OK. I feel I must speak out. Instead of the geek community cowering away waiting for the RIAA to come directly to our doorstep and sue us within an inch of our lives, I propose this... A new generation of application should be written. Designed using encryption end-to-end. Allowing multiple central points for search and still total anonymity to the users. The technology is already there. It just needs someone to be stitched together. Each time Big Media plugs a new "hole" (i.e. removes any innovation), the geeks of the world go back to their compilers. They look at the old methods and make them better. A look at old methods... Napster made an easy target... just kill one central server and the world is back to normal. Gnutella was very slow and offered no swarming downloads. Bittorrent does not use any encryption and requires a central server easily taken down. It does not mask the identity of the downloader or the uploader. Here is my idea... Should be written in a language that does not support "managed code" and should be completely open-source. This way all encryption will be open and less likely to be tampered with. Use port 443 and SSL encryption end-to-end. SSL traffic cannot be stopped on the Internet. After all, closing this port will basically stop all forms of E-commerce. Make all communication (even at the bit-level) look just like data being transferred on the web. Deep packet inspection can also be sidestepped be making the traffic indistinguishable from other forms of communication. Superhubs that aggregate data shared on each machine. These should be surfed like a web page, complete with search, photographs and meta data. This will allow the convenience of surfing for torrents, without having any central server that can be taken down. Over time, the best of these will rise to the top and gather a large group of users. Each hub can either be an actual server, like one at Pirate bay, or just a client with an add-on package that allows it to catalog the data of other users. This catalog can be stored in RAM, so a simple reboot of the server will dump all data. Most modern machines have gigabytes of RAM and could store filenames, some metadata and a link to a photo or two for each file (if needed). This data is textural and does not take up that much space. Moreover, each time a database goes down, it can connect to the users again and re-sync. The Superhubs could also use some form of voting system to keep the best downloads at the top and allow the worst to die from the network. Swarming downloads. Use SHA1 to hash each file an allow multiple users to connect to connect to each other and share parts of a file. Use the bittorent method of distributed downloads. Store no log files or IP addressed to disk. All peer info stored in RAM. Each time a client starts again, it will peer out and find some nodes to connect to. This can be done using a seed file downloaded from a secure website, or a URL to a SSL webserver. For example, Pirate Bay could have a connection peer that catalogs SuperHubs. This would be very difficult to get this seed data removed from the web. Use TOR, or some form of Onion Routing technique. Allow each client to route messages from one node to another, in true peer-to-peer fashion. The allows users to pass messages from one node to another while preserving the anonymity of both the sender and the receiver. (Please see websites about TOR for more info). However, this would not use the current TOR network, because it is not made for this type of traffic. With this feature, it is very difficult to determine who is hosting files. This routing method should be used for both searching SuperHubs and all transfer data. We need to stand together on this issue. There will never be a better time to stamp out DRM and bring equality back to the Internet. It only takes one person to write the application that will bring freedom to millions. FTRIAA!

i second that notion.