Google launched Social Graph API. Using publicly available information, Google will begin to connect people and relationships/friendships from blogs and public social networks. As an example, let’s say you have a MySpace profile with friends. You would like to join Twitter and one of the pains of joining a new social network is building friendships again. With Google’s Social Graph one could check to see if the individual’s MySpace friends are on Twitter, and import those friends. There are some checks and balances to stop malicious activity in forming connections, i.e.: saying you’re Bill Gates when you aren’t. The API would check for references, for example, on a blog linking to a MySpace profile. If the same MySpace profile links to the blog under the “Me” section of the profile the relationship would be validated. Here are some examples of relationship links:

http://tymesaid.com" rel="me"
http://yoscrivs.com" rel="co-worker"
http://domain.com" rel="sweetheart"
http://www.diaryofawebsite.com/" rel="friend"


The process is automatically performed when Google’s spider crawls a site. If one would like to clearly define sites and relationships, the owner of the site can use XFN or FOAF to describe relationships linked on a site. A relationship can be removed by removing the link from the site. Google’s spider will pick up the change the next crawl. The individual can submit for a re-crawl if he/she does not want to wait for the next crawl. This YouTube presentation by Brad Fitzpatrick explains the API in more detail.

The one thing I like about this API: the relationships are starting to have meaning. As we talked about in the last 3by9 podcast, most social networks give equal weight to “friendships”. If social networks implement this API then perhaps they will adopt levels of friendship to allow multiple forms of communication with friends.

One caveat that came to mind: just because I am friends with someone on MySpace doesn’t mean I necessarily want them following me on Twitter as they are different services with different goals (one might use one for professional reasons and the other for personal). In Brad’s presentation he mentioned finding a connection between profiles with completely different names (oopsies if the person did not want the connection known). Because of this, I would strongly suggest deleting old profiles and 3rd party blogs in the event those companies decide to participate.

Keep in mind normal links do not count, so I decided to play around and see what I would find for 9rules for “me” links. The following results came up:

lutrov.com
pownce.com/9rules
twitter.com/scrivs
flickr.com/photos/34943008@N00
randomconnections.com

Pownce, Twitter and Flickr are valid. I raised an eyebrow on Lutrov and Random Connections. Here is what I found in the source code for Lutrov:

http://9rules.com/" rel="me" title="Highlighting the very best web content in the world, and packaging it in a nice bow for you to unwrap."

This is what I found on Random Connections:

http://9rules.com" rel="me" title="This is one guy's collection of websites, all extolling the virtues of CSS and standards compliance. He is also the founder of Moveabletype.org, great guru of weblogs. Sections of particular interest are Whitespace and the CSS Vault."

Who knew Scrivs started MovableType? Note the “me” links are the only ones Google picked up as the guys and I do link to 9rules on our sites. Perhaps the generous use of “me” links are automatically generated by the software they are using, who knows? You might want to check and see who is claiming to be you.

Interesting approach. It will be interesting to see if this catches on.