> Is anyone else kinda hoping that GPG/PGP loses enough respect in the tech community that something fresh comes along that really solves a lot of the UX and security issues they have?
This already exists. It's just not a single, all-in-one tool - the answer is to use a tool that's fit for the specific purpose you're trying to accomplish (secure messaging, encrypted backups, encrypting application data, etc.)
There is not, and never will be, a modern one-size-fits-all approach because the entire industry has moved on from that model - over the last 30 years, we (collectively) have learned that it's inherently insecure to design a tool to do so many disparate things and expect people to use it correctly each way.
All of these use cases are encrypting files, some of them with a few extra steps as sauce. Stuff like whatsapp / signal is the exact UX that GPG has, fixed by instead ignoring everything that's hard (trust). The asymmetric cryptography is not fundamentally novel or interesting, and the end result of it could be applied to literally anything if they allowed you to touch your own things (which they don't). These modern solutions are build on the infantilisation of their own users, nothing else.
That's exactly the wrong way to look at it. Everything is potentially a file, but not all cryptosystems have the same use cases. The needs of message encryption (forward and future secrecy, for instance) are not at all like the needs of backup encryption (deduplication, for instance). This is one of the biggest things wrong with the PGP model of retrofitting cryptography onto problems, and why it has virtually never been successful at any of them.
This already exists. It's just not a single, all-in-one tool - the answer is to use a tool that's fit for the specific purpose you're trying to accomplish (secure messaging, encrypted backups, encrypting application data, etc.)
There is not, and never will be, a modern one-size-fits-all approach because the entire industry has moved on from that model - over the last 30 years, we (collectively) have learned that it's inherently insecure to design a tool to do so many disparate things and expect people to use it correctly each way.