One nice thing about last week's vacation from the plugged-in lifestyle was that there was a bunch of interesting tech news waiting for me upon my return. The most exciting bits:

• An open-source FLV transcoder. This may be old news — it was posted to the echoditto del.icio.us feed a little while ago. But it's still pretty exciting to me. Using it, an enterprising geek could grab content from YouTube (or any other Flash video source) and programmatically remix it however they want. Creating an automatic video montage based on a particular tag might be a fun project.
• The Xbox 360 may have been cracked. There have already been exploits related to the DVD drive's firmware that allowed burned DVD-Rs to be played, but that only facilitated piracy. Running alternate operating systems or homebrew games remained impossible. But now someone is claiming to have accomplished this feat.
  
  The presumed route of ingress? The shaders on a particular game. Shaders are small executable subprograms that are run extremely quickly by graphics hardware — they're not part of the game's main executable. Instead, they're responsible for things like drawing the surface of water, or making hair look more realistic. They also may be less protected on the 360 than the main executable — a demo disc that was distributed to stores over the internet contained shaders that weren't cryptographically protected and that could be modified and reburned to the disc. Once hackers realized this, the search was on to find exploits that let software authors get at the main system memory and proceed to bend the system to their will.
  
  Now someone has shown some sample video at a conference implying that they've succeeded (although the equipment they used implied that there may be a hardware aspect to the crack as well). Is this a real exploit or just a hoax? It's hard to say — when the route of attack is through a graphics-related subsystem, changing what's on the screen doesn't necessarily indicate that you've fully compromised the system in a useful way. But this is all pretty encouraging.
• Most interestingly, AACS may have been cracked, too. It's a little early to know for sure, but it looks like the copy protection on the new HD-DVD format may have suffered its first setback. A user calling himself muslix64 claims to have found a way to get the title keys for discs by extracting them from the PowerDVD software.
  
  Lots of folks are running around forums saying that this isn't a crack — and it's true, muslix64 didn't find a way to beat the AES encryption. But that's just semantics; if his claims are accurate, he's found a way to get the keys to it, which is the same thing that happened to the thoroughly-broken Content Scrambling System of the DVD format (flaws related to its implementation of the cryptography system were also found later, but only added insult to injury — the damage was done).
  
  AACS is better-designed than CSS, though, in that it allows for key revocation. Here's my understanding of how it works, in a nutshell: each disc's data content is encrypted with a title key that's unique to that disc. Along with the data, the title key is present on the disc — but it's also encrypted. Because it's short, a bunch of different copies of this encrypted title key can be placed on the disc, each encrypted with a different player key. If your company wants to manufacture an HD-DVD player (hardware or software), you must apply for and receive a player key from the HD-DVD format's governing body. Your player will use this key to get at the title key, which is in turn used to to unlock the content on the disc. It's sort of like those little boxes that realtors keeps on the doorknobs of houses that are for sale: if you want to look at the house, you enter a combination and get its key out. Each realtor could get a unique combination, but each property only needs one key. I may be drastically wrong about this analogy — I don't know that much about how realtors operate. But this is how HD-DVD does.
  
  The advantage of this system is that if one player is compromised, the format's protection isn't totally boned. Instead, the HD-DVD consortium identifies the player and stops putting the version of the title key encrypted with that player's key on all discs that are pressed in the future. So if a player is compromised, every disc created prior to that date can be cracked, but all the ones produced in the future won't be. And the player manufacturer will have to get a new key (assuming the HD-DVD consortium is willing to give them another one).
  
  Another thing to realize: you own your computer's RAM, and with the right software you can get at any of its contents. So when you run a program that plays an HD-DVD successfully, it means that somewhere in your system is that player's key, and, during the program's run, the disc's unencrypted title key. There are various ways of obscuring these keys and getting them out of memory as quickly as possible, but with enough monkeys nerds and typewriters debuggers, you'll eventually find it.
  
  That seems to be what's happened. With the DVD format it was Xing's player that leaked its key — and under the DVD protection scheme there was no way to revoke the key. With HD-DVD it appears to be the PowerDVD program that's been compromised, and its key can and probably will be revoked. So this crack probably won't work with discs released in the future.
  
  But it's also likely that more software players will have their keys stolen. The only way to avoid that is to stop issuing keys to software players, or to force everyone to use a trusted computing hardware platform, which would prevent users from indiscriminately accessing their systems' memory (this is why geeks hate & fear the idea of a TC initiative).
  
  The real fun will happen when a hardware player's key is leaked and revoked (it's considerably harder to get at hardware keys, but with the right equipment and expertise it can usually be done). Everyone who's bought the player in question will be left with a box that can't play any new movies. Odds are that consumers won't be very pleased with this development. If I had to put money on a single event decisively turning the public against DRM, I'd put it on this.