So I've been thinking about why there's this big push for DRM. The line sold to the public is that the media companies need this to prevent copyright infringement of their works. The story goes that without _effective_ copyright protection, the incentive to produce good artistic works is diminished, and therefore, good art won't get made.

There are several pertinent objections to this line of reasoning; I could cover the fact that most artists produce good art because it is a passion of theirs, or because it is the only thing for which they are sufficiently talented to receive a paycheck, etc... But instead, I will focus on the DRM aspect of copyright.

A few hundred years ago, copyright for musical performances was wholly unnecessary. In order to enjoy a performance or see a particular painting, one had to be physically present at the performance or exhibition. The artist and performer were paid for each painting or work performed, a process which generally rewarded the effort and talent enjoyed. But technology changed all of that - the printing press made it possible for someone to sell books for which he had no personally labored to write. The vinyl record made it possible for someone with no musical talent whatsoever to make a living selling music. And the cinema made it possible for those without any acting talent to make money from the performance of others.

Of course, there were upsides. An artist could receive royalties well beyond their performing years for works performed during their prime.

And then technology changed again. Formerly, the cost of reproduction was substantial - it required physical resources, and was generally limited to those with a substantial outlay of capital. But, with the advent of the internet and high speed computers, reproduction of copyrighted works became feasible for fractions of a penny per copy. There was no longer any need of a publishing company or printing presses, or recording studios or movie theaters.

Naturally, those in the industry felt challenged. It seemed their entire investment in exploiting the works of others was about to be challenged. Enter DRM. This is touted as the solution to the problem of rampant copying and unauthorized distribution. The only problem is that it doesn't work.

Technical folks like you and I know it doesn't work. The movie and recording studios know it doesn't work - DVDs and CDs are copied bit by bit, copy protection and all, by pirates and sold on the black market in developing countries. DRM schemes are cracked, in some cases by nothing more than a microphone or a camcorder, and unauthorized copies are then distributed, free of DRM restrictions, on the internet. So why is yet being pushed?

It is, as I call it, the Problem With The Public Domain. The reason why DRM is being embraced by the media conglomerates is because it effectively restricts works from ever entering the public domain, thanks to the DMCA. Even after a work's copyright has expired, it is still illegal to build and traffic in tools which would remove the DRM restrictions from said work. It isn't illegal to bypass said restrictions, but without a legal means of acquiring the tools to do so, it won't happen for most users. And this is exactly what the studios want. They do not want us to watch those older movies or listen to the older recordings because it cuts into the money that would be spent on their new works. A rich public domain is a real problem for someone trying to sell music or movies, because it represents a free, competing alternative to spending money for entertainment.

And that is what DRM is all about. The content cartels understand well that unauthorized copying and distribution will always be with us - but what they fear is a world in which people can watch movies or listen to music without paying them for doing so.

Whether someone should be paid multiple times for a work they performed only once is left as a philosophical exercise for the reader.

Lately, with Microsoft claiming that Linux infringes on their IP, it has occurred to me that what is widely regarded as free software cannot introduce disruptive change into the world.

Without a doubt, free software has introduced some very clever things to the world. For example, KDE showed file thumbnails in their file browser before Microsoft had ever thought of doing so - they beat Microsoft by a few years. UI features routinely show up in free software several years before making it into Windows. Windows has yet to offer a transparent taskbar, but it has been available in Linux for what seems like ages.

But this is largely irrelevant. The reason is that these features, while unique and interesting, are always offset by some other, larger disadvantage. Just the effort of installing an operating system and finding drivers for their hardware is sufficient to keep many users using Windows. And the Gimp's user interface - and their insistence on using GTK - makes it little more than a toy compared to Photoshop. It utilizes one of the least usable GUIs I've ever seen. As this has been flamed about elsewhere, I'll not repeat the argument, largely because UI difficulty is endemic to almost all free software, not just the Gimp. Need I mention my Mom trying out vi?

But what would happen if the Gimp suddenly "got it" with regard to usability? What would happen if Wine was suddenly better than Windows at running Windows programs? What if Linux was fully automated to the point that the installers could find and configure drivers for all existing hardware?

We are seeing the consequences of the what-if scenarios even now. If the Gimp was a serious threat to Photoshop, or Linux a serious threat to Windows, you can bet that Adobe or Microsoft would set their lawyers to work finding some way of diminishing the threat. They would claim patent infringement. Or, they'd claim copyright infringement, or that said software was divulging proprietary and confidential information. They might even go so far as to let one of their employees contribute to the project, and later fire the employee, claiming that the contributions were without company permission. They could claim the employee divulged proprietary algorithms - which, while they might not be patentable, could be forcibly removed from a project if they represented a company trade secret. Even if said algorithms could have been implemented by any of the other coders, the fact that they came from an employee gives reasonable credence to the claim that they are trade secrets. Such an estoppal could prevent any other coders from implementing the same algorithm, regardless of the copyright or patent status of such.

But these are just a few of the ways that large businesses could interrupt open source work. The real problem lies in that, regardless of the merit of a lawsuit, the fact is that most contributors to open source projects would simply pull their code rather than spend tens of thousands, if not hundreds of thousands of dollars, defending themselves in a lawsuit. It would be very difficult for most developers to justify spending this kind of money to defend something, that for most, is little more than a hobby. Thus, developing software which can change the world must necessarily be a commercial undertaking, for without the revenue provided by commercial software, most developers won't be able to afford the cost of the inevitable lawsuits which would follow. A company can sue without any merit to its claim whatsoever - witness the SCO vs. IBM debacle, where the CEO's statements regarding SCO code in Linux now appear to be deliberately fraudulent. Regardless, SCO has effectively lost against IBM, but had they brought their case directly against the kernel developers, how many of them would have had the monetary and legal resources to stand up to a company with 30 million dollars to burn on litigation? That's more than most developers will make in their entire lifetime.

I like free software; I want to see it flourish. It has brought some great ideas to the world. Yet, I'm disturbed by the fact that the corporate legal climate in this country dictates that it is nearly impossible to bring innovative new software to the world without charging sufficient money to afford legal counsel. It seems that producing the best software is more a matter of having good lawyers than it is of having good coders and architects.

Problem is, bandwidth isn't cheap. For cable internet, I'm looking at paying about $600 a year. I don't need fast access to most of the content, but occasionally there are things I want that get kind of big - like software downloads and linux distros....

So, I thought about how I could fix this. Turns out that postal system has some pretty good bandwidth, very good reliability, but a rather large latency. Put like this: Suppose I started a software distribution list for free software. I burn all my favorite software onto CD's, and send them to the next person on the list. A few days later, they've got a few gigs worth of software, at which time, they copy these to their HD, and send off the CD's to the next person on the list.

Considering that I can mail about 10 CD's for about $2.50 (guessing) or so, my bandwith costs about 0.038 cents per MB - that's 26 MB per penny. That's pretty cheap.

But now consider this: To improve speed for those far down on the list, every time a list gets more than 5 people on it, it splits into a sublist - new people are added as a sublist. The first person on the list becomes the NodeMaster - he has responsibility of forwarding the CD's to the rest of the list, and if anyone breaks the chain, he patches it by mailing another copy of the CD's to the next person. In exchange for this responsibility, he gets the CD set first.

Shipping costs are paid by reciprocal payment. When anyone receives a CD set, they drop a reimbursement in the mail to the person who sent them. This way, a person only pays for shipping once.

Well, you can do the math. A distro list with 5 members per sublist could reach 390,000 people in 8 mailings, or about 16 - 21 days. At 21 days, the average throughput would be 3.58 kB/s, for an average of about $4/month.