OK, it’s a new century. It’s also true that new network architectures are packet switched rather than circuit switched. But is it really time to give POTS the old heave-ho?
I’ve got my doubts. Mostly because I remain unconvinced the Internet is up to the job. VoIP over DSL or Cable broadband is a dicey thing. It’s a lot like the little girl with the curl in the middle of her forehead. When it’s good, it’s very good. When it’s bad, it’s horrid. If you really think that sounding like you’re deep sea diving or chopping off each other’s sentences is acceptable telephony, then have at it. But please don’t call me until you get that fixed.
What’s at the heart of this problem is the elephant in the policy room called network neutrality. Proponents want absolute equality of access for all. That means you can’t do anything to favor one packet over another. Mandating that all packets get equal treatment protects us, at least in theory, from network operators favoring one content provider over another and ruining the democracy of the Internet. Unfortunately, such a superficial approach prevents implementation of any quality of service measures that could ensure that each packet gets the network performance it needs. As a result, TCP/IP guarantees that non-time critical text files are perfectly transferred, while time critical voice and video streams may or may not get the low latency, congestion, and packet loss characteristics they need to maintain signal integrity.
On the Internet, you launch your packets and you take your chances. Contrast that with the dedicated channels of switched circuit networks. There’s no comparison when it comes to reliable and repeatable call quality. For many consumers, that may or may not matter. We’ve gotten use to variable performance on the cellular phone system. People no longer expect to hear a pin drop. They simply want to communicate when and where they feel like at a moment’s notice. In fact, it’s the rush to cellular rather than VoIP that is the main reason so many people are abandoning their landline phone service.
It’s understandable that AT&T and the other traditional telephone companies are feeling trapped between two worlds. They’re stuck maintaining a costly investment in analog POTS lines and their associated interfaces and switches at the central offices. At the same time, there are fewer and fewer users to amortize the cost. They can’t raise prices without risking that the remaining satisfied users will revolt and move to cell phones or VoIP services.
What are businesses doing? The ones with more than a line or two are going digital, but not on the Internet. A business of any size that has a PBX phone system on-premises has options. The legacy solution is ISDN PRI or T1 PRI, a digital telephone trunking system that aggregates up to 23 outside lines. Oddly enough, this is a PSTN technology but it doesn’t depend on analog POTS service being available. The other option is SIP trunking, a packet network technology. The difference between a SIP trunk and an analog telephone adaptor connected to the Internet is that the SIP trunk runs on a private network where quality of service can be assured. The SIP trunk may be set up to provide both telephone and broadband Internet service on the same line without quality issues.
Realistically, the days of POTS may be coming to an end. We got through the analog to digital television transition and we’ll weather whatever chaos ensues when analog telephony is no more. One key issue that will be hotly debated is the decommissioning of copper phone lines at the same time the dial tone is disconnected. Copper is likely to persist long into the future. T1 and ISDN lines are provisioned over standard twisted pair copper. So is EoC or Ethernet over Copper that uses multiple copper pair to transport mid-bandwidth Ethernet service. The replacement for copper is fiber, and there is precious little of that in the ground right now.