From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnout Vandecappelle Date: Tue, 15 Oct 2013 21:55:42 +0200 Subject: [Buildroot] Bumping packages: some comments/suggestions In-Reply-To: <20131015093444.4ef3fa17@skate> References: <20131014113000.73ca82bf@skate> <1642454622.8785613.1381743494806.JavaMail.root@openwide.fr> <525C6845.50803@mind.be> <20131015093444.4ef3fa17@skate> Message-ID: <525D9DBE.7040802@mind.be> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On 15/10/13 09:34, Thomas Petazzoni wrote: > Dear Arnout Vandecappelle, > > On Mon, 14 Oct 2013 23:55:17 +0200, Arnout Vandecappelle wrote: > >>> What I wanted to add in this discussion is this: if there are people >>> out there that have time for some buildroot development but they have >>> no clear goal of their own, they can among others choose between >>> package bumps and fixing autobuild failures. >>> While there certainly is value in package bumping, I hope that not all >>> of these people jump onto the bump-train and instead help with the >>> autobuild failures. We still hope to get to 0 failures at some point, >>> and bumping packages all the time will most of the time only add new >>> problems. >> >> I'm also not too hot on version bumping just for the sake of bumping. >> Version bumps should be done when they're useful for someone, e.g. >> because a feature or bugfix is added. If it's useful to you, then you'll >> obviously test if it still works correctly. > > I understand this point of view, but on the other hand, it's not really > nice to have packages that have very old versions. For "core" stuff > that isn't necessarily very easy to bump (think Qt, Gtk, X.org and > things like that), it's quite important that the core Buildroot > community keeps that up to date, because this is difficult to do for > newcomers, and this is what people will look at to see if Buildroot is > well-maintained and active. Yes of course. The version detection idea is certainly useful. I just wanted to concur with Thomas DS that this shouldn't distract people from fixing autobuilder failures. Regards, Arnout -- Arnout Vandecappelle arnout at mind be Senior Embedded Software Architect +32-16-286500 Essensium/Mind http://www.mind.be G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle GPG fingerprint: 7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F