Fedora Ring 0 definition

[Brendan Conoboy]

On 09/15/2015 07:26 AM, Colin Walters wrote:
> 'On Mon, Sep 14, 2015, at 05:12 PM, Brendan Conoboy wrote:
>>
>> I'm just one person with an opinion, it would be best if everybody
>> with a stake took part in the ring definitions.  Creating additional
>> rings that address communities where self-hosting is a foreign concept
>> may be useful and desirable.  Making Fedora a first class OS for
>> languages where rpm packaging doesn't make sense is great!
>
> One thing I find strange is that while by some measurements
> the rings effort would be a major change, by others it seems to
> be a minor tweak of what exists today.
>
> I haven't seen for example any evaluation or discussion of
> the apparent assumption that Ring 0 will be binary RPM packages,
> maintained how they always have been.
>
> I haven't seen much discussion of "should ring 0 be RPMs".

We talked about a related question at flock: Should packages built as 
COPRs be allowed into low level rings?  The answer from RCM was no, 
due to trust and stability issues.  I think we're assuming ring 0 is 
RPMs because we don't have a second package format that we deeply 
understand and think is suitable.

> To give a random contrast, look at OpenEmbedded:
> http://www.openembedded.org/wiki/Main_Page
>
> As far as being a flexible base layer that is *explicitly* not
> itself a Product, they do this *really* well.
>
> One thing I like beyond the technology is how they have one
> git repository for the core, then explicit "layers" which are also
> git repositories.  These aggregate maintenance of *multiple*
> components and create a very *collaborative* model.  This is not
> generally true in the "big bag of packages" model since the
> core/extras merge.
>
> One small thing we could do to try to emulate this for ring0
> would be to put all of the spec files for Ring 0 into one git
> repository for example.  And have actual peer review
> for patches, just like one sees on:
> http://lists.openembedded.org/pipermail/openembedded-core/2015-September/thread.html

It's certainly an argument for ring 0 being the minimal install ;-) 
How do you deliver updates?

-- 
Brendan Conoboy / RHEL Development Coordinator / Red Hat, Inc.