The GNU Project is taking a shot at photo sharing. On May 2nd, the wraps came off the GNU Mediagoblin project. If successful, the GNU Mediagoblin could solve several problems that haven't been addressed well by existing photo sharing services — namely privacy, data ownership, reliability, and software freedom.
So what's GNU Mediagoblin? The project is starting with the goal of creating a federated photo sharing site that could stand alongside popular services like Flickr, DeviantArt, Picasa, and Facebook. Eventually, the project hopes to tackle other types of media, but the first target is photo/artwork sharing. Right now? It's very much a work in progress.
The idea is simple: Instead of a centralized photo sharing site that requires users to register under (usually) fairly one-sided privacy and copyright policies, Mediagoblin would allow anyone to put up an instance and then share media with friends — whether or not they happen to use the same instance of Mediagoblin. Right now, if I post my pictures to Facebook or Flickr with any kind of privacy restriction, then anyone I want to share with has to be registered with that service in particular. If successful, Mediagoblin would allow users on different servers to share media with one another and avoid having to sign up for new services.
This also means users can, in theory, run their own instance of Mediagoblin — so there's no dependence on a third party and no reason why you have to grant access to your media to a third party.
If this sounds at all familiar, it's similar to the model that's being used by StatusNet and its Twitter-like service, Identi.ca. Users can join Identi.ca with the majority of users, or they can run their own instance of StatusNet — or have StatusNet host a separate instance for their company, etc. — and still chat with their friends on Identi.ca.
You can find code for the project on Gitorious (a FOSS-oriented, git-based, code-sharing site that takes after Github) — but I wouldn't count on it doing much. Right now it's primarily aimed at hackers that want to help push the service forward towards completion. The tentative timeline right now is to have something ready around September or October, but that probably depends very much on how many people get involved and whether it finds momentum.
Mediagoblin is being released under the AGPLv3, which means that if third parties put up services based on Mediagoblin they'll be obligated to share changes even if they don't "distribute" the software in the traditional sense.
Can it work? So far, Identi.ca hasn't made a dent (pun intended) in Twitter's traffic or relevance to the world at large. Whether StatusNet is making a buck offering private instances for organizations, I'm not sure — but the majority of the value in a social service is when it has critical mass. Twitter (and Facebook) have achieved this, Identi.ca hasn't — unless you belong to a very small community of free software focused folks.
Step one is going to be finishing the code, but I hope that the Mediagoblin folks have plans for offering hosting or finding partners to offer Mediagoblin hosting so that it can generate the critical mass that's necessary to allow users to abandon the proprietary and privacy unfriendly services like Facebook. There's little point in signing off Flickr, Twitter, Facebook, or whatever if the rest of your social network isn't on the free service you migrate to. I suspect that building the software, in this case, is going to be the easy part.
However, the Mediagoblin folks deserve kudos for starting the ball rolling. The GNU Project and Free Software Foundation have a history of telling users not to use "the cloud," rather than trying to provide services that fit its philosophy. Mediagoblin is a step in the right direction, providing services that users want while ensuring that it meets the GNU Project's philosophy. Here's hoping it succeeds.