David demonstrated IMS from the prototype, which requires a user name and password. Contact David at uncle_dave@unc.edu if you would like to get into the prototype.
The Web is an information landfill. Educom, corporations, universities, and the military (the military does a lot of training) are gathereing requirements for an Instructional Management System. They are not developing courses, they are developing a high-level platform for developing tools.
IMS lets you manage groups. The University is a group. The School of Dentistry could be defined as a group. A class could be defined as a group.
Groups can have resources, such as chat.
An individual's backpack defines a personal profile and a public profile.
Control panels can be used to add resources to groups. For example you could add announcements as a resource for a group.
For courses, the resources are course materials; right now that's mostly Web pages. You could open applications, simulations, etc. There will also be backend interfaces to Administrative Information Services (AIS) for classrolls, grades, etc.
The plan includes collaboration -- sharing tools and data from campus to campus.
IMS is 100% Java; it is implemented with servlets (jhtml). Java code, imbedded in html, accesses the database. A servlet gets the data, builds an html page, and serves it. The server compiles, so it needs to be speedy. The client side is just viewing html, so you don't need anything special on the client side.
Metadata is a way to catalog data. Tom Wason is working on metadata -- keywords and metatags.
IMS has a search icon. You can do a basic search. You can also search by author or subject; the information coming from metatags. Special interest groups (for example, physics) would define metadata tags for their group. Similarly, a medical database would have its own metadata tags.
Uncle Dave's Web page (http://uncled.oit.unc.edu/index.html) has links to IMS papers, FAQ, and the national project.
Participants in the IMS project are building tools to the same standard.
Tom Wason is working on worldwide labeling requirements. Metatags come from html pages and from repositories, so search engines don't have to go to every page. Controlled, disipline-specific vocabulary is the key.
Judy Hallman (judy_hallman@unc.edu, http://www.unc.edu/~hallman/)
Campus Webmaster, UNC-Chapel Hill