While it's a day-old topic, if anyone's still following, this is my reason too.
I have a significant interest in using KDE. My two biggest apps (overall, not just KDE) are Dolphin and Kate.
Dolphin's reseatable tree and files in the tree are features I find it hard to do without. So if KDE didn't actively push me away I'd use it.
However, the idea that you can force a new paradigm on an existing userbase (and by force I mean those who don't switch don't get bugfixes, security patches, new features and, ultimately, end up with incompatibilities with new versions of their OS) is one I oppose. I see it as similar to bait & switch. Also, just because you work for free and/or on free software doesn't mean your work can't end up being a net negative, ultimately costing users more.
Many of the time-wasting negatives from these tools are often overlooked by developers. For example, if my machine becomes less responsive I now have more things I need to check. Did a recent update accidentally re-enable Nepomuk? I've also spent a bit of time searching trying to find why "Desktop Search Services are Active" is being displayed even when it is inactive - a bug that I wouldn't have had to worry about if I didn't have it installed. Further, people now have to remember when installing KDE apps on other DEs to disable Nepomuk. Many may not even know they've just installed it and will not know if Nepomuk reduces their system performance.
Now, don't get me wrong, I have no problem with providing users with the option of a semantic desktop. But the right way to do this is to make things modular. Akonadi, Nepomuk etc should be removable packages, and apps that can use them shouldn't depend on them. They should have options enabling or disabling connections to these services.
No apps that predate these tools should require them. If an old app now has hard dependencies on these services you've done it wrong.
KDE needs to, sooner rather than later, set up a team to remove the concrete dependencies on these components. I know that's a lot of rework and unenjoyable reversal but IMO it's the only way to save KDE.
One final thing to mention: Even if more people liked the change (which isn't even true here AFAIK), a massive negative for a small number of people can outweigh a minor positive for the majority.