> I doubt it's as extensive as the Python libraries, but then a lot of the > python > libraries are not related to science/engineering/maths.
I agree with this, but I think it's a strength. Part of my research includes scraping BGP data, and there are modules for doing this in python.(Currently by calling libbgp which is a java library in a script, it seems there are some python modules that may do this directly, but this has served my purposes well enough). A fellow grad student is doing work that scrapes data off a webpage of an online game, which is easily automated using python. Another colleague uses it to search through and analyze bug reports in version control systems. Using sage and python I can often do my data collection and analysis in one place, whereas mathematica it seems might be more restricted to the analysis. This adds another layer. Collect the data, convert it into something mathematica can read easily, import it, then do the analysis. I am sorry I don't have any references, I guess I was alluding to my own work. I've used python (and numpy and scipy) for a number of undergraduate classes (both engineering and applied math), and now continuing on in grad school and it always proves to be effecient and easy to use. This was after learning MATLAB and then switching to octave because it was free, then hating octave because it was buggy and slow unless I could figure out the best way to vectorize the computation I was doing. I've just been using sage for the last 8 months, and it remedies the things I hated about python (ie getting everything set up in the first place, not to mention upgrading). After all that rambling what I mean is, mathematica, maple, and matlab are all very good at doing math, and doing so in their specific subfields very well, but they aren't very good at interacting with all the other parts of my computer, the internet, or other things that computer scientists are interested in. Sage on the other hand has python which can interface with anything and do the math once the data is collected. Ben On Sep 9, 3:37 pm, "Dr. David Kirkby" <[email protected]> wrote: > On 09/ 9/10 04:41 PM, Ben Edwards wrote: > > > I might shy away from any personal attacks on Stephen Wolfram, despite > > controversy as him as a scientist. This should be about comparing sage > > and Mathematica, not the people behind them. > > I think that would be *very* wise. > > I'm no fan of Steven Wolfram. IMHO he is a clever guy, who knows it, but has a > *HUGE* ego. I'm under-impressed with Wolfram's book "A new kind of science". > > But ultimately, Mathematica is developed by a lot of people. Some of those > people are very good at their job. Some of you may know Daniel Lichtblau. I > have > an incredible respect for Daniel. He clearly knows he stuff, is not full of > bull-shit and has been helpful solving Mathematica problems when it was clear > I > was really trying to debug Sage issues. > > http://groups.google.com/group/sci.math.symbolic/msg/bf17230f19d67698... > > Daniel's comments were a lot more helpful than RJF's that is for sure!! > > Your comparisons with Sage and Mathematica need to be on technical and > economic > grounds - not what respect you may or may not have for Steven Wolfram. > > Subjects I'd consider would be > > * Ease of installation. > * Ease of use. > * Messing around with license servers in some cases with Mathematica. > * Support > * Reliability > * Cost > * Platforms supported. > * The fact Mathematica license limits you to 4 CPUs without extra cost. > * Openness of source code. > * What might happen if Wolfram Research go bust - not an impossibility, > though > I certainly hope it never happens. > * What might happen if William leaves Sage - his blog indicates he is going > to > concentrate on PSage, not Sage. > * The fact the main public support channel for Mathematica is moderated. > * Bug list for Mathematica hidden. > * If Wolfram Research find a bug, even if it means someone's results could > be > inaccurate, they will not notify users. > > > > > > > One thing that is a natural advantage for me is the ability to not > > only use all the packages included in sage, but any python package. I > > find myself using sage to scrape, reconfigure and analyze large pieces > > of data from a variety of sources. Python makes this easy. Because > > there are so many python modules, it is very likely there is one that > > does what I need. If not I can code it (usually relying on other > > modules), and release it as its own module, building the number of > > things that can be done quickly with python and sage. And as far as > > scientific computation, if there isn't a python module that does what > > I need (that is not included in sage) there is certainly an R package > > that will. > > > Sage is not only good for mathematical research, but any scientific > > and engineering research, just because of the huge number available > > packages to use in python. I can't say I think the same is true with > > mathematica. > > Ben, you say Sage is good for scientific and engineering research, but I don't > see much evidence of Sage used in engineering research. Can you point me at > some? Engineering PhD theses, engineering papers? Engineering departments > teaching Sage? > > Do a google search on these terms > > * Mathematica engineering - 857,000 hits > * Sage engineering - 7,620,000 hit, but I could not find any that had > anything to do with the Sage maths project. > > Also, note there are ways to access Python from Mathematica > > http://library.wolfram.com/infocenter/MathSource/6622/ > > But there are also many Mathematica programs freely available. Wolfram > Research > have a library. > > http://library.wolfram.com/ > > I doubt it's as extensive as the Python libraries, but then a lot of the > python > libraries are not related to science/engineering/maths. > > > Ben > > I think to do Sage justice, you need to check some of your facts. I think > people > will take your points more seriously if they are balanced. > > Dave -- To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/sage-support URL: http://www.sagemath.org
