Traditional phone companies absolutely should do a better job vs spoofing (and I believe there is supposed to be at least some progress there via STIR/SHAKEN in the US at least), but I'm a lot more inclined to blame services and particularly armed response services that still treat them as secure in a way they have long long LOOONNNGGG since proven they aren't. Telephone numbers simply were never ever a particularly secure thing, predate the entire net let alone widespread use of cryptographic protocols, and retrofitting real security onto that legacy has proven non-trivial. They shouldn't be blindly depended on for something like SWAT deployment. Security services live in a world where they can be lied to and they act like it in other circumstances. They're the ones sanctioned by the state to responsibly employ lethal force, and have been given ludicrous amounts of outright military grade gear. They should be investing in tools and techniques to evaluate reports and gather intel on site before "storming the house". Even basic old humint like "asking the neighbors if they heard any shots" seems to have been tossed aside. Like take this very reported case:
>The caller told the dispatcher he had killed his girlfriend. He had barricaded himself inside his home in a quiet, affluent neighborhood on the eastern edge of Palo Alto.
OK, so maybe there is a criminal there, but by his own (suspicious) words nobody is in any immediate danger. The supposed body he's killed isn't going anywhere. Surround the place from a distance, sure, and if someone does start shooting from the house obviously that's clear enough. But if the police have the place covered and nothing happens, maybe ask some neighbors. Send in a drone or two. Get on the line with a judge for a quick order to use IR or terahertz imaging or whatever and see if that reveals anything. Get a megaphone and ask everyone to "come out with their hands up" even! If the police though just go and storm the place though they could well cause the very situation they seek to avoid, what if someone is fast asleep and hears a break in and opens fire to defend themselves? Hardly an impossible thing particularly in America. It's happened.
Everyone in tech knows all about blindly trusting user input. Yes absolutely it'd be good to deal with trivial spoofing but that alone isn't going to mean there can't be malicious input. Or even non-"malicious" per se, there are people out there on bad trips or with untreated disorders who might report fake stuff fully believing it. The human mind is an imperfect thing. Stopping spoofing also won't stop someone from making a real call from a real phone that they've stolen or hacked into some business network and used their VoIP, calls from the victim's number could just as easily be calls from "a neighbor who heard shooting and saw flashes through the window" or "concerned passerby". E911 is supposed to have a variety of standards for including location information and in principle that should cut back on this as well, if someone claims to be in a house and the location doesn't show they're at the house that should be a red flag. But that still leaves holes due to legacy and risk aversion from bad incentives. At the end of the day, I think any service for the general public needs to be careful about considering their inputs. Ask for money upfront for that pizza for the first order. Trust but verify and all that.
Edit: One useful thing the government could perhaps provide here would be something along the lines of a national "known harassed number(s) registry", where someone could report getting this kind of thing at their local police station in person, show proof of identity, and ask for their number to be flagged for a year along with a code word. Any future 911 calls could automatically be flagged in turn, the operator could ask for the code word, or at least let responders know that there was a history of spoofing and everyone should be extra careful.
Yeah, I've suggested this very thing myself on several occasions, but basically people who are against engaging in self-defense and prefer to use police officers as their own personal security are dead set against any kind of reform on this point. You're really not going to get any kind of reform on this point until it's more than just little people who are being inconvenienced by this. Basically unless and until SWAT shoots someone "who matters" it won't change... because some people are REAL sure they need a militarized police force who can kick in your door at any time on the word of an anonymous jackass on the other end of an unsecured line.
I'd agree, were it not for my knee-jerk tendency to argue the contrary - the girlfriend may in fact not be dead, but only in shock. Also how do you know there isn't anyone in the house?
But probably those would be outlier situations, and the steps you suggest would be better than storming in.
What is more important is how society will act with said regulations/decisions, and then you'll have no choice but accept that the best way forward is the one that's hardest to abuse.
The aggressive approach is extremely easy to abuse and there are currently no dangers associated to the abuser, which is why SWATing is on the rise.