On 21 July 2017 at 10:16, David C. Rankin <drankinatty@suddenlinkmail.com> wrote:
On 07/21/2017 02:23 AM, Richard Brown wrote:
As a rolling release, Tumbleweed is developed for users who desire the latest upstream packages as soon as they're 'ready'. 'ready' is a defined in collaboration between upstreams (we typically choose upstream stable versions)
Here is where the unease comes in that may lend itself to the glorified beta moniker.
There needs to be clear definition on what the "latest upstream package" is. For Arch, that definition is crystal clear. "The latest upstream official 'Release' - period. Not a kinda/sorta "we talked with them about it and we think the beta they are working on is stable enough" stuff.
"Current with the latest upstream official release."
That is a philosophy that is hard to poke hole in for a rolling release to follow.
There is no reason future releases, not yet official, cannot be included in testing, but for roll out to TW, it would serve it well to stick to the latest upstream official release. Until upstream has the confidence in its own package out as an official release, it's hard to make an argument that it is good to pull something without that level of confidence into a production rolling release.
Just my .02...
The Arch philosophy does have one hole - there is no common standards from all those upstreams. What they stay is stable might not be stable. What they say is unstable, might be stable. For the interpretation of stability, we trust out maintainers to be able to make sensible judgement calls on the reality of their respective upstreams. Some of our maintainers occasionally go faster than their upstreams. Some go slower. Either way, openQA has proven itself to be sufficient to cover the fallout of those decisions in the vast majority of those cases, and I would argue that our maintainers are proven wrong no more often (and probably less often) than blindly trusting upstreams declarations of stability. The Arch way has it's flaws, and in order to correct them requires the blind reliance on upstreams to improve. Our way has it's flaws, but they are flaws which we are better positions to correct ourselves. I prefer that, personally. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org