3 ms·

In my experience even many decades ago in the US, the "ring" that the caller hears in their handset before the recipient picks up is not actually from the "bell

In my experience even many decades ago in the US, the "ring" that the caller hears in their handset before the recipient picks up is not actually from the "bell" of the recipient's handset; rather, I understand, it is generated at the caller's local switching station. And there's no system-wide synchronization of these ring signals. So, if the recipient hears N actual rings, the caller may have heard N-1, N, or N+1. Does anybody who used these signaling schemes have experiences where the wrong message (or none at all) got communicated?

inetknght 6y ago
That's been my experience too. It "works" for local calls but it completely breaks down across a nation.
toast0 6y ago
It can be generated more or less anywhere along the path. I don't think it was typically generated at the caller's local switching station, because often there's some delay before it starts, and I didn't think that level of information was communicated. However, sometimes you would get some ringing and before a final signal tone (busy, fast busy, call cannot be completed as dialed, etc). The phone system has almost always been a heterogeneous distributed system with lots of different capabilities though, so lots of possibilities. I have a hard time listening to podcasts or audiobooks, but I like to try to listen to Evan Doorbell's Telephone Tapes, a bunch of old (mostly 1970s) recordings of the phone system with recent commentary from the person who made most of the recordings. http://www.evan-doorbell.com/production/group1.htm http://www.evan-doorbell.com/production/group1.htm
unishark 6y ago
In the US, I recall the ringing sound for the caller being different depending on who you were calling. I understood it to be generate by the remote station, powering their ringer via the line. Presumably you could still get different ring counts in such a setup if one side hasn't been connected to the call yet.
HeyLaughingBoy 5y ago
Correct. The remote ring signal (90VAC @ 20Hz IIRC) that causes their phone's bell to ring is generated by their central office. However, I believe (it's been a long time since I did anything phone related) that the "ringback" that the caller hears is only generated when the remote switch verifies that it can complete the call, so it's probably synchronized that way.

inetknght 6y ago

That's been my experience too. It "works" for local calls but it completely breaks down across a nation.

toast0 6y ago

It can be generated more or less anywhere along the path. I don't think it was typically generated at the caller's local switching station, because often there's some delay before it starts, and I didn't think that level of information was communicated. However, sometimes you would get some ringing and before a final signal tone (busy, fast busy, call cannot be completed as dialed, etc). The phone system has almost always been a heterogeneous distributed system with lots of different capabilities though, so lots of possibilities. I have a hard time listening to podcasts or audiobooks, but I like to try to listen to Evan Doorbell's Telephone Tapes, a bunch of old (mostly 1970s) recordings of the phone system with recent commentary from the person who made most of the recordings. http://www.evan-doorbell.com/production/group1.htm http://www.evan-doorbell.com/production/group1.htm

unishark 6y ago

In the US, I recall the ringing sound for the caller being different depending on who you were calling. I understood it to be generate by the remote station, powering their ringer via the line. Presumably you could still get different ring counts in such a setup if one side hasn't been connected to the call yet.

HeyLaughingBoy 5y ago

Correct. The remote ring signal (90VAC @ 20Hz IIRC) that causes their phone's bell to ring is generated by their central office. However, I believe (it's been a long time since I did anything phone related) that the "ringback" that the caller hears is only generated when the remote switch verifies that it can complete the call, so it's probably synchronized that way.