On 19 November 2015 at 23:33, John Andersen <jsamyth@gmail.com> wrote:
On 11/19/2015 01:37 PM, Richard Brown wrote:
So SLES packages being 25% of Leap represents a pretty high percentage of the packages that could be available from SLES ;)
That fact alone is one of the more enlightening facts posted here.
SLES is a server product, most people running SLES don't even run a graphical user interface, or if they do, they just use what ever it comes with as a default.
Yet the link I posted does show a boat load of packages for KDE and Gnome, although its not clear if any of them are installed by default. (I haven't installed SLES since version 9.)
SLES includes and installs GNOME 3.10 by default. The KDE packages you see in the list you provided are base libraries, allowing SLES to run KDE applications - KDE is not included in SLES 12 In Leap, the openSUSE GNOME Team chose to use GNOME 3.16 instead of SLE's GNOME 3.10 - that probably accounts for a few hundred of the packages that Leap could have had from SLE but does not. And so, going back to the point of Leap and stability - a huge part, I'd argue the core part, of an operating systems 'stability' comes from it's base system But First what do I mean by 'stability'? I do not just mean 'does it work?' - that's RELIABILITY, and that's a trait shared by both Tumbleweed and Leap already, and validated every day by openQA I'd define that STABILITY includes reliability, but has the additional attributes of 'it doesn't change dramatically'. You implement it, it works (reliability) and it keeps working the same way until a specific time when an upgrade comes around - and even when that upgrade does come around, the transition is smooth. That's what I mean when I talk about STABILITY. Tumbleweed and Leap both provide reliability in spades (and I really do disagree with those people suggesting it's going to take a while for Leap to 'settle' down - it's settled, we wouldn't have released it if it wasn't ready, and while it isn't perfect, the amount of bugs reported so far are very low, of limited severity, and they're being fixed very quickly) Tumbleweed doesn't promise much in terms of my above definition of 'Stability' - it moves at the pace of contribution, that's fast, it's meant to be fast. In Tumbelweed, we make no promises about keeping things similar. If upstreams do major changes, we reflect it as soon as we can do so reliably. We do our best to make the transitions always smooth as possible, but Tumbleweed offers the latest and greatest in a reliable way, and if we provided more in terms of 'stability' in Tumbleweed, we wouldn't be offering the newest upstream developments to our users, so it would be pretty darn useless as a rolling release Leap on the otherhand, has it's mission of being the more STABLE (as defined in this post) counterpart to Tumbleweed. And with a shared Base System with SLE, we are starting with a Core that it Enterprise stable, the highest grade of stable, perfectly matching the definition I have above. Ontop of that, our community have done a wonderful job of picking and choosing from the whole FOSS ecosystem, the best versions of the packages available to accomplish Leaps goal of being the stable equivalent to Tumbleweed. GNOME 3.16 is a fine example of decisions being made to ensure that those software stacks in Leap 42.1 are tried, tested, proven, and should provide a stable starting point for Leap 42.1 to move on from. And so, while I understand the reluctance of those who haven't jumped onto Leap yet, I also think it is misguided. Leap 42.1 is a quantum shift (a Quantum Leap? [1]) from the previous openSUSE releases, and represent the first openSUSE releases which have actually set out with a clear goal to be the stable choice for its users. It's built properly, tested properly, maintained properly, compared to the previous approach which, despite our best efforts, was often a lot more happen-stance and impacted by external factors - We did the best we could, with the tools, talent, and upstream timings available to us..but doing a time-based regular release always offering the latest of everything each 8 months resulted in us being totally at the mercy of the upstream projects and the quality (of lack thereof) of whatever they pushed out, no matter how close it was to our release date. With the Leap model, we're much more empowered to be in control of our own destiny than we ever were. Even the parts inherited from SLE, openSUSE decides what to do use, or not. Which versions of an upstream stack to include? openSUSE decides which version, or not. Of course this still means working with and alongside upstream projects, but if their latest version isn't good enough, we don't need to use it. The mission is to provide a stable distribution to our users, and we no longer are beholden to that always including the latest of everything - Tumbleweed takes care of that for those users who want that. And so I'd argue anyone clinging to 13.1 or 13.2 should reconsider if it's really the best choice for them, or maybe now's the time to take the Leap, and start reaping the benefits of the work everyone has done to make Leap 42.1 as good as it is - It's only going to get better from here. [1] https://www.youtube.com/watch?v=DjK9GJMBpt0 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org