Re: Fedora Lifecycles: imagine longer-term possibilities

[John Florian <jflorian@xxxxxxxxxxxxx>]

On 11/14/18 7:54 PM, mcatanzaro@xxxxxxxxx wrote:

On Wed, Nov 14, 2018 at 4:42 PM, John Florian <jflorian@xxxxxxxxxxxxx> wrote:

I still don't understand what makes updating these for a *new* release significantly easier than an *existing* one.&nbsp; So let's just say GNOME (or whatever) comes out next month with a new major release we want to showcase.&nbsp; Why is it necessary to have a Fedora 30 to be able to realize this update.&nbsp; What is so difficult about providing this for Fedora 29.&nbsp; I'm trying to understand why these upstream updates can't be decoupled from the Fedora release schedule.

It's all a matter of QA. The freeze, the blocker bug process, and the quantity of users who test the stuff for us. We'd need major changes to our updates process to account for this in a mid-release update. The blocker bugs process would be needed, for a single bodhi update. At leas t a month or two of testing (during which new versions of the update will be released, so the update will have to go through some iterations). And lots and lots of testers: currently we get those for free because tons of people help us test beta releases of Fedora, I think far more than run updates-testing.

I think if we did this right, however it looks, multiple testing repos, rings, modularity, whatever... we might easily attract more testers than we have now.  I think this whole problem can usually be distilled down to, "I want LTS for everything because I hate breakage and I hate tech treadmills because I've already got too much to do.  Except for Foo, the version every other distro has is too old and I'm willing to get dirty if necessary because Foo is what matters to me."

This is all doable and solvable. Not a blocker. But if we don't take it seriously and make some big changes in how we release updates, it won't go well. Not well at all. So I'd recommend against it, unless there is some major benefit available from doing so.

I totally agree, but we are talking about radical changes here and I think we should keep all options on the table.  If some particular path forward is overwhelmingly desirable, that is the time to decide if the push is worth it, not earlier IMHO.  If the proposal, whatever it be, is great and everyone  agrees its great, the seriousness will be automatic.  Fedora has a long history of catering to some niche ideals that parts of our community are dead against.  It's awesome that Fedora is so flexible, but if we're going to fiddle with the release model, lets find something we *all* get behind and be happier with for the next 15 years, however radical it might look like right now.

_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx