Slashdot Mirror
FCC Preparing Transition To VoIP Telephone Network
mantis2009 writes "The US Federal Communications Commission (FCC) published a request for public comment (PDF) on an upcoming transition from the decades-old circuit-based Public Switched Telephone Network to a new system run entirely with Voice over Internet Protocol (VoIP) technology. This is perhaps the most serious indication to date that the legacy telephone system will, in the near future, reach the end of its life. This public commenting phase represents a very early stage in what will undoubtedly be a very complex transition that makes this year's bumpy switch from analog to digital television look relatively easy."
The death of dial-up has been greatly exaggerated. No broadband available where I am in NY, within 50 miles of Syracuse.
Dave
POTS is and has been stable and secure.
VOIP... not and never will be.
...is that the user terminal (the phone) is totally passive - no power needed, it's a totally dumb terminal, and very robust (at least, if it's a Western Electric product!). The POTS system is the result of some careful design and decades of improvements to increase reliability. That's not to say that there aren't benefits to be had from VOIP, just that we should think carefully before deciding that everyone will be converted to VOIP.
Disclaimer: In addition to my nifty 2.4G multiple handset cordless phones with built-in caller ID and voicemail, I have two POTS phones which work fine when the power goes out.
VoIP, while an interesting and disruptive technology, is not quite ready for ALL voice applications. Some thoughts;
It is frequently easy to tell when you are speaking to someone using VoIP. Clipped high and low tones, often choppy like a bad cell call. Most businesses will not want their customers having that experience talking to them. Residential is fine - those customers are just looking for cheap, cheap, cheap. Many businesses are concerned with appearances, and a bad call experience can sour a sale in a competitive marketplace.
Many (most?) alarm companies cannot successfully run alarms (fire, elevator, burglar) over VoIP lines. Not sure if it's latency, compression or what, but I have heard this complaint MANY times from various security (alarm) company people. In some states, doing so is actually against the law.
911 routinng - have all the 911 PSAP routing issues been resolved with VoIP? This is a biggie that most people switching to VoIP don't consider.
Your Internet connection goes down, your voice is gone. One thing you can say about the PSTN is that it is pretty dependable. In all my years (I have some gray hair) it has been rare that I have trouble with a POTS line.
VoIP has its uses - I'm not denying that. But the landline network will not disappear overnight, this year, or even this decade.
Indeed--- as a result, even the poster children of truck shipping, UPS/FedEx, have moved much of their cross-country shipping to rail. If you order something FedEx to Texas from the Northeast, for example, chances are it'll make a stop in Hutchins, Texas.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
My compa y has VOIP an it see t have pro le wit cu out.
When Fascism comes to America, it will call itself Anti-Fascism, and tell you to give up your guns.
POTS works over low voltage DC. As I recall, it's somewhere in the vicinity of 48 volts, but don't quote me on that. It's entirely feasible to have a cheap, dedicated VOIP chip that runs on 48 volts and draws perhaps 50 to 100 miliamps of current - well within the normal range of today's POTS power draw.
VOIP doesn't have to be VOInternet. They coul just as easily have a dedicated IP network for telephony, then run something like PPPOÈ or VPN to gateway to the public Internet and do away with separate SL MODEMs.
You'd still probably need a long distance plan, even though the point of one is technically idiotic.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
It doesn't specify that the IP based service has to start in home. As far as I can tell, it could be a standard RJ11/single-twisted pair to the base station where it then gets routed via IP.
A home user wouldn't notice the difference.
Self proclaimed typo king, and inventor of the bear destroying coffee table (patent not pending).
Google would be well-served by implementing WiFi now, and I think it would be fun if they did it in the same sort of participatory manner that they do everything else - they ship you a cheap or free GoogleRepeater, you put an antenna on your rooftop, and in certain areas Google pays for an Internet connection that they can connect to the GoogleRepeaterGrid. The network spreads as people are willing to install and run GoogleRepeaters, and remains fast based on them adding fiber connections at strategic points along the GoogleRepeaterGrid.
If they can find a channel, the long-haul connection between GoogleRepeaters could be handled on a longer-distance higher-bandwidth frequency or range of frequencies, and the local repeaters could output standard WiFi. But they wouldn't have to pay to put up towers, because there are a good number of people who would be more than happy to install the repeater gear at our houses and help spread the signal. Google? Are you listening? You can ship it to me now. I've got a primo spot on my rooftop antenna tower with your name on it.
As to the rail thing, it's still used for a lot of transportation of goods. It's amazingly efficient compared to any other way of moving product (except maybe floating it downstream on barges, but rail doesn't have to worry about river flow directions). You might be surprised at how much of the stuff you use every day was hauled at least part of the way by rail. It's more efficient than barging it, and almost ten times more efficient than hauling by the next-most-efficient method that's not dependent on current (trucks).
"This post contains words, known to the State of California to cause thought. Wash brain thoroughly after reading."
I wonder what bit rate we can push through the copper at most houses in rural America? My father-in-law's old house used to get very bad static on the line when it rained, but voice was still audible. Would this VOIP be capable of service, or does that house require new wiring? Anything requiring a lot of people to change the wires in their walls is going to face some serious problems. I bet new hardware in the field could get 64kbit or maybe 128kbit digital without much problem. If you're not worried about a computer talking on the line at the same time, that is way more than sufficient. Since the FCC solicitation seems to suggest they're using this as a way to force wider broadband deployment, 256kbit might be the minimum for a connection intended to share with a computer, although I'd hesitate to call that "broadband".
I bet we could help with the reliability of VOIP by putting cheap NiMH batteries in each VOIP device (one per house, at the pedestal? or each device needs its own?). Enough capacity to last a few hours on standby and maybe 15 or 20 minutes of talk time would cover emergencies.
I think it would be very interesting to be on a technical committee to write a new standard to cover bidirectional communication on low quality twisted pair. There would be interesting coupling challenges with using one wire for send and the other for receive, but using a current sense methodology on a differential signal has its own ugliness too. It would be cheating to take turns every 10-100ms using a training sequence, but there would be power and signal benefits to weigh against the increase in latency and cut in available bandwidth (and if each device gets its own CODEC, having more than 3 people on the phone may have ludicrous latencies).