SDR, FPGA, microcontrollers, and even surface mount are sometimes viewed as suspicious newfangled magic.
Look at any amateur radio meetup or organisation and it's almost entirely 60yo white dudes. It has the lowest diversity of pretty much any social institution anywhere, in terms of age, gender or whatever.
The single biggest thing that "happened" to amateur radio, basically, is the Internet.
Young(er) people interested in radio today are typically not interested in "chewing the rag" as their motivation - we have constant access to Twitter, forums, and other Internet services that perform that function.
The main areas of interest, which need to realistically be the focus areas if the hobby is going to evolve instead of going the way of the apatosaurus, are machine-to-machine communications which aren't oriented around the idea of a person talking on the end of the link.
Machines that talk to machines, telemetry, wireless sensor networks, IoT, OSI-model networks generally, software-defined radio, microcontrollers and programmable logic, open-source instruments and tools built around these modern technologies, radar, radio astronomy, satellite communications and remote sensing are probably some of the core areas that people are interested in with regards to RF today.
The concept of a "shack" or station where a human operator sits and keeps all their gear is anachronistic, too.
Emergency communications? Government can communicate with the public via SMS, TV, broadcast radio, etc.
For very fast, agile updates of what's happening in the world, check Twitter.
If all cellphone and internet services are completely knocked out by some catastrophic event, what would that event look like?
And what role, realistically, would ARES play?
Maybe modern EMCOM needs to take a different approach - IP networks, Wi-Fi WANs and femtocells that can rapidly be deployed to restore connectivity that the general public is familiar with, using the devices that they already have.