From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35121) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1addJH-0001R2-Is for qemu-devel@nongnu.org; Wed, 09 Mar 2016 07:33:52 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1addJC-0004aJ-KT for qemu-devel@nongnu.org; Wed, 09 Mar 2016 07:33:51 -0500 Received: from mx1.redhat.com ([209.132.183.28]:43523) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1addJC-0004a9-Es for qemu-devel@nongnu.org; Wed, 09 Mar 2016 07:33:46 -0500 References: <56DFF8DA.2070408@redhat.com> <56E0136E.8050709@redhat.com> <56E014BF.7060507@redhat.com> <56E017AF.2090107@redhat.com> From: Paolo Bonzini Message-ID: <56E01825.60102@redhat.com> Date: Wed, 9 Mar 2016 13:33:41 +0100 MIME-Version: 1.0 In-Reply-To: <56E017AF.2090107@redhat.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] why restrict pull reqs to signed tags? List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Laszlo Ersek , Peter Maydell Cc: "Jordan Justen (Intel address)" , David Woodhouse , qemu devel list , Ard Biesheuvel On 09/03/2016 13:31, Laszlo Ersek wrote: > > Ironically, projects using github pull requests do this. They do code > > review through the website and merge with a button. The resulting > > history is non-linear. > > The website based review is a big minus: > - email is more flexible for formulating a careful, detailed review > - the review discussion is independently archived, not held hostage in > a proprietary system > > The final result is also inferior I think: > - the various feedback tags are not captured in the commit message of > each individual patch > > We've been getting github pull requests for edk2. I'm always in a rush > to reject them, lest another maintainer click the button out of > oversight. I insist on keeping it all on-list. I agree, hence the "Ironically" part. Still, the point remains that github pull requests result in a much more non-linear history than Linux or QEMU. That said, it also doesn't do exactly what David says, because the tags are recorded in the web interface only---not in the commit. Paolo