Captain Chaos Theory and the World of Tomorrow
"I'm not a dentist," but over at the Roadrage Blog (which I sometimes think is a much better blog than this one and deserves more readers), we've been talking about the government's plans for road pricing, and a number of things strike me about it which sort of relate to wider issues, if you have a disordered brain.
Now, I am not a scientist, which of course makes me uniquely qualified to comment on scientific theories of complexity, chaos, and systems. I have read the James Gleick book (and, behold, internetters, note how I have correctly spelled his name), so I'm confident I know all you really need to know. Everything else is just grant applications and me-too academics trying to justify their existence.
One of the most interesting things about the figures being bandied about by the government on road congestion is that a 4 (four) percent reduction in traffic would lead to a 46 (forty-six) percent reduction in congestion.
I'll leave that hanging for a moment.
It's ironic how people often describe traffic jams, hold-ups, gridlock etc., as "traffic chaos", when actually the opposite is true. The traffic is chaotic when it is freely moving, but organised when it is standing still. An extra grain of sugar in a super-saturated sugar solution will suddenly generate crystallisation. A few extra cars/trucks/vans on a city streets and the traffic will spontaneously self-organise and we have gridlock.
This is how chaos theory works. Long-range weather forecasting is notoriously difficult simply because of what they call "sensitive dependence on initial conditions," (aka the mythical Butterfly Effect). Your one-day forecast stands a fairly good chance of being right. Two days is iffy, unless there's a major weather system sitting right on top of us. Three days is entering the realm of the impossible: and five days, as we know, is a total joke. Or ask the Williams Formula 1 team: even forecasting a few hours ahead is fraught with difficulty at the local level.
Like the weather, the roads and the economy are systems, and are as difficult to predict. For example, yesterday morning I drove my wife to the airport early and so left for work 40 minutes later than usual. Now, in normal circumstances, I've realised that 4 minutes at the beginning of my commute can make a difference of 20-30 minutes at the other end. In other words, if I take the scenic route to start with, I'll end up sitting in a half-mile queue at Junction 26 of the M1. If I go the 4-minutes-quicker way, I'll be (at worst) halfway down the slip road before I hit the queue.
And yet, yesterday morning, arriving at J26 40 minutes later than usual: no queue. On any other day, any other Thursday, I'd have been queuing from beyond the half-mile sign. On any given day, for whatever reason, conditions will be totally different. Mondays, I've found, being the nation's favourite Sick Day, are an easy drive. Tuesday, being the new Monday, is a bastard. Thursday, as the new Friday, is also a bastard.
What's interesting there is that, most Thursdays and Fridays*, there seems to be substantially less traffic on the road. Looking at the government figures quoted above, I realise that it's probably only as little as four percent less traffic - but it makes a huge difference.
It seems obvious to me, in fact, that instead of expensive satellite systems and black boxes, that between us we could come up with some way of reducing traffic levels by 4%. How about, for example, office workers starting to work from home, by agreement, one day a week? On any given day, there could be anything up to 10 or 15% fewer cars on the road. Why not experiment with that - lost productivity caused by the inevitable goofing off would surely be compensated by the improved productivity of reduced congestion.
Do that, change school hours, adjust school holidays, encourage supermarkets to start closing at a reasonable time instead of staying open 24-hours. Loads of things we can do, not requiring satellites of love. Unless... they want the satellites for another reason they're not telling us about. To monitor your speed, for example.
Finally, and this is what I wanted to say to begin with, have you noticed how all governments and pundits always talk as if things will go on being the way they are now? Arsenal won the league last year, so they were talked up as the inevitable champions this year. Traffic is like this now, so in ten years time, if we reduce it by 4%, it will be better by 46%.
But there is no evidence, and no reason, to ever think that things-as-they-are-now will continue to be. Things change. Wars happen, oil companies announce that, er, their reserves aren't what they said they were, companies go bust, towns are inundated by floods, people come to their senses, Apple switches to Intel. Whatever: nobody knows. Which is not to say you shouldn't make plans; you just shouldn't make plans as if things will go on being as they are now. Pack the kagoul and the bucket and spade.
===
*Friday night traffic is notorious, but my monitoring indicates that this is caused, not by additional traffic, but by people driving drunk after Friday lunch in the pub.
Now, I am not a scientist, which of course makes me uniquely qualified to comment on scientific theories of complexity, chaos, and systems. I have read the James Gleick book (and, behold, internetters, note how I have correctly spelled his name), so I'm confident I know all you really need to know. Everything else is just grant applications and me-too academics trying to justify their existence.
One of the most interesting things about the figures being bandied about by the government on road congestion is that a 4 (four) percent reduction in traffic would lead to a 46 (forty-six) percent reduction in congestion.
I'll leave that hanging for a moment.
It's ironic how people often describe traffic jams, hold-ups, gridlock etc., as "traffic chaos", when actually the opposite is true. The traffic is chaotic when it is freely moving, but organised when it is standing still. An extra grain of sugar in a super-saturated sugar solution will suddenly generate crystallisation. A few extra cars/trucks/vans on a city streets and the traffic will spontaneously self-organise and we have gridlock.
This is how chaos theory works. Long-range weather forecasting is notoriously difficult simply because of what they call "sensitive dependence on initial conditions," (aka the mythical Butterfly Effect). Your one-day forecast stands a fairly good chance of being right. Two days is iffy, unless there's a major weather system sitting right on top of us. Three days is entering the realm of the impossible: and five days, as we know, is a total joke. Or ask the Williams Formula 1 team: even forecasting a few hours ahead is fraught with difficulty at the local level.
Like the weather, the roads and the economy are systems, and are as difficult to predict. For example, yesterday morning I drove my wife to the airport early and so left for work 40 minutes later than usual. Now, in normal circumstances, I've realised that 4 minutes at the beginning of my commute can make a difference of 20-30 minutes at the other end. In other words, if I take the scenic route to start with, I'll end up sitting in a half-mile queue at Junction 26 of the M1. If I go the 4-minutes-quicker way, I'll be (at worst) halfway down the slip road before I hit the queue.
And yet, yesterday morning, arriving at J26 40 minutes later than usual: no queue. On any other day, any other Thursday, I'd have been queuing from beyond the half-mile sign. On any given day, for whatever reason, conditions will be totally different. Mondays, I've found, being the nation's favourite Sick Day, are an easy drive. Tuesday, being the new Monday, is a bastard. Thursday, as the new Friday, is also a bastard.
What's interesting there is that, most Thursdays and Fridays*, there seems to be substantially less traffic on the road. Looking at the government figures quoted above, I realise that it's probably only as little as four percent less traffic - but it makes a huge difference.
It seems obvious to me, in fact, that instead of expensive satellite systems and black boxes, that between us we could come up with some way of reducing traffic levels by 4%. How about, for example, office workers starting to work from home, by agreement, one day a week? On any given day, there could be anything up to 10 or 15% fewer cars on the road. Why not experiment with that - lost productivity caused by the inevitable goofing off would surely be compensated by the improved productivity of reduced congestion.
Do that, change school hours, adjust school holidays, encourage supermarkets to start closing at a reasonable time instead of staying open 24-hours. Loads of things we can do, not requiring satellites of love. Unless... they want the satellites for another reason they're not telling us about. To monitor your speed, for example.
Finally, and this is what I wanted to say to begin with, have you noticed how all governments and pundits always talk as if things will go on being the way they are now? Arsenal won the league last year, so they were talked up as the inevitable champions this year. Traffic is like this now, so in ten years time, if we reduce it by 4%, it will be better by 46%.
But there is no evidence, and no reason, to ever think that things-as-they-are-now will continue to be. Things change. Wars happen, oil companies announce that, er, their reserves aren't what they said they were, companies go bust, towns are inundated by floods, people come to their senses, Apple switches to Intel. Whatever: nobody knows. Which is not to say you shouldn't make plans; you just shouldn't make plans as if things will go on being as they are now. Pack the kagoul and the bucket and spade.
===
*Friday night traffic is notorious, but my monitoring indicates that this is caused, not by additional traffic, but by people driving drunk after Friday lunch in the pub.
posted by Anonymous at
Friday, June 10, 2005
0 Comments:
Post a Comment
<< Home