Symbian folks inside Nokia at the time lived in a reality distortion field. Your experience is not the only instance where the Symbian division inside the company railroaded the Maemo project, which they saw first and foremost as a threat to their jobs and political dominance. As TFA notes, N900 was preceded by N770, N800, and N810 "internet tablets". N810 in particular was basically the same device as N900, but despite having otherwise state of the art wireless communication tech all of them lacked one very crucial component: a mobile data modem. Internet connection was only possible over WiFi (or even RJ45-USB dongle) but not SIM.
The reason for this was of course not technical - the Symbian folks inside Nokia simply managed to convince the executive level management that adding mobile data would turn these devices from PDAs to a smartphones, and it would be very ill-advisable to launch a competing smartphone platform to Symbian.
When Maemo folks finally got a chance to "give us a try" (and I guess management saw with the competition that they needed something more modern), the result was N900. Considering it was meant as a "niche" device its sales exceeded all projections. However, in the grand scheme of things it was just an experiment designed to fail and further cement the dominance of Symbian.
I think Nokia also developed a fully-working tablet around either Maemo or MeeGo. It was literally ready to ship, but was at last minute cancelled before the launch event due to backroom dealing again by the Symbian folks.
It has all been documented in the book "Operation Elop", which paints a pretty good picture about everything that went on behind the scenes during the demise of the once great Nokia: https://asokan.org/operation-elop/ The book is based around absolutely exclusive interviews of people who served in Nokia's board and top executive positions at the time.
Maemo was the right platform at the right moment to compete with Android and iOS and I was very surprised they closed it down in the time of crisis instead of realising they had built exactly what was needed.
It had a modern OS with memory overcommit instead of the endless drudgery of checking out-of-memory conditions as in Symbian.
The N770 launched with a GUI based on Gtk and convincing OSS developers to move to Qt in later versions of the platform was bump in the road.
(I think Gtk had problems with basing the inheritance model on matching strings instead of vtables as with C++, so performance on this type of devices was problematic. Or - someone please correct me. I know it feels a bit weird given the long history of Gtk.)
It wasn't about performance; GTK (or rather a GTK-based framework called Hildon) continued to be the default toolkit all the way to Maemo 5 ("Fremantle"), which is what shipped on the N900. It was also what Intel used on Moblin, the OS that was supposed to merge with Maemo to become MeeGo.
The issue with QT was all about Nokia: QT was supposed to save the Symbian platform, and also provide an on-ramp to Maemo (at least in theory - in practice nobody was really committed to that...). So the Maemo folks were obligated to switch to QT, which they did fairly easily; the result was what shipped on the N9, effectively Maemo 6 ("Harmattan") rebranded as "MeeGo 1.2".
In short: QT was adopted to please the Symbian people, and it was all for nothing anyway. That suited me just fine when first announced (Hildon was an under-documented mess of C with obsolete and unusable Python bindings, whereas QT had first-class wrappers like PyQt), but many Maemo old-timers never really warmed to it.
The reason for this was of course not technical - the Symbian folks inside Nokia simply managed to convince the executive level management that adding mobile data would turn these devices from PDAs to a smartphones, and it would be very ill-advisable to launch a competing smartphone platform to Symbian.
When Maemo folks finally got a chance to "give us a try" (and I guess management saw with the competition that they needed something more modern), the result was N900. Considering it was meant as a "niche" device its sales exceeded all projections. However, in the grand scheme of things it was just an experiment designed to fail and further cement the dominance of Symbian.
I think Nokia also developed a fully-working tablet around either Maemo or MeeGo. It was literally ready to ship, but was at last minute cancelled before the launch event due to backroom dealing again by the Symbian folks.
It has all been documented in the book "Operation Elop", which paints a pretty good picture about everything that went on behind the scenes during the demise of the once great Nokia: https://asokan.org/operation-elop/ The book is based around absolutely exclusive interviews of people who served in Nokia's board and top executive positions at the time.