Mailinglist Archive: opensuse-project (252 mails)

< Previous Next >
Re: [opensuse-project] Re: Leap 42.1: Positioning / USP
On 17 July 2015 at 22:02, Jay <MyMailClone@xxxxxxxxxxx> wrote:

Yep. But "The Up-to-date Reliability Release" (URR) puts more emphasis on
being "current" which includes hardware-support. That's why for URR to work
the latest LTS-kernel would be essential.

Being more current would set Leap apart from Debian, Centos and other
LTS-"stability"-releases.

Richard Brown's idea of "Best of both worlds" puts more emphasis on
"community-innovations" - if I understand him correctly. And this might work
without the LTS-kernel.

First, I want to share some thoughts about the Kernel as this topic
keeps on coming up :)

While it looks certain that Leap will have the 4.1 LTS Kernel because
of the very valid reasons raised by everyone on these lists, I don't
think it's safe to assume that every minor release of Leap will always
jump to the latest LTS Kernel each time.

The current prediction is that we'll be doing a Leap release once per
year (though that is broadly dependant on SUSE doing the same for SLE
12 service packs, which we understand is the intention)

If you look at https://www.kernel.org/category/releases.html you'll
see that a new Kernel is declared as an LTS some where between once
every 6-13 months

So, while I think we COULD probably move kernel every year (if we're
lucky with LTS kernel announcements), I don't think we necessarily
SHOULD

I think we should make a sensible evaluation for each minor release
what makes sense for us, as maintainers and users.

Starting to build Leap with the 3.12 Kernel from 2013 in 2015 (and
needing to be maintained until at least the end of '2016' which is not
even guaranteed according to this published list) is clearly not the
right decision for us

But sticking with the 4.1 kernel until the end of 2017 might be the
right decision when we come around to making it 1 year from now.

And so, to the point at hand, what makes Leap unique, is exactly
factors like that

We're not building a CentOS where the logical choice is to stick on
the Kernel used in RHEL, no ifs, no buts

But, nor are we building an Fedora/old-style openSUSE where every
release will always automatically have the latest kernel without
considering whether the choice is a worthwhile effort in terms of
risk, stability & benefits to users.

That's the kind of thing I mean by the "best of both worlds", we can
make something that fills a unique role in the Linux ecosystem, a
distribution that provides both great stability, and great features.


But without that, I agree with jdd here - that if you specify just "The
best of both worlds", you might make people think that you were talking
about, for example, the Linux and Windows worlds coming together.

Jim

That's up to Richard to comment on.


But perhaps we should just say

"Leap 42.1 - The Best of ALL Worlds!" ;)

"The Best of Both Worlds" refers to the Enterprise-provided and
Community-provided sources which we build Leap

I don't see how that could possibly be construed to include Windows,
but if it did, then "The Best of All Worlds" is at least 33% worse as
it would also include OS X

;)
--
To unsubscribe, e-mail: opensuse-project+unsubscribe@xxxxxxxxxxxx
To contact the owner, email: opensuse-project+owner@xxxxxxxxxxxx

< Previous Next >
Follow Ups
References