In his latest post Stanislav Malyshev looks at a RFC he's proposed to allow array keys to be objects including some of his thoughts behind the proposal and how he sees it being helpful to the language.
Iām going to put to vote soon another of my RFCs, namely one about āobjects as keysā. So, I want to outline the case for it here and address some criticisms and questions raised while discussing it.
He starts off by answering the "why" question, mentioning specially the introduction of things like GMP numbers and how, despite them seeming to work like numbers, other things can be done with them. He talks about how you'd use this functionality "the right way" and how that'd relate back to value objects. He answers a few other questions about the proposal including why it's better than just using __toString or spl_object_hash instead. He spends the rest of the post looking at some of the implementation problems, disadvantages and some of the possible names (function names) for the handling.