I think this email from Owen summarizes what we need to keep in mind here.
Containers have caught on due to solving some important problems and thus
people are looking at models for what the future operating system would look
like where containers are the primary content delivery mechanism. In Fedora we have
efforts arounds Docker/OCI containers and Flatpak containers and we are looking at image
based OS installs with the Atomic and Atomic Workstation effort. The fact that we are
developing stuff like this in Fedora is a good thing as
it means that if it does turn out to be a better model we are well positioned to take
advantage of the shift in the market. And if the scepticism some people have about
turns out to be well founded we still have our RPM based OS to fall back on.
So in the long term maybe we will start only shipping containers, if nothing else that is
worthwhile goal for the efforts as it informs the design decisions we take for this like
Project Atomic. But having that as a effort goal and it actually becoming the reality is
two very different things. If the model works really well, yes maybe we at some point
into the future decide to stop doing RPMS (at least as an end user artifact), but in the
nobody should feel threatened by the efforts done around containers or modularity.
And who knows maybe the future ends up being a bit of a hybrid, Colin Walters is doing
work around rpm-ostree for instance.
----- Original Message -----
> From: "Owen Taylor" <otaylor(a)redhat.com>
> To: "Development discussions related to Fedora"
> Sent: Tuesday, July 18, 2017 10:15:12 AM
> Subject: Re: F27 System Wide Change: Graphical Applications as Flatpaks
> On Tue, 2017-07-18 at 08:35 +0200, nicolas.mailhot(a)laposte.net wrote:
> > > > Even if it eventually succeeds crash-landing it in Fedora while
> > > > half
> > > > the security and management tools are lacking is a great way for
> > > > the
> > > > distribution to get an awful reputation, while others will rip
> > > > the
> > > > fruits of this work some years later.
> > > I'm entirely puzzled about how you think we could possibly land
> > > Flatpak
> > > support in Fedora well integrated with our infrastructure, and our
> > > security and management tools without starting to work on it, which
> > > is
> > > essentially what this change proposal is about
> > Working on it is fine. Improving it is fine. Proposing Fedora-
> > generated Flatpacks outside of Fedora is fine.
> If Fedora community members are using Fedora infrastructure to build
> Flatpaks that's really by definition part of the Fedora project, isn't
> > Planning to ship stuff as Flatpack only when basic questions such as
> > inter-component dependencies, automated deployment (kickstarts),
> > actual network and disk use (chromebooks), actual user adoption,
> > actual convenience of the security model, etc are not resolved is
> > not.
> I think it's important to think ahead and be transparent about what we
> are thinking about, so in that sense we're "planning" to ship things
> Flatpak only. But I want to be clear that there is no *proposal* on the
> table to ship things Flatpak only, and *no proposed timescale*. And
> there won't be until we know how the tools work out for packagers, how
> Flatpak usage works out for users, and we have a significant body of
> Fedora packages built as Flatpaks to look at things like installed size
> and network usage.
> These are things we can only get to by building out the infrastructure
> so that packagers can start trying building Flatpaks and users can
> start trying installing them.
> > That's the hard and tedious stuff most people on this list care about
> > and GUI app writers, not a lot. That's the point of no easy return
> > where Flatpack is forced on users be it ready or not.
> > There is not a vast amount of trust given past history and the way
> > some Flatpack proponents clearly intend to shaft the methods that
> > built Fedora (and its userbase) to jumpstart something else.
> It's perfectly fine to be skeptical, it's perfectly fine to ask hard
> questions about areas you think need more work.
> But the only way that we get to something that is an evolution of how
> Fedora currently works, that pays attention to the needs of current
> Fedora users, and builds on the strengths of the Fedora infrastructure
> and the people doing all the work in the Fedora community is to work on
> it within Fedora.
> Your earlier mail could definitely be taken to mean that we should go
> off and work on Flatpak elsewhere, and when we have a fully working
> ecosystem elsewhere, and have (on our own, without any engagement)
> fully met all the needs of Fedora users, then we can look at
> integrating that into Fedora. That sounds hard to distinguish from
> ignoring Fedora and starting something else. :-)
> devel mailing list -- devel(a)lists.fedoraproject.org
> To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org