So, here's the thing... and this is entirely IMHO (some may question whether the "H" applies :-) There's been some off-and-on sidebar chatter on this subject for a while and this thread seems a place to insert it publicly.
Sage has a HUGE amount of functionality built in that is very useful to the community attending SciPy '09... but it isn't obvious. I come from the engineering community (we call it that, but engineers are more like cannibalistic tribal hordes than true communities) and had never heard of Magma before encountering Sage. The first time I saw the Inheritance Hierarchy of "1" I thought it was really cool but had no idea that the integer 1 was a group of a ring over a multidimensional irrational transnational space. However, since I'm more of a coder than many engineers, and was a Python person, I saw the Notebook, Cython, R, numpy, Blas and the dozens of other packages and knew that everything I needed was in there and had no problems cuz the stuff I didn't understand / use isn't in the way (most of the time... as long as I put a %python in every cell I avoid coersion which I understand in the abstract but applies to domains way outside my needs / knowledge) Last summer I was going through the presentations / documentation. Most have examples showing something about ZZ. Then there's some kind of factorization which spews a few thousand integers. Very cool and I'm sure there are wildly useful reasons for all that... But I'm also pretty sure that most engineers come away thinking Sage is for some branch of weird crypto stuff thats probably cool, but really just wanted to invert this matrix and plot the Eigenvalues. Thing is, Engineers / physicists (well, maybe not physicists) are easily overwhelmed. The choice of packages available is daunting and the culling process requires breadth first search. My guess is that Sage gets passed over all too often because it speaks a different language. If it were me, and I were trying to show the folks from my past life "why is sage important" I would start with the notebook... I'd do some simple numerical stuff with regular python, some plotting, explain how interact works... and then very gently go through numpy, a few examples with scipy, do a bunch of plots, and sorta finish off with cython. The cython example would emphasize both the incremental approach to improving performance and, very importantly, how it makes integrating with C/C++ very easy (python people have often scanned the C integration documentation and come away knowing they're never gonna do any of that) The cython example might start with native python and by adding a few key words and some strong typing that "a miracle occurs" (BTW, many of these examples exist but are often accompanied by ZZ / prime rational modular form group ring theory). Recently, William has been showing an example he did in real-time which was very simple to comprehend and drives the cython story home. If it were a longer presentation, I'd then discuss some of the underlying mechanisms making it even possible to integrate the 80 or so open source packages included in Sage. To an engineer, thats truly mind blowing and speaks volumes about "that which can't be seen" meaning an architectural approach which also just happens to include some very very hard core mathematics. Engineers like math, and even use a bit of it from time to time... (many sage folks would call it "arithmetic") but its related to their needs and from time to time mathematicians come up with some really cool stuff... that we engineers can truly use. (Of course, virtually all math used by engineers has a hard cutoff date somewhere in the 1800's (laplace and fourier are rockin' dudes to engineers)) So again, IMHO, the issue is entirely about telling the Sage story to an audience that might assume the ZZ() function is sage's version of sleep (get it... zzz's) The key is there needs to be a path for non-mathematicians to see that Sage is way more than something for mathematicians and is, in fact, a comprehensive approach to problem solving which already happens to natively include 99% of the capabilities most engineers / scientists need. -glenn --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
