The personal peer to peer universal media publishing system (PP2PUMPS) is a way for the average user to publish everything they ever create on their computer via a peer to peer system.
What is the purpose?
I personally have created thousands of pieces of content and media but it takes a lot of effort to package up pieces of content for distribution. This software system would handle most parts of the media file publishing automatically. The purpose being to allow users to publish anything and everything they ever create on their computer automatically with limited bandwidth usage. So much content that people create to amuse themselves on their own machines is lost because they either never publish it, they reformat their hard drive, or somehow it is lost.
The main thrust of this application would be to help speed the conversion from a "producer -> consumer" (server/client) model to a "proconducesumer <-> proconducesumer" (peer to peer) model of media creation and consumption. Everyone is a producer and a consumer at the same time.
What is the implimenation?
The user would have a service running on their computer which would watch for file creations and file updates. The service would have a set of defineable file types such as MPG, AVI, JPG, GIF, DOC, TXT, MOV, etc. It would be similar to the "Recent Documents" functionality in Windows. However, instead of being in the start menu it would live in the Windows Taskbar as a "Deskbar". If the Windows Taskbar was at the bottom of the screen it would consist of a drop down box. If the Windows Taskbar was at the left or right of the screen (optimally) it would be a listbox of the most recent watched file changes and creations. The files that you most recently worked on would float to the top of the list.
However, every file in the "watched" list would have an invisible metadata file created and updated for it. This metadata file could have the same name as the original file and could end in .metadata. The file would be set to "hidden" in the file system. The contents of the file would be readable and writeable either in XML or INI format. Optionally the .metadata file could be encrypted with a user defined password. The purpose for encrypting the file would be to keep malicious programs from modifying it. Third party applications could access the metadata files given the user password.
The metadata for the file could consist of the original creation time, a log or each additional file modification time, Creative Commons licence information, other optional metadata such as EXIF data from camera JPEGs, file publishing permissions(?), and a wiki style log of all file changes. The wiki style file change log would or could allow you to revert to any version you ever created of that file. File storage is at a point which allows us to have every version of the file ever created.
The file publishing permissions could be implimented in a number of ways. It could be based on a setting in the metadata only. Where the server application says okay all of these files which are set to "share" in the metadata are automatically packed into archives which include their metadata and published via the BitTorrent? style P2P system. Maybe all of the published BitTorrent? file links are listed on a personal web server. Or the server application could require that if you want to publish a file you must place it in a specific directory. All files in the specific directory are automatically packed in archives and shared via a BitTorrent? style system.
You would be able to right click on each file in the "recent files" list of files in the Windows Taskbar Deskbar component of the service and set this publishing permission (Share or Don't Share). Don't share would be the default of the application. The application default could also be set to Share. Each file could have it's own "Share or Don't Share" permission stored in the metadata file. The file sharing would be built using BitTorrent? or some more of BitTorrent? style technology for the P2P file distribution part of it. I am unclear on if each user would be hosting their own Tracker or if there were a host of public trackers to choose from or even decentralized trackers.
Additionally, you could right click on the files in the "recent files" list and set the Creative Commons licence for that specific file. The default licence for each file would be "Attribution" only. All of the other CC licence types would be supported as well including a "Public Domain" licence setting.
There could be a remote backup component built in as well. All of the shared files would be backed up remotely to a site like OurMedia?, third party remote backup providers, or even to your Google Gmail account. When the main source of a shared file (the users machine) was offline and no other copies of the file were available via other people sharing them the backup source would be used for the download via HTTP.
How it works?
The way the system works is you would use your computer normally. Each file that you create (depending on your default settings) would automatically be published for other people to access. For example, if you create a 3D model and your settings are set to share say 3DS files as soon as you saved the file it would be shared. Each time you save the file the metadata (including the previous version) would be updated and re-shared using your default Creative Commons licence. The same goes for any other file you might create like textures, music, sound, video recordings, or virtually any other media.
