I think Sage is awesome. I've started using it in my research. I've been using it to teach classes, and I've even forced my students to learn Python (and Linux) in the process.
Yes, Sage has lots of bugs, but at least I get to see the code, and if I'm motivated then I can even fix it myself. I used to hit a bug a week with Magma, but I can't see that code, and I'm no longer motivated to report the problems (because now I'm at an institution that can't afford a Magma license so I have to use other people's Magma installations). We should all try to write good code, good documentation, and good tests. When time and money limit what is possible, each developer just does the best she can, and the rest of us should just say, "Thanks for the hard work!" (By the way, thanks for the hard work everyone! I really appreciate it!!) As for the "formal methods" and "proof of correctness" approaches: I just don't think that they are viable given our current resource limitations. I've worked on military projects requiring that level of verification, and even with multi-million dollar annual budgets it was painful to achieve high-assurance on projects much smaller than Sage. So, even though it would be nice to be able to believe the results of every computation, I think we just have to let the users know, "Sage makes a good attempt to give you the right answer, but bugs happen, so double check your computations if you need to rely on them." Keep up the good work!! (Sorry, I guess I didn't add anything to the conversation, but it was more fun than grading finals.) --jason --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
