back to tim's mas.123 page

Sam, the StoryMat, and the Reflectory all provide interesting environments for kids to tell their stories. These stories may be real (in the case of a kid writing a letter to another about his day at school) or imaginary (as the stories told on the StoryMat) but in each case an explicit environment has been created to allow kids to tell their stories. Each provides a different technological extension to its story-telling environment.

Sam provides a virtual character to share the story-telling experience with. The use of shared space provides an interesting medium for the virtual character to tell a story with the child. Of the three, this might most obviously fall prey to the same objection which many raise with television-that the machine should not be a substitute for a human playmate or human interaction. The limitations of the current system make this even more apparent. However, a simple way to avoid this problem might be to utilize a technique used with the StoryMat-have two real children interact together with Sam. This would allow the two children to develop their own stories through interaction with each other as well as elements incorporated from Sam's suggestions. As we discussed in class, the current system's method of selecting new story segments leaves much to be desired. I suspect that some meaningful context-matching will be necessary before Sam can truly function as a third playmate.

I was quite impressed with the ability of the StoryMat to fulfill this function without any human embodiment at all. The amount of context-matching that the system was able to provide by simply knowing the path traveled through the physical space was quite impressive; however, this is certainly fostered by the variety evident in the images on the mat itself. It seems quite remarkable that the kids were willing to accept the contextual suggestions of another kid's disembodied voice and continue their own story from there. However, I was disappointed to see that even if the story was continued, the system had no way of realizing this and splicing the two segments together so that a collaborative story could be asynchronously created. Admittedly, automatically determining that the story is a continuation could be very difficult. However, it might work to simply ask the child after hearing the other story whether he or she wants to continue the story they heard or begin a new one. On the other hand, this introduces "the computer" as an element in the interaction. The transparent nature of the interaction in the current iteration (in which the child currently using the interface hears and sees only other children's creations) seems to add to the system's effect. I particularly liked the way that multiple children could share the physical and technological space to tell a story in collaboration both with each other and the stories previously told by other children.

Finally, the Reflectory provides a different kind of asynchronous collaboration-a technological augmentation to the pen-pal concept which allows the child to explore historical events related to their writing. This seems like a good concept-it strikes me as being in a similar vein as the Image Maps, both in their focus on historical events and their idea of taking the material from which historical materials are selected from the users' own experience, making the experience using the tool much more personally meaningful and effective. However, the projects differ markedly in their intended use; the Image Maps system is done as a specific activity in which the users embark out into their environment with the intent of exploring it and its history. The Reflectory brings a similar idea to the everyday activity of writing a letter to a friend. I wonder if the intrusive nature of the historical information in the interface is appropriate to this activity. After all, the purpose of sitting down to write a letter to a friend is not to learn about the American Revolution. It seems that the continual, changing presence of such information taking up half of the screen is somewhat oppressive for an interface used primarily for writing, not research. A less intrusive approach might be to have a few lines at the bottom of the screen which contain the first few lines of a diary entry that the system thinks is related to the last sentence you typed. If you clicked on the lines, you would be given the full text of the diary entry and the ability to browse further. This admittedly makes the interface for writing markedly different from that for browsing the historical documents, but the distance between the activities is far enough already that that might well be more appropriate, allowing the user to explore further when interested in the material, and letting him or her simply continue to write otherwise. This would preserve the personal nature of the explorations without the feeling that your writing is being interrupted. Mike mentioned that there was more work to be done on the context-matching algorithms to find historical information to display, and I expect that this will make the tool very effective to explore historical information that might otherwise seem stale if presented in a history classroom.

back to tim's mas.123 page