MBone lost its reason to exist when the internet backbone turned out to be much easier to upgrade than the edges. There's not much need to implement a complicated proxy system to save bandwidth on the backbone when almost everybody is constrained by their last mile link.
For years I thought TV stations might connect to the MBone to do simulcasts for people on the Internet once broadband became widespread, but the world moved on before it could become a reality. Part of me still thinks this is a missed opportunity but it's too late to cry over it now.
My internship, in 2006, involved working on an IPTV prototype for Philips. I remember working on a multicast server and the set-top-box client -- the server had a DVB-T card, and rebroadcast all the BBC channels over the LAN. Since the stream was always there, changing channel was extremely fast. DVB broadcasts information packets regularly, containing the TV schedule and so on. These were also forwarded over IP, and then cached by the STB.
It was neat, but presumably most of the multicast stuff was abandoned not long after I left.
Someone else worked on a system to help the user schedule their TV around the broadcast times.
The problem with using multicast for content delivery is more down to the subscriber end and how ISPs manage their networks. I worked on a (recent) project that uses multicast in the broadcast context at the mezzanine level. When you've got full control of the network, end to end, it works a lot better.
The subscriber end...and the goddamn Internet backbone. You can't route multicast across the core, which is a huge impediment to its adoption.
There was a vicious circle where the core didn't do multicast so the big iron routers didn't put multicast support in hardware, which made it impossible to support multicast on the core...
For years I thought TV stations might connect to the MBone to do simulcasts for people on the Internet once broadband became widespread, but the world moved on before it could become a reality. Part of me still thinks this is a missed opportunity but it's too late to cry over it now.