The RAID/LVM removal from Palimpsest followed a similar pattern (code deleted during a rewrite of an underlying library), but the developer was willing to work on reinstating it. Here we don't really know what Persch thinks — though damn the users is a tempting guess.
It's not a bug, it is the removal of a feature. Is a bug tracker really the most appropriate place to discuss in which direction a software is to go ?
'No' and a link to the relevant reason and workaround should be fine in that case but apparently that had already been discussed before (and a reason was given).
I don't know if removing it through so much commits is good practice but if the feature was to go it doesn't make a difference in the end, he sculpts his piece of the software any way he likes it as long as it's feeature oriented, right ?
Note: I have no advice on wether that feature should have been removed or not in the first place.
Also, inevitably some people in good faith will believe they find a bug when something doesn't work the way they expect it to behave. You can't expect everyone to follow all discussions about features getting cut, so a link to whatever discussion there was (if ever) would have been appropriate.
Do users post to the firefox bug tracker when features are removed like in the gnome case ? How is it handled ?
Apparently, this behaviour was considered a bug, if a very long-lived one. So a recent update 'fixed' that behaviour, although the justification is stronger there, and the workaround (SuperStop addon[2]) is actually useful, although still on a different keybind (Shift-Esc)
[1] https://bugzilla.mozilla.org/show_bug.cgi?id=825486
[2] https://addons.mozilla.org/en-US/firefox/addon/superstop/
The reason you have these discussions on the issue tracker is so when the next person comes along to file the issue, he/she now has the relevant context. Randomly guessing at which group to search and post to hardly helps in discovery. If you don't want discussion, disable the comments. But that's hardly conducive to good community teamwork.
And I'm really not aware of any project that only uses a "bug tracker." They almost all track issues, ranging anywhere from bugs to new feature requests. Maybe Gnome is different in this regard, but I don't think it's crazy for someone to believe otherwise. Redirecting conversation smacks of a "bugger off" mentality.
I might be naive about collaborative software development or the way gnome dev handle things but I think bugtrackers should be about bugs and technicalities, not feature approval or UX discussion if user feedback is wanted otherwise everything that a given user finds annoying would be considered a "bug" and drama would occurs.
This "talk" is more shocking in my opinion than the one that made the headline.
Yes, the original answer is "just" missing the link. As is also the answer to the why comment.
I asked why he wants to remove the feature. He said it’s loosely connected to other features that are buggy and he doesn’t see the point of the feature anyhow. Whether I had a justification for the feature to stay.
I replied and gave him two good reasons, one of personal merit, one of technical merit.
Then nothing else.
Then I discovered that it indeed had been removed, while I was following git snapshots. So I checked the NEWS and Changelog, ReleaseNotes and git. No mention. Could’ve been temporary or unintentional.
So I wrote the guy again to ask what’s up. No reply. Then he replied with a non-answer only pointing to a FAQ entry he made the same day with a distracting workaround.
So I checked out git locally to see where the code went.
It turned out among unrelated commits where the code was #if 0’ed out only to be removed a few commits later with "delete dead code" etc.
So I wrote him again to ask why he did this. No reply.
So I downgraded to the known-good old version and waited for release time. Same story repeats with other people. Apparently nothing personal because of my tone.
So. Conspiracy or justified puzzlement about this behaviour?
Then this trainwreck happened and people were asking questions.
So I thought it worthwhile to add what I knew. I don’t think I disclosed anything personal. It was strictly technical matters relevant to the bug at hand.
So why did you ban me with a false accusation?
If anything the "removing type-ahead search from nautilus" bug should’ve taught the lesson that an information campaign works better than outright censorship.
When that bug got attention Jon McCann came in, explained, and then made a full blog post[1] to expound the directions they were taking. I still didn’t like it but I was satisfied with the explanation.
Stop digging in and realize that just slightly amending your policies and interactions with users would fix everything.
"He hid the removal among very unrelated changes (gtk to gsettings switch) and spread the rmeoval across more than one commit. Such that you cannot easily revert it. I strongly assume this was done intentionally."