From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from r-finger.com (r-finger.com [178.79.160.5]) by yocto-www.yoctoproject.org (Postfix) with ESMTP id 1E886E01467 for ; Fri, 12 Oct 2012 00:37:04 -0700 (PDT) Received: from [192.168.0.2] (host81-153-82-212.range81-153.btcentralplus.com [81.153.82.212]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by r-finger.com (Postfix) with ESMTPSA id 7C28D99A4 for ; Fri, 12 Oct 2012 08:36:56 +0100 (BST) Message-ID: <5077C897.6000800@r-finger.com> Date: Fri, 12 Oct 2012 08:36:55 +0100 From: Tomas Frydrych User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.5) Gecko/20120624 Icedove/10.0.5 MIME-Version: 1.0 CC: meta-ti@yoctoproject.org References: <1349912433-21880-1-git-send-email-ebutera@users.berlios.de> <1349912433-21880-5-git-send-email-ebutera@users.berlios.de> <5076C554.6060200@mlbassoc.com> <5076D6CC.5010303@mlbassoc.com> <50771DC8.3090903@r-finger.com> In-Reply-To: Subject: Re: [PATCH 5/6] gstreamer: bbappend to use TI-patched versions needed for gst-plugin-ducati X-BeenThere: meta-ti@yoctoproject.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Usage and development list for the meta-ti layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Oct 2012 07:37:04 -0000 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit On 12/10/12 00:00, Daniel Stone wrote: >> I don't think there is anything at all gained by supporting >> simultaneously installing two versions of gstreamer; it makes no real >> sense on distro level. More so adding a version into a package name >> creates a world of pain and dependency hell. Would it not be enough to >> have _0.10 and _1.0 recipes alongside each other, one of which would >> have DEFAULT_PRIORITY=-1; a distro then can choose an appropriate >> version as it wants to? > > Well, given that all the client and plugin code both needs to be > explicitly ported from 0.10 to 1.0, I don't see the problem tbh. Keep in mind that we are not talking about desktop-like distros aspiring to support gazillions of packages of random provenance (because, as you well know, Linux is about choice!). We are talking about carefully crafted distros for specific HW configurations. IMO it is a poor policy to ship a device with two versions of anything on it, it just increases QA complexity. Tomas