It occurs to me that magic, by definition, doesn't exist. The definitions presented by the dictionary are distinctly unsatisfying; most refer to the occult or supernatural. But when it comes down to it, something is "magic" if it simply cannot be.

All kinds of things (for example, predicting the future, influencing the weather, or instantaneously appearing and disappearing) are dubbed magic. We name a thing magic if we cannot tell how it could possibly happen. Either we know it to be impossible on the face of it (levitation, invisibility, etc.) or we know that it can be done, but not without certain prerequisites that aren't currently present (moving from one location to another without travelling the space in between, for example).

It has been argued that we need magic. I won't disagree; one of my favorite quotes is "Logic gives man what he needs; magic gives him what he wants" (Tom Robbins in Another Roadside Attraction). But in terms of what stock we put in it, it's important to recognize just what we categorize as magic. It seems that, for all practical purposes, magic is the same as imagination. It just now occurred to me that they even seem to share a common root... I'll have to crack open the OED on that one sometime. We do, I think, need the ability to conceive of and comprehend the impossible; otherwise, new things never become possible. But knowing the difference between imagining a way to cure cancer with crystals and actually believing that you can do it, right now, can be a life and death matter.

There is still a lot in this world we don't understand. Most of it has to do with how we work; we're still quite foggy on just what makes us self-aware, intelligent, or even alive. We can say "this is alive" and "this is not alive," but we don't know how to go from one to the other (without using a currently living organism to process the non-living matter, anyway). But calling such a thing "magic" can be misleading. It implies, to many people, that it not only isn't possible for us to do now, but that it never can be understood; that its very nature is supernatural, which of course is a paradox.

Ok, done babbling, back to calculating mortgage payments for my math homework that was due last week... and I'll be dreaming of some magical way to have all my homework done while I read /.

People have often argued that OS is the "solution" to the rather broken application of copyright to software. A common response to this is that "Open Source is not commercially viable." It's hard to argue with that. But, on the other hand, it's not entirely relevant.

In the US, we're used to thinking of everything from the market economy standpoint. We bought into Adam Smith's efficient allocation of resources (without noting the caveats about perfect information or perfect mobility... oops). So we have trouble conceiving of any other system of distribution.

In yesterday's journal entry, I mentioned the free rider problem. In a market-regulated economy, free riders bring the whole system down, because people see that someone's getting something for nothing, and they stop paying also. In the end, you have a suboptimal solution for everyone, even if every individual is acting in their own best interests (the prisoner's dilemma). That's the justification for government intervention in market failures.

Open Source is another kind of response to a market failure. It has the assumption of free riders. This is something that boggles the American mind. Instead of making something good because you can sell it, people make something good so that they have something good, and then let other people use it. This can indirectly increase their wealth; they may get a great job offer based on their OS work, for example. But since they don't immediately, directly see financial benefit from their work, there's an assumption that they won't do as good a job as they might if they were being paid for it.

The corollary to this assumption is that open source cannot produce as good a product as proprietary software. People say this, and ignore the fact that it doesn't appear to be true. With no large-scale marketing, no bundle licensing agreements, and no evil empire, Linux has a 26% market share in corporate back-office servers. It's not just the price, either; sysadmins are switching to Linux because of the stability, the ease of administration, and the security over Windows-based solutions.

Generally speaking, the OS model seems to produce a product that is superior in mechanism, but inferior in interface. The superior part isn't hard to figure; people are making this stuff because they want to use it, so they're motivated to make it work well. Add to that the concept of peer review -- your code is out there for all to see, so you want to make it look good. Did your mother ever say "Wear clean underwear in case you get in an accident?" Sort of the same thing, but less foreboding.

The interface is where OS falls behind, so far. Again, looking at how it's made makes this obvious: people who are good with software are making it to be used by people who are good with software. There's not much point to spending the effort to make it appeal to the lowest common denominator... not when you could spend that time and effort making it run .01% faster. ;-) Of course, the attempt to make it commercially viable by selling packaging and support is changing this, slowly. In time, the gap will probably close to a point that there's no real difference between using Linux and using a proprietary desktop or server operating system.

So, basically, they're right: open source can't work in a free market economy. It works outside of it. It's making a new economy, one that's desperately needed. /me cheers.

So this morning I was sitting in Introduction to History and Theories of Planning, and the topic of public goods came up. It's a frequent cause of discussion in planning circles. The definition of a public good that they've pounded into our heads has two parts:

1. It is impossible or unfeasible to control access to the good.
2. It is not consumed by use; it can be used by an unlimited number of people without being used up.

Economists sometimes describe public goods as an example of a market failure. Since it is not possible to consume or to control access to a public good, the market cannot regulate it through supply, demand, and pricing. People are liable to underreport their desire for the good, in order to avoid paying their fair share (the free rider problem) or it may be that no individual values the good enough to pay for it to be set up, though it may be beneficial to everyone (such as a symphony orchestra). In some cases, no individual *can* buy/create/establish a particular public good; imagine if United had to build their own airport in every city they wanted to fly to, or that you and your neighbors had to get together and lay a new sewer pipe.

Planners and other social science types often argue that market failures are areas where government can and should step in and regulate the situation, even in a free market economy. Through taxes and other assessments, governments can evenly distribute the financial burden of providing public goods, so as to ensure they are available to all that want or need them.

So it suddenly struck me that, according to the definition I provided above, digital information is a public good. Since the quality doesn't degrade as you create new copies, and the cost of copying is practically nil, it is not consumed by use. It is currently impossible to really prevent information from being shared, as well. Access can be controlled only through relatively extreme measures; for example, the military has installations that literally have no connection to the outside world via anything but the front door, so information can only be hand-carried out. Palladium is an enormous effort being put forth to control access to information, and there's no sure bet that it will actually work. It may simply make information harder for people to use legitimately to the same degree that it prevents illegitimate use.

So, what? Well, then, I would argue that government needs to take an active role in provisioning the public good of digital information. Exactly what form this role should take is a little more difficult; I just had this thought today, give me time. ;-) But certainly this would be a new role for organizations such as the National Endowment for the Arts, or a justification for regulating some software companies like public utilities. In any event, going on pretending that we can control access to digital information isn't going to solve anything. We need to think of it in a new way.