So why has the industry accepted IMS so suddenly?

Well, it's primarily due to the fact that IMS has been adopted by the VoLTE initiative as the call session engine for voice in LTE networks.

While early LTE deployments are more about bandwidth and data connectivity, voice will continue to be a key service within the service provider domain. After all, people won't stop talking to each other. So, you need a robust architecture that will support millions of concurrent voice and multi media sessions. With consistent quality, all the time. But, of course, that leaves a big question unanswered: it may be voice that has finally cemented the position of IMS within the industry (although the lack of credible alternatives has something to do with it as well), but can voice continue to be a revenue generator for carriers?

Since voice revenues continue to decline, that's a very big question.