On Tuesday, October 15, 2019 5:01 AM, Kerim Aydin <ke...@uw.edu> wrote: > On 10/14/2019 7:56 PM, Jason Cobb wrote: > > However, I would still argue that the rules should, at least ideally, avoid > > circular dependencies, even if only as a matter of idealism. I've attached > > to this email an svg file that I didn't include in my original submission. > > But are the dependencies you've identified truly circular? Many rules have > multiple independent clauses that could very well be separated into smaller > rules. If the "circularity" is created by a link to two entirely separate > clauses within a rule, that could easily be separated without any change of > function, is it really a circular reference to be concerned about?
Agree. Another way of thinking about it is, if all the rules were combined into one rule (disregarding the necessary rephrasing of various metarules like R1030 and R2240), there would in theory be no mechanical difference, yet all the circular dependencies would disappear. The main reason this wouldn't work in practice is that we sometimes have higher-powered rules depending on lower- powered rules - so I argue it's those relationships we should be trying to eliminate, not circular dependencies themselves. -twg
