Benjamin Gruenbaum wrote: > Correct, I mean value types (or objects) that have value semantics and support operator overloading. We'll probably talk about that tomorrow in a break-out. At least I hope to! More after this week's TC39 meeting. /be