Open Source Geek Toolkit (i.e. Personal Social Media Aggregation)

The intention of this session was to the interest in creating a low-overhead, open source CMS that aggregates all of your personal social media that you've generated across multiple websites onto your own website.

There was discussion of whether to use existing open source content management systems like Drupal, but the group's consensus was to roll their own solution.

A google code project has been created at ???, and further discussion will happen on the wiki at ???.

USER STORY

We're generating all of this "user-generated content" and posting it to other sites, but that's actually pretty difficult to automatically aggregate it to your own website, categorize it and resyndicate it. It should be a lot easier than it is.

The desire is to build a set of aggregation tools for geeks to set up and use that will pull in all of the content that you've created across the web -- from LiveJournal posts, blog posts, Flickr photos, statuses from Twitter, Facebook, Pownce, del.icio.us links, Videos from blip.tv or YouTube, etc -- and have a one-stop shop for everything you've created on your own website.

This system could also be extended to aggregating your friends media as well. In some views, you want to see the entire firehose of information. In other views, you may want to just see a set of photos from a friend -- or a group of friends.

You may also want to just see all of the media from a particular friend -- or clusters of friends. So the RSS feeds would be automatically tagged by source, media type and/or username..

It would also be really nice to have one text box that will automatically cross-post status-updates to the APIs of twitter, Facebook, pownce, etc. And ideally, the system would be intelligent enough to delete duplicate status postings -- but also duplicate photos if you're already cross-posting Flickr photos to your LiveJournal blog.

This system would also allow you to strip out extraneous information that is added to RSS feeds. For example, usernames or whatever other cruft has been added. It could potentially add CSS classes that would allow you to theme the various data in various different ways depending upon the source.

[NOTE: This is from memory. Please feel free to add any more information here...]