[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Axiom-developer] Openmath, Singular

From: Michael Brickenstein
Subject: Re: [Axiom-developer] Openmath, Singular
Date: Tue, 4 Oct 2005 07:28:13 +0200

Hello Bill!

I will forward the mail about the bug report to the singular mailing list (address@hidden). But you shouldn't care about it, Singular should work fine without this dynamic module "". In fact, dynamic modules are quite new in Singular and still experimental.

The only problem which could arise: At the moment gcc-4.0.x does not support boost::python on my Mac, which is needed for the python interface, on which openmath depends. So for that, it might me necessary to use a different compiler. But time will solve this, boost is one of the
projects with biggest compiler support.

The most important I wanted to write you is, that while you cannot use Singular directly on a python level, I can licence my python openmath library to you. At the moment, I did not yet decide for a certain licence. Using GPL would simplify things very much.

In principal however I tend to use a very liberal licence and have yet to check, if this would exclude me from using Singular (which is GPL), which I have strictly to avoid.

It is generic (doesn't depend on any CAS), uses only standard python libraries, is designed to be customizable (at least offers the possibility to improve this point), compact.

The limitations are: I have to clean up code and add some more documentations than "FIXME:" comments. While there exists some Singular plugin and everything is strictly seperated, there does not exist a sophisticated plugin mechanism at the moment. Also, it does not yet provide network support.
The cleanup will need a little bit time.

Since I am in contact with SAGE, which also uses Python, the library could be the base for a common work, using the CAS as plugins, so it will be improved in time. If also MathAction would use it, this would be a joy of my heart: I write my software for being used (very surprising). ;-)

Michael Brickenstein

reply via email to

[Prev in Thread] Current Thread [Next in Thread]