Cwm acts as a rules processor, using information written in N3 rules to guide it in manipulating the RDF/N3 information it has stored. While rules processors are not exactly commonplace, and understanding them is not manditory for the working programmer, they do have a long and solid history. Where does cwm fit into that history?
The field has sometimes been called Knowledge-Based Systems or Expert Systems, but now people often just says something "uses rules."
There seem to be four different ways people think about and want to use rules:
Automated reasoning using first-order became generally feasible in 1965 with Robinson's resolution and hyperresolution algorithms. Today a raft of automated theorem provers continue this tradition, but they see little use in general computing. The strength here is general expressive power: the machine does perform classical logic operations; the weakness is that such systems generally become too practical, real-world application problems.
A focal point for this research is Thousands of Problems for Theorem-Provers (TPTP), which includes links to provers and a conversion utility between logic languages. You can play a little with its web form (try problem ALG001-1).
In the RDF/Semantic Web community, people have used at least OTTER and SNARK.
In 1970-1972, Prolog introduced Logic Programming, which took a restricted form of first-order logic (Horn clauses) and offered to prove things with them in a deterministic order, very much like running a program. Prolog always chains backward from a query.
(1970-1975 also saw the introduction of C, Pascal, Scheme, Smalltalk, and Microsoft BASIC.)
Tabled Prolog (as in XSB) allows rules to be written without worrying about looping, much like with cwm.
A different approach became popular with OPS5 (which is since continued into CLIPS and JESS), presenting rules as being "triggered" and causing actions.
Unlike Prolog, these are usually (but not always) chaining forward from the givens, like cwm.
Modern reasoners and rule systems often use a complex hybrid of strategies, or are simply developed for a focussed application domain.
Some worth mentioning include: