Tuesday, June 30, 2009

DGBL Primers, the New Sesame Street, and Elon University Students

There is so much literature out there about computer games, DGBL, and education, it is sometimes a bit overwhelming to find a good overview that can quickly provide an orientation point for future research. In the past I have suggested Richard Van Eck's Digital Game-Based Learning as a starting point, but a few days ago I came across Moving Learning Games Forward by Eric Klopfer, Scot Osterweil, and Katie Salen of the MIT Education Arcade, which is also an excellent introduction and provides a broader overview. Katie Salen is the author, along with Eric Zimmerman, of Rules of Play, a hefty tome and definitive guide on game design. I heartily recommend this book, although its attention to detail sometimes makes for difficult going. Weighing in at only 56 pp., the essay is a much easier read. It also seems that the Sesame Street folks are getting in on DGBL. Last week I found Game Changer, put out by the Joan Ganz Cooney Center at Sesame Workshop. If Sesame Street has finally gone DGBL, shouldn't higher education as well?

Finally, a few photos of Elon University students hard at work on the DigiBahn Project:


Jack Garratt, discussing the finer points of digital gameplay. Jack (who has gone on to graduate studies at Georgetown University) was instrumental in developing the game narrative.

Natalie Lampert on-site at Stuttgart Central Station taking photos for the 3D modelers to use as visual reference.


Monday, June 29, 2009

Initial Blender Experiments

For the last few weeks we have been working on getting meshes created in Maya and Blender. Just to see what they look like together, I threw some together this morning for a quick render:

In the next few days I will focus on getting the handrails on the escalator ironed out; the mesh here is a bit choppy and no amount of smoothing in Blender will get rid of the lines. After playing around with the model, I think that it is simply a matter of extruding a some new, continuous edges and connecting the vertices where necessary. The less connections I do by hand, so it seems, the better. And I gotta add some railings around the pit; wouldn't want anybody falling in. So that you can see the "business end" of the software, here is an image of the Blender interface:


Looks like the cockpit controls for a Airbus A380, nein? And, finally, here are some photographs of the real place, the Klett-Passage in Stuttgart, for comparison:




Coming soon: Importing more models, removing unwanted vertices to speed up rendering (does this mean good-bye to the vertex-rich tree?), applying textures to the meshes, a video walk-though of the 3D environment, and -- eventually -- tinkering around with the Blender game engine.

3D-DGBL and Instructional Simulations for the Humanities

Three dimensional digital game-based learning (3D-DGBL) environments and instructional simulations provide several advantages for the teaching of the humanities and foreign languages, including the ability to foster constructivist learning environments and provide personally-tailored and highly motivational instruction; the promotion of student-directed learning, free inquiry, and exploration; and the power to emulate remote or inaccessible real-world environments, recreate vanished environments, and lend substance to purely imaginary or literary ones.

The Cathedral from Sebastian König on Vimeo

The above video, completely modeled and rendered in Blender, is a virtual recreation of the cathedral of Halle an der Saale as it might have appeared in 1526. Although not an interactive instructional simulation due to the large amount of processing overhead required for image rendering, the video is interesting in that it hints at the possibilities that 3D-DGBL can offer the humanities. What about, say, an educational simulation of a medieval cloister or a game that allows a student to assume the role of a nun at this cloister?

Modding a COTS Game

In October 2008 the Elon University Digibahn Project began modifying The Elder Scrolls IV: Oblivion, a commercial off-the-shelf (COTS) game with a freely available game editor (TES Construction Set) in order to build a proof-of-concept game prototype. By April 2009 we had imported a few basic 3D meshes, developed an instructional blueprint, and programmed some basic NPC dialogue and interactivity:



Although creating game mods TES Construction Set considerably lowered the barriers to game 3D game development by freeing the programming team from developing difficult game engine components (e.g., physics engine and collision detection), the platform also imposed serious limitations. First, each development team had its own industry-standard software and proprietary file formats, which made it difficult to import model meshes created by the 3D art team into the game development kit used by the programming team. Files needed to be reformatted via intermediary software and then exported in a different file format, which in turn could introduce errors into the model meshes. Second, Oblivion-based mods retained the graphical user interface (GUI) of the commercial off-the-shelf (COTS) game, with its concomitant fantasy role-playing game elements, which the content and instructional development team thought would be distracting to player/learners. Finally, although the final game mod could be freely distributed, playing the mod would require player/learners to purchase The Elder Scrolls IV: Oblivion.

Based on these concerns, in May 2009 development on the DigiBahn Project shifted from modding a COTS game to developing our own stand-alone product. After much debate, the development teams decided on using the Blender 3D content creation suite as a game development platform as it is free and open-source, can run on all major operating systems, and comes with a built-in Python-based game engine containing audio support, collision detection, and game physics. Game Blender, which uses a system of logic bricks with link lines to develop logic within a game, allows non-programmers easily to develop 3D computer games, although more complex games will require programmers fluent in Python programming language. In addition, Blender can easily import Wavefront (.obj) and COLLADA (.dae) file formats, which would allow the 3D art team to continue developing in their industry-standard Autodesk Maya platform and then to export their work into Blender. The move to Blender will significantly simply the project workflow by reducing the number development platforms to only two. Future game development may see an eventual transition to a more powerful graphics rendering engine such as OGRE.

Learning from Computer Games?

I discovered this video of Battlefield 2142 gameplay a few weeks ago and found it interesting for the amount of learning that actually is taking place in the game:



The video depicts a teams of players, each who has individual and unique responsibilities on a combat assault team. The players can access the game form remote sites and may never have met each other in person. In this version of the game, players must work together to coordinate an attack in order to capture an enemy flag. A direct attack is not feasible due to overwhelming enemy forces, so the squad leader modifies plans and instead opts for an assault along a circuitous route. While communicating with squad members via VoIP, focusing on the immediate tactical aspects of the current operation, the squad leader is also managing the broader strategy of his team's performance in the game via a heads-up-display of the battlefield and prepositioning combat resources accordingly. What the game provides, therefore, is an engaging 3D experience with numerous complex layers that interact with each other during real-time while demanding the player to instantaneously evaluate emergent challenges in the environment, create a hypothesis on how to deal with these challenges based on existing game resources and player capabilities, test this hypothesis for validity, and, if necessary, make any changes to the hypothesis and allow for retesting. Clearly, there is a lot of learning going on in computer games.

By the way, you can view my Battlefield 2142 statistics here: Doktor_Frag

A Journey of Ten Thousand Vertices

It was during a temporary job that one takes between academic positions when I was reintroduced to computer games. I say “reintroduced” because I grew up when the Atari gamestation, Pac-Man, and the mall video arcade were coming of age. Although my junior-high school friends spent numerous hours – and quarters – in dimly-lit rooms mastering the intricacies of Battlezone, Tron, and Space Invaders, I was less interested in these pastimes. On the one hand, I never could earn enough quarters to support and cultivate a serious gaming habit. On the other hand, I also did not find the narrative of these early computer games to be exceptionally captivating or persuasive. What ever the reason was, economic or aesthetic, I did not get hooked on computer games as a teenager and I continued to eye them suspiciously as a waste of time and money as I grew older. So when a fellow worker at the aforementioned job invited me to the LAN (local area network) party he was throwing, I begrudgingly accepted – but only for the sake of collegiality. I would simply drop in, play a few rounds of whatever they had running on their computers, and then quickly return home.

It has been a few years since that first LAN party, and in the meanwhile I have built my own gaming “rig” (a term used to describe a PC with added hardware functionality that can support graphically rich 3D computer environments), try to keep up with recent game releases, and have expanded my academic research topics to include game studies, digital game-based learning (DGBL), computer-assisted language learning (CALL), and situated cognition. Why the change of heart? Well, I would have to say it was the vastly improved game narrative and computer graphics; it was an unmistakable "sense of place" that gave me vertigo in Half-Life 2 or knowing exactly what to do in Battlefield 2142 as I had interacted with the virtual landscape repeatedly.

It is this sense of place and knowing what to do in this place that, in my opinion, makes computer games a valuable instructional medium for learning a foreign language. And with technology and open source software being what it is today, it is absolutely reasonable to assume that even small language departments and universities can develop a DGBL environment on their own -- or at least a small prototype. I'm surprised (and perhaps a bit dismayed) that language departments have not yet leveraged this technology to bring themselves squarely into the 21st century.

So, this blog will document the efforts of the Elon University DigiBahn Project to create a
DGBL environment based on the sociocultural spaces of a German railroad station, namely Stuttgart Central. In the end, we hope to have a 3D graphic adventure game requiring students to navigate this station while meeting specific instructional goals such as purchasing a train ticket, locating the appropriate track, making sense of arrival and departure tables, and interacting with non-player characters (NPCs). We've already begun to get the 3D models (meshes) together and in the coming weeks (months?) will work on texturing them and getting them into a game engine. Interesting to think that all these vertices, simply coordinates in virtual 3D space, can create an educational experience larger than the sum of their individual parts.