ABSTRACTTraditionally, an Avatar has been viewed as a representation of an individual. This approach applies unnecessary restrictions to the Avatar's usefulness. By extending the range of Avatar capablities, we find a new area of interactivity and expressive power. Like a genie, they can become any type of model, graphic or animation. They can do this in sync with any type of oral or text based presentation. Avatars can become content delivery platforms. They would inherently blend the best elements of three dimensional interactive enivronments with human sensitivity and human flexibility. INTRODUCTIONEducation and Training courses are migrating to the Internet. "Distance Learning" is the buzzword in cutting edge online college programs. Because of its three dimensional nature, Multi-user VRML has a unique role to play in this developing field. VRML leaves the flat 2-D screens behind and becomes a true environment, a true classroom. In this vision of the VR classroom we can marry the best that technology has to offer with the human interaction and sensitivity needed for a rich educational experience. The use of Multi-user VRML as a classroom environment provides the educator with a number of unique and intriguing opportunities. This paper deals with only one: the use of an Avatar to deliver educational content. In typical VR environments, Avatars are human representations with some human gestural qualities. This traditional approach has a good deal of merit, but it is not the only approach. We can and should empower that representation to do much more. When an Avatar is a representation of a teacher, it ought to have sweeping powers to present and manipulate educational information. Turning away from concept of Avatar as humanoid representation, we find the Avatar can also represent educational information. With the addition a simple switch node the teacher's Avatar can act as slideshow. A series of informational graphics, models or even worlds can be presented in sequence and completely in sync with the teacher's spoken or written text. For the first time a teacher can go beyond merely presenting some body of information to the students, she or he can literally become the information presented. In viewing an Avatar as a Content Delivery Platform, we empower the teacher to provide information in uniquely effective manner. The inherent value of this technique becomes apparent when extended into a real world setting. Imagine yourself in a real world classroom. Teacher states that the day's topic is on marine mammals. Then in an instant your teacher becomes a whale. She continues the discussion, answers questions and then move on the whale's diet. She becomes a single plankton. To represent a meal she becomes hundreds of them. Then the teacher continues through the discussion becoming two-dimensional charts, other marine mammals, and even the ocean itself. Any type of visual information can be presented, even with audio enhancement. By using the Avatar as a Content Delivery Platform, we effectively deal with a number of issues. Avatars are flexible: through standard controls the teacher can show visual information in sync with their spoken or text based lecture. The teacher can stop and ask questions of the students at any time and students can ask questions of the teacher as well. After the questions have been fielded the teacher simply push a button and advance to the next visual. Control of the presentation is entirely in the hands of the educator. They and they alone advance the presentation. The amount of control and interactivity the students have is completely at the educator’s discretion. The implementation of this is determined by the nature of the behavior ascribed to the individual pieces of the presentation. An object (a slide-show image, say) can exhibit non-shared behavior. In the presentation of this image with which the teacher wishes to provide some time for the student to explore individually ( such as the activating a small animation or the ability to rotate a model to different angles or different perspectives). The behavior would be non-shared: a single object could be examined differently by a number of students, at the same time. Essentially every student has his or her own copy of the resource to learn from. If, on the other hand, the teacher did not wish to provide the object to the students for individual manipulation, the behavior would be Shared (Avatar Specific): only the teacher can manipulate the item. Each manipulation is observed by every student. The general function of advancing the presentation is of this type. With the teacher becoming the information, the student no longer has to worry about navigating an unknown environment in order to find what is needed. The information can be presented directly in front of the student. In essence, the teacher does the navigating for the student, enabling the information to be presented in a clear, linear manner. Should a student not understand how to operate a model, the teacher is present to talk them through it. The teacher will also have the ability to present a large block of information in smaller portions, providing greater clarity for the students and manageability for the teacher. The concepts we are dealing with here have one particular attribute that makes it singular in the fields of VR research: the notion of Avatar as Content Delivery Platform is within the scope of today's technology, not tomorrow's. Nearly all of the current VRML multi-user environments allow for behaviors to be added to an Avatar. On most multi-user platforms even multiple behaviors can be programmed into an Avatar. In a very practical sense we are proposing nothing new. No new technology needs to be invented, and no changes to the VRML code are necessary. All that is needed is a new slant on the function of an Avatar's behavior. Rather than limit ourselves to humanoid representations with humanoid expressive qualities, we can explore the use of all expressive qualities. Through the abstraction and extrapolation of acts of expression we can refocus the concept of teacher as presenter of knowledge to teacher as embodyment of knowledge itself. INFORMATIONAL STRUCTURESWe have suggested that the teacher as Avatar can present students with a slide-show of facts. While accurate this suggestion is not the only viewpoint available. Each 'slide' can be much more than a flat graphical representation. These slides can be subdivided into the following types of informational structures: graphic, animated graphic, model, animated model, world, or characters. THE GRAPHICThis is the classic form of information medium. Just like old fashioned slide-shows we can display photographs, illustrations, text information, bullet lists, charts, graphs or graphic imagery. These can support the text/narration the teacher provides. By using them in a series one can build an simple straightforward presentation. THE ANIMATED GRAPHICFlat graphics are often animated in multimedia presentations. This option is also available in VRML, either by a flip-book style (via a switch node) or as an MPEG movie. It is useful for describing a process or procedure. This is also similar to conventional presentations, A new twist, however, can be added by specifying one of three options: animation runs continuously, is initiated by the teacher, or is initiated by the individual student. THE MODELConventional teaching methods have often used models to illustrate certain types of information, or to simply show examples of a specific topic of discussion. In VR environments, 3-D models are the norm. One additional benefit is that (via a sphere or cylinder sensor) each student can manipulate an explore the model independently. In a very real sense, every student gets his or her own 'model' to observe and interact with. Given the common complaint of "not having enough to go around" this is an excellent way to give every student an equal chance to explore. THE ANIMATED MODELWe are able to present 3-D manipulatible models which can be built to simulate a real process such as a combustion engine or an abstract process. Often a real world process like the combustion engine does not allow for easy viewing of internal actions. Additionally, there is a level of danger involved with the close inspection of this action. VR models offer the ability to view any process clearly and without concern for personal safety. THE WORLDIt is possible to view the teaching Avatar as a world. In a literal sense, the Avatar would be a large model surrounding the students. Avatars can exist inside other Avatars. There is no technical issue that demands an Avatar be "human" sized. . In this environment, learning becomes an immersive process. This would be particularily helpful in dicussions about oceans, the cosmos, or any type of environment. To show more a practical example of this concept, we return to our earlier example of a combustion engine. It would be appropriate to show the model small and whole and allow the students to get an overview of the engine in relation to the automobile. After that overview, would it not be appropriate to explore an engine from the inside? To view small details more closely? In a very real sense, by virtue of scale alone, our model of the engine can become an entire world with its population on the inside. THE CHARACTERWe will not deny that a physical manifestation of the teacher may be useful. We therefore include that possiblity as well into our larger scheme. Perhaps even expand that concept. Imagine Leonardo DaVinci dicussing his "flying machine" and then becoming it. We simply seek to remove the arbitrary constraints on how Avatars are used in VR. In our experience, it is has been shown to be true that having a person "become" the information rather than show it to the side is in most cases a more effective means of making a point. CLOSINGIndividually, each of the above informational structures is a useful, educational form. Our concept is to use various combinations of these in educational presentations. Here we draw upon the strengths and the appropriateness of each form to further the learning process. Given that each model, graphic or animation appears before the student, we can provide the educational resources directly in front of the class. Not only do we have issues of control, flexibility, ease of use addressed, we the widest range of options in presenting information of any media format. Because the teacher is the presentation, she or he has complete control over the pace of the presentation and is able to stop at any given time and answer questions. This interplay between teacher and student is a cornerstone of any thoughtful educational program. In our VRML format the teacher can set the level of student control at the appropriate level (without ever losing control). By this we mean whether the controls for animation are accessible to the student or whether a SphereSensor or TouchSensor has been attached. Even a Billboard Node could limit the students view of a graphic or object to the "front" or the view chosen by the teacher. In conclusion, we have found this process of sharing visual information through the Avatar to be quite effective, There are those that assume that the concept of Avatar as Content Delivery Platform is a merely temporary solution to many VRML issues. Temporary, because they consider teacher and information as separate and assume it must be the most effective means to present knowledge. The basis for this belief is rooted in the limitations of the reality one is observing from. Because of the new freedom VR brings us _the new reality_ these limitations are subject to review.
DEMOSTRATIONThe following is an working model of an Avatar being used to deliver content. We have created a lecture presentation on Comets. Like a slideshow the teacher's Avatar becomes the information he is discussing. Each of the informational structures is represented, as is the handling of questions and navigation. The pictures were taken at Jeff Sonstein's "Town Square" one of the Vnet MU environments. We would like to thank Miriam English for the use of her "dog" as the student and Beth Shipper for her illustration of the teacher.
VR ASTRONOMY 101Introduction: An Avatar in the form of a teacher greets a student and sets up his orientation.
The "Next" button has been pressed. A new graphic image is displayed. ( The "Next" button is used to advance the presentation forward )
A second graphic, as viewed by the student.
A third graphic, with teacher & student interaction.
The teacher's avatar becomes the galaxy and surrounding black space. The galaxy is animated. It is slowly rotating counter-clockwise.
A model replaces the galaxy. The points are rotating at differing speeds and directions. ( Note the small internal model )
The "Move" button has been pressed. The cloud sweeps past the student. This is done via a transform node. ( Any model can moved by the teacher, this is useful for large environments where "trekking" about would be arkward. )
The model is placed in front of the student. It is moveable and rotateable. The planets and Comet are on an animated loop.
A model without the surrounding black space. This animation is triggered by the individual student. It is also moveable and rotateable.
The model has been rotated. Each student see's only their own interactions. ( Note teacher / student interaction )
A Flipbook style animation ( As we are Mac Users, Mpeg Video is not available to us.)
We return to a simple graphic
We return to the original "character" type Avatar
This concludes our demo page. We hope you find the new direction we have placed on Avatar usuage of interest. As technology continues to advance we feel this approach will be commonplace.
VNET An open-source open-platform multi-user environment.TOWN SQUARE This is the world used in this Demo.To try the Avatar yourself go to the "Square". Choose a "Custom" Avatar. Type in the following:
http://noel.pd.org/~thatguy/b/11_poet.wrl As of Jan 31st the Avatar is not fully optimized, nor have instructions been written for its usePoetry & Education Screen grabs of education and a poetry demo - courtesy of Miriam English.Poetry & Education More screen grabs of education and a poetry demo - courtesy of Cati Laporte.CREDITS
Steve Guynup Mr. Guynup is the leading VRML developer on the Macintosh OS, an NEA funded VR Artist and occasionally writes on the subject of 3-D web technlogies for NetProfessional Magazine. He is also an award winning digital artist and recovering poet.
Many thanks go out to: the VNet Community (in particular, Stephen White and Jeff Sonstein) for developing VNet ( a Java based Open Source - Open Platform Multiuser VRML Environment ). Without which Mac Users like ourselves would be unable to contribute to this cutting edge field. Also to Malinda McCall for her friendship and her help in this and many other projects. We would like to thank the following organization for their support in this and other VR projects
|
