Community and social computing
with Tags: social_network_analysis X
Every time I see another piece or book on social network analysis, I have to say something, particularly because the term is used so often these days.
There is definitely a science behind social network analysis that has existed for years. In simplest form, it maps the relationships between different people in a group, and draws the connections between folks. This contrasts against the hierarchical structure in most organizations and companies. In other words, the actual way people connect and interact in a group can often be very different than reporting structures and formal relationships between different teams. SN analysis examines these relationships and helps to point out bottlenecks, misdirections, and point out key people in your organization.
However, the shortcoming of this approach is, first, it requires that the members in the group all participate and respond to a questionnaire (and honestly) on who they interact with. The truth is that we may remember the first set of fewest people we interact with the most often (e.g., those folks on your speeddial, "myFaves" in Tmobile, etc.), but there are many others who we do not interact with so often but still hold value. So what SN analysis uncovers is frequency more than accuracy. That's the second part, accuracy implies quality, and the quality of relationships between folks are hard to determine, especially when you consider that people rarely record and evaluate every single interaction they have with others.
The diagram on page 99 of this month's Fortune magazine is pure fantasy--it is not on the online version of the story. It is unlikely you will ever uncover this kind of information, mostly because it is very invasive and raises many privacy concerns. Don't get me wrong; there are probably people who really do try to create such a map, but it can get horribly complicated over time and with larger populations. The reality of what you can get is shown in the article, luckily.
The article does however, point out the reality: there is a hidden network of how people interact in a workplace. You cannot directly measure this to a really fine degree and probably shouldn't even try. Every manager knows that they should get to know their employees to some degree; some managers can remember more about the work relationships and history of their employees than others; in that way, they build their own mental map of their surrounding hidden workplace.
How does this relate to "social networking" in the Web 2.0 sense (i.e., blogs, forums, wikis, tags, etc.)? When you get frequent and regularly participating members in any community, you have this same kind of invisible relationship network. I won't say it is impossible to map this out; someone might eventually prove me wrong. Software makes it easier to track interactions since there is a "written record" somewhere. Smart software may be able to automatically analyze your interactions with other folks to help you keep track of your own social circle. I have seen this at work already. It can look at how you email, instant message, chat, online event, blog comment, forum reply, etc. that you make with each individual and create a map for you. This is harder than it sounds because it means the software has to look in many places. It also means that the analysis software needs to be compatible with each of these applications which you use; since most people don't use the same set of such tools that may not even be in the same site or organization, it is really hard to track.
This points out to keep track of relationships more than messages. While we still tend to think of email as the main way we communicate online, this is already starting to spread to many other types of applications. Eventually email may no longer be king; even if that day is a long way off.
This also points out another behavior that is an unfortunate side effect of the availability of peer networking social tools like LinkedIn. The value of social network comes from deep or meaningful realtionships with others, not quantity. It may be a while before this sinks in so reiterating it frequently is a public service. It's not how many people you connect with but how many of these are meaningful. Quality over quantity.