scales.
As described in Prediction Needs Regulation this characteristic PSD is evident in weather prediction and couples the possible resolution (where it happening), accuracy of the state measure and the time interval that allows more change with more time. These are the RAT constraints that define the scale levels to which we can apply our limited bandwidths.
Is there any prescription in this? Absolutely. Our social hierarchies or management and governance scales are coupled to this RAT constraint. Your ecology is little, precise, short and of little change when you stick to what you really perceive and can act on with assurance. The State cannot and must not spread itself over all these distributed ecologies and decisions. The State should and must deal with the higher, aggregate level that is powerful but also statistically aggregated in space, accuracy and time. The State should “think big” and worry about the powerful threats to ecological form. Nor should it presume to “improve” what emerges in an orderly form.
There is nothing extraordinary in this. It is pure constitutional principle. The Constitution is our political form. How it has been ambiguated and ignored traces the change from regulating our ecology to excusing projects and other risky adventures. The failure of regulation is the sin of omission. The projects that disrupted our urban form and promote consumption—of space without place or information without meaning—are the sin of commission by our governance.
Regarding projects we too easily believe that more mobility or bandwidth excuses any “unintended consequences” on the ecology and now ideology. All we hear about is “streamlining” of those nasty regulations that inhibit getting More. NEPA and its EIS are a failed experiment of ecological accountability. No project or its analysis has the bandwidth to be accountable for all outcomes. A regard for our ecological/ideological form is accountable to no project or individual. It has to scale up and when it does it is not a matter of wanting and getting something, as we do. It is a matter of protecting form from its consumption. That is the role that Hardin’s “Tragedy of the Commons” (TOTC, 1968) begs.
The prescription of LTM elaborated in previous and future posts is about our social hierarchy of power of government and corporations. We must have regulation to preserve form instead of letting that power loose on getting More by consuming the ecology and dominating channels with its propaganda.
The increase of political power, concurrent with the intensity and extent of our ecological interactions—because we have acquired more mobility and bandwidth—begs the regulation under a constitutional form. Inherently that has to limit the use of power to its responsibility and accountability. Instead we now have a perfectly personified “I want, I get” limited to the bandwidth of the reigning egomaniac. The “regulation of commerce” clause of the Constitution somehow became an unconstrained power for mobility and bandwidth projects. The irony, and tragedy for the TOD, is how it came to suppress the railroads and left the State above and beyond similar regulation. We get the economic nonsense of “free” roads.
The rise of economic power invited regulation, at least against monopoly power, but otherwise has ignored ecological risk as long as money is made, and the most is made by the corporations most extensive on the ecology. The management hierarchy gives us the want-get egomaniacs who now dance on the political stage as well.
Bandwidth directly invites the current creation of cyber wealth. Dollars came out of exploiting physical resources. Crypto tokens are from sequences of bits in the “proof of work” of those entropy producing, and polluting, server barns. These proofs, although constrained by some deductive logic, still produce artificial nonsense and not ecological truth. The lies propagated for political or economic ends are equivalent. There will just be more of same with more bandwidth. We should rather attempt to have meaningful communities of people accountable to each other and the ecology. There are a lot of ideas for that if only you filter them from the Library. However, the volumes are slim, and rare. The meaning is up to you.