Tailorable blog

Limited options for applications in Second Life

For our new project on Future Workspaces, I have been looking into the possibilies of integrating applications in the Second Life environment. Frankly, I was a bit disappointed by the options and the limitations of the Second Life script language (LSL). First of all, in-world applications have to be constructed out of the basic building blocks of Second Life (prims). You can define limited functionality inside such a prim using the LSL language or let it communicate with the outside world using e-mail, URL requests or XML-RPC. So far, so good…

However, the LSL scripts have some serious limitations (as I discovered when I tried to program an SL poker table in my free time). First of all, a single script can only use 16K memory (welcome to the 21st century!) and there are all kind of unpracticalities, such as no nested lists and no library functions, so code reuse is done by copy-pasting code. Moreover, there are serious memory leaks and build-in delays, for instance when sending an e-mail. Of course it is possible to overcome these barriers, but this complicates any programming effort so much, that it quickly takes the fun away.

Because of these limitations, some operations that seem pretty standard become extremely complex, as illustrated by the story of displaying an external image by Dav Yaginuma, who even went as far as to create a series of cubes to represent individual pixels.

A few days ago, I saw an integration of Basecamp in Second Life and I was very curious how they did that. It turned out that it was a static screenshot, placed as a texture on a prim, with a link embedded to a Basecamp webpage. Clicking on the object just opened your browser to a fixed webpage. That is not quite the integration I had in mind.

Part of my trouble is of course that I don’t now yet what I would like to make to support knowledge workers in Second Life, but I figured that it makes sense anyway to explore the options. Anyway, if you know of any cool examples, I would be happy to hear from you.

A standard for avatars

Yesterday, I attended a talk by <a href=”http://en.wikipedia.org/wiki/Peter_Ludlow”>Peter Ludlow</a> at Twente University. After his talk we had a brief discussion on the need of a standard to represent avatars. In my opinion, such a standard is needed because of the following two reasons:

• More and more virtual worlds are being created. We already had Second Life, There, Entropia Universe, Active Worlds, Habbo, Croquet, Ogoglio, and a million more… and now even Coca Cola is creating its own virtual world.
• It takes a considerable amount of time and effort to customize your avatar, build its social network of real life and virtual friends, and build a reputation.

I would most certainly hope to see a standard arise that lets me transfer my avatar and its social network between different virtual environments. Of course there will be aspects of an avatar and her posessions that are platform-specific, so my avatar may look different in different environments and may have different possessions, but I am curious how much we can cover in such a standard and still keep it acceptable for the developers of virtual worlds.