[ojAlgo-user] ojAlgo's future

[Anders P. <an...@op...>]

Hi,

This mailing list hasn't been used in a long time. Thought I'd post one last message before looking into closing it permanently. (The recommended alternative is to use Discussions at GitHub.)

The only reason ojAlgo exists is that, a long time ago, one of Optimatika's clients needed a (pure Java) QP solver. None were available, so we decided to build one. Since then ojAlgo has grown immensely in scope and capabilities. Over the years, maintenance and further development has in part been sponsored by various client projects, but mainly it has been a hobby project (sponsored by Optimatika or Anders Peterson personally). This is no longer possible - it's simply too much.

There are plenty of abandoned open source projects, who cares.

Well, if you use ojAlgo's suite of optimisation solvers (LP, QP & MIP), then maybe you should care. The truth is, there aren't that many alternatives. Sure there are plenty of Fortran or C solvers around, both commercial and a open source, but if you prefer code that runs on the JVM your options are very limited.

One of the very few alternatives is that Apache Commons Math contains an LP solver. Here's how that compares to ojAlgo:

https://www.ojalgo.org/2021/10/pure-java-lp-solver-benchmark/

BTW if you haven't yet upgraded to the latest version of ojAlgo (v51.3.0) you should do so. There's been significant improvements to the optimisation solvers in several of the recent updates. Also, the finance related stuff has been merged back to the main library, and there's a bunch of new stuff. 

Optimatika believes there is value in ojAlgo, and has decided to refocus and build a business around providing ojAlgo support. We believe the best way to do this is to form long term partnerships with organisations (whos code base) depend on ojAlgo.

Many organisation and development teams find it difficult to recruit talent.

If you have, or plan to get, a dependency on ojAlgo AND is in need of additional developers, Optimatika can help you!

/Anders


ojAlgo contains much more than the optimisation solvers. At its core it is the fasted pure Java linear algebra library available:

https://www.ojalgo.org/2022/02/java-matrix-benchmark/

That can be useful when building a wide variety of things.