Wolfram Research Releases Mathematica 7

mblase writes "Wolfram Research has released the seventh version of Mathematica, and it does a lot more than symbolic algebra. New features range from things as simple as cut-and-paste integration with Microsoft Word's Equation Editor to instant 3D models of mathematical objects to the most expensive clone of Photoshop ever. Full suites of genome, chemical, weather, astronomical, financial, and geodesic data (or support for same) is designed to make Mathematica as invaluable for scientific research as it is for mathematics."

Comments

Refund please

[bargainsale]

I want a refund on my copy of "A New Kind of Science" before thinking about paying more money to the Wolfram organisation.
Much handwaving, little meat, astonishing arrogance.

One of the most overhyped books I've ever actually been suckered into buying.

I found particularly offputting W's treatment of important parts of his own thesis (computational completeness of some automata) as commercial secrets

Re:I love mathematica

[djupedal]

>Can't wait to see what new stuff they put into this.

Other Recently Added Features:

Visualization & Graphics
High-Impact Adaptive Visualization
Automated Computational Aesthetics
Fully Automated Graph Layout
Real-Time 3D Graphics
Automated Table Layout
Dynamic Interactivity

Mathematics & Algorithms
Integrated Geometric Computing
Combinatorial Optimization
Constrained Nonlinear Optimization
Equational Theorem Proving
High-Level String Computation
New Generation Numerical Integration

Computable Data
Financial Data
Astronomical Data
Country Data
Particle Data
Graph Data
Mathematical Data

Data Manipulation
Exploratory Data Analysis
Symbolic Sound Support
Symbolic Report Generation
3D Printing & Scanning Support
Symbolic Statistical Computing

Core Language
Unification of Graphics, Text & Controls
Language for Data Integration
Dynamic Graphical Input
Instant Multimedia Programming
Real-Time Code Annotation
Instant High-Level Debugging

Interface & User Experience
Symbolic Interface Construction
Integrated Graphics Editing & Drawing
Built-in Gamepad & HID Support
Streamlined Presentation Framework
New Documentation Framework Dynamic Interactivity

Skill

[Anonymous Coward]

The amount of skill and programming know how to make a program like Mathematica is amazing. I would love to see the code on how they do things.

Re:Skill

[muuh-gnu]

>The amount of skill and programming know how to make a program like Mathematica is amazing.

You mean, as amazing as the amount of skill and know how required for practically every large scale application?

>I would love to see the code on how they do things.

You can any time start looking at and learning from completely free systems like GNU Octave, Sage or SciLab.

Re:From my point of view

[muuh-gnu]

>This just seems like its got so bloated that it will likely be priced beyond the budget of most students.

It isnt aimed at students.

>what's wrong with small tools at low cost which work together?

Wolfram does not want you to work with any competitor's product. He wants you to raise a mortgage in order to be able to pay for his "complete solution".

>which isn't really what we want.

Except it really is what most of us want. Why shouldn't it?

Fuck Mathematica

[Anonymous Coward]

and fuck Matlab too, while we are at it. I got a free hit of Matlab in university and then found out how much they charge for licenses only after I was an addict (had a pile of useful code that I didn't want to throw away). I am not going to keep paying for the privilege of running my own code and am busily learning Python.

Mathematica code belongs to Wolfram Research, Matlab code belongs to the Mathworks, but Python code can be MINE! (and yours too, if I want to give it to you.)

I don't buy into the virtual machines they are pushing now either; they might be free as in beer, but it is only a short-term solution and is nothing more than "free hits" to generate more addicts that need licenses.

Free Alternative? Sage maybe.

[TiberSeptm]

The closest thing to a free alternative I've been able to find is Sage: http://www.sagemath.org/ Compared to MatLab, Maple, and Mathematica (yes I know MatLAB is differently purposed than the other two) the usability of Sage blows. It's pretty powerful sure, but when even Maple is easier to use then you've got a problem. I may give the new Mathematica a try. Integration with Word will make some of my lab writeups go a bit faster. Well, maybe as long as Mathematica doesn't take too long to figure out. Too bad our University doesn't sell it to students for $5 a pop anymore.

Re:Fuck Mathematica

[sakdoctor]

Maple 12 life!

Maxima

[Brain-Fu]

Maxima is released under the GPL.

Re:Maxima

[Anonymous Coward]

This is a development version of Maxima. The function bug_report() provides bug reporting information.

Yeah, I know, it's not a real excuse - they probably could do with some improvements to their automated pre-release testing, even for development versions.

But then, it IS a development version, not a stable one, so things like this aren't entirely unexpected, either. And you really shouldn't use development versions for production, anyway: you actually got lucky that you just got a segfault. It might just as well have been the wrong result, without any indication that anything went wrong.

Re:Maxima

[TheodoreGray]

For what it's worth, the equivalent input in Mathematica is a bit simpler:

q = RotationMatrix[t, {0, 0, 1}];
Simplify[q . Array[t, {3, 3}] . Transpose[q]]

The output is:

{{Cos[t]^2 t[1, 1] - Cos[t] Sin[t] (t[1, 2] + t[2, 1]) + Sin[t]^2 t[2, 2], Cos[t]^2 t[1, 2] - Sin[t]^2 t[2, 1] + Cos[t] Sin[t] (t[1, 1] - t[2, 2]), Cos[t] t[1, 3] - Sin[t] t[2, 3]}, {-Sin[t]^2 t[1, 2] + Cos[t]^2 t[2, 1] + Cos[t] Sin[t] (t[1, 1] - t[2, 2]), Sin[t]^2 t[1, 1] + Cos[t] Sin[t] (t[1, 2] + t[2, 1]) + Cos[t]^2 t[2, 2], Sin[t] t[1, 3] + Cos[t] t[2, 3]}, {Cos[t] t[3, 1] - Sin[t] t[3, 2], Sin[t] t[3, 1] + Cos[t] t[3, 2], t[3, 3]}}

And it doesn't crash.

Re:Fuck Mathematica

[rcallan]

Octave is a free version of Matlab, practically all your Matlab code will work in Octave.

Re:Slashvertisement

[SQLGuru]

http://en.wikipedia.org/wiki/Comparison_of_computer_algebra_systems

Take your pick. Some will obviously be better suited to your needs (or lack of needs) as appropriate.

Layne

Re:But does it

[fm6]

Actually, it does. The Itanium-Linux version costs an extra K, though.

Re:Fuck Mathematica

[rahuja]

You don't have to throw that code away or port it to an entirely different language (though Python rocks, and I wish my day-to-day job let me use more of it) Try GNU Octave - that's what I used to back in college because my department didn't have licensed copies of MATLAB installed/available, so-called student versions were insanely impossible and expensive to get hold of (Indian students can't afford $100), and I didn't want to pick a pirated one like the rest of the class.

Possible the first open-spurce software I practically used (except playing with Linux).

Code was very cross-compatible between Octave and MATLAB, except say constants like "e" and "exp" (and of course the MATLAB-specific toolkits). The toughest part at that time was explaining to the professor (who had no idea what "open-source" was) that I did *not* use MATLAB, but it would run on MATLAB fine if he wanted to check that my assignments work fine.

API sucks

[pzs]

I had to write some code using the Mathematica API once, and it hurt. It provides a pipe of tokens, but if you ask for the wrong token, it hangs. You can peak at the front of the queue, but it's still the case that every time you want to read in a token you have to write code to expect any of a million different types of token for all the crazy error messages you never knew you might get.

Also, the GUI is awful. That notebook metaphor just does not work. You want to remove a buggy line of code somewhere but it might be attached to another block so it's really hard to get hold of it. The navigation keys (pg up, end and so on) don't work as you'd expect in an editor so you become very mouse reliant, which is awful for anybody used to working in a programming environment.

In my experience, Matlab is far superior although as others have pointed out, I'd still rather be working in Python. Numpy anybody?

Re:Slashvertisement

[mblase]

No advertising here, just a happy math nerd who was recently investigating alternatives like Maxima and SciLab himself recently, and was impressed that the new version of Mathematica leapfrogged them all by doing much more instead of just doing what it does faster.

(This despite the fact that Mathematica is, and nearly always has been, far more number-crunching power than I've ever needed in my academic or professional career.)

Re:Slashvertisement

[Xamusk]

Sage http://sagemath.org/ is coming pretty good. Version 3.2 will come out in just a few days.

And you can use Mathematica, Matlab, Maple, Magma, Maxima, etc from inside Sage if you have those programs available.

rich enough to use Mathematica?

[mkcmkc]

does it matter that it's open source or not?

It does if you don't have $2400 to spend on a copy of Mathematica.

Re:Slashvertisement

[TiberSeptm]

For what it does, Maxima is pretty good. It's fairly easy to use compared to the other big free alternative. That being said, it is fairly limited compared to Mathematica, Maple or Sage. If you need it to be free and need more features, check out sage (www.sagemath.org) but don't expect to produce anything useful in the first minute. If you're looking for a basic accessible CAS, then Sage wouldn't be the answer. In that case, Maxima might do it for you. Sage is more useful for people who need a more robust system, but I have often found I can write my own tools faster than I can do it in some of the free alternatives.

Re:Fuck Mathematica

[dextromulous]

Octave is a free version of Matlab, practically all your Matlab code will work in Octave.

... if you don't use any of Matlab's GUI stuff... or their toolboxes... some of your code MAY be runnable by Octave if you're lucky. I do like Octave and use it myself, but porting Matlab code to Octave doesn't always work.

Re:From my point of view

[Anonymous Coward]

Mathematica is truly one of the most impressive programs I have used. (Unlike other large scale programs like Office, Matlab, Labview, etc, it contains a lot of code that I just wouldn't know how to even start rewriting if I had to....stuff like Solve[] or Integrate[])

That being said, Mathematica 6 really annoyed me. They completely changed the way MultipleListPlot[] works, which broke a lot of my older notebooks. (The old way was not very good, but backwards compatibility is crucial.)

Re:Slashvertisement

[ctennenh]

I use GNU Octave. I'm a grad student in Discrete Mathematics and Octave's ability to read my MATLAB code lets me work from home and test out light algorithms without the mess of SSH-ing into my server at the office. It's been great.

Re:Slashvertisement

[lexDysic]

Another vote here for Sage. On the open-source side of things, nothing comes close, because everything else that's any good (Maxima for example) is included within Sage, in a fairly transparent way. (I.e., the user doesn't need to know she's using Maxima.) Secondly, the (free) support is awesome. If you spend a little while learning Python and the basics of Sage, and you still have questions, the response time at sage-support at googlegroups is incredible.

Re:Slashvertisement

[Anonymous Coward]

Engineers use Matlab.

Re:Slashvertisement

[bcrowell]

The Wikipedia list is very long. For anyone who's specifically interested in OSS that runs on Linux, here are some of my impressions:

• Maxima - In my experience, it's very mature and bug-free. It's only suitable for interactice use; e.g., if you do certain integrals, it will ask you whether a particular constant occurring in the integrand is positive.
• Yacas - Unlike Maxima, is designed to be suitable for both interactive and noninteractive use. Somewhat buggy, and fails more often than Maxima does.
• Axiom - Has a complete implementation of the Risch algorithm, so it can do some integrals that other programs can't. E.g., it can integrate 1/(x^4-8*x^3+8*x^2-8*x+7), and so can Maxima, but Yacas can't.
• Sage - Pros: Sage lets you program in python, so if you want to mix in some general-purpose programming, python libraries, etc., you can. Sage implements arbitrary-precision arithmetic much more efficiently than programs that use the GMP library. (E.g., sage computes (2^123456789-1)%(2^12345678-1) in about 10 s, whereas ordinary python takes longer to evaluate (2**123456789-1)%(2**12345678-1) than I was willing to wait.) Cons: It's basically a hairball of other math packages, and the interface to other packages often doesn't seem to be very good. It's not packaged properly for debian/ubuntu. The tutorial shows you how to do lots of fancy things using examples from abstract algebra, but doesn't tell you ordinary, useful things, like how to integrate x^a.

Re:Slashvertisement

[mblase]

Let me know when it leapfrogs them in openness.

Sorry, but as a mathematician and a teacher it's more important to me that a CAS application be (1) instructive and (2) correct.

Re:Slashvertisement

[Niten]

Octave is a great program (I switched from MATLAB too, after my bought-and-paid-for copy of MATLAB was broken by a simple OS X upgrade). But Octave is not a symbolic computer algebra system like Mathematica, Maple, Maxima, etc., so it cannot properly be called a Mathematica alternative.

Re:Slashvertisement

[node 3]

Let me know when it leapfrogs them in openness.

Sorry, but as a mathematician and a teacher it's more important to me that a CAS application be (1) instructive and (2) correct.

Which aren't mutually exclusive with openness.