As in any campaign, reconnaissance of the battlefield is essential. We have outlined in the “Where do they live” post the difference between the four tools for community interaction and building, and how they are complementary to each other.
Every community will therefore select where it wants to live, and the first task of the observer is to figure that out.
Worried about Privacy?
It is important to remember that the observation we are conducting is completely respectful of the group members privacy. What we are reading is nothing but what a group member has decided to share with the world; in a way we are doing exactly what these people were hoping someone would do when they posted: listen to them.
This applies also to all information we can gather about specific individuals (see the “Leadership mining” tutorial)
Some practical advice for the cartographer:
Be aware of the Long Tail – “Long Tail” is the specialists’ jargon to indicate those websites, blogs, forums that form the bulk of any Google search results, but are usually hidden deep away from the first pages. The vast majority of sites fall into this category, but people usually pay very little attention to them because of the old traffic-centered view of the Internet (see next point). The reason for this is that by the time a story reaches a major Core property, chances are it’s too late to do anything to manage the issue and you can only turn to full-blown Crisis Management to limit the damage.
It’s the links, stupid ! – by now it’s clear that the new shape of the Internet does not use traffic as a key gravitational metric, but rather links. Of course, the two are connected more or less directly, but there are also important differences. A high link density may turn into heavy traffic under special circumstances, but if you are not under these circumstances, links are going to be just that, links that nobody follows. Traditional web stats used to measure traffic going to a site and infer its importance from that metric. In the new web traffic is a rapidly varying parameter. An unknown site might become the center of the Universe for a few days and then plunge back in obscurity.
How many people had heard about Mark Russinovich before Halloween, 2005? How many sites fold under the beating of a sudden surge in traffic because they’ve been slashdotted?
A moving target – traffic captures the static nature of the web, links capture the traffic potentiality; it’s a bit like dry gulches in the desert: 99 out of 100 times you look at them they’re dry as a bone, but this does not mean they’re always that way: flash floods happen, and people die from them. So if I want to evaluate how safe I am, I have to look at the problem from a more holistic point of view. How much water is flowing where is not enough: I need to know how likely it is there’s going to be a storm, how many dry beds converge where I am, how deep they are, etc.
In general, remember that the link-centered network is rife with monodirectional relationships: each blogger, each community member watches someone in particular. It would be enough to watch one’s bookmark folders to figure out readily such relationship, but of course we are not able to do this and must instead rely on observation. Microsoft Research also made available a powerful tool for social network analysis called Netscan which is able to track cross-posts between Usenet groups, thereby mapping proximity relations between communities.
What we are looking in this phase are data about each site population and likely readership: we must bear in mind that there are no hard and fast rules for this: contrarily to traditional websites, for example, newsgroups have no logfiles, hence making traditional statistics impossible to evaluate; however, one can observe forums and other places which usually publish data e.g. on post reads and derive an informed estimate about the order of magnitude for the lurker-to-poster ratio: this is metric that varies wildly with topic, community and moment in time: when Steve Jobs presented the 5th Generation Video iPod, the room was electromagnetically shielded to avoid bloggers posting away during the conference, which meant they had to physically walk out of the room into the hallway to beam them up – as a result, iLounge posts on the event received ten of thousands of reads each!