As much as Feeding the troll is bad, I think it brings to light a lot of the public criticism towards GNOME; and I believe this is a problem of communication rather than you guys actually being evil developers.
The problem is, all the public hears about GNOME is the project removing things. Nautilus 3.28 removed Desktop Icons, and then there's this blog post about "Themes are bad". In both cases, the intent wasn't to say "we're removing things because we don't like them", but rather "this thing is bugged, awkward to work with, probably has a bunch of spaghetti code accumulated over the years, and there is a way to make it better - but that means first bringing it down."
Simply saying "System tray is bad, we don't want anybody to use it" won't make anybody change their minds - especially when looking at the GNOME project from some distance, where it fits in the Linux ecosystem along with KDE, XFCE and
other Desktop Environments that will gladly say "hey we support the system tray!". Developers not targetting the GNOME ecosystem (especially with technologies like Electron which allow easily crafting cross-platform applications, mainly made by macOS users, and distributed to all platforms because it's an option, and that comes with them not caring about the existence of a GNOME HIG.
The fact is, all recent headlines about GNOME have been about "GNOME planning on removing X" or "GNOME's new version won't have Y". Which you might argue isn't even true, as the System Tray was merely replaced by the AppIndicator support (although as an extension), the Desktop Icons were never planned to be removed, just their feature set moved from Nautilus to its proper module (although the Desktop Icons replacement is still in beta and last time I tried would crash the Shell when the screen turns off, or any other extension updates). And the themes aren't even going
to go away yet, this is just somebody's opinion on the current state of customizability of the GTK look and feel.
If I can go back to one point of the blog article though, it has to be the part about "Developers are the bloodline of the project". That is definitely true, but I believe the analogy conveniently leaves the users out of the picture. Rather the better analogy would be supply and demand: Developers absolutely are the ones who make the project possible and need to have an environment in which it is possible to work at their most efficient. But without users, all the effort, money, time, sweat, time debugging..., is in vain. If no users are going to use GNOME, no developer will want to work in it for too long.
I also believe that most Linux users came here for the customization if not for the tinkering, and I do push forward the ability of Linux to fit itself to the users' needs and desires when I am out c
onverting friends and family to Linux.
GNOME does have a visual identity of its own, regardless of theme. The top bar with centered date, activities overview, clutter-less design and integrated set of applications very much defines GNOME, and not that it uses a particular style of icons or a white, round relief design theme around it.
But removing customizability because it "would look more unified" isn't it. If you remove customizability, you lose all the people who don't like the Adwaita theme (and whether we like it or not, there are a lot), you lose all the people who are here because they feel at home. GNOME's ease of use will not outweigh this hypothetical rigid black-box configuration model.
I know that with every setting comes a cost and an increased configuration space that needs to be tested, and I understand the idea of cutting back on customizability in order to promote stability and ease of use (by it being harder to
break, in this case). But those are features that Linux users are used to having; those eye-candy features are what attract new users to Linux in the first place.
Now I do understand Sam's argument in his blog post - but that doesn't mean that GTK themes shouldn't be avoided altogether. With theming being such a popular feature I believe something should be done to enable theming in its proper sense, and not just drop it altogether.
Because that will be a big blow to an already worsening reputation of GNOME in the eyes of the public.
