On Tue, Nov 15, 2016 at 11:27 AM, Andreas Tunek <andreas.tunek@xxxxxxxxx> wrote: > 2016-11-15 17:12 GMT+01:00 Josh Boyer <jwboyer@xxxxxxxxxxxxxxxxx>: >> On Tue, Nov 15, 2016 at 11:06 AM, Andreas Tunek <andreas.tunek@xxxxxxxxx> wrote: >>> 2016-11-15 15:35 GMT+01:00 Stephen John Smoogen <smooge@xxxxxxxxx>: >>>> On 15 November 2016 at 01:35, Andreas Tunek <andreas.tunek@xxxxxxxxx> wrote: >>>>> 2016-11-15 1:06 GMT+01:00 Zbigniew Jędrzejewski-Szmek <zbyszek@xxxxxxxxx>: >>>> >>>>>> You seem to confuse something being in the release criteria, and >>>>>> something being possible. But that's secondary. If you would like this >>>>>> scenario to be supported, then commit to testing it before the next >>>>>> release (in advance, at least a week before the beta freeze). >>>>>> >>>>> >>>>> If something is in the release criteria I expect the feature to be >>>>> present in the release. If the feature in question has worked for ~20 >>>>> Fedora releases it is not my first priority to test, unless there is >>>>> specific communication otherwise, like specific "Mac install testing >>>>> days" (which to my knowledge there hasn't been). >>>> >>>> You seem to be under a misapprehension on how distributions are >>>> built.. even ones which are paid for. In any distribution, if >>>> something is important then the stakeholders (the people who think it >>>> is important) make sure that it is funded so that the resources are >>>> there for testing. That funding in a corporate distribution is in >>>> paying for equipment, staff and hours to get that item tested and >>>> developed. In a 'free' distribution, that funding is volunteering the >>>> time to make sure it is tested. [And even in paid distributions, when >>>> things aren't paid for and people assume that X was going to be >>>> there.. if X isn't working on release day it comes back to the >>>> stakeholders that they needed not to assume it would be.] >>>> >>>> >>> >>> I was under the (mis?)assumption that, due to being in the release >>> criteria, mac "support" was important for Fedora. It seems to have >>> been that way for 10 years. >> >> It is as important as any other platform that the community wants to support. >> >>> If that is not the case anymore it would be good if that would be >>> communicated in advance so that all users on mac hw could either >>> switch distros or gang together to make a remix or something. >> >> You are confusing Fedora with a company. There is no top-down >> communication on what is or is not supported. There is no hardware >> support list or hardware certification list. It is literally what >> people show up and test. >> > > Maybe the release criteria should reflect that then. Right now it says > that mac support is a criteria for release. That's exactly what some of us have been saying :). >>> As a user and very occasional contributor (mostly of bug reports) it >>> is very hard to know what is most important to test. AFAIK there >>> wasn't any "Installer test day" or anything similar. So I and a lot of >>> people probably tested other stuff*. >> >> As a user and contributor, it is important to test what YOU find >> important. If that is Macs, then test every aspect of Mac support >> (install, runtime, etc). If it doesn't work, report it. If it does >> and you want to move on to testing something else, do that. >> >> The cumulative efforts define what works and doesn't work in Fedora on >> any given release. >> > > As everyone else I have limited time and resources and lot of things a > care about in an OS. It is unfortunately impossible for me to test > everything. Which is why I said to test things you really care about. >>> If I would have paid for RHEL 8 and it did not work on hardware that >>> RHEL 7.3 supported without any communication from Red Hat I would get >>> pissed and most probably find another vendor. >> >> That's the primary difference between Fedora and RHEL. RHEL is made >> to support what people pay for and the Service Level Agreements >> dictate what is and isn't supported. Fedora is made by people that >> want to see something work, and that thing working is entirely >> contingent upon people showing up to make sure it does. >> > > Yes and the people who showed up and did the work has made a release > criteria. Do you want to change that? No, but they need to realize it's not a one-time thing. You don't do the work and then disappear assuming it's going to continue to work forever at the expense of someone else ensuring that. It requires sustained commitment. josh _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
