That is one of its key benefits: the specification is separate from the implementation. That encourages good implementations, and readily allows for multiples.
-John Sent from my location > On Dec 5, 2014, at 7:20 PM, Dominic Hamon <dha...@twopensource.com> wrote: > > If it ends up being not open, then we'd have to. But they're clearly not > going in that direction given they've already published it publicly and have > requested feedback. > > I would strongly urge us to avoid this, however. A common spec is so much > stronger than any individual one. > >> On Fri, Dec 5, 2014 at 6:49 PM, Arunabha Ghosh <arunabha...@gmail.com> wrote: >> Should CoreOS decide to adopt a governance model which is not open or what >> the Mesos community wants, does it make sense to adopt the Rocket spec and >> evolve it along with Mesos ? Having a documented container spec for Mesos >> which can evolve along with Mesos would in some ways be better than perhaps >> adopting Rocket's spec. >> >> Arunabha >> >>> On Wed, Dec 3, 2014 at 11:49 AM, Dave Lester <daveles...@gmail.com> wrote: >>> Thanks, I'll take the discussion to the GitHub issue. >>> >>>> On Wed, Dec 3, 2014 at 11:44 AM, Tim Chen <t...@mesosphere.io> wrote: >>>> Hi Dave, >>>> >>>> I actually can't remember is it on the github issue, email list or twitter >>>> feed, but definitely one of them. >>>> >>>> Tim St Clair just brought up Apache on #139, and I'm a +1 on that as well. >>>> >>>> Feel free to chime in on that Github issue you linked. >>>> >>>> Tim >>>> >>>>> On Wed, Dec 3, 2014 at 11:40 AM, Dave Lester <daveles...@gmail.com> wrote: >>>>> Hey Tim C, >>>>> >>>>> Out of curiosity, which GitHub issue are you referring to when you say >>>>> "Apache is one of the options mentioned"? >>>>> >>>>> I don't see it it in the discussion thread for >>>>> https://github.com/coreos/rocket/issues/139, but I would love to find it >>>>> and +1 the idea. Moving to Apache would be great to see. >>>>> >>>>> Dave >>>>> >>>>>> On Wed, Dec 3, 2014 at 10:02 AM, Timothy Chen <t...@mesosphere.io> wrote: >>>>>> Hi Tim, >>>>>> >>>>>> Definitely I agree, i think what I am getting at is that it's clear from >>>>>> the conversation that a open governance is what they want from day one. >>>>>> Apache is one of the options mentioned one the Issue, and I believe >>>>>> something along that line is most probable. >>>>>> >>>>>> As long as that's true it won't be as difficult as other options to >>>>>> maintain as an containerizer option for us. >>>>>> >>>>>> Tim >>>>>> >>>>>>> On Dec 3, 2014, at 9:42 AM, Tim St Clair <tstcl...@redhat.com> wrote: >>>>>>> >>>>>>> inline below >>>>>>> >>>>>>> From: "Tim Chen" <t...@mesosphere.io> >>>>>>> To: u...@mesos.apache.org >>>>>>> Cc: "dev" <dev@mesos.apache.org> >>>>>>> Sent: Wednesday, December 3, 2014 11:20:47 AM >>>>>>> Subject: Re: Rocket >>>>>>> >>>>>>> Hi Tim, >>>>>>> >>>>>>> I see you've already commented on the rocket repo about this, and from >>>>>>> their messaging it aims to be independent which should be the whole >>>>>>> point of the open container spec. >>>>>>> I'm all over this like white on rice. >>>>>>> >>>>>>> I think the best way is just to be involved in the spec early on and >>>>>>> continue to do so while we move forward, and we have relationships with >>>>>>> the rocket people which should help also being in the loop as well. >>>>>>> Relationships alone won't cut it. >>>>>>> Friends one day, enemies the next, isn't that the way it worked with >>>>>>> Docker...? >>>>>>> >>>>>>> Governance, such as Apaches model, is of critical importance. >>>>>>> >>>>>>> Tim >>>>>>> >>>>>>>> On Wed, Dec 3, 2014 at 8:26 AM, Tim St Clair <tstcl...@redhat.com> >>>>>>>> wrote: >>>>>>>> >>>>>>>> Not to put too fine a point on it, but how are folks planning on >>>>>>>> establishing governance around the App Container spec? >>>>>>>> >>>>>>>> https://github.com/coreos/rocket/issues/193 >>>>>>>> >>>>>>>> If the mesos community decides to leverage our own, how do we ensure >>>>>>>> that we have say in the spec going forwards? >>>>>>>> >>>>>>>> Cheers, >>>>>>>> Tim >>>>>>>> >>>>>>>> ----- Original Message ----- >>>>>>>> > From: "Tobias Knaup" <t...@knaup.me> >>>>>>>> > To: u...@mesos.apache.org >>>>>>>> > Cc: "dev" <dev@mesos.apache.org> >>>>>>>> > Sent: Monday, December 1, 2014 11:39:58 PM >>>>>>>> > Subject: Re: Rocket >>>>>>>> > >>>>>>>> > An important point to clarify is that two things were announced: a >>>>>>>> > spec >>>>>>>> > (App Container) and an implementation (Rocket). >>>>>>>> > Here is the spec: >>>>>>>> > https://github.com/coreos/rocket/blob/master/app-container/SPEC.md >>>>>>>> > This separation of spec and implementation is important. It makes it >>>>>>>> > much >>>>>>>> > easier to integrate in Mesos. systemd is also just the >>>>>>>> > implementation of >>>>>>>> > the runtime part of the spec that CoreOS chose for Rocket. Mesos can >>>>>>>> > use >>>>>>>> > something else or come with its own. >>>>>>>> > >>>>>>>> > >>>>>>>> > On Mon, Dec 1, 2014 at 1:29 PM, Dominic Hamon >>>>>>>> > <dha...@twopensource.com> >>>>>>>> > wrote: >>>>>>>> > >>>>>>>> > > Instead of considering the Rocket runtime as implemented, we should >>>>>>>> > > instead consider how we can implement their specification. A >>>>>>>> > > community is >>>>>>>> > > always healthier when there are multiple implementations of a >>>>>>>> > > specification, and through implementing it we may find ways to >>>>>>>> > > improve it. >>>>>>>> > > >>>>>>>> > > Also, this allows us to be a strong voice in the community and >>>>>>>> > > provide >>>>>>>> > > value through a C++ implementation. >>>>>>>> > > >>>>>>>> > > I've created a JIRA ticket >>>>>>>> > > https://issues.apache.org/jira/browse/MESOS-2162 to track any >>>>>>>> > > thoughts on >>>>>>>> > > this. >>>>>>>> > > >>>>>>>> > > On Mon, Dec 1, 2014 at 11:10 AM, Tim Chen <t...@mesosphere.io> >>>>>>>> > > wrote: >>>>>>>> > > >>>>>>>> > >> Hi all, >>>>>>>> > >> >>>>>>>> > >> Per the announcement from CoreOS about Rocket ( >>>>>>>> > >> https://coreos.com/blog/rocket/) , it seems to be an exciting >>>>>>>> > >> containerizer runtime that has composable isolation/components, >>>>>>>> > >> better >>>>>>>> > >> security and image specification/distribution. >>>>>>>> > >> >>>>>>>> > >> All of these design goals also fits very well into Mesos, where >>>>>>>> > >> in Mesos >>>>>>>> > >> we also have a pluggable isolators model and have been >>>>>>>> > >> experiencing some >>>>>>>> > >> pain points with our existing containerizers around image >>>>>>>> > >> distribution and >>>>>>>> > >> security as well. >>>>>>>> > >> >>>>>>>> > >> I'd like to propose to integrate Rocket into Mesos with a new >>>>>>>> > >> Rocket >>>>>>>> > >> containerizer, where I can see we can potentially integrate our >>>>>>>> > >> existing >>>>>>>> > >> isolators into Rocket runtime. >>>>>>>> > >> >>>>>>>> > >> Like to learn what you all think, >>>>>>>> > >> >>>>>>>> > >> Thanks! >>>>>>>> > >> >>>>>>>> > > >>>>>>>> > > >>>>>>>> > > >>>>>>>> > > -- >>>>>>>> > > Dominic Hamon | @mrdo | Twitter >>>>>>>> > > *There are no bad ideas; only good ideas that go horribly wrong.* >>>>>>>> > > >>>>>>>> > >>>>>>>> >>>>>>>> -- >>>>>>>> Cheers, >>>>>>>> Timothy St. Clair >>>>>>>> Red Hat Inc. >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> -- >>>>>>> Cheers, >>>>>>> Timothy St. Clair >>>>>>> Red Hat Inc. > > > > -- > Dominic Hamon | @mrdo | Twitter > There are no bad ideas; only good ideas that go horribly wrong.
