Tedium - What Was ISDN? 📞

Today in Tedium: If you live in the United Kingdom and still rely on traditional copper lines to get telephone calls or get on the internet, we have some bad news for you. In 2025, British Telecom plans to shut off its ISDN (Integrated Services Digital Network) and PSTN (Public Switched Telephone Network) services entirely, in favor of modern technologies like Voice over Internet Protocol (VoIP). It is within its final year of existence, which means if you do not have an upgrade plan by now, maybe you should have one. This, of course, means ISDN is getting dangerously close to Tedium territory, which as you may know involves “a pattern of stillness.” Which means now’s a good a time as any to write a retrospective on one of the most interesting technologies the phone company ever gave the masses, ISDN, and why relatively few people have actually used it. Today’s Tedium considers ISDN’s legacy. — Ernie @ Tedium

Editor’s note: This issue is dedicated to the memory of Adam K. Olson, a longtime reader who I sadly only just found out passed a little while back. He always had a friendly note to share in my inbox, so I hope to share his memory in all of yours.

A British Telecom ISDN ad. Yes, BT actually promoted ISDN on television.

Where did ISDN come from, and what was it used for?

In many ways, the legacy of the T1 line and ISDN are intertwined. While somewhat unclear to the average consumer, telecom companies like AT&T had spent decades developing technologies to improve switching technologies, so they worked on digital protocols.

So, while you were making analog phone calls, the phone company delivered those calls through digital lines. While the average consumer thinks of T1 as a one-time way to get fast internet access, it was a pipeline for the phone system that had additional uses.

The next step, clearly, was to figure out how to bring this digital infrastructure to the average consumer, the “last mile” as they called it. Potentially, businesses, and maybe even consumers, could leverage data distribution over dedicated lines. This is what ISDN became.

But it wasn’t how it started, and the concept didn’t even start in the United States. For years, before the Integrated Services Digital Network became an actual technology that people associated with the telephone system, it was something of a buzzword, a signifier of the future of telephony. Per the Communications Museum Trust in the United Kingdom, the term was created by the Consultative Committee for International Telephone & Telegraphy (CCITT), the predecessor to the International Telecommunications Union, a modern-day standards body.

Originally named Integrated Digital Access, a name that British Telecom continued to use for its own endeavor, it gained its current name back in 1971. The technology itself didn’t actually reach the public until the mid-1980s at the earliest, for obvious reasons. After all, they had to do the hard work of developing the standard, making sure it worked, and ensuring the infrastructure could reach far and wide.

(One factor for this is CCITT’s approach to decision-making, essentially making decisions on new technology in four-year periods—first firming up the goals for the technology in 1981, and agreeing on some of the earliest standards in 1985, as noted in a 1991 guide to ISDN standards.)

One question that I’m sure was on the mind of big telecom during this period: Were copper wires enough? Sure, phone lines could handle calls or data, but could they really do both particularly well? It turns out that by applying T1 lines’ alternate mark inversion encoding scheme onto copper ISDN lines, it was possible to get more out of the old boy. The theoretical limit jumped from like 64kbps of data to 160kbps of data—not much now, but a lot in the 1970s and 1980s, and enough to make it possible to get high-quality voice calls out of a copper phone line. (In practice, a standard ISDN line was split up like this: Two bearer lines of 64kbps for voice conversations, each, plus an additional 16kbps line dedicated to data. You could theoretically combine the bearer lines to get a full 128kbps data connection, before any compression was added.)

If you have nothing to do for an hour and twenty minutes, here’s an AT&T training video on ISDN for your viewing pleasure. You can thank me later.

Even in the early 1980s, when ISDN was a concept barely heard beyond high-tech circles, it was spoken of in ways not dissimilar to more modern buzz concepts, like the Internet of Things, 5G, or cloud computing. The language was vague, and focused on real world uses, rather than the ins and outs of the protocol. Longtime Bell Labs leader John S. Mayo, then the vice president of that organization, talked to Computerworld at the end of 1980 and said this of ISDN:

The Integrated Services Digital Network is more of a concept than an actual network, and it means different things to different people. Generally it is a combination of digital switching systems interconnected by digital transmission facilities and linked to customers by digital subscriber loops. This network would carry high-speed digital pulse streams to homes and offices equipped with digital terminals.

In a sense, we already have an integrated services digital network. Digitization of the telephone plant has been proceeding according to a general plan set up a long time ago.

Essentially, it was a broad, nonspecific idea of where the industry wanted telecommunications to go in the 21st century. But then, one day, it was a firmed-up standard, represented by a wire, that went into businesses around the globe.

What businesses? One of the first to sign up was McDonald’s, which embraced the technology at the corporate headquarters it was building in the late 1980s. A Chicago Tribune story from the era paints Illinois Bell as banking on a high-profile client whose fortunes were so high that it might just inspire a bunch of copycats. Here was the promise, as the article put it:

ISDN’s chief advantage for the office of the future: It not only provides a consolidated network capable of handling virtually every transmission requirement without the cumbersome “wiring spaghetti” necessary today, it also allows for the integration of the diverse array of computer, telecommunications and office equipment, much of it incompatible, that has sprung up in virtually every workplace.

That means not only that McDonald’s can build its new headquarters without planning space for huge bundles of coaxial cable, but that the company will not have to face skyrocketing costs every time it wants to move employees around.

But even given all that, the potential of this technology they built was kind of vague. Sure, the article discussed data transmissions and video streaming, but the “why” of this equation somewhat lingered.

“I think it is safe to say that today we don’t understand all of the things we can do with this. A lot of new applications will come out as the desktop digital set develops,” said Perry Erhart, the phone system’s senior manager of major accounts.

In some ways, this is true. But in others, it was a technology without a distinct, obvious use case.

An ISDN modem at a BBC studio. (Cory Doctorow—yes, that Cory Doctorow/Flickr)

In a world of 56K modems, who would want an ISDN line?

When I was a kid, I don’t believe I knew a single person who had access to an ISDN line. Odds are, you probably didn’t, either. The technology, representing a massive investment by the telephone industry, was basically ignored by the market.

In part, it’s because ISDN was expensive and poorly explained to the masses, largely relegating it to the B2B realm. It seemed like a solution in search of a problem, something decided on by the phone industry and pushed onto the market as a standard, rather than the market making the choice themselves. To be fair, when the phone industry started developing the technology in the 1970s and 1980s, there weren’t many competitors to their dominance in big telecom. But a lot had changed in the meantime. In the U.S., for instance, this process started when the Bell system was in a single piece. That was not the case by the time ISDN finally reached consumers. And in the UK, where British Telecom heavily backed the ISDN initiative, the British government had started to privatize BT, meaning the technology it had cheered on was emerging on the open market.

And in the meantime, cable television had gone mainstream, which meant many consumers already had high-speed coaxial cables in their homes. The first online services had started to crop up. And while nowhere near as good as an ISDN line, dial-up modems had gotten faster and were pretty common among computer users.

ISDN attracted numerous funny acronyms that pointed at its relative uselessness to the average consumer and high cost. Among them:

• “I Smell Dollars Now”

• “I Still Don’t Know”

• “It Still Does Nothing”

• “Innovations Subscribers Don’t Need”

A half-solution that was the target of sarcastic snickers wasn’t the promise of this new technology. It was meant to be the system for the next 50 years.

“ISDN’s strategic value is tied to its goal of providing low-cost voice, data and video access to nearly every home and office in the United States and, eventually, the world,” wrote authors James C. Brancheau and Justus D. Naumann, in an article for Data Base, describing the value of ISDN for corporate managers.

The problem with ISDN was simple: We already had effective options for everything it was trying to do. And one of those options, the traditional telephone line, was already in homes the world over. Inertia is strong, and we preferred the inexpensive option that didn’t require us to pay massive installation costs. The equation was different for businesses, but the dream of “nearly every home and office” was clearly a pipe dream by the early ’90s.

Not helping were the phone companies. Nynex, the firm that served most of the Northeast U.S. in the ’90s and later became part of the modern day Verizon, charged aggressive fees both to install ISDN (between $130 and $205, per one estimate), along with per-hour usage charges, the very kind that people were worried would affect them on traditional dial-up lines. While other phone networks elsewhere in the country had more favorable deals, it was clear that this emerging technology was not being priced to sell.

But ISDN was not without its advantages. It was faster and more reliable than modems, and the voice quality was pristine, making it a great solution for audio production needs. As a 1999 Sound on Sound article noted, it helped musicians collaborate across oceans.

Another big advantage was its connection speed. Data connections did not need to be converted into audio and back, which made the connection much more consistent and reliable. The handshake period, which could take half a minute or longer on a traditional phone line, was near-instantaneous with ISDN, making it a desirable option when time is money. Hence, if you were a remote worker in the ’90s, you might invest in getting an ISDN connection in your house—or if your company had real money, a T1 line.

Meanwhile, other technologies were beginning to emerge that actually solved the problem ISDN’s creators identified. The first of those was coaxial cable, which was disregarded in the McDonald’s build as adding significant infrastructure costs, but was already in tens of millions of homes.

Digital subscriber line, or DSL, attacked the data part of the ISDN equation from the bottom up, leveraging already existing phone lines in a new way. And then, there were fiber optics, which had significantly more headroom than copper wires could offer, and were already seeing mass consumer implementations in some parts of the world a little more than a decade after the first ISDN line went live.

ISDN’s best ammunition against these emerging technologies was telling: As a 1996 CNET article put it, it was already on the market at the time when cable modems and DSL were exceedingly rare, effectively placing the technology in this position where, only a decade after its debut, it was already falling out of favor for internet use.

So much for 50 years. Based on the UK’s planned shutdown next year, it’s only getting to 40.