Planning as Misguided Faith in the Impossible
As part of an ongoing attempt to define what planning actually is, some six years after I started learning about it, I've begun a reading group with some equally misguided collaborators. To that end, we plan to work through assorted key texts from the past thirty years, beginning with the low-point of rational planning, that, in a way, marks the beginning of alternatives.
Wildavsky, A. 1973, "If Planning is Everything Maybe it's Nothing", Policy Sciences, v.4, p.127
Rittel, H., Webber, M. 1973, "Dilemmas in a General Theory of Planning", Policy Sciences, v.4, p.155
Published in the same journal, these two papers present two fascinating criticisms of our ability to plan, one expressing hopelessness, the other bordering on contempt.
Wildavsky begins by trying to define planning, distinguishing between attempts to plan - to create a plan, or to make rational decisions - and the ability to have control over the future. The semantic confusion over what planning is, in many ways being the key theme of the article, but one that needs resolving if one is to judge the efficacy of planning as an activity. Is it an activity to be judged according to its inputs - the quality of its planning - or its outputs - its success. And more importantly, when are we actually engaging in planning, in order to judge it; as Wildavsky says:
"Since practically all actions with future consequences are planned actions, planning is everything, and nonplanning can hardly be said to exist"
The remainder of Wildavsky's work consists of dead-ends, each pursuing some conception of planning to show that, in fact, planning is always something else. Thus, planning as causation - the ability to predict the effect of actions, and therefore choose them rationally - is, regardless of whether there is a plan, merely another form of decision making.
Planning is, therefore, merely a form of power - "the probability of changing the behavior of others against opposition" - or politics. But power is necessarily limited, and planners, not being dictatorial governors, are also limited, perhaps irreparably. The plan itself shares a similar limitation. Objectives must necessarily change with circumstances, but too many changes imply a lack of planning whatsoever. To plan in an adaptive way avoids the problem of "future control", and becomes, by Wildavsky's reasoning, indistinguishable from any other form of decision making. A logical pattern repeated for planning as process - as goal directed behaviour, indistinguishable from goal directed decision making.
But if planning is just decision making, then Wildavsky argues, perhaps instead it can be judged against its intention - did it succeed? Here too lies a problem: when a plan fails, was it the fault of the plan, or was the plan itself merely a conduit for the decision making process. Plans stop being intentions to deliver and start being symbols of a policy process that is rational, efficient, coordinated and consistent. These goals though, are, again mere platitudes, indistinguishable from other forms of decision making contained within the machinations of bureaucratic governance.
The true meaning of the tile is thus derived. Planning is either indistinguishable from any other governance activity, or, it is a badge of honour, worn by professionals as a means of arguing for their specific forms of governance, and a (possibly costly) article of faith for those who believe that decision making should be rational and planned.
While unarguably true in some ways, Wildavsky clearly over-reaches in others. Planning may be merely dressed up decision making, but it is not clear whether the costs of those plans do indeed outweigh some improvement in the decision making process. In the past I've argued against a formal bill of rights, on the basis that they rarely seem to matter when the rights are being questioned, that indeed, like plans, they are no more than well intentioned articles of faith. An interlocutor disagreed, arguing that those symbolic words meant something - they affected the relations of power, merely by existing. Plans too, may do that, provided they point somewhere - which is not necessarily the case these days.
There is some hope to be derived from Rittel and Webber's equally disparaging article on the technical problems of planning. Rather than stretching across the gamut of planning definitions, Rittel and Webber consider planning from a position of power, where planners are capable of directing and solving problems from within a sympathetic government. Here the problem of planning is not governance but ability.
Planning problems, they argue, are wicked, meaning they: are ill-defined problems ("the formulation of a wicket problem is the problem); are never ending; have ill-defined solutions; have innumerable potential solutions; are essentially unique; are merely the symptom of some broader problem; are immune to logical hypothesis testing; and require solutions that work.
Effectively, Rittel and Webber argue something now taken for granted: that social problems are intractable by purely scientific means. Although their formulations are somewhat repetitive when taken together. Providing a solution that works is only necessarily if a solution could be judged objectively, which they have already dismissed. Being essentially unique is irrelevant if the problem is defined by the actor. As it is often said, "if all I have is a hammer, all problems look like a nail"; as above, such an approach is only a failure if somehow viewed objectively.
Stepping beyond mere problem solving, Rittel and Webber come to a similar conclusion to Wildavksy. In a pluralist society, planners are incapable of solving problems objectively, and are therefore only political players, not value-free experts. But as with Wildavsky, they overstate their conclusions, taking out the nuance whereby some solutions might still be considered "better", even in a pluralist society, and seemingly dooming planning completely:
"We are also suggesting that none of these tactics will answer the difficult questions attached to the sorts of wicked problems planners must deal with. We have neither a theory that can locate societal goodness, nor one that might dispel wickedness, nor one that might resolve the problems of equity that rising pluralism is provoking. We are inclined to think that these theoretic dilemmas may be the most wicked conditions that confront us."
14th August, 2008 23:55:33
[#] [1 comment]
Monday Melbourne: CLXVI, August 2008
The city, from the Yarra Boulevard, Fairfield. Taken August 2008
12th August, 2008 21:49:13
[#] [0 comments]
Assorted old transport notes
No blogging recently, distracted by teaching and marking instead, but some stories are too good to pass up. The Brumby government has been copped a beating a while back for spending money on spinning their transport initiatives instead of on the system. The spin is annoying, but otherwise I can bear it, and the money is, as stated, fairly piddling. But then you come across a non sequitur like this:
"The reason we have done the ads is, in all of the research we have done with commuters, they say they want information about what is happening on the system," she [Lynne Kosky] said.
Now, I may be wrong, there may be a commuter survey out there saying that people want information on new projects and initiatives being undertaken by the department, rather a raft of actual projects and initiatives. But I doubt it.
What this comment represents is a deplorable disconnection between the department and/or the government with what commuting actually entails. If I ticked a box or made a comment saying I wanted "information about what is happening on the system", then I wouldn't be talking about the next few years, I'd be talking about while I was traveling". I'd be wanting to know if I was wasting my time freezing my arse on the bus stop when it would be quicker to walk to the station, or a different bus. Or I'd want to know the alternative routes and times from where I was, at the time I wanted to travel (a service already provided incidentally, but only online, not by mobile).
It is unbelievably sad, that a quite reasonable and sensible request by commuters for information that would actually take pressure off the system has been met with a stale and wasteful advertising campaign. Especially when that same information, distributed widely and reliably, would in the current climate of invention and added customer value, allow people to build useful systems for real time route finding.
But it is also a bizarre decision in light of recent practice. The quality and quantity of information presented on the transport system has improved by an order of magnitude in the past few years - remember when you'd have a one in five chance of having a route map on board a train? The aim should really be to take the next step beyond those initiatives, but as with most little transport initiatives, they are piecemeal, even when they do some good.
Meanwhile, the minister's department is proposing to waste money putting bicycles on buses. This is not to say there isn't half a dozen commuters who are both a) travelling a significantly distance across town and b) in need of a bike at either their origin and destination.
But two things need remembering. Firstly, the major strategic purpose of both bicycles and buses is to connect commuters to the radial railway system. Very few commuters would use either a bus (even the smart buses) or a bicycle for a trip longer than thirty to forty minutes. And secondly, for trips of that length, a bicycle is roughly as fast as the bus anyway - faster once you include delays from congestion and waiting times.
There is hardly a significant demand for such a service, unlike on trains, where bikes are still either belittled or inadequately provided for: is it that hard to install some vertical, space efficient hooks in carriages?
The money would be far better spent improving bicycle lanes and off-road routes parallel to smart bus lanes, to get cyclists there quicker and easier.
It would also be remiss of me not to round out this summary of transport articles from a month ago without mentioning The Age's campaign to bring back conductors. I am in faovur of the idea, but apparently, at least according to Chris Berg this makes me nostalgic for human interaction and a nicer society, rather than a disgruntled opponent of ticketing systems.
I'm not though. I have spent far too many years designing human-computer interfaces not to recognize too fundamental truths. Firstly, somewhat regardless of how clever you are, and how good your interface is, until a computer passes the Turing test, you can't beat a human's ability to be flexible in performing a task. Yes, a computer can sell tickets, and provide some directions, and other commuters can help people on and off the tram, but not to the same level of quality, and not in a way that improves running efficiency and reduces far evasion. Secondly, no matter how intuitive your interface is, customers have a remarkable ablility to both break it, and be baffled by it. The current ticket system in Melbourne is completely baffling to the uninitiated (ie. tourists). Myki will probably be worse, because it tries to be more clever, and clever is confusing.
Which is not to say Myki isn't a good idea - actually it is probably a pointless waste of money, but politicians love the idea of automation, because it looks like an easy way to save money; at least until the IT bills come in. Fare system improvement is a good idea. Re-introducing conductors is a good way of improving the fare system.
12th August, 2008 19:43:05
[#] [0 comments]
Displaying Elevations on Google Maps
For the past month or so I only seem to have been doing three things: marking assignments, attending weddings and going for bike rides. Unfortunately there was far too much of the first and not near enough of the latter. Marking is practically finished now, and I won't be teaching for a while to focus on finding that elusive thesis question.
The bike riding will hopefully continue though, with the plan to have more long rides, especially now that I've ridden around most of the paths the city has to offer. Finding places to ride can be extremely annoying though. Google Maps is really good at giving you a route and a distance, and has a pretty cool terrain viewer to give a sense of the hills, but not perfect for other important bike things.
Firstly, the route finder gives the fastest route, which means the main-est road route. Since I'd prefer to avoid main roads, what I'd really like (and can't get) is a sense of road quality. A route finder that confines itself to sealed roads would be handy. Information on the map detailing whether a road has a shoulder, or a bike lane would also be helpful. Doubly so if it could tell me what roads are smooth, instead of long wearing too. I'm not sure what it is like for cars, but I notice a 10-15% difference in effort between the asphalt of highway road surfaces, and that normally used for bike paths.
Route finding for bikes is possible with the proper GIS information, and I'm working on that in a more general way, mfor more research type reasons, already. Writing my own route finder to plug into Google Maps is possible (even likely), but not quite yet. I'd need to look into a dedicated server to handle the processing.
Secondly, although there is a pretty handy terrain viewer, it is not easy to read contour lines to calculate hills Particularly winding roads where the distance is hard to calculate, it needs to be at full zoom to see them properly.
I can do something about this though. Google Maps has a pretty good API for playing with its stuff. Very good actually, and it isn't until you get into it that you realise how cool it is, and how easy it is to do some funky stuff (though a few things I wanted to do meant rewriting slabs of functionality). Unfortunately, Google doesn't tell you the elevation at different coordinates. As with route fnding, with a dedicated server and proper GIS information I could find this easily enough, but not yet.
Fortunately, Topocoding has a clean and hard working API of their own, providing data to 90m world wide. Combine that with Google's route finder, some Walter Zorn's drawing tools and some (possibly also server intensive) PHP back-end coding, and the result is (I think) worth a few days of effort:
I am not sure what the image processing is doing to my hosting service (it is very slow, but that might be a coding error). Topocoding limits the number of requests it allows anyway, so the full working model might get reserved for my personal purposes in the event of unexpected popularity. Please note: working out the route, and setting the scale before getting the elevation makes a big difference to your wait times.
It is a funky plaything though. That's the Tour de France's brutal tour decider over L'Alpe d'Huez above. For the benefit of significantly more of my readers, here is the link to the significantly easier, but more familiar, climb up Northcote Hill
12th August, 2008 17:44:32
[#] [0 comments]
A Change of Methodology
A brief respite after the conclusion of the summer schedule has given me a chance to consider better ways of calculating the ratings, and to address one of its long term (and deepest flaws). As the ratings were calculated, teams will improve their rating with a win, and regress with a loss, with a draw impacting slightly on the better team.
There were several factors in calculating the change (in declining order of importance): the ratings difference between the two teams, adjusted for a home team advantage (expectation); the margin of victory or loss (result); the number of tests in the series (significance); and whether the series was alive, meaning it could still be drawn.
Two problems have persisted. An over-sensitivity to unexpected results, and a lag in the date of decline of very good teams. Something that could be seen most clearly in the period 1948-1954 with the Australian team. Although the Australian team is generally considered to have peaked in 1948, they continued to win for several years afterwards, and therefore continued to improve their rating.
Old win/loss based ratings: the mid 20th century.
Although great teams will continue to win for a while after they lose their great players, it would be nice if the ratings reflected the declining quality of the side before they start to record losses. And vice versa, for losing teams (New Zealand during the same period for instance) to start recording improvements before they begin to win.
A change has been made therefore, to adjust not on the result itself, but on the actual and expected margins. Setting this was complex. There are essentially two linear fits, a noisy win line of about 0.6 runs per point of ratings difference, and a flat line of draws. Because a draw is stil a result of sorts - those not the result of prolonged rain - the expected margin has been set at <0.35> runs per point of ratings difference, with the home advantage still 100 points. These values are always relative, and therefore marginal, but it is nice to have them accurate.
The other factors above have been retained. As has the result, because it should matter whether a team actually won. A modifier of 2/3 is therefore used on both draws (which can be merely unlucky) and victories by margins smaller than expected.
The effect of the change can be seen by comparing the past two decades of ratings. Whereas before the Australian team is marked by continual improvement with a few significant drops in 2001 and 2005, the closeness of the losses they suffered, and the size of the victories has meant they've been fairly stable since about 2002-03, with perhaps just the hint of a drop in the last two series. Overall it has had the effect of smoothing changes out, with the best side being more clearly delineated, and in most cases, sides having to show consistent quality to improve their ranking.
Old win/loss based ratings: the past 16 years.
New margin based ratings: the past 16 years.
At some point I should blog more extensively on the historical rankings. For now though, here are the new rankings for each team prior to the Australia-West Indies and New Zealand-England series.
12th August, 2008 17:22:20
[#] [0 comments]