* The XHTML Friends Network
Here’s an interesting idea: The XHTML Friends Network or XFN, a simple way to represent human relationships by embedding attributes in hyperlinks.
The focus of XFN is for categorizing relationships between the owner of Web content (such as a site or a blog) and the owner of some other Web content. For example, each entry in a blogroll (the list of blogs that a blog or Web site owner lists as being important or meaningful) can be modified to describe whatever relationship exists by simply adding a ‘rel’ attribute to the each tag. Thus, you might have an entry in your blogroll such as:
The rel attribute above a “contact” is defined as, “Someone you know how to get in touch with” and the proposal notes that this relationship is “Often symmetric.” The “met” attribute indicates, “Someone who you have actually met in person. Symmetric.”
The idea of symmetry in relationships is equivalent to the commutative property in mathematics (e.g. a+b = b+a), thus, meeting someone is symmetric (if you met me then I met you) but knowing how to contact someone isn’t always symmetric (if you know how to contact me I don’t necessarily know how to contact you). Relationships can also be transitive (as in math, if a=b and b=c then a=c) and inverse (father <–> son).
The range of relationships offered in XFN 1.1 include identity (me), friendship (contact, acquaintance, friend), professional (co-worker, colleague), geographical (co-resident, neighbor), and family (child, parent, sibling, spouse, kin).
XFN 1.1 also includes romantic relationships (muse, crush, date, sweetheart), which implies more than an eye towards social networking as it applies to the MySpace crowd.
This work has an obvious link with Tim Berners-Lee’s Semantic Web (see descriptions on Wikipedia and W3C). In addition, an explanation of the ideas behind XFN can be found on the project’s background page.
Lest you think that XFN has no traction in the market check out the XFN Tools page. It shows that a number of blogging systems support or can be configured to support XFN, and that there are several XFN-specific editing and browsing tools available.
The problem with XFN in the consumer Web is that there isn’t an obvious value to using it. Consider MySpace: On that service you can explicitly list your connections, and categorizing them has no value.
Over in the business world, companies such as LinkedIn only require a minimal amount of relationship classification because the context (business connections) provides all of the relationship detail needed. Moreover, it is to LinkedIn’s advantage to not use an open standard that could dilute the value of their service.
As the value of the semantic Web becomes more obvious and necessary, standards such as XFN will become far more common. Indeed, I expect that their use will most likely be driven by personal productivity applications through built-in services (imagine your copy of Outlook finding all links related to personal contacts on your Web site and automatically updating those links according to the data in your name and address book and your e-mail exchanges). Until then, XFN and similar initiatives bear watching to see where the business value and commercial opportunities will come from.
