Slashdot Mirror

Have you tried using sage?

http://www.sagemath.org/
[...]
Sage is a free open-source mathematics software system licensed under the GPL. It combines the power of many existing open-source packages into a common Python-based interface.

Mission: Creating a viable free open source alternative to Magma, Maple, Mathematica and Matlab.
[...]

I think it's great that this is happening and look forward to the adoption of open textbooks.

However, I looked at the PDFs that they post on their website, and they look awful. The formulas and plots were hard to read and pixelated. They should have used LaTeX for their typesetting and formulas and maybe sage or matplotlib to make the graphs.

BJ

Mathematica could be rebuilt in 5 years with a good focus.

Some projects such as Sage already have made large strides:

http://www.sagemath.org/tour-quickstart.html

Sage has similar capabilities to Mathematica including the separation of client and server for example.

I'm sure somebody will point out that there's areas in which Sage and other open projects lag behind the capabilities of Mathematica, Maple, Matlab, etc. Maybe some people absolutely need the capabilities in these proprietary products to do their work, but I'll be damned if I'm going to let myself get shoehorned into a research niche that "requires" dependence on proprietary products, because I don't want to find myself held hostage to somebody's generosity.

At least I know that if I use Sage, SciPy, NumPy, matplotlib, octave, Maxima, etc., to generate some result for a paper, and properly cite my use of said software packages, I'll probably get a, "Gee, thanks for telling people you used our stuff!" response, instead of a cease and desist letter from some money-grubbing fucktard that thinks he owns my research results.

Well,

    I can tell you one thing. If it ever is held up in court and program output becomes copyrighted in any way, I am basically going to quit the industry and open up an Italian restaurant.

    I have no intention of participating in a field that is seething with greed and sowing the seeds of its own darkness.

    The restrictions of IP are so catastrophic right now, that real advances in computer usability are essentially being delayed and in their place, anything that you can create with pretty bitmap graphics is declared a HUGE ADAVANCE or some how "cool".

    This whole mess is because we do not make anything worth a damn any more. In my opinion everyone wants to live like a king and do little if any real work, which is what the whole idea of extending copyrights and IP to ludicrous ends is all about.

      Computers suck right now, and I do not see it getting any better if this sort of restriction is placed on the industry. Can't f'in own anything any more because some rich arse has a army of lawyers to bribe congressional leaders and grease the rails for new extensions to IP laws.

    Perhaps we should target Wolfram in earnest, and simple remove the incentive to buy Wolfram products. We did it with UNIX, (we=open source community). Mathematica could be rebuilt in 5 years with a good focus.

Some projects such as Sage already have made large strides:

http://www.sagemath.org/tour-quickstart.html

Sage has similar capabilities to Mathematica including the separation of client and server for example.

-Hack

http://www.sagemath.org/

Sage is a free open-source mathematics software system licensed under the GPL. It combines the power of many existing open-source packages into a common Python-based interface.

I agree with tb()ne. While I am impressed with FORTRAN and its durability, Python is the tool for my scientific programming. The Sage notebook to be precise, with its access to a raft of mathematical tools and libraries. If you are going to teach programming to science and engineering students, you could do a lot worse than teaching them to do it in Python.

That being said, I have punched cards in FORTRAN IV and slaved over a vt100 with FORTRAN 77 and most recently written some code in FORTRAN 95. When the times comes to do some heavy-lifting with MPI, FORTRAN will be the choice. [With a bit of work on Sage, perhaps writing something to use MPI from Sage that played nicely with schedulers such as PBS and derivatives would be worth looking into.]

http://sagemath.org/

You can learn the syntax of a language that is similar to other languages you already know in a few days, sure. But understanding a new approach to programming (dynamic vs. static, procedural vs. OO vs. functional vs. whatever), understanding the idioms and subtleties of a language, and learning the details of the available libraries, takes months.

Agreed. I think I said so in my post.

I've spent a substantial proportion of my career fixing the mess made by people who thought average programmers could be up to speed for that kind of work within a week or two, and another substantial proportion training those programmers afterwards until they were fit to operate autonomously, so I'm going to have to respectfully disagree with you on that one.

Huh. So, how do you expect programmers to learn? Do they get experience by sitting near an old guy with a big beard? Not really. People learn by making mistakes. The take-home lesson is, don't assign a new programmer to a new project. Assign a senior programmer to a new project, and assign a couple of lackeys to the senior programmer. Ideally, the new programmers will be up and running in a couple of days, able to hack the basics, fix bugs, etc. With a senior programmer checking their work, the damage they can do is pretty limited. Of course, if the senior programmer is a jackass who just wants the code written, then the damage the team will do is unlimited.

I don't claim that average programmers can be fit to operate autonomously in a couple of weeks, but I do think that they can be fit to operate in a couple of days. Also... you never let a new hire thrash around, unchecked, on your codebase -- I don't care how many of decades of experience (s)he has in whatever. Or better yet, implement a peer review system -- every line is double-checked, or the code doesn't go in. This is how Sage has been operating for quite a while now, and we've probably stopped dozens of bugs in their tracks.

Not to be confused with the non-profit maths platform, also called Sage.

I'm a Sage developer, and our only GUI is a web interface. Run Sage on your local machine, and you serve to localhost. As Sage developers use the GUI, we're getting more attached to it, and we keep adding more IDE-like features. Recently, there have been discussions to make it easier to edit Sage directly from the GUI. With a little care and extra work, it seems as though we'll be able to make the system such that multiple developers can collaboratively edit the source, making messy merges a thing of the past.

I won't claim that Sage will become a great IDE -- that's not our plan: we want to make great math software. But, the way that people write software is changing. Local editing tools are the best right now because they've had the most time to develop, and today's developers have grown with them. In 10 years? I'm not sure that the younger generation of developers is going to stick with a local copy of emacs. More and more tools are migrating to the web; I don't care to predict that the world isn't going to change.

You can try SAGE http://sagemath.org/ which is a Python-based CAS which describes itself as trying to build an alternative to Magma, Mathematica, Maple, Matlab, etc.

I use it and it's pretty good. Also it's under heavy development and you can get to directly influence it's direction (if you can code).

http://www.sagemath.org/. Sage rocks. It's python based, brings together many of the useful libraries already mentioned (numpy, scipy, matplotlib, etc.), and has a very active mailing list. Can't recommend it enough.

There's Sage, for advanced mathematics. Built with Python.

Have you considered using sage (www.sagemath.org)? It is FOSS and has highly active community and developer support. I'd suggest reading the tour http://www.sagemath.org/tour.html and seeing what you think.

When that code needs to be modified to run on a supercomputer instead of a workstation, it is usually converted to pure C or Fortran.

Why is this done? If your Python program is spending almost all of its time inside a C module, why is it worth converting the whole thing to C? Not a flame, I just don't understand. Is it because you run the whole computation over and over in a loop, and the Python overhead starts to be noticeable?

P.S. You are a mathematician... have you seen Sage? It's basically a whole bunch of math libraries, glued together with Python.

http://sagemath.org/

steveha

Sage is pretty awesome. [disclaimer: I'm a Sage dev]

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.

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.

Mathematica 7 has launched, as noted in Stephen Wolfram's blog post. Among the new features are huge equation typesetting, transcendental roots, and discrete calculus. Looking back at the version 6 discussion, it's perhaps inevitable that comparisons will be made to CAR, CGsuite, GAP, Geogebra, Geometer's Sketchpad, Geometry Expressions, Geonext, LaTeX, Magma, Maple, Matlab, nauty, noneuclid, Pari, Sage, or SeifertView. In other news, the Wolfram Demonstrations project now has over 4000 interactive math demos.

I hope some math professors are reading this. They always seemed to think that they only needed to teach the "how", as "why" would already be obvious or would become clear. It didn't, not for me. More like that was the excuse, because actually "how" alone was much easier to teach. I studied PDEs in calculus classes, but never used them for anything. When they did come up with example uses, they were pretty contrived, and often could be solved with plain old algebra. Or they were so small that hand application of numerical methods could pin down the answer. Took only a few iterations of the Bisection method to get that zero, or you'd hack up a quick and dirty program to push some data into a linear algebra library function and get back results, something like that. And what's a student to think on hearing that although faster, Newton's Method, which is based on calculus, isn't as reliable as Bisection, which is simple algebra. Not good examples when trying to show students how useful and valuable calculus is.

Books? There's more than books alone out there. Lots of material on the web. Lots of combined material. Here are some books associated with Sage. Are you making use of mathematical software: Sage, Matlab, Mathematica, Maple, or some such? Or are you at least able to code up something in a general purpose language if needed? Much math is to the point where you can't advance without computers. Maybe I'm a bit behind. These days, I suppose all math students use such software.

I've noticed also that people with backgrounds in pure math don't have a good basic understanding of Computer Science. You know all about Fourier Transforms, you've heard of the Fast Fourier Transform, you've heard of big O, but you don't see what the big deal is about the FFT-- to you FFT is just one of many ways to do a Fourier Transform, one specific to computers which a person would not use if working out such a transform on paper. Do you have an appreciation of the algorithmic complexities of the math problems you are encountering? The way multiplication is done in grade school is just fine for relatively few small numbers, but when you want to do millions of multiplications of large numbers (1000 digits, say), you'd better use a computer, and you'd better program the computer to use FFT. A textbook on Numerical Methods could be worth checking out.

Someone should tell this guy about SAGE http://www.sagemath.org/

Microsoft also recently donated $32,000 to support the Sage Mathematics Software Project, which produced GPL'd free software. See the financial contributors list.

  -- William (Sage developer)

Microsoft also recently donated $32,000 to support the Sage Mathematics Software Project, which produced GPL'd free software. See the financial contributors list.

  -- William (Sage developer)

You might want to take a look at SAGE as a platform option for that cluster. Then you should be able to farm out jobs from Matlab, Mathematica etc.

If you haven't already, check out Sage: http://www.sagemath.org/ Here's a review of sorts from a heavy Matlab user: http://vnoel.wordpress.com/2008/05/03/bye-matlab-hello-python-thanks-sage/

Hmm, sounds like a job for the Sage math package.

How about SAGE http://www.sagemath.org/ or matplotlib?

Both Python based.

> What the community needs right now is a Python distro with enough of a
> numerics and graphics package rolled in to do 90 percent of what is in
> Matlab.

Good idea. This is what both Sage and the Enthought Python Distribution are
shooting for.

> (Are the Python people still hashing out that Numerics/Numpy divide?

No that is done. And the lead developer of Numpy -- Travis Oliphant --
now gets to work full time on Python scientific computing, as an
employee of Enthought.

> Is there an engineering graphics library that is Numerics/Numpy compatible?

There is Matplotlib for
matlab like numpy graphics, and Chaco for more dynamic 2d graphics. MayaVi and Sage both provide powerful 3d graphics.

> What the community needs right now is a Python distro with enough of a
> numerics and graphics package rolled in to do 90 percent of what is in
> Matlab.

Good idea. This is what both Sage and the Enthought Python Distribution are
shooting for.

> (Are the Python people still hashing out that Numerics/Numpy divide?

No that is done. And the lead developer of Numpy -- Travis Oliphant --
now gets to work full time on Python scientific computing, as an
employee of Enthought.

> Is there an engineering graphics library that is Numerics/Numpy compatible?

There is Matplotlib for
matlab like numpy graphics, and Chaco for more dynamic 2d graphics. MayaVi and Sage both provide powerful 3d graphics.

David Joyner has a book which explores some of the math behind the Rubik's cube: http://www.amazon.com/Adventures-Group-Theory-Merlins-Mathematical/dp/0801869471

If you are interested in playing around with the symmetry group associated to the Rubik's cube, Sage (http://www.sagemath.org) has good support for it; the documentation can be found at http://www.sagemath.org/doc/html/ref/module-sage.groups.perm-gps.cubegroup.html . Sage also includes a number of efficient solvers for the Rubik's cube.

This article is related to Sage ( http://www.sagemath.org/ ), a free open-source math project. The article is about a computation (not using Sage) of an L-function, a computation about that L-function (using Sage), and a major new NSF-funded initiative to compute large tables of modular forms and L-functions that William Stein (director of the Sage project) is co-directing, which will have a large impact on Sage development.

• Octave
  
• Scilab
  
• SAGE
  

Or, you could use a computer algebra system which has easy-to-use distributed computation built in already. Oh, did I mention, it's open source, so every single point above (with possible exception of software updates) is completely invalidated?

Or, you could use a computer algebra system which has easy-to-use distributed computation built in already. Oh, did I mention, it's open source, so every single point above (with possible exception of software updates) is completely invalidated?

Sage also has an AJAX interface.

I've been making an effort to use Sage in place of Mathematica lately and so far I'm impressed. Although, right now I prefer using the CLI rather than the web interface.

Why not use both?

Exactly. That's why sage will win in the not-too-distant future.

Documentation 101

http://www.sagemath.org/SAGEbin/apple_osx/

"**
    These are only for OS X 10.4. They will not work on OS X 10.3.
**

1) Download the file here to your Desktop (or wherever -- put it
      in a directory with no spaces in it):

      http://sage.math.washington.edu/SAGEbin/apple_osx/

2) Double click on it.
3) Once it extracts double click on the "sage" icon.
4) Select to run it with "Terminal":
          Choose Applications, then select "All Applications" in the
          "Enable:" drop down. Change the "Applications" drop down
          to "Utilities". On the left, scroll and select "Terminal".
          Click "Open", then in the next dialog select "Update".

5) SAGE should pop up in a window.

6) For the graphical notebook, type
          notebook()
      and follow the directions, which are to open firefox or safari (your choice)
      to the URL
              http://localhost:8000/

-- William"

Um. No. Not really.

I wish you the best of luck but this why I use MATLAB.

> I find the packaging of SAGE to be rather arrogant and self-important.

The current packaging of Sage was the most technically efficient way to
accomplish the goals of the Sage project quickly: (1) create a distribution
of math software that builds from scratch on all modern OS X and Linux
installs, (2) create a new library of functionality that ties it all
together and more, and (3) create interfaces to most existing mathematical
software. Creating .deb, .rpm's etc., is something that we've always planned
to do: see http://wiki.sagemath.org/DebianSAGE

> At first glance it looks like SAGE is millions of lines of source code.
> On closer inspection I find that SAGE is really just several dozen open
> source mathematics packages bundled together in a tarball with the SAGE
> name slapped on it.

Sage is not "just that". It is nearly 70 packages, which took many people
a huge amount of time to get to all build together correctly -- and in many
cases (e.g., linbox, genus2reduction, mpfi, pyrex/cython, Singular, etc.)
Sage developers fixed significant bugs in those packages or made major
contributions; in some cases taking them from being nearly-orphaned research
only systems to serious projects. And Sage is also a huge amount
of new code.

> On even closer inspection I find that there is actually
> SAGE code that appears very worthwhile, additional functionality is provided,
> a consistent interface, etc. However, I'm not going to use it seriously
> because I can't 'apt-get install sage-math'.

That will come later when people who want to apt-get Sage actually put in
the work to make it happen. This is of course happening now and I strongly
support it. I just don't have the time to do it myself.

> SAGE wants me to download
> and install more than 200MB of stuff that isn't going to be handled by
> my OS package management, and it duplicates many of the components I
> already have installed that are handled by my OS package management.

In fact, because SAGE builds completely self contained it will not
conflict with or cause any headaches with anything you have installed.

> Don't make me use your own forked and patched versions of Pari or GAP.

I am certainly not making you use anything.

> I can 'apt-get install pari-gp gap' today(and I already have). I'm
> not going to install your alternative versions and deal with any
> inconsistancies between them. I'm not throwing out the ease of use that
> OS level package management provides to get SAGE. I know many other
> people who aren't going to do it either.

Look, there is nothing whatever about the Sage project or me that is
against mainstream packaging. It's just that Sage is a volunteer project
for which most developers are naturally mathematicians. We simply don't
have the time to maintain Debianizing dozens of packages. Arrogance has
nothing to do with it. I very much hope http://wiki.sagemath.org/DebianSAGE
takes off.

  -- William (a Sage developer)

> I find the packaging of SAGE to be rather arrogant and self-important.

The current packaging of Sage was the most technically efficient way to
accomplish the goals of the Sage project quickly: (1) create a distribution
of math software that builds from scratch on all modern OS X and Linux
installs, (2) create a new library of functionality that ties it all
together and more, and (3) create interfaces to most existing mathematical
software. Creating .deb, .rpm's etc., is something that we've always planned
to do: see http://wiki.sagemath.org/DebianSAGE

> At first glance it looks like SAGE is millions of lines of source code.
> On closer inspection I find that SAGE is really just several dozen open
> source mathematics packages bundled together in a tarball with the SAGE
> name slapped on it.

Sage is not "just that". It is nearly 70 packages, which took many people
a huge amount of time to get to all build together correctly -- and in many
cases (e.g., linbox, genus2reduction, mpfi, pyrex/cython, Singular, etc.)
Sage developers fixed significant bugs in those packages or made major
contributions; in some cases taking them from being nearly-orphaned research
only systems to serious projects. And Sage is also a huge amount
of new code.

> On even closer inspection I find that there is actually
> SAGE code that appears very worthwhile, additional functionality is provided,
> a consistent interface, etc. However, I'm not going to use it seriously
> because I can't 'apt-get install sage-math'.

That will come later when people who want to apt-get Sage actually put in
the work to make it happen. This is of course happening now and I strongly
support it. I just don't have the time to do it myself.

> SAGE wants me to download
> and install more than 200MB of stuff that isn't going to be handled by
> my OS package management, and it duplicates many of the components I
> already have installed that are handled by my OS package management.

In fact, because SAGE builds completely self contained it will not
conflict with or cause any headaches with anything you have installed.

> Don't make me use your own forked and patched versions of Pari or GAP.

I am certainly not making you use anything.

> I can 'apt-get install pari-gp gap' today(and I already have). I'm
> not going to install your alternative versions and deal with any
> inconsistancies between them. I'm not throwing out the ease of use that
> OS level package management provides to get SAGE. I know many other
> people who aren't going to do it either.

Look, there is nothing whatever about the Sage project or me that is
against mainstream packaging. It's just that Sage is a volunteer project
for which most developers are naturally mathematicians. We simply don't
have the time to maintain Debianizing dozens of packages. Arrogance has
nothing to do with it. I very much hope http://wiki.sagemath.org/DebianSAGE
takes off.

  -- William (a Sage developer)

>> ... and there's actually another SAGE project at the
>> University of Washington (which I can't find a link for)
>> which does something entirely different.

>The System to Administer Grants Electronically?

I've had accountants from University of Washington call me about
learning Sage (their Sage, not mine) in one of my classes, or
in one of the Sage Days coding sprint workshops like
http://sagemath.org:9001/days7.

I also have a really cool "SAGE" (the grant system) refrigerator
magnet that my accountant here gave me.

  -- Willam (a Sage developer at University of Washington)

For the Sage wiki it is faster to use http://www.sagemath.org:9001/

For the Sage trac server it is faster to use http://trac.sagemath.org:9002/sage_trac

For the Sage wiki it is faster to use http://www.sagemath.org:9001/

For the Sage trac server it is faster to use http://trac.sagemath.org:9002/sage_trac

There is a lot of interest in getting Sage packaged for Debian, but as you correctly guessed this involves a lot of work due to package dependencies. There is a google group coordinating the effort at http://groups.google.com/group/debian-sage/ and a wiki page at http://www.sagemath.org/DebianSAGE .

--Mike ( a Sage developer )

The site is already very slow, so posting the actual links.

http://www.sagemath.org
http://sage.math.washington.edu/sage
http://modular.fas.harvard.edu/sage
http://www.opensourcemath.org/sage/
http://www.cecm.sfu.ca/sage
http://sage.apcocoa.org
http://echidna.maths.usyd.edu.au/sage
http://sage.scipy.org/sage

(be kind: use a mirror: http://sagemath.org/mirrors.html) :)

Downloadable for Linux, Mac, and the other one:
http://sagemath.org/