Fedora 25 Not Scheduling A Mass Rebuild Is Raising Some Concerns

[bug77]

Sorry you have bad experiences with whatever "hybrid agile" means but the method we have in Fedora works very well for Fedora for its needs. As simple as that.

And that's the issue here: you always have a roadmap, you never stick to it (because "release criteria is considered more important than a strict schedule") and when people tell you you should go one way or another, you always have the "it works for us" line ready.

I don't even care, I haven't used Fedora in 2-3 years. But the changes you'd have to do are so small, I cannot fathom why you are so staunchly refusing to do them. Just remove the public roadmap (make it available on request if other projects really need it) and publish the list of features and state they are in instead (if indeed features are more important than time, then it's more important to put the feature status upfront and not a timeline you know you won't keep anyway).

[RahulSundaram]

And that's the issue here: you always have a roadmap, you never stick to it (because "release criteria is considered more important than a strict schedule") and when people tell you you should go one way or another, you always have the "it works for us" line ready.

I don't even care, I haven't used Fedora in 2-3 years. But the changes you'd have to do are so small, I cannot fathom why you are so staunchly refusing to do them. Just remove the public roadmap (make it available on request if other projects really need it) and publish the list of features and state they are in instead (if indeed features are more important than time, then it's more important to put the feature status upfront and not a timeline you know you won't keep anyway).

If you have a serious proposal, take it up to the Fedora council. Posting here won't accomplish that.

You are the only person who has ever proposed hiding a schedule in an open source project and you haven't explained why you think it is so critical to hide the planned schedule. Yes, it does work for the project and that is the primary purpose of it. It is a massive change to hide it (not a small one like you think) because it affects internal and external collaboration like I already explained. You clearly haven't thought about the logistics of hiding something as important as a schedule when nearly all communication is via public mailing lists and bug trackers. Btw, you are still conflating release criteria with new features. They don't have anything to do with each other.

[bug77]

If you have a serious proposal, take it up to the Fedora council. Posting here won't accomplish that.

You are the only person who has ever proposed hiding a schedule in an open source project and you haven't explained why you think it is so critical to hide the planned schedule. Yes, it does work for the project and that is the primary purpose of it. It is a massive change to hide it (not a small one like you think) because it affects internal and external collaboration like I already explained. You clearly haven't thought about the logistics of hiding something as important as a schedule when nearly all communication is via public mailing lists and bug trackers. Btw, you are still conflating release criteria with new features. They don't have anything to do with each other.

Look, I really don't care. I have only told you what makes Fedora the oddity among Linux distros. If you don't care, there's no proposal/ticket/whatever I can come up with to change anything.

[RahulSundaram]

Look, I really don't care. I have only told you what makes Fedora the oddity among Linux distros. If you don't care, there's no proposal/ticket/whatever I can come up with to change anything.

Yes, Fedora has different goals from other distros and different scheduling process to meet its goals. There is *zero* relationship between my care (or lack thereof) and whatever proposal/ticket you may file because I am not involved with Fedora leadership teams anymore and I don't affect the outcome of it in any real way. Up to you.

[Guest]

No it doesn't, because we've already *done* the GCC 6 mass rebuild. In February. It will be in Fedora 24.

Great! Thank you for clearing this up, Adam

[lsatenstein]

If the GCC version 6 compiler has no serious bugs in the code generation part, then a massive recompile should produce the same logic in the executable. The executable will differ because of an updated date field that is embedded within the object(s)