I respectfully acquiesce that making the kernel better is indeed the more elegant and sound goal, engineering-wise. But I really think that it's just not *realistic* to expect users to wait for kernel developers to fix the very complex and demanding issue of battery life before they can have a Fedora that works with good battery life out of the box.
Users simply don't care. If Fedora depletes what used to be an 11-hour battery on Windows/OS X in 6 hours instead, they'll simply switch back to their former OS within a week. I have years of experience with this scenario and I strongly think you should take my word for it.
Maybe it doesn't have to be TLP, but until the kernel becomes better at power management, there has to be *something* that quenches the undeniable thirst that users have for Fedora's battery life simply being competitive with Windows, OS X, and even Ubuntu, which is also slightly superior in that regard.
Once the kernel indeed improves to solve this problem, then great -- we can remove the short-term solution that Fedora adopted and stick to what the kernel provides. But until then, you have to understand that leaving users with comparatively very under-performing battery life, for reasons they don't care about or understand, means that Fedora will not be able to compete and win more hearts as the free desktop operating system we all love.