Wonderblog

I’ve just posted a new tutorial on the Wonderland 0.5 wiki site. It explains how to program
interactive 3D objects in Wonderland 0.5. The URL is:

http://wiki.java.net/bin/view/Javadesktop/ButtonBoxTutorialPart1

I participated in the Taxonomy of Virtual Worlds for Education workshop on Thursday of last week (I was supposed to be there on Friday as well, but unfortunately I got sick and could not attend the 2nd day). You can find a complete agenda of the workshop on the Taxonomy of Virtual Worlds for Education Workshop web site and can read about the Friday sessions, focused on educational applications, on this excellent blog by Erin Murphy.

Thursday was devoted to virtual world technology for K-12 educational applications. I gave a short Wonderland presentation followed by a 2-hour live demo. As you can see from the photo, I had Wonderland v0.4 running on one of the University of Pennsylvania’s classroom machines, and Wonderland v0.5-dev3 running on my laptop. This allowed me to show features from both versions.

As an aside, in the photo near my hands you’ll notice a Solo USB echo-canceling microphone. I highly recommend investing in one of these if you’re going to be doing Wonderland demos. You simply plug in a set of powered stereo speakers to the back of the device and everyone in the room can hear the in-world audio clearly and no echo is introduced for the remote participants.

Reflections on Developing Wonderland for K-12 Education

In reflecting on what I heard in my informal conversations with the mainly K-12 educators who were attending the workshop, I came to a new understanding of what it will take to make Wonderland a viable platform for this audience. They certainly liked many of the features of Wonderland. In particular, the educators appreciated the security aspects, the ability to run live applications in the world, and the fact that the software is free and open source. For some reason, they particularly loved the Cone of Silence. That said, almost all of the people I talked to were intimidated by the technology, feeling that the barrier-to-entry was too high.

Given where Wonderland is in its evolution, this reaction is not too surprising. The two worlds they cited most often as being easily accessible to educators were Whyville and especially Active Worlds, which was by far the most beloved platform. I found this Whyville demo video and this Active Worlds demo video on YouTube in case you’re not familiar with these systems. Some of the features people said they liked in these systems were the strong focus on educational content in the case of Whyville, and the ease of building with pre-fab components in Active Worlds. In addition, both environments have strong child safety controls.

I think the key is that the Wonderland toolkit is not appropriate for most K-12 educators. What they need is an educational environment that someone else builds using the toolkit. These environments can range from what I think of as "produced" activities in the form of pre-scripted, slick educational games, to much more open-ended, teacher-directed worlds. While teachers like the produced content if it fits exactly into their curriculum, many of them would like to create their own educational spaces.

Some Examples

Our job as developers is to figure out what primitives teachers need for building worlds. For example, could we create a history bundle with libraries of historic building models and period avatar costumes? Teachers could then set up a scene from ancient Rome or the Wild West, have students dress up their avatars, and re-enact scenes. Students re-enacting the US Revolutionary War period, for example, could use in-world tools to write and share "broadsides" or pretend to be wax museum figures that come to life and recite their biography when another avatar approaches.

What about a 5th grade science kit? Perhaps it includes a few different, simple eco-systems of "living," interdependent plants and animals, life-like weather systems based on real data, flowing bodies of water, and so forth? The teacher could then send teams of students on missions to observe various aspects of the eco-system and use in-world data collection tools to record findings that could be reported back to the class. Perhaps the teacher could introduce disease or natural disasters or pollution into the system, or have the students build a simple town and see the impact the construction has on the animal population, the waterways, and the pollution levels.

Assessment was another large area of concern for these educators. I told them that one of the great advantages we have with Wonderland is that the system is still under construction. This means that educators can have an important voice in the development of the tools and features that are created by the open source community.  Even if the Wonderland platform is not ideal for K-12 educators in its current form, I strongly believe it has the potential to be a powerful educational tool. We’re already seeing some interesting work in this area, including the iSocial project at the University of Missouri. I look forward to seeing more work in the K-12 space once Wonderland v0.5 is released.

Developer Release 3 of Project Wonderland v0.5 (aka dev3) has been released by the Wonderland team today. This is the third in a series of early, psuedo-monthly "releases" for developers to work with the new platform, leading up to a user release this summer. For the nitty-gritty on getting started, have a look at the 0.5 web page.

Disclaimer: Before we get onto the new and cool things available in dev3, we wanted to briefly mention the standard disclaimers: Project Wonderland is experimental, free, open-source software. It is not a product. The Sun Labs team working on it is relatively small. We are focused on delivering the core platform features so that the community can build interesting and dynamic worlds with the toolkit. Because of the team’s size, we do not have a QA or a documentation group. Particularly in this early-access developer release, there are a substantial number of bugs and missing features. Okay, so now that you’ve been adequately warned, onto the cool stuff.

What’s New in Developer Release 3

There have been a whole slew of new features and architectural improvements to the software since the last developer release. In some cases, this may require you to do some different things in your module code. Here’s a brief list of what is new:

1. Persistence in Wonderland 0.5: As Jonathan eloquently described in his blog post, we have turned on features to make the worlds you create persistent across server restarts (short-term persistence), so that the state of the world is as you left it. For longer-term persistence, we added the ability to take "snapshots" of the world. (Note: The short-term persistence mechanism is not yet fully-implemented so please read the blog for further details. It is also important to distinguish between stopping and restarting the "ant run-server" process, and the Darkstar server via the web administration UI. If you kill the "ant run-server" process and restart, the world is not persisted.)
2. Avatars: Our brand-new avatar system is making its first appearance in dev3. All of the avatars will appear the same for now, but you can test out how they work on your system and play around with some of the new gestures.
3. Optional Modules: In this developer release, there are also two optional modules that you can compile and deploy to a Wonderland server: an audio recorder and an SVG-based whiteboard. The whiteboard also features some early peaks at the new HUD (heads-up display): a HUD is designed using Embedded Swing components, like any other Java Swing application, and can appear on the glass or in-world.
4. World Assembly: Also appearing in dev3 are a new set of in-world "assembly" tools to assemble your worlds. We call this world "assembly" rather than world "building" since you are creating the art work in external tools, importing them into Wonderland, and using these assembly tools to arrange them alongside other components. Currently, the assembly tools consist of a Cell Palette, to create instances of new cells in your world, a Cell Properties menu to move a cell, rotate a cell, or resize a cell via text fields, and a Context Menu to bring up visual affordances (or manipulators) to move a cell, rotate a cell, or resize a cell.
5. API Changes: We also made some core API changes that will impact cell developers. In many cases, the changes were simply naming changes to clarify the concepts involved. In other instances, we simplified the code developers need to write by making use of Java annotations. If you are working based on the Extending Wonderland tutorial series, refer to the porting guide for an overview of the changes that affect you.

And Now for Some Pictures…

What would be a blog post from our team without pictures from our testing sessions?