As some will remember, I complained a month ago about several bugs/shortcomings with the technical computing software Mathematica. To my astonishment, I was contacted in very short order by their customer relations and then technical support staff. And that is about where it ended. I got a nice phone call from one of their technical support staff in which I explained in more detail the problems I was having and the specifics on the system I was using, he said he would look into it and contact me the next week, and that was the end of my contact with them. The only contact they have maintained is sending me invitations to online Mathematica training courses, which to me is focusing on entirely the wrong problem (though it is nice of them to offer to help me overcome my Mathematica inexperience, that is not my main problem with the software). Since my research exchange ends next week, it is a little late to get in contact with me to resolve these issues, but on the off chance that someone from Mathematica continues to stalk my blog and truly wants to help people get the best out of their software, I can refer you to other researchers who are here on a more permanent basis with the same or very similar issues.
For the record, I think Mathematica does have some remarkably useful aspects as a piece of software. It has a wide variety of graphical outputs, and supports both symbolic and numerical computation. There are times that I find it obtuse and difficult to fully control, particularly when it comes to keeping track of which variables are instantiated (and to what) in a given kernel. I also recognize that when designing software for a variety of computer configurations and operating systems there are bound to be some bugs that are difficult to diagnose and track down (such as the window resizing bug and stability issues). One change that I think would go a long way to mitigating some of the frustration inherent with Mathematica's use would be to extend the editing environment so it supports auto-saving/document recovery capabilities or at least multiple undo/redo.
Subscribe to Computing Intelligence
Thursday, July 30, 2009
Subscribe to:
Post Comments (Atom)
0 comments:
Post a Comment