One argument supposedly favoring online communities like the BIT320 Blogosphere is that they can scale beyond what a “real world” community can.
One datum is that real world communities can stay cohesive up to about 100 – 150 members. The idea supporting the scalability of online communites is that they make contacts between community members more efficient. However, I really wonder if that is all there is to it. The real issue to me seems to be keeping up with what is going on in the community, capturing the zeitgeist.
How do you do that? I've been having some discussions with Andy Seidl, one of the MyST boys, on that topic. He generally is of the opinion that what you want to capture is threads of ideas. He spends lots of time figuring out how people and machines should organize information to make it more easily archiveable on the web and discoverable by web robots.
That definition of relevance works for impersonal search. I don't think it works for the notion of zeitgeist which is inherently social. When capturing zeitgeist, you want to know what a certain group of people is thinking at the moment. You're not looking for a specific thing related to what your doing or hoping that help will come to you in the task you are trying to accomplish. You're looking to see what the buzz is.
Capturing that idea in a blogosphere may be as simple as creating an aggregated web page of recent posts in reverse chronological order. It could also be captured by organizing news feeds in an aggregator. However, this second step requires that the user has a fairly high level of sophistication. Further, it may run counter to group building efforts where you are trying to capture a whole class. People using aggregators are free to choose what to aggregate and what not.
Update: Andy Seidl has posted a comment extensively stating his position on this topic, which is rather nuanced. The comment makes for interesting reading.
I feel compelled to clarify what I actually believe. I am not "generally is of the opinion that what you want to capture is threads of ideas." While this is an obvious information use case, the statement is far too limiting. I am of the general opinion that we are after the Holy Grail: effective knowledge capture, persistence, enhancement, and transfer.
I believe the road to the Holy Grail must be built on fundamental and very abstract elements. Specific use cases--capture of idea threads, social interaction mapping, discovery methods, and many, many more--are all not only possible, but practical, but only if you have a sound underlying information architecture. The first rule of thumb for any KM architect is that the use case requirements will change; the underlying architecture must anticipate change; it must be agile.
I do spend a great deal of time thinking about the best ways to organize information, but not "to make it more easily archiveable on the web and discoverable by web robots." I worry about information agility--will we be able to easy realize KM use cases that we realize are important now and those that nobody has thought of yet? [1] As it turns out, focusing on the fundamentals of agility has lead to architectures that do make information easily archivable on the web and easily discoverable by web robots, but that satisfy many other use case requirements as well.
Capturing the zeitgeist--the general intellectual, moral, and cultural climate--of a blogosphere is a great example of a difficult KM problem; one whose requirements will continue to shift. One approach is (for you, a human) to actually read every post in the blogosphere and develop your own subjective sense of zeitgeist. For this approach, it would be helpful to have tools (e.g., aggregators, composite "recent post" summaries, a unifying blogsite, etc.) that draw your attention to, and perhaps even summarize, new postings and simplify navigation to related postings. But, in this scenario, the real burden of knowledge processing remains a human task (which, of course, makes it difficult to scale).
To have a machine not only capture, but help humans recognize the zeitgeist is much more difficult; some might even call it an AI problem. To achieve this in any meaningful way, the underlying information architecture must provide a rich semantic representation of blogosphere activity--not just the text of each post, but many relationships (formally, typed associations) between individual posts and the various other concepts relevant to the blogosphere (i.e., its ontology).
For more info see:
A Holy Grail: Effective Knowledge Capture, Persistence, Enhancement, and Transfer
http://myst-technology.com/mysmartchannels/public/item/53612
Social Transactions
http://myst-technology.com/mysmartchannels/public/item/53727
Agile Content
http://myst-technology.com/mysmartchannels/public/item/5675