On 11 July 2018 at 14:44, Levi Morrison <le...@php.net> wrote: > > My logic is quite simple: > > 1. Something as big as Typed Properties shouldn't be a last minute, > rushed > > RFC. Really - any RFC shouldn't - but in particular major language > changes. > > I have seen this sentiment expressed elsewhere. I want to personally > affirm that Nikita and Bob have been working hard on this for a long > time. This is not a last minute change being rushed. They have been > quite thorough and have discovered and fixed many edge cases that I > suspect other authors would not have found or would have ignored. Yes, > it is being voted on in the narrow gap between first alphas and > feature freeze but that does not mean it is rushed. > >
They have indeed discovered many edge cases, and the RFC is extremely thorough. But by the same token, they have made judgement calls about the best way to handle some of those edge cases, and the RFC is 8000 words long. That's a lot of information for the rest of us to digest, and it's not just trivia: accepting any implementation of this feature means making some far-reaching decisions for the future typing functionality of the language. For instance, I think the initialisation requirements for non-nullable properties deserve a discussion in their own right, but I haven't yet had time to write up my thoughts. This isn't about rejecting the hard work that has gone into this; it's about making sure there isn't some tweak we want to make before people start building code that relies on it. [Incidentally, I just noticed that we have both https://wiki.php.net/rfc/typed-properties-v2 and https://wiki.php.net/rfc/typed_properties_v2 - should this actually have been "3.0"?] Regards, -- Rowan Collins [IMSoP]
