Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:120788
MIME-Version: 1.0
References: <CAFMT4Nq3y+XKUP7aG7F7aht1DdM7dpqxMT-hcS+Nyvc9FaT=+A@mail.gmail.com>
 <CAFMT4Np=n7ENbFvNrygc+vRRiBhtKVPFx5pN1gBMyRhMdsCVBw@mail.gmail.com>
 <CAAfnsFUo+QU_QwiQivH9dfUH0i82tRKo-M9_Xy-n16ac9ytVmQ@mail.gmail.com> <B7E2C645-79F4-45F3-9F60-4D9BF34973E1@woofle.net>
In-Reply-To: <B7E2C645-79F4-45F3-9F60-4D9BF34973E1@woofle.net>
Date: Wed, 12 Jul 2023 01:20:27 +0200
Message-ID: <CAPTD5yE1-fS_L4nKs5BK5m882PNV9jvAVJG85AcJGFeFLNDUaA@mail.gmail.com>
To: Dusk <dusk@woofle.net>
Cc: Peter Kokot <petk@php.net>, Levi Morrison <morrison.levi@gmail.com>, 
	internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary="000000000000dc115006003e5869"
Subject: Re: [PHP-DEV] [VOTE] Interface Default Methods
From: jorg.sowa@gmail.com (Jorg Sowa)

--000000000000dc115006003e5869
Content-Type: text/plain; charset="UTF-8"

> So, here's a crazy thought: why do inheritance, interfaces, and traits
> need to be separate at all?
>
> At some level, they're all ways of composing a class with different
> restrictions, e.g. inheritance requires a single parent, interfaces must be
> abstract, traits don't interact with typing, etc. This RFC was an attempt
> to lift some (but not all) of these restrictions on interfaces; in that
> sense, it almost seems like too small of a step, and one which leaves the
> language in a confusing intermediate state.
>

I agree with this statement. Although I can't vote, I would be for this
feature. However, abstract classes would become orphans with the approval
and should be more clarification in the RFC what to do with them and the
whole inheritance design in the language, because this is significant
change.

--000000000000dc115006003e5869--