In what appears to be an oddity among modern liberal democracies, the United States generally has state legislators draw up federal boundaries. Not only is this a massive violation of the principle of dividing powers, it practically invites redistributing boundaries in partisan manner. Unsurprisingly, this leads to abuse of the boundary-drawing power, and spectacles such as Democratic legislators in Texas fleeing the state in order to prevent Republican legislators from redrawing House of Representative boundaries.

In contrast, a better method would be to have a committee which is at least nominally non-partisan (while you could compose it of political hacks, that might well lead to feuding on the committee). Take, for example, the Canadian experience: the federal boundaries are drawn by committees consisting of a nominee of the Chief Justice of a province, plus two other members nominated by the Speaker of the House of Commons, with the explicit intention of being politically neutral. Similarly, in Alberta, the provincial boundaries are drawn by a chair who is either the province's Ethics Commissioner, Auditor General, the president of a post-secondary education institute, or a judge; two people nominated by the government; and two people nominated by the opposition.

Sure, judges may have political biases, but given that modern liberal democracies already trust them with so many other tasks (read: interpreting constitutions and legislation), if they're going to grind their axes, they've got plenty of other opportunities.

There are a few options to restrict the process of gerrymandering. We can always introduce multiple-member constituencies (as suggested by Matt Yglesias) that reduce the effects of boundary manipulation. Or, one can make electoral ridings more "compact", such as those proposed by Joseph Schwartzberg. In essence, his standard measures the ratio of a shape's perimeter to its area, while mathematically normalizing the result.

But let's stop bringing out the computers and mathematical formulae before our eyes glaze over. In summary, yes, gerrymandering is a problem, but it's very, very difficult to eliminate the process. This is much like your neighbour who plays loud Eminem albums at 2 A.M. in the morning -- he's certainly a problem, but you can't eliminate him (though, you wish you could). However, to mitigate the harmful effects of gerrymandering, certain strategies can be implemented. (Much like those used to mitigate the harmful effects of your neighbour, but without the earplugs.)

Well, Mme. Speaker, it seems obvious to me that Alberta is richer than the other two Prairie provinces because of oil (well, oil and natural gas and coal): if you look at Alberta pre-Leduc, there was little distinction between the three provinces --- in fact, Alberta was at times the least-populous and poorest. As an astute student of history, Mme. Speaker, you will be aware that Alberta was the first province to default on its debt, in 1936.

But then, why is Ontario better off than Manitoba? Ontario is, at first glance, better off because it has a more diverse resource base, is better situated to export its goods and services, and was settled earlier, leading to earlier infrastructure and denser settlement. In short, it appears that it's due to extrinsic factors, rather than anything the province of Ontario itself did.

How can a province seek economic development, and if it can, should the government do so, and if it should, how ought it to? It seems to me that there are three means to development (excluding, for the moment, strictly intra-provincial trade): first, to extract more resources from the earth and sell them outside the province; second, to refine resources into finished goods, thus increasing the value of such goods; and third, to provide services to entities outside the province.

Given that natural resources are finite, there's a clear limit on how much further Alberta can develop by means of extraction: every barrel of oil removed from the ground is a barrel that can't be removed later, when oil prices may well have increased, every tree cut down is a tree that can't be cut down next year, and every hectare of wheat grown is a hectare of land that can't be used for housing, forests, or wildlife. While it may very well be the case that Alberta is not near the limits of its development in the primary sector, the primary sector alone is vulnerable to fluctuations in world prices, making it unsuitable to base an entire economy upon.

Second, refinement. While this is superficially viable, Alberta is a long way from the markets for most goods made from its resources. In many cases (for example, sulfur), it's much cheaper to simply ship the large quantities of raw product elsewhere, and then ship the small quantities of finished product needed, back. This could be, to some extent, alleviated by means of improved transportation links, but there's a limit on how far this can take you.

That leaves the tertiary service sector for further development. More on that later, Madame Speaker.

The other level is this: why should the end matter? In achieving a goal (defined by the set of desireable criteria), there are many possible ends (solutions that achieve that goal by meeting all the desired criteria). Which end is ultimately selected and achieved should be determined through a process of good decision-making and action.

You should decide on the desired criteria (the goal) and then try to acheive those criteria in the best way possible; a good process will achieve the criteria the best way possible. If you instead take the criteria, select a solution (an end), and then try to achieve that solution (the end), you may not have chosen the best solution often because your biases are at play or because you're taking an opportunistic route.

What's actually happening when you focus on the solution instead of the process is that biases and opportunism are adding additional criteria to your goal (e.g. preferred ideological stance, least work needed, etc.) If these are truly desired criteria, then they should be defined at the outset and they should be part of the goal. By not defining them up front, it means that they are not sufficiently important to warrant consideration in finding a solution. That you then allow them to "sneak" into your solution's criteria without your knowing is the fundamental problem to focusing on the end and then selecting the means: you're unconsciously allowing criteria deemed to be unnecessary to factor into a solution.

Focusing on process ensures that the solution meets the set of criteria deemed desirable. If the focus is, instead, on the solution, the solution may meet additional criteria to those deemed desirable—by meeting additional criteria, you're further constraining the set of possible solution and perhaps excluding the "best" solution.

By focusing on the means, you do the morally right thing and you are more likely to achieve the "best" solution.