wpegg: This is NOT a piracy thread in itself. We've had those. I want to focus on the original question. How do you remove DRM without justifiying it?
One game at a time.
I imagine there is likely some curiosity, even within the high-up decision-making layers of EA and the other 'bigs', of what would actually happen if they took a step back from DRM. And there's no reason it couldn't be tried, on a predicted-to-be-popular but less then AAA (read: around $20) title to see what the results really end up being.
Actually, just thought of an interesting (to me, anyway) experiment:
As a test, for a few titles serialize each installer, so every single instance has a unique identifier at the time that it's purchased and downloaded the first time by each discrete consumer account. Meanwhile, release to the pirate / torrent world a slightly different version with a different set of self-generating serial numbers but with a unique marker within the serial number that identifies that instance as coming from one the planned-for-piracy copies. Track the use of the serial numbers, not necessarily by WHO is using them, but simply by how many different times that serial number shows up online, or maybe how many times it shows up at the same instant (if serial number 123abc is in use at 3pm by 47 different people, then it's assumed that it was pirated at least 46 times). With a mechanism like this, there should be generated a data set with some correlation to the actual level of piracy, and also where it comes from: individual users handing out an installer to buddies here and there, or organized piracy on torrent sites, for instance.
That should create at least some data with which to compare pirate versus purchase. For obvious reasons it would have to be done in secret else the public would decide that since the publisher was putting out the pirate version itself then it must be prepared to face a 100% loss, and thus the consumers would have no problem with stealing the game.
The key would be to do nothing to track personal data, and also not take legal action against any who download the versions seeded by the publisher itself, since this would be purely for the purpose of gathering data. Not sure if the consumers would allow it, but they could also add a mechanism to identify individual PCs (essentially tracking the 'who'), and see how many pirated copies eventually become purchases. Many people
say they do this, but nobody knows if it's 2%, 15%, or 83%. This would give them some indication, not 100% accurate, sure, but at least an idea of what the numbers are. And try it for multiple titles so one fluke doesn't skew the data.
The problem is that it would eventually leak before the data gathering was complete, and people would scream to high heaven about big brother and all that, even though the goal would be to get some real-world data of the scope of piracy, and with the aim of determining the pros and cons of DRM for the seller and consumer.