On Mon, 18 Sep 2017 22:09:08 +0200
Arnt Karlsen <arnt@???> wrote:

> On Mon, 18 Sep 2017 17:03:47 +0100, Antony wrote in message
> <201709181703.47988.Antony.Stone@???>:
>
> > On Monday 18 September 2017 at 15:54:16, Arnt Karlsen wrote:
> >
> > > On Mon, 18 Sep 2017 14:35:24 +0100, Antony wrote:
> > > > On Monday 18 September 2017 at 14:27:04, Arnt Karlsen wrote:
> > >
> > > ..and we still haven't proven it good enough for many of them.
> > > And that can _only_ happen on Devuan Jessie LTS' own merit.
> >
> > So, we focus on getting people to use Devuan Jessie - not Wheezy.
>
> ..correct, we disagree on whether or not a minimal Devuan Wheezy
> will be an useful tool.
>
> > > > I don't think there's any point in Devuan attempting to support
> > > > the outdated packages in Debian Wheezy when we are already
> > > > offering an escape route to those syadmins who've chosen to
> > > > stick with Wheezy because they don't want systemd.
> > >
> > > ..you missed my "_minimal_." ;o)
> > > I agree supporting outdated stuff is a waste of time and good
> > > manpower, it'll only be useful to help people update to Devuan
> > > Jessie etc.
> >
> > I don't see the purpose of having a Devuan Wheezy.
>
> ..then we disagree, I only see it as an useful stepping stone
> tool to help (convince) people upgrade to Devuan Jessie etc.
>
> > After all, Debian Wheezy is systemd-free by default (yes, you can
> > install it yourself if you want to, but people who've done that
> > aren't looking for Devuan), so what's the benefit in us advertising
> > a "systemd-free Wheezy" to the world?
>
> ..see above.
>
> > Anyone doing a review of it would say "this is Debian with some
> > optional packages taken out".
>
> ..see above.
>
> > > ..according to http://popcon.debian.org/ 30809 had
> > > popularity-contest 1.56 (wheezy) installed, about 11000 more have
> > > even older versions, and according to our own
> > > http://popcon.devuan.org/ and to http://popcon.debian.org/ we
> > > still don't match the 2693 who might still have 1.46 (lenny)
> > > installed.
> > >
> > > ..plenty good manpower out there we could use here. ;o)
> >
> > What do you expect that "manpower" to do?
>
> ..some of them will help build Devuan in the same way they
> have helped build Debian.
> How many, and who, depends on how we welcome them.
>
> > Not everyone running Debian as a sysadmin (or end user) is going to
> > contribute to the Devuan project.
>
> ..correct.
>
> > > > 2. Those who don't want to upgrade something that's still
> > > > working.
> > > >
> > > > Devuan Jessie is the solution for group 1, and there is no
> > > > solution for group 2 because they're happy with what they've
> > > > already got.
> > >
> > > ..once we have proven e.g. Devuan Jessie LTS etc on their own
> > > merits, they might get curious and wanna try our stuff and wind
> > > up helping out. Just give it time. ;o)
> >
> > 1. Anyone who's running Debian Squeeze or earlier isn't failing to
> > upgrade because of systemd, therefore Devuan isn't an argument in
> > their case.
>
> ..correct, they only need upgrades if they somehow are exposed to
> internet-like threats thru usb-sticks, old floppies, lan wiring,
> dying BIOS chips (BTDT), UEFI firmware with timeouts etc.
>
> > 2. Anyone running Debian Wheezy is either systemd-free, in which
> > case Devuan Wheezy would be identical, or else they've installed
> > systemd themselves, which presumably means they want to run it, and
> > therefore aren't interested in Devuan.
>
> ..you forget the kinda people who got stuck with systemd because they
> trusted Debian to be the good guys. BTDT.
>
> ..the only real difference between your vision of Debian Wheezy
> and Devuan Wheezy is the upgrade path away from systemd.
>
> ..the difference between your vision of Devuan Wheezy and mine,
> is I only want it to contain the bare minimum people might need
> to _safely_ move on upgrading to Devuan Jessie.
>
> ..that bit "_safely_" is what most people hesitating, may wanna
> see a track record on. And, either way, good or bad, we will
> build it.
>
> > 3. Anyone running Debian Wheezy who wants to upgrade to Jessie can
> > upgrade to Devuan Jessie to avoid systemd. If there are problems
> > with that upgrade process, *that*'s what we need help with debugging
> > and fixing.
>
> ..that too, yes.
>
> > 4. Anyone running Debian Jessie who wants to get away from systemd
> > can also upgrade to Devuan Jessie. Again, if there are problems
> > with that upgrade, that's where we need help, not with taking Debian
> > Wheezy, stripping out the optional systemd bits, and relabelling it
> > as Devuan.
>
> ..that's why I said "_minimal_", yank out "everything",
> keep only what people may need to survive upgrading
> from Debian Wheezy-and-earlier, to Devuan Jessie.
>
> ..the smaller it is, the easier it will be to support
> and maintain.
>

The problem is that wheezy is only supported until May next year and by
volunteers. You are suggesting that Devuan uses some of its precious
resources to support something that is probably going to be EOL by the
time Devuan has it fully up and running. This would take away time from
Jessie and Ascii. Which do you think Devuan should support, a dying
Debian version or the supported Devuan version and its replacement ?

I know which I would vote for, especially as you can upgrade from
wheezy to Devuan Jessie very easily.

Rowland