To play Devil's Advocate, choosing a DE makes no statement about the capabilities of the designer. My company chooses to pay me for other work I do that results in profits, and pay Red Hat for the support for DE's. I've been coding a very long time, and if I thought that coding my own DE and supporting it forevermore was the way to go, I would do it.
And while I know you work on Fedora and aren't personally responsible for all RHEL / Fedora issues, you need to understand that some of us are your RHEL / RHEV customers. When we're talking about silly things not working -- such as the System Monitor in Fedora and RHEL not showing all 16 cores (it's way too wide for the screen and can't be made narrower), which should have been vetted in Fedora before shipping in RHEL -- we can become frustrated. We were told by Red Hat that Activity Monitor was broken and they were aware of the issue, and we should use KSysGuard with all its KDE dependencies instead. The point is, sometimes a usable value isn't set as default, and common configurations don't work as expected. That's GNOME 3. The key to using a system is to provide reliable features with little surprise and even less irritation wherever possible. When you're buying expensive licenses you expect the built-in tools to work as expected, barring inevitable bugs. When a bug is brought up through the normal support channels and after an unsatisfactory support response you include your VAR, assigned Red Hat sales staff, as well as Red Hat technical support, you expect real answers.
GUI KVM settings don't save if you use the command-line kvm tools? Well then, don't use the GUI, we were told. It will be fixed soon.
Windows 2008 R2 timing settings result in CPU spikes on Nehalem, while idling, under RHEV 3? *No one* at Red Hat support -- and I had more than 4 reps involved with that one over a couple of months -- could solve it. I solved it and reported the solution so it could be incorporated into a bugfix.
And don't get me started on what we were promised with RHEV 3 vs what was delivered either, or the fact that I was told our problems would be fixed "in a couple months" all the way through October, then told it was ready but there was no upgrade path yet, and finally when there's a semi- sort-of upgrade path, it's too risky to justify. Companies like the one I work for don't like the risk and it was embarrassing to be kept waiting when Red Hat sales said a bugfix was imminent.