On 10-12-2013 19:33, Michael K. Johnson wrote:
On Mon, Nov 18, 2013 at 08:40:38AM +0100, Mark Trompell wrote:
So we're down to the crossroad I feel, where we have to quickly decide
where we are heading. Do we want do revive foresight and jump on the
fl3 train or do we want to go on with our lifes and let foresight die.
I've been silently pondering this thread for a while. I apologize for
being so quiet, but I didn't want to be too hasty.
I still think that my old proposal (as Rune referenced) is a general
good start.
I'd suggest taking bite-size pieces would make it easier to get to
something that works:
1. Import Fedora 20 (it is just about to be released) using mirrorball.
Just do a binary import to start with. We might be able to start
small by starting with one of the smaller "spins" and then add
more packages as we go, in order to make the pieces more bite-sized.
Thanks for chipping in!
Is it feasible to take just a few groups, like the "Minimal Install"
group, to begin with? Because that looks reasonably self-sufficient to
me? If so, then once we get the base working, we could also import Xorg
as Mark T suggests?
Or are you referring to a proper spin, anaconda and all?
Also, would this happen on SAS-sponsored infra or locally?
2. Optionally, developers who would like to explore rebuilding from
SRPMs with mirrorball could start a track doing that work, bootstrapping
from the binary import, in a different repository.
Again, would this happen on SAS-sponsored infra or locally?
3. Building Foresight as a conary-native layer that can run on top of
either the binary import or the bootstrapped source build, that
has only things different from the underlying platform, would let
Foresight return to its roots.
Its roots as in ... what? From where I sit, the world has moved on. I
think the only value proposition we are realistically in a position to
care about at this time is delivering a developer-friendly Linux
distribution by means of Conary.
When it comes to UX ('because your desktop should be cool!') my guess is
that we'll have our hands full with just keeping the binary imports and
the SRPM rebuilds running smoothly, making this the primary focus for
the next 3-6 months or so. Especially so now that António appears to
have become engrossed in his family project.
What comes after (if indeed there *is* an after) is anyone's guess.
4. Import newer versions of Fedora as they are released, and as they
are ready, move the Foresight layer on top of them.
Here's my problem: I know I will personally have limited time
available to participate in this process. So I would suggest doing
this only in concert with developers now at SAS who were previously
at rPath. In order to make this proposal, I'd need to know who of
the Foresight developers would be willing to run the import process,
noting and investigating failures in the import process, and working
through updates and fixes with those developers. To propose this
internally, I'd need to know who was volunteering to work on the
import process.
So: If you care enough about this that you'd be willing to help
drive the import process and help isolate bugs in the import process
so that we can build on top of an existing base, please respond,
either to the list or privately to me, so that I can propose this
internally at SAS.
If it means that I get to have Fedora + Conary, then yes, I'd be willing
to help.
Just being very clear: this is expressing my own ideas, and does
not represent any views of my employer. But if we (Foresight)
have enough commitment to working on this, I'm willing to propose
it internally (SAS) and then follow up with the volunteers regarding
the result.
Realistically, how many volunteers do you feel we need for this to
become a success?
For my own part, I have a 16GB quad core box sitting right next to me
waiting to get started. However, I don't quite know where/how to
begin. Is the old rPath documentation available anywhere these days?
Or am I asking the wrong question in this regard?
/Rune
_______________________________________________
Foresight-devel mailing list
[email protected]
https://lists.foresightlinux.org/mailman/listinfo/foresight-devel
