Kirby- This is the reason I raised the issue of cross-platform support on this list on Feb 4, 2000 ("Common Graphical Framework for Python Tutorials?"). From what I read in comp.lang.python, TK support on the Mac has historically been questionable (although the base command line system otherwise works well). My personal feeling is that widespread acceptance of CP4E absolutely requires addressing seamless cross-platform support at the GUI level (like Squeak or LearningWorks), as well as incorporating numerous advanced features found to varying degrees in other systems (like Squeak, LearningWorks, or Dr. Scheme). http://www.squeak.org http://learningworks.neometron.com/ http://www.cs.rice.edu/CS/PLT/packages/drscheme/ Some discussion on edu-sig of needed features in Python to support CP4E has been discouraged as "vaporware". There is much truth to comments made here that people posting suggestions may have no plans or desire to implement them any time soon. But I feel that it is equally true that it is useful at the beginning to have some level of collaborative discussion of system needs and features (and prior art) before it make sense for anyone to make a big commitment to CP4E-specific implementation. Many people who are not implementers can contribute to such discussions based on their previous experience in education or with educational software (in a sense, as "requirements analysts"). There is also much value in just discussing what one can do easily on the PC with the existing implementation. I have greatly enjoyed your posts on your personal experiences and examples doing so. In many open ended systems with a body of useful scripts there are three categories of participants. Users: 90% use what is there (and complain/suggest and tell their friends). Authors: 9% write simple things and make minor improvements in existing scripts. They know what the important/feasible issues are, and do much of the support of the 90% in answering common questions, and making simple fixes and changes. Gurus: 1% understand the system through and through and do the bulk of the implementation, and do much of the support of the 9%. All three types of participants contribute in their own ways to making a system a success. Each level recruits from the previous level (World->User, User->Author, Author->Guru). In the case of CP4E, this would imply: Users: 90% of educators would use Python based materials that already exist (scripts, tutorials, and exercises). This group explains to the rest of the world why Python makes sense for education and where to get materials. They make creative suggestions they can't implement (like "Python should have a cross-platform GUI"), but in effect, these suggestions and complaints are producing "requirements". Authors: 9% of educators might write simple new materials, or improve a script or extend an exercise. This group helps support the 90% with minor customizations, and encourages some users to become authors. Gurus: 1% of educators are going to make significant changes or additions to Python and related educational materials to support CP4E. Much of these improvements are driven by observation of the 90% and more direct interaction with the 9%. This group encourages some authors to do bigger projects (especially if they complain or suggest too much :-) and take over aspects of system maintenance (becoming gurus). In the case of the Mac issue, which is a deep issue, only 1% of educators could do something about this, and only a fraction will be interested. So, this means we probably need about 1000 educators involved with CP4E to get this Mac support issue resolved (900 users, 90 authors, and 10 gurus, of which one guru will be a Mac enthusiast and also interested in resolving cross-platform issues). -Paul Fernhout Kurtz-Fernhout Software ========================================================= Developers of custom software and educational simulations Creators of the Garden with Insight(TM) garden simulator http://www.kurtz-fernhout.com Kirby Urner wrote:

In a thread starting at: http://forum.swarthmore.edu/epigone/geometry-research/pefrorsken

Clifford J. Nelson advises me to forsake a Python-based approach for another language called Clean, because "Python is not supported well for the Mac."

I thought I'd run that by comp.lang.python, as well as the Python edu-sig. Given the importance of Mac technology in the schools, limitations re Python on that platform are of some relevance to the "Computer Programming for Everyone" (CP4E) thread. [snip]

Kirby