More from defections to other (more spam-averse) technologies. Remaining, if it can still compete with cable, as a data backhaul.
Slightly more fleshed out here:
Don't forget the PSTN, includes cell phones, SMS/MMS, and the entirety of the 911/999/112 systems too. It's not just a POTS line hooked up to a 2500 set anymore.
By "dead", I don't mean PSTN won't exist at all. But its status will be like Usenet or IRC: a system which technically still functions, but is a wasteland of deserted spaces and/or spammers and scammers, while the hub of discussion and commerce is found elsewhere.
There's a tremendous foment of alternative systems, though it's not clear if any one of these will emerge ascendant. It's quite possible that you'll see splits among major usage bases, particularly in business/commerce and/or personal messaging. The usurpation of direct realtime voice comms by texting for many (though certainly not all) uses is also a part of this.
A key point to remember is that once a large network with network effects enters into a death spiral -- positive feedback loop of decreasing value and increasing costs -- the end comes surprisingly quickly. And in both news and conversations I'm having with any number of people and organisations, dissatisfaction with the existing phone system is rampant.
The fact that what most people consider as "phones" are actually relatively general purpose computing, information, and comms devices, on which any arbitrary protocol can be layered, may actually be the kiss of death for PSTN. You won't need to buy a new device, only install a new app (or use the one that's been pushed to your device by the vendor).
For home and office users, the final merger of comms to the desktop (or laptop) system is likely to occur. Some mechanism for patching existing infrastructure to the new system(s) may work for older / hard-wired legacy installations.
What the replacement system offers, its specific capabilities, and how directories are handled, remain interesting questions. There should be both past examples of similar transitions (telegraph and TELEX to voice and computer messaging, particularly) which offer clues. We're probably staring at most of the pieces if not the complete system presently.
When you can't rely on the fact of any arbitrary business, organisation, and/or person having PSTN service through which they can reach you or you them, the end will be well in sight.
> As someone with a heavy background, it would break a fair bit of the PSTN
As it stands, I think quite a few consumers are ready to just walk away from PSTN unless something pretty drastic is done quite soon.
If they still made an iPod like device that ONLY included data, I'd switch to it today. I can get my calls via any of the various online services.
As it stands, my phone is in do not disturb mode so that only calls in my address book get through.
I previously thought of this as something old people do because they can't type on a phone, but I tried it, and it does provide a more rapid, conversational way of creating text.
I'm sure it will live on in many places that aren't concerned with customer satisfaction, though.
We really need federated communicator.
Federation has enormous downsides: since anyone can join the network and send messages, there aren't good ways to police bad actors.
The two largest federated communications networks today are email and PSTN. They're both infested with spammers.
There are many ideas to stop spam like proof of work.
Main reason for spam in email and PSTN is legacy protocols, which were created long time ago and cannot be significantly improved without breaking compatibility.
1. https://en.wikipedia.org/wiki/Public_switched_telephone_netw...
Now, most calls, the BTN/ANI and CID are the same number string, but when you start looking at call center operations, calls from PBX's, anything with backup trunking arrangements - that gets less and less true. Another wrinkle in this whole mess, is that with LNP (local number portability) you can have a number that on paper is owned by carrier Y (by assignment) but has actually been ported to carrier X (there is a database however that carries this information).
The issue as it comes down, is there is no central database to show that Caller ID string X is valid for trunk/line Y, for the most part customers were trusted to not be malicious and follow the rules. Once we got into a world where a SIP trunk could be had for free, and usage was a minor incremental charge, that whole system broke down.
The good news in all of this is that the FCC has mandated a move to authenticated CID (https://www.fcc.gov/call-authentication), but much of the equipment in play here is decades old, and will take time to be upgraded - but it's happening and happening pretty quickly for an industry that normally moves at a near glacial pace.