It works!...But what does that mean?
Creating the Second Life (R) GeoWiki was a multi-step process that of course had some unanticipated speedbumps.
First of all, it turns out that the SL(tm) map API, which is basically the only real map available, was way too small for my purposes. It was Amanda who came up with a brilliant and elegant solution:
make the map myself by going in-world, flying to a sufficient height, and taking a snapshot (using the built-in feature) while looking straight down. As she put it, "You are your own satellite."
With the map images created, the next major step was actually getting the GeoWiki to run. The application is tremendously flexible, running in Flash and XML, but the documentation is way too succinct.
It took a lot of experimentation and frustration to get it to work properly. Part of that was not just customizing the application, but collecting the data to use for the inital set of annotations.
This involved several steps: 1) collecting references to SL locations online; 2) translating the related coordinates from the SL system (which is an X,Y,Z system with the 0,0 X-Y origin at the bottom-left
of the space) to the GeoWiki system (which is latitude, longitude - that is, Y,X - with the origin in the center of the space); and 3) putting all the data together in an XML data file. This was quite the
calculation-intensive process. Not to mention time-consuming, which means that a distributed-labor approach, in which all of the community becomes involved in the process, is quite attractive.
But when it was all complete, the maps had a lot to say. For Healthinfo Island, it turns out that every reference but one was either a wiki entry or a blog entry. This reflects that the types of
people making the entries are documenters, people who keep track of events and information. It makes sense for a region that is explicitly about information, but it becomes startlingly clear with the
large number of blog icons on the map. Additionally, there are visible clusters of entries. It is evident that the critical GeoWiki can be a traffic-analysis tool, as it offers a spatial dimension
to traffic-volume data. This information can also be used to show where events tend to occur, which might be useful for sociologists and anthropologists, region designers, virtual realtors, or marketers,
among others.
In the case of Kerlingarfjoll, what the map said was a whole lot of the proverbial nothing. The fact is that there were no data to be had outside of a set of co-located UCSB references and a bunch of broken,
outdated links. Which brought up the question of what to include in the data set. In the end, I opted to leave out broken links, outdated information, and non-specific references (those that might mention
a region by name without referencing a specific location therein). Time will tell whether or not that was a reasonable choice.
There are countless other things to be learned from this applied methodology, and with thousands more regions to map, the prospects are both daunting and exhilarating. It is, of course, too large a project
for one person to undertake, but a community effort - including a legion of intrepid explorers to take bird's-eye photos, online researchers to collect and verify data, and detail-oriented organizers to format and enter
it - could get it done with relative efficiency. The idea has a lot of promise.
Back to main.
END OF LINE.