Mailinglist Archive: opensuse-project (422 mails)

[Per Jessen <per@xxxxxxxxxxxx>]

Greg KH wrote:

> On Thu, Dec 16, 2010 at 04:50:41PM +0100, Per Jessen wrote:
> > To me there is not enough new material in 11.4 to make it a major
> > release, so 11.4
>
> Well, the original message stated that there had to be:
> - drastic changes in user experience during installation or the
> way linux works
> - drastic changes in the base system that make it much harder
> than usual to do live updates.
>
> But there's probably not going to be a "drastic change to the way
> Linux works" in any near year timeframe ever.  Linux is all about
> constant little improvements all the time.
>
> If you go back and look at a distro from 5 years ago, yeah, it looks
> hugely different, faster, and nicer.  But we aren't doing 5 year
> releases, we are doing them in months.
>
> So I don't think these requirements are _ever_ going to be met in the
> next 3 years at the very least for openSUSE.

If openSUSE, like Linux, is "all about constant little improvements all
the time", you're absolutely right.  In which case we should just
continue with 11.4-5-6-7-8-9 etcetera.  (this sounds a lot more like
Tumbleweed though?)

Otherwise, if we desire to have major and minor releases, we should
gather significant improvements/changes into one release, which will
then make it major. 

> So, that means we stick with the 11.X series for a long time?  Or we
> should redefine what the rules should be :)

I have a distinct feeling of deja vu - didn't we have this version
numbering discussion a few months back?

> However, we do have things in 11.4 that seem much "larger" than
> normal: 
>
> - systemd, a major way the boot process works, speeding things 
> up massively
> - all wireless devices supported by open drivers
> - major 3d open driver advancements
> - large KDE advancements from previous releases
> - Tumbleweed providing "rolling" updates
> - possible MeeGo(tm) "spin" for netbooks

Putting on my reality/functionality-filtering glasses :-) , that
becomes:

- systemd, a major way the boot process works, potentially speeding
up things massively 

possibly plus:
- major 3d open driver advancements
- large KDE advancements from previous releases

I'm classifying those as "possible" because I don't know if they are
significant enough to be noticed by the end-user?

> p.s. This is why I feel the whole major.minor numbering scheme for
> software is broken, and just use 1 number for projects that I was/am
> in charge of naming (udev, usbutils, etc.)  I think it's worked out
> much better that way over the long-term.

A working major.minor versioning scheme requires a purpose and active
management. 


-- 
Per Jessen, Zürich (-4.3°C)

--
To unsubscribe, e-mail: opensuse-project+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse-project+help@xxxxxxxxxxxx