My first paper on a Language for Legal Discourse (LLD) was published at the International Conference on Artificial Intelligence and Law in 1989 [24]. I used the language subsequently for several small projects: [48] [28] [30] [31], and it motivated much of my theoretical work on Knowledge Representation and Reasoning in those years. At the time, no one was attempting anything as ambitious as the “Rules as Code” movement, and thus I never wrote an interpreter for the entire language or used it to encode a complete statute. But I think this is a feasible project today. Even without a full-scale implementation, I think the design choices embodied in LLD provide useful guidelines for anyone trying to translate legal rules into executable computer code. I will describe these choices in this extended abstract.