1) Social Relationships Must Meet Four Preconditions: Willingness to establish a relationship with someone presupposes the existence of mutual trust, respect, context, and self-disclosure between the parties. These qualities are exceedingly difficult to achieve without face to face contact. Social Software tries valiantly but for the most part unsuccessfully to provide a basis and framework for these qualities to emerge, using systems of recommendations and endorsements to create reputation, and degrees-of-separation referrals to create verification. The best these tools seem to provide is a rough filter of inappropriate candidates for relationships, a 'negative assurance' that there is no reason to believe you should not enter into a relationship with someone. But we tend to be distrustful of people's online identities because they are so easy to fake and exaggerate...
2) Relationships Require a Conversational Ice-Breaking: Conversations are much denser and hence more efficient and effective ways of transferring a lot of information while filling in context, than asynchronous communication mechanisms. Likewise, interviews are much more powerful ways to glean information from a subject than surveys...
3) First Impressions Matter: Our quick Blink judgements about a person, an event or an idea (provided they fit with our frames of understanding and belief) are critically important, and seem very difficult to dislodge with later information, no matter how compelling. Perhaps that's why visual clues about strangers are so highly valued in establishing trust, context, disclosure and respect. "Sniff, sniff, it's OK, he's one of us". Once we have established an impression or initial judgement, what we seem to seek most is reassurance that this initial assessment was valid. This introduces some obvious dangers: ideological echo chambers, groupthink and the proliferation of conspiracy theories for example. And just to make the situation worse. we tend to ignore and turn off information that we cannot (or don't want to) change, which further entrenches those first impressions and judgements. Physical appearance hence becomes an icon of our identity.
4) Information Conveyed by Observation Counts More Than That Conveyed by Language: We instinctively give enormous credibility to our direct sensory observations, while we tend to consider what we're told skeptically. Observation is a rich source of fast understanding and shared context. Our senses simply have much more information-processing bandwidth than the part of our brain that processes the spoken and written word. As a result, it may actually be more important to a relationship to see the other person's environment than to see them. As Dave Snowden says: "We can always know more than we can tell, and we can always tell more than we can write down". Observation allows us to mine data that the person to whom we are speaking cannot convey in words, or may not even be aware of. "Don't tell me, show me."
5) Collaboration is the Miracle Glue of Relationships: Doing something together, the more participatory and challenging the better, immediately establishes deep trust, respect, shared context, disclosure, even a shared identity (e.g. Lennon-McCartney). Collaboration is also an essential precondition to real community, far more effective than any amount of shared gossip, online chat and head-nodding agreement. Until we've worked on something together, our agreements may be simply politeness, insufficient to form the basis for a strong and lasting relationship.
6) Every Interaction Carries the Burden of Our Entire Networks: Throughout any relationship, conversation or interaction our brains are processing a series of social constructs: (a) Information (what am I learning from this interaction?), (b) Meaning (so what -- what does this mean to me?) and (c) Implications (what should I do, decide or expect as a result?) Information, meaning and implications are rarely just personal -- in each interaction we are considering how to forward or explain information and its meaning to others in our networks, how to justify and discuss its implications with others in our networks, and how to motivate others to take what we think are appropriate actions or decisions as a consequence. So there is a huge invisible 'audience' for each network interaction beyond the direct participants in it. Social Networking Applications need to recognize and involve this audience (by recording and forwarding the interaction, by inviting others affected to join in etc.) "I appreciate what you're telling me, but how am I going to explain and work this out with A, B and C?"
7) Social Networks are Complex Systems: Social Software is designed as a solution to a complicated, rather than a complex, problem. We hope that we can one day in some way completely diagram, understand, and optimize use of our complete social networks, but the best we can hope for are possibly dangerous oversimplifications. The complexity of our networks simply cannot be fully known, explained or applied in any predictive way -- there are too many variables and nuances of relationship to fully know, depict, or expect complicated-system software to handle. We should probably therefore be humbler about what we hope Social Networking Applications will be able to do for us. Their goal should perhaps be simply to better understand how our networks benefit us, how we can use them more effectively, and how to improve the quality and value of conversations and other interactions.
The rest of the post (Dave is nothing if not thorough) consists of a series of questions that greatly illuminate a development path for the next-generation of social software. If you're looking for a world-changing programming project, check this out. Seeing as how We are one of the most sociable societies in the world, who better?
The Jenius Has Quoted.