Exactly. So far RP in the article sounds a lot like data-flow engines (spreadsheets, various visualization tools, DB triggers, even make builds). It has spanned decades and fields, too. Many artificial intelligence systems used this type of reactive engine; for example, the CLIPS engine "reacted" by matching conditions to a subset of currently-asserted facts to trigger actions (which can then cascade by asserting new facts and causing other patterns to match).
The common aspect to all these applications of a data-flow engine is that a Result has Dependencies, and those Dependencies may be "atomic" (like a file timestamp in make), or a Result from an earlier conclusion. At any point in time, the entire scenario can be paused, and each pending Result has a list of Dependencies that may or may not be satisfied at that point. Spreadsheet calculation 101.