Replying to myself:

It sounds a lot like the Lego Mindstorms robotic kit programming environment -- but a truly visual Python IDE.

Speaking of which, the last few weeks I've had a strong desire for a system that would integrate with my IDE and with my UML design tool, a system that could track interactively what classes, methods, and attributes I'm typing in the editor, store them in a meta-code repository, and present the changes in the UML diagram. Of course, such a system would generate new classes, methods, and attributes in the repository and out to the editor from changes made in the UML model. "Round trip engineering" solutions currently on the market aren't anywhere close to being this integrated (but I'm be curious to hear from users who love their RTE solution...) The IDE-for-kids system mentioned in my previous post is similar to the above description, but rather than using UML abstractions, the focus is on kid-friendly representations of lower level code steps. However, the interactive, integrated nature is the same. A great idea mentioned on this list earlier was to present learners with partially completed solutions, or simple working solutions (classes), which they would modify and extend as part of learning about the system. It strikes me that, stealing a model from Mathematica, an IDE-for-Kids could load and save project "worksheets", which could contain multimedia data and meta-data about the project in addition to the base source code. This model has been used to create some amazing eduware in the Mathematica market. Teachers could assemble worksheet projects and submit them to the PythonForge for others to download, use, modify, and resubmit. (Blue sky: the network-friendly nature of Python could be leveraged to build automatic evaluation and grading features into the worksheets...) I hope this discussion isn't perceived by some readers as vaporware. I for one intend it as an initial requirements generation and scoping discussion for project(s) that can advance the cause of CP4E. If enough developers are interested and enough educators feel one of these proposed projects would be useful, the next step is to begin analysis and design phases, with iterative prototypes to give the community something to actually play with and evaluate. Eventually (we're talking personal life goal here), I'd love to merge my interest in education with my interest in building persistent virtual worlds. The systems we are building in Ultima Online 2 would make extremely engaging educational micro-worlds. That's not our (Origin's) goal or audience right now, but that doesn't mean it won't ever be. :-) Cheers, Jason Asbahr Origin Systems, Inc. jasbahr@origin.ea.com