> Of course creating first Maemo on top of GTK and after that switching to QT sounded weird and with Meego there was other parties like Samsung and Intel meddling around so they lost lot of steam in that transition too.

From my perspective one of the problems was that there wasn't just one major transition, but instead there were two. First was the GTK->Qt transition, and then just as that was stabilizing came new transition from Maemo to Meego which never really succeeded. I think the reason for the latters failure was that while there was a lot of reasons to be exited about Qt (over GTK), there was very little to be exited in Meego (over Maemo).