Wikipedia luckily provides some context, as the linked page is bit spartan:
> In 2007, Axiom was forked twice, originating two different open-source projects: OpenAxiom[3] and FriCAS,[4] following "serious disagreement about project goals".[5] The Axiom project continued to be developed
So FriCas is one of the three main branches of Axiom. But what is Axiom?
> Axiom is a free, general-purpose computer algebra system. [..] which defines a strongly typed, mathematically (mostly) correct type hierarchy [..] and a sophisticated (dependent) type system
> The primary philosophy is that Axiom needs to develop several fundamental features in order to be useful to the next generation of computational mathematicians
This helps make sense of the description of FriCAS
> FriCAS is a general purpose computer algebra system with a strong focus on mathematical research and development of new algorithms
So in summary sounds something that is more aimed to actual mathematicians rather than engineers or other scientists who would be more interested in practical applied methods.
Of the forks, FriCAS on a surface level seems the most alive. Original Axiom had its last release in 2014, but appears to still have some ongoing development. OpenAxiom had last release in 2013 and last commit in 2015, so it feels pretty dead to me. FriCAS meanwhile has had 9 releases since 2014 and at least some activity in GitHub.
The last Axiom update was just this week (March 21, 2019). In general, Axiom is updated about once a month.
The current Axiom effort is directed toward proving the algorithms. The research combines computer algebra with proof theory. A survey done last year shows that there is almost no overlap between these two fields of computational mathematics.
Computer algebra IS mathematics. The present state of the art is "works-for-my-case" implementations. That is fine for the short term but Axiom focuses on the "30 Year Horizon". Early computer algebra systems did integration "by heuristics". Systems like Axiom use the Risch algorithm.
The next generation systems should use proven algorithms. At this time this is still an open research area. Axiom was originally a research platform (at IBM Research) and, as open source, still retains its research focus. Research takes time.
This is a difficult research problem. Not everyone agrees it can or should be done. "Works-for-my-case" implementation is easier.
Axiom and Fricas are not in competition. We simply have different goals. The Fricas people are doing excellent work. I would encourage you to try the software.
> In 2007, Axiom was forked twice, originating two different open-source projects: OpenAxiom[3] and FriCAS,[4] following "serious disagreement about project goals".[5]
> In 2007, Axiom was forked twice, originating two different open-source projects: OpenAxiom[3] and FriCAS,[4] following "serious disagreement about project goals".[5] The Axiom project continued to be developed
So FriCas is one of the three main branches of Axiom. But what is Axiom?
> Axiom is a free, general-purpose computer algebra system. [..] which defines a strongly typed, mathematically (mostly) correct type hierarchy [..] and a sophisticated (dependent) type system
> The primary philosophy is that Axiom needs to develop several fundamental features in order to be useful to the next generation of computational mathematicians
This helps make sense of the description of FriCAS
> FriCAS is a general purpose computer algebra system with a strong focus on mathematical research and development of new algorithms
So in summary sounds something that is more aimed to actual mathematicians rather than engineers or other scientists who would be more interested in practical applied methods.
Of the forks, FriCAS on a surface level seems the most alive. Original Axiom had its last release in 2014, but appears to still have some ongoing development. OpenAxiom had last release in 2013 and last commit in 2015, so it feels pretty dead to me. FriCAS meanwhile has had 9 releases since 2014 and at least some activity in GitHub.