Political Entropy

Entropy is a strange word. If we look at some definitions, the strangeness becomes apparent through a list of synonyms: chaos, uncertainty, equilibrium, stasis, homogeneity. I'm going to use it to show that public participation in the policy process is inevitable. Unavoidable. A sure thing.  We need to get ready for it.

In thermodynamics, where it all started, entropy is a measure of the uniformity of energy distribution within a system - higher entropy means more uniform distribution. John von Neumann is reckoned to have told Claude Shannon to name his measure of uncertainty in information theory 'entropy' because (among other things) 'nobody really knows what entropy is, so in a debate you will always have the advantage'.

Wikipedia offers:

'An everyday example of entropy can be seen in mixing salt and pepper in a bag. Separate clusters of salt and pepper will tend to progress to a mixture if the bag is shaken. Furthermore, this process is thermodynamically irreversible. The separation of the mixture into separate salt and pepper clusters via the random process of shaking is statistically improbable and practically impossible because the mixture has higher entropy.'  - Wikipedia

This highlights another key part to our understanding of entropy - a closed system will increase in entropy both inevitably and irreversibly. We can see this in cosmological entropy, which argues that our universe is a closed system, and will thus reach a state of maximum entropy where all energy is evenly distributed and (consequently) all parts of the universe will be the same temperature.  It's a theory that doesn't bode well for us in the (very) long run.

For me, when I hear the word entropy, I don't perceive a unit of measure - I perceive the irrevocable march toward homogeneity. Keeping heterogeneous things that are in contact from becoming homogenous takes a lot of effort. We see global systems become more alike as they come into contact, loss of biodiversity, loss of cultural diversity, loss of political diversity, loss of economic diversity, and loss of the protections that come with diversity. There's not much to be done about it either - as we go global, as our culture becomes a single closed system, rising entropy is inevitable. We see our attempts to keep our heterogeneity alive taking a lot of energy, and generally failing.

In light of this, we might observe three options:

1. Get some negative entropy - find some new cultures
2. Start embracing entropy - hooray for homogenisation!
3. Create closed systems - don't put salt and pepper in the same bag

Point 1 only delays the inevitable. It is highly interesting that elements of both point 2 and point 3 are generally championed as solutions to the problems we face today. Is retaining some heterogeneity while allowing some homogenisation the right approach? Can it be possible to maintain both heterogenous and homogenous elements in a closed system? What is the right combination and how do we control it? According to the laws of entropy, it would appear that we can neither stop nor reverse homogenisation. Of course seeing our world as a closed system is short sighted: it is part of our solar system, which is itself part of our galaxy, and our universe. We get energy exogenously from the sun, and all life ultimately uses this source of energy to endogenously maintain diversity - to swim against the relentless tide of rising entropy. From this perspective we apparently have great potential to choose between homogeneity and heterogeneity. The trap, however, is that whenever our attention wavers, the tide sweeps us a little further toward homogeneity, and the way back may never appear. We must fight perpetually for heterogeneity if we want it. Once we perceive diversity, it is at permanent risk of fading away.

The term 'political entropy' is interesting:

“The entropy measurement gives the average social uncertainty about what will happen for event sets in the social system. An entropy value for a unitary social system is analogous to a temperature reading for thermodynamic system, such as a volume of gas . In a state of temperature equilibrium one temperature measurement describes the whole volume of any part of it. If a social system is in an entropy equilibrium, a single entropy measurement describes the state of the system or any subsystem. For a system in partial equilibrium, the entropy values of its subsystems must be known.“ - Stephen Coleman

Coleman is saying that when we reach maximum political entropy, we will have maximum uncertainty over what is happening - in a democratic system this might mean many candidates with similar popular support - calling the result is very difficult.  Further research supports this interpretation: Coleman felt that the lowest entropy system was one where the certainty of the political outcome approached 100% - e.g. a one party democracy.  He also understood voting patterns as a means to measure political entropy - at minimum entropy any vote sample will identify the outcome, while at maximum entropy we must sample the entire vote to reach a conclusion.

One key aspect of the thermodynamic system is the inevitable tendency toward homogeneity, and Coleman identifies this is in his discussion of political entropy - we will head towards political systems with less certain outcomes. Also highlighted is the role of heterogeneity - the presence of subsystems, each of which must also be undergoing changes in entropy, and which influence each other to reach an eventual state of entropy equilibrium.  This subsystem relationship must also be recursive, with subsystems containing subsystems to an undefined degree of complexity.  The conclusion here, then, is that at maximum entropy a democratic political system is homogeneous - every citizen is a candidate with the explicit support of themselves alone.

Of course, we don't have the mechanics to support such a homogenous system - it is not possible for political entropy to reach that equilibrium.  It doesn't make sense at many levels - what are the means of election? what are the means of governing?  In fact, a maximum entropy democracy sounds a lot like anarchy. That's ok though - it's a theoretical maximum, an ideal - it serves as a bookend in the entropy discussion. We can observe, however, that public participation in policy making provides a pressure to increase political entropy - more people, more involved, more often. And therein lies a small paradox - our quest for transparency, for involvement, to have a say in our own government will actually deliver less certainty.

Less certainty? We don't want that do we? One might assume such at first glance, but if we look at some recent history of certainty [Iraq War][Copenhagen][Business deals][Credit Crunch][Iran election] we may see that it is in fact our our ignorance and impotence that drives calls for a more participatory and open government.

So now, with a little imagination, we can begin to see our political and cultural landscape through the lens of thermodynamics - as bubbles of gas inside each other, determined to coalesce into a single bubble of uniform temperature. On this landscape, humanity helps, hinders, increases, reduces and divides these bubbles - often unintentionally, and often without understanding the outcomes and implications.

When we look at the future of government, something becomes clear in the context of this discussion - it is inevitable that citizen involvement will increase and, barring monumental upheaval, we can't stop it, and we can't go back.  We're going to need better tools to manage our cultural and political entropy - because government as a platform will deliver mechanisms that allow us to move ever closer to the theoretical maximum.

Measuring Opportunity

In a recent post, I discussed 'compulsory voting' as a powerful tool for measuring opportunity. It is important to note that 'compulsory voting' only mandates that you demonstrate your opportunity to vote, and does not compel you to vote.

Some theorists suggest that democracy is not something that everyone should participate in :

"The effective operation of a democratic political system usually requires some measure of apathy and noninvolvement on the part of some individuals and groups." - Samuel P. Huntington

There is a perception [Noam Chomsky][Andrew Gelman] that Huntington was proposing a ruling elite - a minority rule in which (as Chomsky puts it) '... the peasants cease their clamor'. From such a perspective it is hard to extract the idea that Huntington saw that everyone should be given the opportunity to participate. The observation on its own, however, could appear to be a pragmatic expression of the statement - 'One should not be compelled to participate, but one should have the opportunity to participate'. To my mind, the latter holds more importance than the former, and thus we arrive at a middle ground ('compulsory voting'), where your participation is required only to the degree that your opportunity is reliably measurable. Alternate mechanisms of measuring opportunity may reduce this burden of compulsory participation.

To this point we have been talking about representative, liberal democracy - where an elected minority have the power to implement policy, under the assumption that they will do so in accordance with the policy platform on which they ran to obtain the mandate of the people. This mechanism is a solution to the scalability problems of direct democracy - i.e. not everyone can have a say on all things all of the time.

Samuel Huntington might observe the solution to another perceived problem - not everyone should have a say on all things all of the time. He might argue that this is a positive side effect of our solution to the scalability problem. There are many reasons why we might want to solve the second problem - some decisions need to be made quickly, some decisions need to be made with expert advice, and some decisions have subtle implications that are not easily perceived without a detailed awareness of the problem.

Largely, in modern democracy, the people do not make policy decisions, for both scalability and pragmatic reasons. For example, an Australian citizen gets two opportunities to do so - voting in elections (Federal, State and Local), and voting in referendums. The opportunity to participate directly in policy decisions beyond this is very difficult to perceive [media][lobbying][big business][nepotism][etc.] - and the transparency of those policy decisions is very often questionable.

It is in this space that Gov 2.0 provides new opportunities for our democracy - public participation in policy. It increases our opportunity to participate at a fine grained level in a far wider range of policy decisions.  It offers us new opportunities to solve the scaling problem.

Of course we also have the problem of pragmatism - who should be eligible to participate? Eligibility to vote is generally determined by citizenship, age and residential address. Becoming eligible to participate in policy at a finer level may not be so trivial - a formal qualification or previous experience might be required. Expert labs offers some insight in to this aspect of policy making. To reliably extend this participatory model to everyone, we must be able to measure the opportunity to obtain the necessary qualifications or experience. And here we see a key element of 'opportunity to participate' - measuring opportunity is hard, even when the rules are very simple, such as for voting. Measuring opportunity when the eligibility criteria are more complex will be even harder.

Almost by stealth, we have begun to discuss two elements of opportunity -

1. Opportunity to become eligible to participate - women were denied this opportunity until recently
2. Opportunity to participate once you are eligible - afghani citizens were denied this opportunity under threat of violence

One might observe that these are essentially the two elements of suffrage. However, suffrage applies only to the right and opportunity to vote, and not to more complex processes such as policy making.  Point 2 is largely unchanged in both situations, however point 1 takes on a different meaning - eligibility might be earned, rather than an inviolable right, and we must now measure a group or individual's opportunity to become eligible - we can no longer wave the wand of universal suffrage to satisfy eligibility.

Of course policy making doesn't happen once every four years - it happens in real time, all the time - which is why Web 2.0 is such a great fit. Surely we can't show that everybody had all their opportunity, all the time - in fact we can largely guarantee that they didn't. With government as a platform - a system available on the internet, in the cloud - we have the potential for a new means of 'turning up', of proving your opportunity: log in periodically. This mechanism is a little strange - and highlights vividly the difference between turning up to the polling station on the one hand, and submitting your vote on the other - the former demonstrates that opportunity, while the latter exercises that opportunity. It is a convenient coincidence that they are one-to-one. By logging in to the government platform you could demonstrate your opportunity to do many things - everything the platform offers - a one-to-many scenario.

This also highlights the converse situation - the reason why people are opposed to 'compulsory voting'. Requiring every citizen to log on to a government system periodically is an extreme version of turning up to the polling station - and feels much more restrictive - or does it? If, instead of turning up to the polling station on a designated day for each compulsory election, you could go to your local library, or sit at your desk at home - wouldn't that be easier?

There are a great many additional considerations, especially related to identity [Andy Oram][Gadi Ben-Yahuda], but one of the things we will need to consider is the opportunity to participate in Government as a Platform, and how we measure that opportunity - especially when the requirements for participation are more complex than how old you are and where you live.

Public Participation in Policy - PPP

For this year's Web 2.0 Summit, Tim O'Reilly and John Battelle produced a white paper defining 'Web Squared' - their next evolutionary step beyond Web 2.0. It's about the opportunities for harnessing our collective intelligence. Data in context, in real-time. It's about seeing the web as a conduit for making real things happen in the real world, in ways that couldn't possibly occur without it.

This is the first example of web squared from the above white paper:

'The election of Barack Obama has demonstrated how the Internet can be used to transform politics. Now, his administration is committed to exploring how it might be used to transform the actual practice of governing.

The US Federal government has made a major commitment to transparency and open data. Data.gov now hosts more than 100,000 data feeds from US government sources, and the White House blog is considering a commitment to the 8 Open Data Principles articulated by a group of open data activists in late 2007. There’s a celebration of the successes that many are now calling "Government 2.0." We’d love to hear about Government 2.0 success stories from around the world.

But in his advice on the direction of the Government 2.0 Summit Federal CTO Aneesh Chopra has urged us not to focus on the successes of Web 2.0 in government, but rather on the unsolved problems. How can the technology community help with such problems as tracking the progress of the economic stimulus package in creating new jobs? How can it speed our progress towards energy independence and a reduction in CO2 emissions? How can it help us remake our education system to produce a more competitive workforce? How can it help us reduce the ballooning costs of healthcare?'

There are a number of points to take from this -

1. The provision of data in context, with semantic meaning, is the main achievement to date e.g. data.gov . This delivers transparency on what's already happened, and to a lesser degree on what's happening.
2. We are now looking at how we can go further than simple data provision, to solve problems - to act on this information. Not only transparency, but application.
   
3. We are looking at application in terms of achieving actual policy outcomes.
4. If you read between the lines - we need to create the means to crowdsource involvement in the  policy process - discussion, definition, implementation, communication, review and measurement. Transparency not only on what's happening, but on how and why things happen.

Provision of government data and services is really just the tip of an iceberg - the real value will come when we can provide the capacity for transparency of, and participation in the entire policy process. The complexity of the policy process is immense - it makes sense to start with data and services, but the vision for Gov 2.0 will inevitably be PPP. There is no doubt that Tim O'Reilly is heading in the right direction when he talks about government as a platform.

It is no surprise that we currently use a representational system to manage our democracy - fine grained public participation in the policy process is a deeply complex vision.  It is a whole new paradigm for our democracy - the ultimate crowdsourcing endeavour, and one that is only now becoming feasible as we enter the world of 'Web Squared'.