There has been a lot of activity and discussion recently among online social networking practitioners and thinkers concerning how to describe and mark up types of relationships between people. The FOAF initiative rumbles on and gathers support, as does the simpler XFN, whilst two researchers have proposed a lengthier RDF controlled vocabulary called RELATIONSHIP.

Clay Shirky makes a clear case for rejecting RELATIONSHIP and similar attempts to narrow down our multi-faceted inter-relations to a few binary terms, and I think he is quite correct to do so. danah boyd continues from this to point out that explicit descriptions of types of online relationship miss out on three vital elements: context, culture and power, none of which can be adequately captured or represented using RDF. Foe Romeo discusses this point as well, but is more amenable to the basic concept of relationship markup, quoting Ideas Bazaar in support of the idea:

"Anthropologists understand that kinship operates at three levels: terminology, rules and practice, and the inter-relationship between the three of these. This means at the categorical, jural and practical level: how are people related, what terminology is used to describe their relatedness, what behaviour is 'meant' to obtain between them (joking / avoidance?), and what behaviour does obtain in practice. Shirky seems to confuse the existence of a terminology with static relationships and fixed behaviours obtaining between people in this relationship. Anthropologists understand that a dynamic interplay exists across these 3 levels."

Despite this reading of anthropological practice, I fail to see the value in trying to apply common terms to online social networking relationships. We are in the early roll-out phase for a large, diverse knowledge community that seeks to join up people with a serious interest in the improving the delivery of mental health services and whilst this system supports the construction of online social networks to promote knowledge sharing, we have yet to find a convincing argument against people being allowed to describe their linkages with other people, nodes, organisations, etc in free text terms that reflect their own perspective and terminology. You can still conduct searches of material rated highly by your network and other value-added social networking features, but until we find a compelling rationale for machine-readable (as opposed to human-readable) markup of these relationships, we will stick with the simplest, least threatening or exclusive solution.

However, we are watching the experience of larger, more well-established online communities for insights into managing and moderating user/member input, which is always a fine balance to get right. Clay Shirky also wrote recently about the online community Kuro5shin and its attempts to solve the age old problem of a few bad apples spoiling the barrel. Rusty, the site's creator, has taken a leaf out of the book of older, more conservative social networks such as golf clubs and gentlemen's clubs in general and opted for a system of user sponsors - you can't get in without a sponsor and if you misbehave then they are held responsible. Ironically, whilst young hipsters are discovering golf club etiquette, some middle class kids are moving in the opposite direction, discovering bling-bling kudos in a small gift economy they are creating within FunHi.