> I feel about cyclic data about the same way as you feel about lists, and if 
> you could find some reliable way for the compiler to detect and reject all 
> cyclic data (other than perhaps at the global level as Elm allows), I would 
> greet it with a sense of relief, as we would never have to consider it again 
> other than to help library writers who have used it to replace it with 
> something better!

Well B/D does mention that cycles are really rare but can come up. In Nim the 
situation is worse than that because we really want to support `swap` well. 
However, the cycles that cause trouble in practice for Swift etc. are those 
resulting from closures pointing back to some "upper layer" and afaict these 
cycles are prevented. Statically.

Reply via email to