Clay Shirky's article on weblogs and power laws has stirred a lot of comment. Dave Winer thinks that Shirky doesn't get it:
"The scaling equation for weblogs is, emphatically, not like BBSes, mail lists, not like the Well. The popularity of this weblog does nothing to interfere with the growth of lawblogs, or warblogs, or bizblogs, medblogs, governmentblogs, divinityblogs, you name it. Perhaps within each there may be some hierarchy because humans build hierarchies like other primates. No big news there."
I don't necessarily see an argument here. There may be a power law distribution of links across all weblogs and at the same time local clusters of links, each with their own power law distribution (probably until you get down to very small clusters such as a group of friends).
In response to these comments, Ross Mayfield has a useful framework for looking at the weblog ecosystem. He starts with two observations:
- not all links are created equal, and
- conversational relationships are not scale-free
He then summarises Shirky's three types of weblogs:
"(1) Blogs-as-mainstream-media: a point to multi-point distribution of weak ties that realizes economies of scale.
(2) Blogging Classic: The Magic Number 150, a multi-point to multi-point distribution of weak ties. "The figure of 150 seems to represent the maximum number of individuals with whom we can have a genuinely social relationship, the kind of relationship that goes with knowing who they are and how they relate to us." Robin Dunbar
(3) Blogs-as-dinner-conversation: The Strength of 12, a point-to-point distribution of strong ties. When most people are asked to list whom they would be deeply affected if they die, a measure of strong relationships, the average list is of 12 people."
Why are there different types of weblogs with their patterns of weak and strong ties?
"Relationships take time. A strong relationship requires a continuous investment in time to stay current. Trust is built from this investment. A weak relationship requires no continuity, an affinity where time costs are optional.
A time investment is a requirement for defining a relationship. And according to Duncan Watts, as you start to ratchet up the requirements for what it means to know someone, connections diminish. Which changes the distribution of the network. The Distribution of Choice maps to three distinct networks, each optimized for a different time investment to realize relationships and each with a different distribution."
These three distinct networks, mapping to Shirky's three types of weblogs, are:
- Political networks based on representative weak ties instantiated by a link.
- Social networks based on functional weak ties instantiated by an investment in time.
- Creative networks based on functional strong ties with an active and continious time investment.
Mayfield discusses each of these [excerpted with some discussion snipped]:
"The Political Network is based upon representative weak ties instantiated by a link. A hub designs itself as an institution, optimizing the transaction costs of information flow for point-to-multipoint distribution and feedback. This allows it to scale - creating a Scale-Free Network, or Power-law distribution.
But within the Political Network each hub also has its own Social Network. This Social Network of stronger ties has a lower transaction cost of passing information, and consequently sways the activities and decisions of the hub with greater influence than the readership.
The Social Network is based upon functional weak ties instantiated by an investment in time such as conversational inter-linked posts. A Social Nework is transactional by nature, with the means of establishing a relationship commoditized. Close to the Law of 150 in scale, a time investment is made by each node to be at least peripherally conscious of the other nodes and the information flow between them.
The Creative Network is based upon functional strong ties with an active and continuous time investment. Instantiated by real world relationships with a firm foundation of trust with dense inter-linking. This is the core of a person's network and serves as the basis for regular collaboration and production, leveraging the Strength of 12. The requirements for a relationship of dense interconnections are so high that what remains is a bell curve in distribution."
In discussing Social Networks (~150 people), Mayfield says: "One design challenge for social software is extending the capabilities of people to hold a higher number of meaningful conversations and cultivate relationships." I'm not convinced that this is or should be a design challenge for social software. Research by people like Robin Dunbar suggests that for genuine social relationships, a maximum of around 150 people is more or less wired into the human brain. Significantly extending this number may well mean weakening the ties between the participants, moving rapidly from the Social Network to the Political Network. A more pertinent challenge is that, given this human scaling constraint, how do we help larger communities and organisations to operate effectively across and between both the small tight-knit groups (~12 people) and the mid-range loose-knit groups (~150 people).
11:03:24 AM 
|
