A Wayback Machine for virtual worlds?

One of the joys of exploring virtual worlds grids is finding new fantastic new regions and builds. Then comes the pain of virtual worlds — you bring your friends back to visit, and the builds are no longer there.

Maybe it cost too much to keep the region up and running, or the owners decided to put up something else, instead.

As Second Life and the various OpenSim grids evolve into a true 3D Web, these early builds will become items of historical interest. It would be a shame if they all disappeared forever.

Today, for example, I visited the Steelhead Shanghai region of Second Life, after learning about it in a Designing Worlds program.

I’m sure that this and other “steampunk” regions will help contribute an aesthetic to virtual world design moving forward, and design students will someday want to come and visit.

If this region was a Website, it would be archived by the Wayback Machine run by the non-profit Internet Archive, which currently archives over 150 billion pages.

Somehow, the Wayback Machine is able to operate despite the copyright issues involved.

Eventually, of course, the Wayback Machine will be extended to cover 3D content, or a new archive will arise specifically dedicated to virtual worlds.

Meanwhile, abandoned builds will be lost forever. Everyday, we get news that someone has left Second Life, and shut down their region. Or, more often, the region is shut down quietly, with no word at all.

It doesn’t help that there’s currently no easy way to make a backup of a Second Life region, as there is of OpenSim regions.

And there’s the fact that regions have to be up, live and running, in order for someone to visit them — a very large drain on resources.

One solution could be to save regions into an archive and only bring them up live when someone wants to visit them — after all, an archive doesn’t have to have all the same functionality as a live world. In addition, just as the Wayback Machine doesn’t archive all the live functionality of a Website, so a grid archive doesn’t necessarily have to save all the scripted behaviors and special effects of a region.

Would it be too difficult to imagine a similar project archiving early virtual worlds?

Meanwhile, the technology necessary to index virtual worlds and archive them can have important commercial offshoots. There is currently no universal way to search grids, for example.

And the techniques used to quickly bring up live regions could be useful for worlds that need to model sizeable areas of land — planetary surface surveys, for example, or three dimensional maps of ocean floors — but don’t necessarily need all the regions up and running all the time.

Bringing up regions on demand would also be useful for companies that need extra conference rooms for meetings, or for adventure games that want to have large territories for players to explore or battle over, without wasting resources when those territories are unpopulated.

Maria Korolov