Newsgroups: php.internals Path: news.php.net Xref: news.php.net php.internals:19422 Return-Path: Mailing-List: contact internals-help@lists.php.net; run by ezmlm Delivered-To: mailing list internals@lists.php.net Received: (qmail 36017 invoked by uid 1010); 5 Oct 2005 16:56:33 -0000 Delivered-To: ezmlm-scan-internals@lists.php.net Delivered-To: ezmlm-internals@lists.php.net Received: (qmail 36001 invoked from network); 5 Oct 2005 16:56:33 -0000 Received: from unknown (HELO lists.php.net) (127.0.0.1) by localhost with SMTP; 5 Oct 2005 16:56:33 -0000 X-Host-Fingerprint: 204.11.219.139 lerdorf.com Linux 2.4/2.6 Received: from ([204.11.219.139:40974] helo=colo.lerdorf.com) by pb1.pair.com (ecelerity 2.0 beta r(6323M)) with SMTP id 03/AA-54476-0C504434 for ; Wed, 05 Oct 2005 12:56:32 -0400 Received: from [10.0.54.117] (host-212-158-241-115.bulldogdsl.com [212.158.241.115]) (authenticated bits=0) by colo.lerdorf.com (8.13.5/8.13.5/Debian-2) with ESMTP id j95Gu54O030946 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Wed, 5 Oct 2005 09:56:08 -0700 Message-ID: <434405A5.3040901@lerdorf.com> Date: Wed, 05 Oct 2005 17:56:05 +0100 User-Agent: Mozilla Thunderbird 1.0.6 (Macintosh/20050716) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Andi Gutmans CC: Edin Kadribasic , Derick Rethans , Christian Schneider , internals@lists.php.net, Jani Taskinen , Zeev Suraski , "ilia@prohost.org" References: <43413474.90802@cschneid.com> <4342371E.8040604@cschneid.com> <43423CB8.2050503@cschneid.com> <434243A0.1060901@cschneid.com> <43424EC4.6010801@emini.dk> <6.2.3.4.2.20051004200445.03535040@localhost> In-Reply-To: <6.2.3.4.2.20051004200445.03535040@localhost> X-Enigmail-Version: 0.92.0.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [PHP-DEV] Re: $ref =& $this; From: rasmus@lerdorf.com (Rasmus Lerdorf) Andi Gutmans wrote: > Regarding BC breakage. I'm not saying we shouldn't break BC in some > cases, but in many cases, there's no big advantage (except for some > people's warm and fuzzy feelings), and it can cause a lot of heart aches > and disruption in PHP use. It's always a tricky balance and I agree with Andi and Dmitry on this particular one. There are plenty of places where a BC break might be warranted, but we really have to pick our spots very carefully, communicate the change and provide clear and simple examples showing people how their code should be fixed. The number of times we can do this is limited and the number of areas we can touch is limited. People seem to have latched onto this $ref =& $this issue for some reason. Is this really a place where you want to use up some of our BC breaking karma points? I really don't see this particular point as being much of a problem. It's somewhat dangerous code in PHP5, but certainly not illegal in itself. And while much PHP 4 OOP code will need to be fixed to run in PHP 5 that is a different issue from the ability to write OOP code that has a chance of working in both. For dangerous, but technically legal code, throwing a warning seems like the best option to me. So, let's pick our battles in a sane and organized manner and give people a feasible migration path. -Rasmus