Hacker News Reader

Comment by zackmorris | Hacker News Reader

It's because:

1) Caller id was conceptually flawed from the beginning so was implemented as an honor system instead of something required to communicate (like an IP address)

2) Caller id is unregulated, so there is no penalty for spoofing

My previous comment on it:

https://news.ycombinator.com/item?id=19303931

From the linked article:

Unlike phone numbers and domain name service (DNS), wherein there are internationally recognized databases that are authoritative sources, there is no central authority or regulation for caller ID. No FCC guidelines regulate carrier accuracy, and although federal regulations exist regarding telemarketers and spoofing, carriers can maintain their caller ID databases as they see fit.

https://www.onsip.com/blog/how-caller-id-works-why-it-might-...

View on HN

Exactly this. The last time I had a chance to talk to an FCC commissioner (they were in the Bay Area at an IEEE meeting) I strongly suggested that they could do a world of good by creating such a database, requiring anyone who wants to call into the US have their numbers registered in such a database, and to put 'DO NOT TRUST' on any number that wasn't in the data base as its default Caller ID.

Pair that with creating an access restriction to the US PSTN for any entity that originates a call with a spoofed number that is enforceable with technology rather than by voluntary or compelled action, and much of the problem will be mitigated.

That doesn't solve the problem where legitimate numbers are spoofed, does it? It seems my number is among the many that are being used by robocallers, because I sometimes get texts from strangers saying "who are you?", or "sorry, can't talk now".

That is part two, adding real teeth to spoofing a legitimate number. Basically on confirmation that a provider allowed such a call to be made on the US PSTN would result in all their numbers being blocked from the US PSTN until they could prove they had fixed the problem.

There's a system like that in the works now. Although it's not implemented by the FCC, the agency has threatened "regulatory intervention" against any providers who don't implement the SHAKEN and STIR protocols. https://arstechnica.com/information-technology/2019/03/att-a...

So why dont they display what serves the purpose of the IP address instead? They have to be able to route calls somehow

Originally the telephone network was directly switched, that means that it effectively connected your phone directly to the caller's. There was no need to route a return path because, well, it was as if you had a copper line going from your phone to the caller's. In such a setup the caller might not even need a functional phone number, just an entry point into the network.

It's very different from a packet switching network where data is cut in small chunks with a header meant to be routed across the graph (and in which case you need to be able to find your way back for the reply somehow).

Of course I expect that modern phone networks have evolved massively from the early days but I'm sure that this original design decision still makes what you're proposing difficult. There's a lot of historical baggage.

It used to be a felony to plug unauthorized equipment into a telephone line. You had to get your telephone from Ma Bell.

Caller Id came about in '68. It was built around an honor system because that's the environment that's how telephone lines used to be operated. They weren't able to predict the future.

"Have you ever received a call from a robot?
10 times per day?

You will. And the company that will bring it to you is AT&T."

It's sad how well that fits into their line of old commercials from the 1990's. And yes, I was inspired by this post from yesterday [0].

[0] - https://news.ycombinator.com/item?id=19504544

Isn't that called something like a "Carterphone Rule" ?

And that also came about in '68. The wheels of progress turn slowly.

The wheels of *politics turn slowly.

Reply on news.ycombinator.com