On Tue, Apr 21, 2009 at 11:36 PM, Arthur Pemberton <pemboa@xxxxxxxxx> wrote: > On Tue, Apr 21, 2009 at 1:45 PM, Mark <markg85@xxxxxxxxx> wrote: >> On Tue, Apr 21, 2009 at 4:20 PM, John5342 <john5342@xxxxxxxxxxxxxx> wrote: >>> 2009/4/21 Callum Lerwick <seg@xxxxxxxxxx>: >>>> On Tue, 2009-04-21 at 03:04 +0100, John5342 wrote: >>>>> What i would be interested in though is some kind of standardised api >>>>> implemented in all the major bug trackers so other things can be >>>>> easily implemented on top of them such as semi automated up-streaming >>>>> of bugs. Making use of something like OpenID would reduce the multiple >>>>> login issue. >>>> >>>> +1 the Open Source world fixing the Single Sign-on problem would go a >>>> long way towards solving much irritation. I suppose it's a matter of >>>> getting everyone to support OpenID... >>> >>> The good news with getting everyone to support openid is that most >>> people use some version of one of a limited number of bug tracker >>> systems such as bugzilla or trac. A quick google shows there is >>> already some work being done on making bugzilla support openid. If a >>> few more trackers work on it then we will be well on our way towards >>> convincing individual projects to use it. >> >> OpenID is indeed a solution to one of the problems (multiple bug >> reporting accounts). But that would only work if: >> 1. All bug trackers support OpenID >> 2. all currently using a bug tracker update to a version that uses OpenID. > > > What exactly is the intended point of having a single IMS? Lets assume > that all IMSs support OpenID, what is the point of a single issue > tracker then? For all your questions: read the first post. On Wed, Apr 22, 2009 at 12:39 AM, Basil Mohamed Gohar <abu_hurayrah@xxxxxxxxxxxxxxxxx> wrote: > On 04/22/2009 04:58 AM, Callum Lerwick wrote: >> >> On Tue, 2009-04-21 at 22:37 +0200, Mark wrote: >> >>> >>> I'm also playing a bit with the idea that there is one central bug >>> tracking system with ALL (ideally) foss projects in it and that place >>> is the main and upstream place(that's the general idea you all know by >>> now). >>> >> >> No. This is not desirable. Forget it, it's never going to happen. Clear >> your mind. Think git, not CVS. Think distributed bug tracking. >> >> Interoperation, not consolidation. >> >> And there's the basic fact that no one wants a single point of failure. >> What if the One Bug Tracker To Rule Them All goes down? The entire Open >> Source world screeches to a halt? Who is everyone going to trust to run >> this thing? What if the OBTTRTA gets hacked? >> >> The beauty of Open Source is that people are *not* forced to work >> together. >> > > I think the same argument could be made to apply to FreeNode for IRC, and > yet, it seems to be working out just fine for the vast majority of FOSS > projects. So, the "One X to Y them all" concept can theoretically work for > some aspects of FOSS development. However, everyone's (almost, at least) > are already using FreeNode, so we don't really have to convince anyone. > > Having said that, I think you're on to something about the git vs. CVS > concept - distributed bug tracking. There's already something minorly akin > to this with feature of remote bugs in Bugzilla, such as in Ubuntu's > launchpad or Gnome's Bugzilla. Maybe it's not single-sign on that's needed, > but a standard interface for bugs across different implementations, such > that data can be gleaned from them via web services. Yeah, doesn't sound so > yummy, but it's something that can be implemented gradually, as well. Ehm.. this is not a new piece of software that's going to compete with existing ones. It's a piece that will join all existing ones, offer a database where the existing ones "hook" in, without losing the existing ones (or that's what would be best if you ask me). -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
