Newsgroups: php.internals Path: news.php.net Xref: news.php.net php.internals:101072 Return-Path: Mailing-List: contact internals-help@lists.php.net; run by ezmlm Delivered-To: mailing list internals@lists.php.net Received: (qmail 65208 invoked from network); 6 Nov 2017 12:38:35 -0000 Received: from unknown (HELO lists.php.net) (127.0.0.1) by localhost with SMTP; 6 Nov 2017 12:38:35 -0000 Authentication-Results: pb1.pair.com smtp.mail=johannes@schlueters.de; spf=permerror; sender-id=unknown Authentication-Results: pb1.pair.com header.from=johannes@schlueters.de; sender-id=unknown Received-SPF: error (pb1.pair.com: domain schlueters.de from 84.19.169.162 cause and error) X-PHP-List-Original-Sender: johannes@schlueters.de X-Host-Fingerprint: 84.19.169.162 mail.experimentalworks.net Received: from [84.19.169.162] ([84.19.169.162:52896] helo=mail.experimentalworks.net) by pb1.pair.com (ecelerity 2.1.1.9-wez r(12769M)) with ESMTP id A3/3B-09857-9C7500A5 for ; Mon, 06 Nov 2017 07:38:35 -0500 Received: from kuechenschabe.fritz.box (ppp-46-244-186-130.dynamic.mnet-online.de [46.244.186.130]) by mail.experimentalworks.net (Postfix) with ESMTPSA id CDFFE682DB; Mon, 6 Nov 2017 13:38:30 +0100 (CET) Message-ID: <1509971910.22577.22.camel@schlueters.de> To: Michael Morris , PHP internals Date: Mon, 06 Nov 2017 13:38:30 +0100 In-Reply-To: References: <9331841e-ea9a-652b-3b54-5dc24dbd273b@gmail.com> <1509917800.22186.15.camel@schlueters.de> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.18.5.2-0ubuntu3.2 Mime-Version: 1.0 Content-Transfer-Encoding: 8bit Subject: Re: [PHP-DEV] Proposal: Rename image functions From: johannes@schlueters.de (Johannes =?ISO-8859-1?Q?Schl=FCter?=) On So, 2017-11-05 at 20:18 -0500, Michael Morris wrote: > Then, for backward compatibility, add a php.ini setting that runs > "import PHP\Legacy" at the start of each script. We did that mistake with magic_quotes and the ze1 compatibility mode in PHP 5.0. We should learn from that and reduce number of ini settings which have impact on the script behavior, not increase this. This makes, among other things, writing portable code hard (While the environment changed, back in the days shared hosting was a bigger thing, while nowadays with containers users often have more control over their environment) *If* we do something in those areas this should always be that we provide superb options to do stuff in a "clean" (as defined in contemporary understanding) way, which makes the old way obsolete. Introducing a new set of APIs without substantial gain won't get traction (as developers have to relearn, books, tutorials, ... are suddenly outdated, libraries have to make compatibility choices, copy&paste-programming keeps legacy alive, ....) As an example: If we provide a new full set of unicode aware string manipulation, routines these might render byte-based string functions obsolete for many use-cases and offer a gain. (how else could an application handle emojis ... so this is nowadays is even relevant for English-only use cases ... ?) Or if we provide an image manipulation framework which supports higher level constructs than pixels and lines ... johannes