What public media’s first digital experiments reveal about the next transition
amtitus / iStock
When I started working in public radio in 1985, the most important technology in the building was a razor blade. Editing audio meant cutting magnetic tape with a grease pencil and splicing it back together. Newsrooms ran on reel-to-reel decks and cart machines, and if something went wrong on the air, you grabbed a flashlight and drove up the hill to the transmitter site.
Radio was defined by geography. Your audience was whoever lived inside the contour of your signal. If your transmitter reached 60 miles, that was your world. The idea that someone across the country — or across the planet — might listen to your local station in real time was almost unimaginable.
Yet by the late 1990s, a small group of us inside public radio began asking a question that would quietly reshape the industry: What happens when radio is no longer limited by towers? The answer eventually produced experiments like Jazz24, which in its early years became one of the most listened-to internet radio streams in the world. Looking back, it’s easy to assume that streaming audio and podcasts were inevitable. Yet in the late 1990s, they were anything but; most of the infrastructure didn’t exist yet.
Learning radio the traditional way
My career in public radio began at KPLU in Seattle/Tacoma (now KNKX). Like many people in the industry, I started on the air and gradually moved into operations, programming, development and management. Over time, I helped launch more than a dozen public radio stations and later served in leadership roles at WABE in Atlanta, Northwest Public Broadcasting, and today as chief operating officer at KUOW in Seattle.
Launching a public radio station is part engineering, part community organizing and part improvisation. You deal with FCC licensing, transmitter sites, programming strategy and fundraising — often all at once. For decades the architecture of radio was simple: build the station, put up the tower and serve the region around it.
By the mid-1990s, however, the internet was beginning to hint at something different. Most broadcasters treated the web as a marketing tool — posting schedules online and maybe a few audio clips. A few of us began wondering whether the internet might eventually carry live radio itself.
Streaming before streaming worked
Trying to stream audio in the late 1990s was an exercise in persistence. Most listeners used dial-up connections, bandwidth was scarce, servers crashed regularly, and audio compression technology was still evolving. Even maintaining a continuous audio stream was difficult.
But we kept experimenting. We built early systems that could take a radio signal and deliver it through the internet in real time. The technology was fragile and improvised, but occasionally it worked. And when it did, you realized something extraordinary had just happened: a radio signal had escaped the tower.
Those experiments attracted attention from streaming technology developers like RealAudio and even led Microsoft to produce a technical white paper examining the approach we were using. At the time, it felt like tinkering at the edge of possibility. But every once in a while, you would hear a stable stream coming through your computer speakers and realize you were listening to something that might change radio forever.
A hallway conversation that changed two stations
One pivotal moment came during a conversation with Al Bartholet, then general manager of WKSU in Kent, Ohio. Both of our stations were experimenting with internet streaming and trying to understand what it might mean for radio. The technology was new, the audience uncertain, and nobody really knew what the right strategy was. But one thing seemed clear: If the internet removed geographic limits, stations could build on their format strengths in entirely new ways. Serving a community began to mean something different than a physical place; it was about shared audience interests.
KPLU had deep roots in jazz. WKSU had built a strong reputation in folk music. In that conversation, we came to a simple conclusion: Rather than competing in the same space, each station should lean into what it did best. WKSU would pursue an all-folk internet stream, and KPLU would build an all-jazz stream. Out of that idea eventually came Folk Alley and Jazz24.
At the time, it was just a practical decision between two colleagues trying to make sense of a new technology. Only later did it become clear how significant those experiments would be.
The birth of Jazz24
Our experiment at KPLU became Jazz24, a continuous jazz stream designed specifically for internet listening. Today, format-based streams are everywhere. Open Spotify or Apple Music, and you’ll find hundreds. But in the late 1990s, the idea was revolutionary.
Radio formats had always been geographically limited. A jazz station in Seattle served Seattle. A station in Atlanta served Atlanta. Streaming changed that. Suddenly a niche format like jazz could reach listeners anywhere.
What happened next surprised even us. In its early years, Jazz24 became one of the most listened-to internet radio streams in the world. Listeners tuned in from Europe, Asia, South America and beyond — many in places where jazz had nearly disappeared from local radio. A 2006 Seattle Post-Intelligencer article noted that KPLU was already receiving pledge contributions from listeners outside its broadcast coverage area, including China and Japan.
For a station built around a regional signal, that realization was profound. For the first time, public radio’s audience could extend far beyond the limits of a transmitter.
Podcasting before podcasts
Around the same time, we experimented with another idea: downloadable spoken-word audio. Working with Audible, we explored ways for listeners to download programs and carry them with them. Using RSS, we posted newscasts on our website to listen to later.
This was years before the word podcast entered the vocabulary. Portable digital audio players were still rare, and smartphones were a decade away. But the concept was already there: audio that traveled with the listener rather than requiring them to be present when it aired. Those early experiments foreshadowed the podcast ecosystem that now dominates spoken-word audio.
Changing the conversation in public broadcasting
As these experiments multiplied, it became clear that public broadcasting needed a place to talk seriously about digital media. In the late 1990s and early 2000s, several of us helped create the Integrated Media Association (IMA). The organization brought together people across public broadcasting who were working on emerging digital technologies — web development, streaming, early forms of online distribution, and digital revenue models.
At the time, the phrase “public media” hadn’t even entered the industry vocabulary. Stations thought of themselves as radio stations or television stations. IMA helped change that conversation. It created a forum where engineers, programmers, web developers, fundraisers and managers could share ideas about how digital technology might reshape public broadcasting. Many of the questions discussed in those early meetings would later become central to the industry’s strategy.
Lessons for the next transition
Over four decades in public broadcasting, I’ve seen the industry move through two major revolutions. The first was the traditional broadcast era — towers, transmitters and local signals. The second was the transition to digital distribution — streaming audio, podcasts and global audiences.
Today, public media is facing another moment of uncertainty. Stations are grappling with audience changes, funding pressures and the rapid emergence of artificial intelligence. It’s understandable that many people feel uncertain about what comes next.
Looking back at the early days of streaming, I’m reminded that the future rarely arrives in the way people expect. In the late 1990s, streaming audio looked fragile and unreliable. Many assumed the internet would simply supplement traditional broadcasting rather than transform it. The breakthroughs that did happen often began quietly — engineers experimenting with new tools, late-night troubleshooting sessions with early streaming servers, or something even simpler: a conversation in a hallway.
The idea that eventually produced Jazz24 and Folk Alley didn’t emerge from a strategic plan or consulting report. It came from two station managers talking about what their stations did best. That experience has stayed with me because the next major shift in public media — whether it involves generative AI or something we haven’t imagined yet — will probably begin the same way: with a curious producer trying something new, with an engineer experimenting with a different approach, or with a casual conversation that nobody realizes at the time might change the direction of the industry.
Public media has always been strongest when it trusts that spirit of experimentation. Because sometimes the future of radio is already there — you just haven’t heard it yet.
Kerry Swanson is Chief Operating Officer of KUOW in Seattle. With a career spanning over 40 years in media, Kerry has held leadership roles at Northwest Public Broadcasting in Pullman, Wash.; WABE in Atlanta; and KNKX (formerly KPLU) in Seattle/Tacoma. He has served two terms on the NPR Board of Directors, is president of the University Station Alliance Board, and has held leadership positions with Western States Public Radio and the Northwest News Network. Kerry was also a co-founder of the Integrated Media Association.
Terrific article. A wonderful historical snapshot of all the transitions public broadcasting–radio and television have gone through, and now combinations of that into new distribution options, including those just evolving. And still free of controlling media giants and in spite of political efforts to wipe it all out. Belief, critical thinking and creativity combined to continue delivering an important alternative.
Kudos! May the force be with us.