Skip to content
Biz & IT

Winamp’s woes: How the greatest MP3 player undid itself

15 years on, Winamp "still lives"—but mismanagement blunted its llama-whipping.

Cyrus Farivar | 421
Tens of millions of Winamp users are still out there. Credit: Flickr user uzi978
Tens of millions of Winamp users are still out there. Credit: Flickr user uzi978
As many of us are busy crafting the perfect playlist for grilling outdoors, most likely such labor is happening on a modern streaming service or within iTunes. But during the last 15 years or so, that wasn't always the case. Today, we resurface our look at the greatest MP3 player that was—Winamp. This piece originally ran on June 24, 2012 (and Winamp finally called it quits in November 2013).

MP3s are so natural to the Internet now that it’s almost hard to imagine a time before high-quality compressed music. But there was such a time—and even after "MP3" entered the mainstream, organizing, ripping, and playing back one's music collection remained a clunky and frustrating experience.

Enter Winamp, the skin-able, customizable MP3 player that "really whips the llama's ass." In the late 1990s, every music geek had a copy; llama-whipping had gone global, and the big-money acquisition offers quickly followed. AOL famously acquired the company in June 1999 for $80-$100 million—and Winamp almost immediately lost its innovative edge.

Winamp's 15-year anniversary is now upon us, with little fanfare. It’s almost as if the Internet has forgotten about the upstart with the odd slogan that looked at one time like it would be the company to revolutionize digital music. It certainly had the opportunity.

“There's no reason that Winamp couldn’t be in the position that iTunes is in today if not for a few layers of mismanagement by AOL that started immediately upon acquisition,” Rob Lord, the first general manager of Winamp, and its first-ever hire, told Ars.

Justin Frankel, Winamp's primary developer, seems to concur in an interview he gave to BetaNews. (He declined to be interviewed for this article.) “I'm always hoping that they will come around and realize that they're killing [Winamp] and find a better way, but AOL always seems too bogged down with all of their internal politics to get anything done,” he said.

The problems began early, since Nullsoft wasn't interested in being a traditional corporate unit. For instance, in 2000, just a year after the acquisition, Frankel released (and open sourced) Gnutella, a new “headless” peer-to-peer file-sharing protocol that understandably steamed the bigwigs at AOL corporate headquarters in Dulles, Virginia.

By early 2004, Rolling Stone dubbed Nullsoft’s founder the “world’s most dangerous geek”—but companies like AOL aren't good fits for dangerous geeks. That same year, Frankel resigned, writing on his website a few lines that were later removed: "For me, coding is a form of self-expression. The company controls the most effective means of self-expression I have. This is unacceptable to me as an individual, therefore I must leave."

Today, Winamp continues to be updated; AOL released its first Android version in 2010 and a Mac version in 2011. Amazingly, given all the time elapsed, AOL still makes a decent amount of money on the site and on the program—while the company has declined to release official figures, former employees who worked on Winamp estimate its current revenue at around $6 million annually. And Winamp still has an estimated user base of millions worldwide, a small fraction of which live in the United States.

However, references to the application’s storied history, both good and bad, have been scrubbed or omitted from the Winamp current site—it’s just a big blank page. (Update: An Ars reader points out that the history page has moved, but the link to the alumni page still reverts back to the home page.) So how exactly did Winamp squander the incredible head start it had on most other Internet music businesses? Here's how.

Winamp lives—AOL released a version for Android in 2011. Credit: AOL

Out of the desert

Like most companies, Winamp was created to solve a “pain problem.” That problem? Two decades ago, it was pretty difficult to organize and play compressed music, which had just started to enter mainstream usage.

The German scientists behind the MP3 format released their first encoder in July 1994, but for the next three years, it remained difficult to share and find the resulting music files. Frankel has always created software “because it was software that I wanted to be able to use,” as he told the Digital Tools blog in 2008. “Winamp grew out of wanting a good, enjoyable way to listen to MP3s on a computer. It wasn’t the first MP3 player, but the MP3 players around before it were hard for me to want to use.”

Prior to Winamp, there wasn’t much available beyond Windows Media Player or RealPlayer. But none of those players could, in the mid-1990s, do something as basic as playlists, much less visualizations and custom skins, nor were they as tightly and efficiently programmed as Winamp. Even today, the Mac version of the Winamp installer is only 4.2MB; by comparison, the iTunes Mac installer comes in at a whopping 170MB.

The Windows Advanced Multimedia Products (WinAMP) player was released to the world on April 21, 1997. The next year, when its parent company Nullsoft formally incorporated, Winamp became $10 shareware. But no one pays for shareware, right? Wrong.

“Nothing ever was broken [if you didn’t pay], there was no feature that was unlocked,” Rob Lord told Ars. “In that year before we were acquired, we were bringing in $100,000 a month from $10 checks—paper checks in the mail!”

At the time, Lord already had staked his first claim to fame: co-founding the (recently revived) Internet Underground Music Archive, the world’s first online legal music repository, while he was a student at the University of California, Santa Cruz. Today, Rob Lord works for his own startup, Sherpa.io, based in the Hatchery, a SoMa (South of Market) co-working space in San Francisco, just a five-minute walk from Twitter’s original offices, alongside a plethora of tech and tech media companies. Developers and entrepreneurs rub shoulders on sunny days in South Park, an adjacent outdoor space that features a gourmet grilled cheese outfit, a taqueria, and a French restaurant with killer quiches. In short, it’s a long way from Sedona, Arizona, Frankel’s hometown and the birthplace of Nullsoft.

Seated on a couch in a Hatchery conference room and dressed in a T-shirt, shorts and sandals, Lord told me about his role at Nullsoft that began in May 1999.

“My title was director of online strategy, although we didn't know what that meant,” he said. “It was a lot of strategic and tactical decisions, around everything except the coding of the app itself. Justin was the one coding on it and he was the only one who had access to source code. I was being helpful by [looking at] product features [on a] business level—I initiated the first monetizing of the website.”

It was Lord’s job to figure out how to make the company money. Like most 1990s startups, the plan involved banner ads and possible partnership deals with other startups. Lord, then in his late 20s, moved from California to join the 18-year-old Frankel, still living at his parents’ home in Sedona, Arizona—where his father, attorney Charles Frankel, acted as the company’s counsel and chief financial officer.

This 10,000-person community in the Arizona desert attracts millions of tourists per year, and for decades has been home to people seeking to enjoy the state’s natural beauty and to take in the “spiritual” side of the place.

“You have this rugged individualism and a very rugged spiritual set and [Frankel] came from that,” Lord said. The Sedona ethos came through while making business decisions; Lord recalled one moment when there wasn’t enough information to confidently come to a decision. “'We'll just do x and let the universe decide,'” he recalled Frankel saying. “I think it describes the mode we were in—things happen for a reason.”

Rob Lord was Nullsoft's first hire, and worked with Justin Frankel in Sedona and San Francisco.
Rob Lord was Nullsoft's first hire, and worked with Justin Frankel in Sedona and San Francisco. Credit: Scott Beale

Finding a buyer

The universe decided that Winamp was awesome. In the first two years that Lord worked with Nullsoft, Winamp’s user base quadrupled from 15 million to 60 million users—and the acquisition offers arrived. “There was no meeting that was ever turned down,” Lord recalled.

Few companies said it outright, but they would couch their queries in phrases like, “Do you guys have future plans?” Before long it became hard to resist valuation offers in the neighborhood of $100 million.

In June 1999, Nullsoft was acquired alongside Spinner (in a deal worth $400 million for both), a streaming media player startup based in San Francisco. Charles Frankel, Justin’s father, took in the second largest portion of the AOL acquisition deal, receiving around $15 million in stock at the time. Frankel himself got AOL stock worth nearly $60 million at the time.

Lord now says Nullsoft was “acquired at the sweet spot of insanity,” less than a year before the January 2000 merger announcement of AOL and Time Warner, which eventually failed in spectacular fashion.

AOL consolidated both properties into Spinner's offices, a single warehouse-turned-office on Alabama Street in the Mission District. The move to San Francisco certainly provided a change of pace for tiny Nullsoft. “I felt this juxtaposition of [our new environment] with [previously being in Sedona with Frankel and] operating out of his bedroom and us having meetings at his dad's offices, on the one road in town,” Lord said.

Everyone involved in the deal expected big things to result.“The thesis at that time was that AOL could be really big in music, create within its four walls the next MTV—and that meant something in music in the 90s,” Fred McIntyre told Ars last month. McIntyre started off as a vice president at Spinner in 1998 and worked directly with Winamp until 2004, staying on at AOL. He returned to Winamp in 2007.

“The general logic was that Spinner had built a service and had a pretty well seasoned management team in place,” McIntyre added. “Winamp had built a product and a platform that was capable of generating meaningful user adoption.”

“AOL, without telling either of us, bought us both,” recalls Josh Felser. Felser was co-founder and president of Spinner from late 1997 until mid-1999, and later became an AOL vice president in charge of both units. Today, he's a venture capitalist at Freestyle Capital. “They threw us together and I was in charge of both brands. It was obvious that our cultures were different, but it wasn't obvious that they would clash.”

At first, Winamp kept riding high. “[Winamp.com] was doing hundreds of millions of page views a month in late 1990s,” McIntyre said. “We were just pushing display ads through those. That was in and of itself a lucrative business—not a sexy business, but a lucrative business nonetheless.”

But problems quickly became apparent.

Culture clash

Former employees say that a deep chasm existed between the corporate cultures of Spinner and Nullsoft. Worse still, with their simultaneous acquisition, sometimes it was unclear exactly who was in charge of what. The Nullsoft team, which at the time of the acquisition numbered only four people, was literally surrounded by Spinner staff, who had already been operating at desks along the perimeter of their old Potrero Hill warehouse space. When Nullsoft arrived, the crew set up camp in the middle of the large space with the help of a massive tarp.

Flush with cash, a confident Nullsoft had Lord and Frankel at its helm and oversaw the hiring of several post-acquisition employees, including Chamath Palihapitiya to work on product and business development. Seven years later, Palihapitiya went on to become a Facebook executive and part-owner of the Golden State Warriors. Last year, the New York Times lauded him as an “unconventional venture capitalist.”

“On my first day, I walked into the office—there was this huge tarp in the middle,” Palihapitiya recalled in a conversation with Ars.

“We had a converted sweatshop in Potrero Hill. There was a lot of light—this glass panel roof panel would tilt up to let air and sunlight in, but was really bright. The Nullsoft guys had this massive tarp, so it was always dark [for them]. I wanted to know: what's going on in that tarp?”

Palihapitiya, a Canadian whose family is from Sri Lanka, studied electrical engineering at the University of Waterloo; upon graduation, he worked for a year on the Interest Rate Derivatives Group at BMO Nesbitt Burns, a Canadian investment firm. But after a year working on Canada’s financial frontier, Palihapitiya got bored and applied for work with “every [company] I thought was doing interesting things.”

“I applied to Winamp, eBay, Google, and was rejected everywhere except Winamp,” he added. “I had decent technical skills. I have no idea why they hired me. I didn't really have any practical skills. It just seemed like a great company. To be honest, I knew so little, but I thought that these guys seem to be figuring out as they’re going along, so that's the best shot I have.”

Despite a lack of formal business training, Frankel, Lord, Steve Gedikian (now a senior manager at Apple), and Ian Rogers (now the CEO of Topspin) managed to grow Winamp’s business significantly. (Gedikian and Rogers also declined to be interviewed for this article.)

But friction with Spinner posed more problems than any tarp. One of the early problems between the sibling companies came after an AOL decision. AOL management decided that, despite Nullsoft's larger user base, Spinner was the more mature company with more traditional corporate leadership. As such, Spinner was effectively given administrative and financial control over the two units.

“There was a clash of parsing out the budget. [Rob Lord] was furious that Spinner was given control and budgeting was being given to Spinner more than Winamp, and Winamp was bigger than Spinner,” Josh Felser admitted to Ars.

“[Winamp] went from 15 million [in 1999] to 60 million [in 2001],” Rob Lord said. “It was perverse. We were in an office with Spinner, they had 40 to 60 people, we had four. We were 10 times the user base and our growth rate was much faster. It was a weird situation. We had bootstrapped the whole thing.”

Fred McIntyre worked directly with Spinner, and then Winamp, from 1998 until 2004, staying on at AOL. He returned to Winamp from 2007-2009.
Fred McIntyre worked directly with Spinner, and then Winamp, from 1998 until 2004, staying on at AOL. He returned to Winamp from 2007-2009. Credit: Fred McIntyre

"It could have been Pandora"

While Nullsoft seemed to play second fiddle to Spinner, all the companies in AOL’s own “Interactive Properties” group (including ICQ, MapQuest, and others) collectively played second fiddle to AOL’s traditional bread-and-butter moneymaker: “the Service.”

According to AOL’s own June 30, 2000 10-K filing with the Securities and Exchange Commission, its legacy Internet access service had 23.2 million users, compared to Winamp with its 25 million users. At the time, the Service made a huge amount of money, and AOL wanted to promote it heavily to a crowd with limited use for it.

One particularly bad corporate decision came in 2000, when Palihapitiya and McIntyre wrote the business case for what would have been the first music subscription service at AOL. But the idea wasn’t released as a product until 2003, under the “MusicNet” brand, two years after Rhapsody launched its own music subscription service in 2001.

Fred McIntyre claimed that one of the biggest reasons why AOL failed to capitalize on this music subscription plan was that the company insisted on using its own indigenous billing system, the one used for the Service.

“AOL was religious about this idea that we had to do billing through the same infrastructure that AOL did billing for service for,” he said, comparing it to the ill-fated decision by Yahoo to force Flickr users to use a Yahoo ID to log into the photo-sharing service.

And that pointed to a bigger problem for Winamp, and for the "Winamp culture" inside AOL—its primary users were music fans, geeks, and people who cared about what bitrate their MP3s were encoded at—in other words, the key users of Winamp in the early 2000s were allergic to AOL as a company. (Or perhaps were just put off by all those AOL promo CDs.)

“Winamp would have a larger US audience today were it not the fact that AOL tried to get people to install Netscape or AOL or something else when they installed Winamp,” McIntyre concluded.

Jascha Franklin-Hodge started out as a Spinner engineer and later became the director of software development in 2003. Later, he co-founded Blue State Digital, the company behind Barack Obama’s 2008 social media strategy.

“When you think about what AOL had in early 2000,” he told Ars, “the only thing that they were missing that [would be] essential to today’s media system is a hardware device. They had the number one software for playing [in Winamp], and in theory, although not in practice, the [Time Warner] content library that could have been a pioneer in streaming. And a radio service. It had all the elements. AOL could have been Spotify, it could have been Pandora.”

But it wasn't; it was AOL. And so, despite having the love of hardcore geeks, music fans, and millions of Winamp users, AOL’s main strategy was to try to convert those users towards the Service, ignoring or marginalizing other monetization strategies. As the years passed, both the Service and Winamp stagnated. (Today, AOL has just 3.3 million users.)

McIntyre summed up the problem bluntly. “Between 2002 and 2007, Winamp was an asset that AOL knew was valuable but didn't know what the fuck to do with."

Alternative histories

The other problem, of course, is that by late 2001, the first iPod appeared. As Steve Jobs himself famously pointed out at the product’s launch, there were existing MP3 players—but they all sucked. By 2003, Apple had sold a million iPods and launched the iTunes Music Store.

“Apple so thoroughly dominated and crushed that space,” Franklin-Hodge added. “I think it set the stage—those [26 million AOL users in the United States at the end of 2002] were a huge chunk of Apple’s initial customer base for the iPod. They were people who got why digital music was a more advantageous way to consume things. They created that door; we just weren’t positioned to walk through it in the way that Apple was.”

Other former employees agreed, blaming Winamp’s slow decline on AOL’s inability to capitalize on Winamp’s user base.

"AOL did more to negate progress than any company I've ever seen,” said Palihapitiya, Winamp’s early head of business development. “These bureaucrats viewed every decision as a political decision. Really good ideas would die on the vine.”

Even though MusicNet took time to start up, the limited download (and later subscription) music service was “crushing it” at first—that is, before iTunes took off over the course of 2003 and 2004. Had Winamp, Spinner, and AOL made different decisions, the history of digital music might have read quite differently.

“Nothing replaces execution,” Palihapitiya concluded. “I think the biggest general thing that happens in most acquisitions [is]: the acquirers tries to exert too much control on the acquiree. The host organism suffocates the young company. Some of it is not ill-intentioned. But that's just how most acquisitions end up. That was true in this case. If you had acquired [Nullsoft] and let it run as an independent operating unit, would it have done more? Probably.”

In the meantime, the core application, known for being thin, light, and fast, was suffering. In August 2002, Winamp version 3 was released (Nullsoft now had 15 employees) and had been re-designed. Some users felt it was bloated; many even reverted to older versions. By 2003, as iTunes and iPods were rising fast, Winamp struggled.

WASTE

But Frankel, the original creator of Winamp and Gnutella, wasn't out of ideas. Unfortunately, his ideas all made AOL furious.

Just as the RIAA was prepping lawsuits against Gnutella users, Frankel kept butting heads with AOL corporate headquarters in Dulles, Virginia—trying to get Nullsoft and himself out from AOL’s transcontinental reach. Besides Gnutella, which he released in September 2000, Frankel coded a program that blocked the ads on AOL Instant Messenger. After that, Dulles told him not to blog without approval.

"AOL as a company should not just sit on their asses and try to keep from losing as many subscribers as it can," he told Rolling Stone for its January 13, 2004 issue. "I mean, I'm a stockholder of the company. I want them innovating. I want them doing things that are good for the world and being socially conscious."

The disputes came to a head by mid-May 2003, when Frankel released WASTE, an encrypted, invite-only, file-sharing and chat network—on the fourth anniversary of Nullsoft’s acquisition. Not surprisingly, AOL pulled the plug within 24 hours.

Frankel described the tension to Rolling Stone in 2004:

“‘We fought off the AOL bullshit as much as possible," he says. When the company tried to insist that an AOL icon instantly appear on a user's desktop during a Winamp installation, Frankel hit the roof. "I'd be like, look, our users don't want to use AOL!" he says. "They think AOL sucks!"

Frankel agreed to stay on to see Winamp 5 through production; the program appeared in December 2003 and Frankel resigned from the company in January. Within a year, Gedikian resigned and AOL shuttered Nullsoft entirely, moving the remaining development staff to Dulles. At the time, Slate lamented the closure of the last “maverick tech company.”

The Dulles-based Winamp team, as of 2012.
The Dulles-based Winamp team, as of 2012. Credit: AOL

Winamp on life support

Probably the most surprising piece of the story is that Winamp didn’t die right then and there in 2004. For the next three years, the software seemed more or less in stasis—no new version appeared.

Ben London, an AOL senior technical manager, took the reins of Winamp in Dulles in August 2004. He admitted that  AOL’s own corporate strategy, which kept shifting, slowed down development.

“There’s just not a long-term investment tolerance,” he told Ars. “We carve out a six-month road map, and then boom, re-organization into a different group, and new, different drivers would ask us to focus on a different aspect of the product.”

But Winamp ticked along. When the new, 10-year-anniversary edition of Winamp finally dropped in 2007, Winamp hit its peak in the post-Frankel era, with a 90 million active user base, according to London.

“Winamp had no business staff,” said Sam Weber, a longtime AOL manager who eventually became Winamp’s business development director in 2006 and its overall director from 2007-2008. “Ben London was managing five developers that were keeping shoutcast.com, winamp.com, and the client on life support.”

Weber told Ars that he and his team figured out something surprising about Winamp. While its numbers eventually dropped to all but negligible levels in the United States, the international market was a different story.

“What do you do if you have 50 million users around the world and 90 percent are outside the US and you have a six-person team in DC?" Weber wondered. When the team looked closer, it found that it had five million Turkish Winamp users and two million Brazilian users, among others.

The Winamp team tried to come up with a new business plan for Winamp, wondering if they could tack on a download service, a streaming service, or even just sell advertising against the installs.

“I think bit-by-bit what we mostly did was try to revitalize the community by re-releasing the product, updating it, both in terms of user interface and the codecs and technology with it, making Shoutcast easier and friendlier to use,” he added. “We started monetizing users who came to the site, and more importantly, we translated it into a number of languages.”

In short, the new Winamp team in Dulles began to figure out what AOL should have realized years earlier.

“The first is don't screw up a good thing,” Weber said. “Don't use it as a mechanism for delivering AOL-branded services. Help make it more relevant to people who do use it.”

So Winamp 5.5 added official German, Polish, Russian, and French support for the first time. But more importantly, the company started to figure out how to make money from users through what came to be known as the “freemium” model.

“You sell a premium version of the product,” Weber said. “You've got this freemium play and it gets big enough, if you can get one percent [of your customers to buy it], you can build a real business.”

Currently, Winamp Pro sells for $20, which, when multiplied by hundreds of thousands of paid users, works out to millions of dollars in revenue. Another way to make money is by selling ads directly on Winamp.com—and with the site doing several million total unique visitors a month, it's an easy way to earn some cash.

But there’s an even better way to make money: put a Winamp player inside a browser toolbar. As Weber explained, this generates “a hell of a lot of money—search makes a lot more than banner ads.”

“Google has a [search] deal with AOL, so we're distributed AdSense, in a branded, Winamp sense,” he explained. “Now you've got millions of queries a day, and we would promote eMusic. Anytime someone searches for something, you're getting a piece of that. That's real money for a little startup. If the model hasn't changed since I was there, it’s advertising against a big audience.”

Geno Yoham is the current man in charge of Winamp Credit: AOL

A second coming?

15 years after its founding, all of Winamp's original employees are long gone (most have moved on to other positions in Silicon Valley, largely in the tech and music sectors) and the product has now existed longer without Frankel than with him. Its user base continues to stagnate as competitors like Rdio, Spotify, and Pandora—to say nothing of iTunes, which last year sold its 15 billionth song—continue to thrive.

But Geno Yoham, Winamp’s general director since October 2008, argues that Winamp will continue to do well as a media player, particularly in emerging markets where Apple hasn’t penetrated as well.

“iTunes is number one, and we’re number two,” he said. "There's a lot of value in the Winamp brand, the media player that's on your side."

Winamp says that it has around 30 million users worldwide currently (a figured based on comScore Web traffic analysis), with less than one million in the United States. The company is also starting to target specific platforms (read: Android) rather than specific markets. Winamp reports that since the Android launch of Winamp in October 2010, it has seen over 19 million installs. Could Winamp rise once more?

“We're thinking about [buying Winamp] again,” said Josh Felser, the former Spinner executive and current venture capitalist, who said he and other investors also tried to buy Winamp from AOL in 2003. He's still in love with the idea of Winamp, and with the community it spawned, and believes that someone could still capitalize on all that potential.

"Winamp had the start of something social in music," he said, dismissing the music sharing that Spotify is now doing on Facebook. “That’s just seeing what people are listening to. It doesn't feel right.”

“[Winamp] had a very distinct, edgy, tech-savvy community, and that's a valuable community to lots of people. That would be a big thing to understand. What are the set of features that we can attach to it to make it become more relevant again? I don't have it. I’d have to get serious about that.”

But, he admitted, no significant steps have been taken in terms of talking with AOL about a possible acquisition.

“Although the idea of starting with that basis [of an existing community and a history] is exciting,” he said with a grin, “I haven't given it more than five minutes thought.”

Listing image: Flickr user uzi978

Photo of Cyrus Farivar
Cyrus Farivar Editor at Large
Cyrus is a former Senior Tech Policy Reporter at Ars Technica, and is also a radio producer and author. His latest book, Habeas Data, about the legal cases over the last 50 years that have had an outsized impact on surveillance and privacy law in America, is out now from Melville House. He is based in Oakland, California.
421 Comments