The irony is that Symbian + Series60 did utterly dominate in many important markets [1]. This gave Nokia a false hubris and made them blind to the deep flaws of their UI framework and their software design process in general.
[1] In 2007, Symbian had 80%+ marketshare in smartphones in all markets other than USA, China and Japan.
The mistake there was to assume smartphone is a separate category of cellphone, when really it's not. (This is the same trick used to suggest Windows Mobile was ever remotely successful, by pretending it was a category of one). By that measure Symbian only succeeded in Japan where it was free of Series 60 and had whole different UIs stuck on it.
What Nokia achieved with S60 was a UI no one wanted, and unnecessary fragementation and antagonism with the other Symbian licensees (and with Symbian itself). UIQ predictably never went anywhere (too complicated), but had Symbian produced a UI closer to S40 in spirit then we'd be looking at a very different situation today. To this day iOS is closer to S40 than S60 for end users, and this is no coincidence.
It's a real shame because Psion, the company who designed the original version of Symbian OS (EPOC32), had a knack for sensible, simple UIs on their pocket organizers -- or whatever those '90s PDA-like things were called.
If that original Psion spirit had been carried to a smartphone design ten years ago, it could have been a great foundation for Symbian. Instead Symbian got a menagerie of confused UIs developed by the phone vendors whose primary aim was to tick more boxes in feature sheets every year.
[1] In 2007, Symbian had 80%+ marketshare in smartphones in all markets other than USA, China and Japan.