What Is The Database of Intention In SecondLife?

Richard Wallis caught my attention with the post 'There be nuggets in them there data!'; his references to the Database of Intentions was lot lost on me. I've read The Search: How Google and Its Rivals Rewrote the Rules of Business and Transformed Our Culture as well¹, and it's pertinent to many things - from the future of libraries to finding things in a virtual world, such as SecondLife and beyond.

Being able to find data that we are looking for is an important aspect of the internet, otherwise search engines would not have enjoyed as much success as they have. And in keeping track of what people search for, they create the database of intentions - and, in turn, the database of intentions tells us, in a quantitative matter, what people are looking for. The trick is to provide quality results to what is being looked for - and that is the battle which Google and other search engines fight every day behind their textbox portal to the internet. Some of it is automated, some of it is done with human eyes, but all of it is done to attempt to give quality results for searches. It makes sense that some triage would be involved based on what the most popular searches are for. Over time, patterns form - and defining those patterns is only possible through a database of intentions and a lot of hard work.

But even as that's happening, let's consider something like SecondLife. People log in and look around - using the search tab. No data on the search tabs seems to be released as a database of intentions - what people are looking for within SecondLife. And people also visit from outside SecondLife using SLurls. And then people teleport their friends over when they find something interesting.

So how does anyone know what is popular? Well, Linden Lab uses a traffic calculation within SecondLife to determine how popular a piece of land is, and from what I gather this has something to do with how long a quantity of people stay on the land. I'm not sure, but it's possible that one person staying in a spot for 3 hours might have the same effect as 3 people staying in the same place for an hour each. Hopefully it is more complicated than that, but I have no insight into how they calculate traffic. To figure out how many people visit, you can use a visitor counter (there are scripts all over), and find out how many people visit in unit time. It's possible to even see how long people stay with such a script (using the same scans for avatars that a visitor counting script uses). But even that doesn't tell you why they were there. If you sell something, you can tell if they bought it or not - and that could be useful.

As a bit of an attempt to bring about finding things easier, Second411 came about - but I've never found anything of use with those search results.

So the real problem of a virtual world is that it's difficult to figure out what people want - which is a problem for marketers. But it's also a problem of people who are looking for things - and that's the real problem. Tricks like loading ads with keywords helps in classified ads. Putting keywords in land descriptions helps if people search using 'places'. But nowhere is there a metric to check how effective these things are.

Handling the database of intention in a virtual world is not an easy task - but it will be required for scalability and growth. A bunch of lost people looking around, as happened over the last week, doesn't help at all. But it also doesn't help that we're not learning from what people are looking for.

Somehow, like the web, I expect sex related terms are pretty high up there...

¹ I reviewed The Search: How Google and Its Rivals Rewrote the Rules of Business and Transformed Our Culture here.