Get real.
One doesn't need extra data to know how it's impossible to boot all firmware below 1 second.

PS: ... I'm even using coreboot.
Do prove the rest of planet earth wrong.
I'd love to see the hardware you're using for that.

What I say is that from my experience VMs boot faster, or at least feel to boot faster, but I have not made actual benchmarks. What I not say is that VMs definitely have to boot faster, unlike your claims that a VM definitely has to boot slower.

I voted sysVinit, I mostly err on the side of caution. I had no idea systemd was born out of the REHL cauldron, etc, not that it really matters, cross platform is always a good thing. In all honesty, I really don't care what they use, as long as it keeps the system stable and secure...and not a nightmare to maintain. Just some things to consider.

I vote for Ludovic Courtès's dmd. Okay, really I vote for sysvinit kept around for those of us interested in running GNU/Hurd or GNU/kFreeBSD while they continue their systemd migration. Is support for systemd and upstart realistic, whatever conciliatory words come from TC people along those lines? I suppose even the sysvinit will fade in time and those of us looking for weird will have to part ways with Debian.

It's interesting coming back to debian (recently) to realize how the "Universal Operating System" must be taken to mean some particular thing different from what I would make up in my own head, something achievable and wanted by the people doing the work. For instance, tell me if I'm wrong, but there's no sensible series of apt-get commands that will give me a system where I can avoid having to learn about pam (having a umask issue and if I'm getting the bash and base files profile script comments' intent, the answer is to learn to configure pam_umask -- currently frustrated by this, but will read the linux pam sys admin guide soon). This is a side track on my road to get to the Hurd I'd rather not bother with. Or am I wrong? Would Debian GNU/Hurd also use pam?

If you ask 5 Debian users for the the definition of "Universal" you will end up with 6 different explanations. Funnily I wasn't able to even find one somewhat official definition which part of the OS should be universal, should it run on as many platforms as possible, should be be suitable for everyone, should software choice be as wide as possible, ... .

Regarding support for sysvinit/upstart/whatever, as I understand the postings on the mailing list support for other init systems should be available as long as there are developers creating scripts/unit-files/whatever for those init systems. How many people will actually do that, and to which extent, is something currently no one can answer.

It's an artifact from the time when Debian was still the official FSF distribution ... (meaning the GNU/Free_Software philosophy).

Section 1.5 - fourth point: https://www.debian.org/doc/manuals/d...c_defs.en.html

In fact it's a slogan which goes back to the 90's, but was never really ratified by the project. It has always been in the site description (browser title bar) as long as I can recall but only reappeared in the graphical banners at debian.org when Debian squeeze was released and the site was revamped.

The "universal OS" has no real official meaning as such but from reading the above link can probably be equated to the "multipurpose/multiplatform" nature of the distribution. The efforts to support as many architectures as possible and even alternative kernels also point to this.

Debian
1. Free means free, free stuff only depends on free stuff, all defaults must be free, users must have a choice if non-free is to be installed. That includes firmware!!
2. Stable - stable release should have 0 bugs or as little as possible, no vulnerability, no spyware, there must be someone to fix the bugs!!!
http://lkml.iu.edu/hypermail/linux/k...4.0/01331.html