The man behind the RealPlayer races ahead with RealJukebox, his aggressive scheme to dominate the downloadable universe.
Here's what Rob Glaser knows: The dawning Era of the Download will change everything. So Glaser, founder of RealNetworks and the king-of-streaming who brought radio and TV to the Internet, is busy convulsing the status quo again, this time with a superficially innocent melody machine called RealJukebox. Introduced in the spring, RealJukebox allows users to download songs from the Internet to a hard drive, upload - or rip - whole compact discs at exceptional speed, and shift tunes seamlessly to a new generation of desktop, set-top, and palmtop devices that appear destined to send your CD changer to the Great Turntable Graveyard. RealJukebox makes your individualized soundtrack available from den to dashboard, from living room to jogging path. It's being hailed by some in the music and tech industries as the first step toward a literal "jukebox in the sky," meaning a networked device that can hook a listener up to any tune, anywhere, at any time.
Here's what else Glaser knows: The downloadable revolution isn't simply about music. The current imbroglio over MP3 - the compression algorithm that has turned digital downloading into a populist enterprise - is a preview of a larger battle over the convergence of all media, a looming phenomenon that promises to overturn the economic underpinnings of manufacturing, distribution, retailing, communications.
Anything so disruptive is, ultimately, about power, and Glaser is perfectly comfortable with that: His goal is nothing less than ruling the multibillion-dollar future of broadband. No wonder even his friends - record labels that use his technology to stream their songs, webcasters and portals that built their companies with his software - are convinced Glaser is going after them. Label leaders are accusing him of crossing the line into content kingmaking, major customers are blasting him for monetizing their property, and his former employer and onetime investor, Microsoft - Microsoft! - is denouncing him as a monopolist.
"Real is playing Microsoft 1985, and we've learned that lesson," says John Paddleford, a program manager in Microsoft's Streaming Media Division. "This isn't about technology. It's about politics."
Rob Glaser built his streaming empire - 1998 revenues $65 million, current market cap somewhere near $4 billion - in no small part upon a mastery of the politics necessary to turn both giant entertainment conglomerates and scores of small Internet startups into a unified, interdependent "ecosystem." Today, as broadband turns from fantasy into reality, he and his minions still hope to keep their partners happy, but are forging defiantly ahead regardless.
"It is a risk; there are entertainment-industry executives completely threatened by what this new world order will look like," concedes venture capitalist and RealNetworks board member Jim Breyer. "But there will be a new world order. The innovative leaders adapt. The others disappear."
The furor over digitally downloaded music has as much to do with music as the Lewinsky scandal had to do with sex. Etunes are just the latest flash point in the convergence power struggle.
Because music crosses all ages, classes, races, and national boundaries, it looks likely to be the most potent magnet for attracting the average consumer to broadband, jump-starting the marketplace in a way neither AOL's chat rooms nor Time Warner's Road Runner ever could. As Glaser says, "With music, the mass-consumer product has been digital for years. It's a product form where consumers have a well-established pattern of wanting to collect and own it. The capacity of devices to hold enormous amounts of music will just leap. All the factors are lining up to drive acceleration of broadband."
The recording industry is simply the first to be overwhelmed by the phenomenon. Online music sales are expected to account for fully 8 percent of industry revenues within the next three years, according to Jupiter Communications. Jupiter's analysts believe that most of those revenues will come from the online sales of physical CDs, with pure digital distribution - hindered by bandwidth limitations and competing standards - accounting for only about $35 million in sales by 2002. Others, such as Web music distributor Internet Underground Music Archive, project a far steeper adoption curve, with revenues from digital distribution exceeding $600 million in three years and reaching $3.7 billion five years later.
Money on this scale - combined with the likelihood that a chunk of all "hard media" sales and rentals, currently worth about $60 billion annually, will shift to digital download - has drawn a diverse pack of competitors: tech giants IBM and AT&T, the five major record companies, conventional media operations such as MTV, tiny companies like MP3.com and Liquid Audio, and, of course, Microsoft.
Glaser has been on a tightrope each step of the way, attempting to placate and at the same time outpace the other players.
When I first meet him, Rob Glaser seems an unlikely aerial artist. At 37, he is a heavy man with a thick handshake and a habit, much joked about inside his company, of showing up later in the morning than tech-industry workaholism deems appropriate. Yet three weeks before the release of RealJukebox, he displays the kind of physical dexterity and mental agility that is, quite literally, breathtaking.
Glaser darts from meeting to meeting, stopping long enough to dictate a press release from his head and flip through a deck of graphs to rewrite a forthcoming speech, also on the fly. "You have the same data on three different slides," he complains to an aide. Although I have been warned about his temper - "intense" is the most frequent word associated with him, but one former colleague more frankly labels him "a screamer" - I never hear him raise his voice, but I frequently hear him (especially later in the day) shift his already rapid cadences into warp speed, with an occasional profanity signaling deep displeasure. In one session, where a development team is showcasing an alpha version of RealJukebox sans a search function it is supposed to have, Glaser starts complaining. "It's not supposed to be a stale page. It's a fucking dynamic page. We're doing press tours at the end of this week. Anyfeatureyoudon'thaveforthelaunchtourdoesn'texistasfarasthepressisconcerned!"
On this day in April, he is not only supervising the forthcoming launch but nimbly trying to keep his fractious ecosystem from devouring him. He has just disclosed a partnership with IBM to facilitate the secure distribution of digital music, a venture lauded by the major record labels, so right now, the entertainment conglomerates are happy. His task is to keep them content when they learn, today or tomorrow, that he is acquiring Xing Technology, which makes players and encoders for the notoriously insecure, deeply hated MP3.
"Are we clear on the messages to the various people we're talking to?" Glaser asks at a morning meeting of his top lieutenants, convened to position the various download deals. On a whiteboard are scrawled the names of partners who need a talking-to: Sony, MP3.com, IBM, Warner, the Recording Industry Association of America, and so on. Most are members of the Secure Digital Music Initiative (SDMI), a coalition headed by the major labels to bless a single pirate-proof standard by which music can be digitally and profitably delivered.
The diplomacy is delicate. The deals are vital parts of a strategy aimed at ensuring that Real and its Jukebox are compliant with the SDMI standard. Provoke the labels too much, and they might just as readily withhold their blessing - or, worse, bestow it solely on Microsoft. "Our line here," Glaser continues, "is that unsigned artists are a legitimate sector." Translation: Stress that digital downloading is nothing more than a great and unthreatening way for would-be rock stars to get themselves known; don't stress that it is also a great and threatening way for people to distribute pirated music.
Although Glaser admits only to "bumps along the road" in Real's relationship with the labels, in fact more than a few people are furious with him. "Their Jukebox is a great way for people to make personal collections of pirated music," one recording executive says. His warning to Glaser: "Don't bite the hand that feeds you." But Glaser, who for months has spoken in conciliatory tones about the piracy problem, now, on the eve of the RealJukebox launch, has grown more aggressive. "Are car manufacturers responsible if someone drives over 50 miles an hour?" he asks testily.
The product of a suburban New York prep school favored by apostates and liberals, a prominent campus rad who penned a Yale Daily News column sympathetic to labor, Glaser may have seemed like an atypical Microsoftie. But having managed during his undergraduate years to gain three degrees (two in economics, one in computer science) and run a videogame company, he was primed upon his graduation in 1983 to blend with an ethic that was all about workaholism. In a period when Steve Ballmer, then as now Bill Gates' chief subordinate, would survey the parking lot on Saturdays to see who was present, Glaser invariably was. "Even though Rob is an independent guy, it was really important for him to fit into that culture," recalls Mike Slade, who came to Redmond at the same time as a newly minted MBA. "There was a while when his attitude was almost 'Microsoft, right or wrong.'"
But in 1993, after rising to become the company's youngest vice president and gaining a reputation as a demanding boss, Glaser lost a bureaucratic tussle over control of the company's multimedia operations to Nathan Myhrvold, prompting Glaser to resign.
With a Yale friend, David Halperin, Glaser hatched vague plans for a company that would link television to the nascent Internet in the service of progressive causes. He'd heard of a new technological construct that would allow the transmission of real-time and on-demand audio signals over the Internet without making users suffer through interminable waits. Glaser, with some of his Microsoft millions, quickly hired a trio of engineers to develop the software. "Streaming" media would, he assumed, help make his new company, then called Progressive Networks, "a place where people who wanted to change the world could go," recalls Halperin, now a speechwriter for President Clinton.
Not long after, the two demonstrated the system to an informal group of liberal advisers at a Washington, DC, hotel. "We've developed a proprietary technology that I think will be critically important to our goals," Glaser told them. He opened up a laptop, called up a server in Seattle, and clicked the mouse. Within a few seconds, a baseball game started coming through loud and clear. Then and there, his friends convinced him to drop the politics, focus on streaming, and donate the resulting profits to their favored causes.
But streaming media had no clear, linear business model, no explicit strategy showing how it would generate consistent, ongoing profits. Glaser and his cronies veered through several notions - content models based on books-on-tape and premium cable television, for example - before realizing that they had, in fact, created a software company.
The idea - giving away the client player software and charging would-be new media moguls for the server software - entailed no small risk. In 1995, multimedia was, to many, a specialty application, not the future of the Internet. Glaser's friends were betting more on him than on the product. Lotus founder Mitch Kapor and Mike Slade, who'd left Microsoft to join Starwave, were early investors. Kapor also introduced Glaser to the venture capital firm Accel Partners. Glaser, who retains 40 percent of his company (currently valued in excess of $1.5 billion), sold more than 10 percent to Accel for $5 million.
Though the company's revenues remain paltry and its profits nonexistent, the faithful have been rewarded both by shares that have risen almost tenfold since the public offering in late 1997 and by the satisfaction of seeing Real's technology achieve near ubiquity. RealNetworks (the company changed its name in 1997) estimates that more than 300,000 hours a week of streamed content are available on the Web, and its own RealGuide aggregation site lists 4,700 sites deploying live or on-demand streaming. In fiscal 1998, software licensing fees - primarily from server software - were $47 million. Eighty-five percent of streaming media broadcasts, Real says, use its technology - a claim no one disputes.
Bill Gates, who had spent most of Real's first few years ignoring the Internet, eventually came to understand that his former protégé was on to something - something he wanted.
The tip-off came in 1997, when video seemed the next streaming opportunity and VXtreme, a California startup that had developed an early video-streaming technology, released its client software. Although widely considered the best system around, VXtreme had trouble attracting financing. "The issues were always the same: How many clients have you distributed?" recalls Pete Mountanos, then VXtreme's CEO. "We were winning every technical shoot-out; we were being distributed for free on the CNN site; we got to a million clients in a week. But Real had about 13 million. We were always going to be number two." Mountanos hinted to Glaser that he should buy VXtreme. Glaser, only months from releasing his own video product, turned him down.
Then, just before Father's Day, Glaser and Accel's Jim Breyer independently heard that Microsoft was on the verge of a VXtreme deal. Microsoft had been a laggard in the streaming space, but Real execs understood that with VXtreme's technology and the distribution reach ofInternet Explorer, Microsoft was becoming a terrifying competitive threat. Breyer and Glaser quickly arranged a Friday evening meeting with two Microsoft senior executives, Paul Maritz and Greg Maffei.
"We knew that Microsoft would eventually become a major competitor, but without a relationship we thought we were in danger of not achieving ubiquity," Breyer says. "It was a question of which was the riskier strategy - completely removing ourselves from the Microsoft world, or trying to work with them."
Real decided that working with Microsoft was marginally less risky. A strategic alliance was quickly cemented, which allowed Microsoft to license, for $30 million, Real's version 4.0 source code and bundle the client with Internet Explorer. The source code would enable Microsoft to make software capable of playing and serving the enormous amount of Web content available in Real's format. Microsoft spent another $30 million for a 10 percent stake in Real.
Inside Real, the union was no cause for celebration. "It was a shotgun marriage," Kapor says. "It was tactically important to prevent all-out warfare."
What happened next is murky. Microsoft insiders say the company made an honest effort to work with Real, but that the quick flurry of deals caused organizational chaos. Real insiders describe the rise of a faction at Microsoft intent on using the technology license to undermine its new partner. Whatever the truth, shortly after the license took effect, Microsoft released its own, free version of the Real server - but with no limit on simultaneous streams. It was a killing blow to Real's main revenue source. "Once they did that," Glaser says, "they were signaling, not necessarily in a rational way, that they were focused on cutting our oxygen supply."
Glaser says Real had no choice but to retaliate by releasing, about six weeks later, its RealSystem 5.0, an updated format incompatible with the technology that Microsoft had just licensed. "It was a 'ha-ha' to Microsoft - 'you've got our source code, but we have something different,'" says a customer of both companies.
If nothing else, Glaser's action was shrewd, setting Microsoft back several months and giving Real a boost in the race toward ubiquity. By introducing an incompatible system, Glaser was ensuring that multimedia distribution would turn into a VHS-vs.-Beta standards battle, with Real, at least temporarily, holding the VHS card. In other words, the company was guaranteeing that any content introduced in its new format wouldn't play on the Windows Media Player - which made it less likely that paying server customers would defect to Microsoft, even for free server software.
With the release of the RealPlayer G2 in July 1998 the tensions came to a head. Built to conform to a protocol called Synchronized Multimedia Integration Language (SMIL, pronounced "smile") - replete with a channel bar and a search tool, and capable of calling up streams in the absence of a browser - G2 brought the company's vision of transaction-enabled multimedia closer to reality. But some PC users who had both the G2 player and Microsoft's equally new Windows Media Player 5.2 would be defaulted straight to Media Player. If the content called up was in Real's 5.0 or G2 format, the user would get an error message. In a few cases, Windows disabled G2 entirely.
As Senator Orrin Hatch, chair of the Senate Judiciary Committee, pressed Glaser to testify about Microsoft at his upcoming Competition in the Digital Age hearings, a debate raged inside Real. Some weren't really sure that what appeared to be happening was systematic enough - or clever enough - to be intentional; Microsoft's attitude, these people believed, was bureaucratic inertia, not willful subversion.
But Glaser grew increasingly convinced that Microsoft was intentionally slamming his clientele. He agreed to testify. He sent a last-minute email to Gates, offering "to share with you in advance what I'm likely to say." Gates' reply: "When you are in Washington, I suggest you visit the National Gallery and the Smithsonian."
Glaser was furious. For several people close to the situation, this exchange epitomized the complicated, residual personal relationship between Glaser and his former employer. Some say that among the factors driving Glaser, even now, "approval by Bill is one," as a partisan puts it. Gates' blithe response also strengthened Glaser's belief that Microsoft was intentionally punishing its defector. "There's a kind of aspect to Rob and Microsoft that you have to go back to Greek drama for," Mitch Kapor says.
Glaser himself raises the psychological issue frequently. "We're still dealing with this on an Oedipal level," he says at one point. He starts another salvo against Microsoft with, "In the annals of Oedipal or other obsessions ..."
"Does 'Oedipal' work both ways?" I interrupt.
Rob grows indignant. "No! Our business strategy isn't reactive. Everything we do is to move forward, to drive our business forward, to drive the medium forward."
On July 23, 1998, dressed in a gray suit and speaking in sober yet angry tones, Glaser told the Senate, "I believe Microsoft is taking actions that create obstacles to the freedom and openness of the Internet." Windows Media Player, he claimed, "breaks our product." He demonstrated how attempts to call up RealVideo streams from CMPnet and RealAudio from NPR were met with error messages when both RealPlayer and Media Player were loaded on the same PC. "What Microsoft is doing is wrong and must be stopped," Glaser insisted.
The reaction, though, was not what he or his executives expected. Microsoft officials responded that the first time they had heard of the problem was during Glaser's testimony. Moreover, Microsoft claimed, the flaw was with Real's technology, not its own. The breakdowns resulted from Real neglecting to inform Redmond about changes between the alpha and beta versions of the G2 player. Further, in upgrading RealPlayer, Real didn't change the file extensions. So a Windows player configured for earlier Real data would attempt - and indeed fail - to play the newer material. "That's why we were flabbergasted, because it worked the way we shipped!" says Gary Schare, lead product manager for Windows Media Technologies. "They broke it."
Glaser has put out a few faint feelers to Microsoft since the meltdown. He sent an email to Gates lauding his support for Third World vaccination programs, but received no response. He also sent Ballmer a note congratulating him on his appointment as Microsoft's president, and offered to chat about "moving the RealNetworks-Microsoft relationship in a more positive direction." Ballmer thanked him but, says Glaser, begged off.
Meanwhile, Glaser's Senate testimony, test results, and supporting statements from other technology executives remain on Real's Web site. But Kapor, for one, concedes that Real made a strategic error: "We did not vet and check out the technology issues fully enough, and therefore went out with an exposed position that, while not wrong, wasn't defendable in the terms we presented. That's where we got hung out to dry."
And how. Real's stock plunged from a split-adjusted 22 1/8 a share on the day of Glaser's appearance to 16 5/8 the next day; by last September, it had sunk as low as 8 3/4.
"It was a difficult period," says Kapor, "but I wouldn't say it was a low point. People weren't demoralized." By then, the Realists understood that they were involved in something larger than a battle over clients and servers. They were engaged in a war for platform supremacy.
There had always been a method behind the madness of Rob Glaser's original software giveaway. His intent, inchoate at first, was to create a virtual platform.
"A virtual platform," Glaser explains, "isn't necessarily an operating system, although that can be an embodiment of it. It's looking at all the elements that are associated with a system in a networked world - in this case, a media-delivery system - and trying to assess how you coalesce a sufficient set of them to get critical mass."
Glaser had learned the importance of such standards on a Microsoft project. The Optical PC was a joint venture between IBM and Redmond to craft a cheap desktop computer with sound, video, and a CD-ROM drive built in. But the $1,500 machine that was supposed to ship in the summer of '90 turned into a $6,000 machine that didn't ship until a year later - "a total irrelevancy." Faced with abandoning the project and, with it, Microsoft's hopes for multimedia, Glaser's team took a different path. They seeded the marketplace with CD-ROM software (most prominently the Encarta encyclopedia) that they thought might drive consumers to demand inexpensive multimedia computers, as well as the software standards and development apps to support them.
"We got enough people drinking the Kool-Aid at the same time to create a virtual standard,"Glaser says. "That's why computers today have sound cards, video cards, and CD-ROM drives as standard equipment."
When he started Real, Glaser was obsessed with getting his users to drink RealNetworks' Kool-Aid - even to the point of doing something many inside his firm deemed heretical. He required users downloading the free player software to provide their names, addresses, and email addresses. The idea provoked fierce debate inside his tiny company. The engineers, in particular, considered it rude, a violation of netiquette. Glaser went ahead anyway - "by fiat, not by logic," he says today. "I was influenced by Esther Dyson's argument that in a networked economy, you don't sell; you build a relationship. I believed that if people were inviting us to take up a megabyte on their hard drives, they were inviting us into a relationship."
That thought was validated in '96, when Glaser, again defying most of his senior advisers, tried to "upsell" his audience on an enhanced RealPlayer - essentially the same product, but with a few extra features - for $29.95 a pop. More than a million copies have been sold.
The sales helped Real's profit margins and crystallized Glaser's thinking about business models. To him, a business model is not linear or static, but an almost fluid, three-dimensional object. It may start with a single product, but it rapidly envelops the consumers who use it. As other companies form to serve this audience, opportunities arise for the original firm to service the newcomers. Glaser likes to call this process "bootstrapping" off the "substrate" - that is, leveraging an existing market ever so gingerly into a new one by assembling a growing set of partners around your platform.
True to the model, hundreds of businesses have been built on the foundation of Real's client-server system. In May, RealNetworks' annual developers' conference brought together scores of them - OZ.com, developer of a 3-D streaming animation system; ComedyNet, with its streaming library of stand-up routines; Sonic Foundry, with its multimedia application tools; Virage, the creator of software that can index and search video and still-image data; the GAYBC Radio Network, a webcaster devoted to gay and lesbian topics. Most Real-based businesses are small and struggling. Some are anything but. A firm called AudioNet started up to host streaming audio for sporting events. Its relationship with Real was symbiotic; it spent millions on Real servers and drove countless listeners to download Real's client, which in turn enabled AudioNet to become the first significant Web portal for multimedia content. AudioNet changed its name to broadcast.com and, this spring, was sold to Yahoo! for $5.7 billion.
Toward the end of 1998, Len Jordan, Real's senior vice president for media systems, gave this business model a name: He called it an ecosystem - a term Glaser enthusiastically adopted and voices incessantly. "It's just pushing the ecosystem forward," he says during one of my visits, discussing an acquisition. At another moment, considering a pending deal, he muses, "The ecosystem is shaping up in a very clear way."
It's a canny term that camouflages the darker side of Real's strategy. For while some creatures flourish in an ecosystem, others must die. To keep RealNetworks among the survivors, Glaser has had to ensure that, within its ecosystem, his company sits at the top of the food chain. "No one wants to leave their destiny in anyone else's hands, no one really knows how it's going to sort itself out," explains Bruce Jacobsen, Real's former president. "So, do I want to have no play in a place that might be a huge strategic benefit?"
In its early years, Real was hindered from invading others' turf by its single-minded focus on streaming and its need to protect its server-side revenues. While Glaser began to look longingly at, for example, the audience broadcast.com was attracting with his technology, broadcast.com president Mark Cuban was able to "keep the company at bay" by using its server purchases as a shield, says a former Real staff member. "Whenever Real said, 'We're thinking of going into aggregation,' Mark was good at saying, 'Don't compete with me. I can go over to Microsoft.'''
The introduction of the RealPlayer G2, however, put new energy behind Real's casual rapaciousness. Len Jordan frankly calls the G2 player "the TV Guide and the channel changer for 400,000 sites; when the world becomes 4 million sites, that role becomes more important" - not least because Real intends the G2 to become an essential part of the architecture of broadband multicasting. "Any convergence that takes place will eventually depend upon a standard platform," says Martin Dunsmuir, Real's general manager of emerging technologies. "We want to have G2 players in television sets. We want to have G2 infrastructure components in the ISPs, whether DSL or cable."
These plans have allowed Microsoft to claim a kind of moral high ground. "Real isn't even a portal; it's a cul-de-sac - they don't want you to leave," says Windows Media Technologies' Gary Schare. "Microsoft's model is more pure. Our view is, 'You want Fox News, go into the browser, type foxnews.com, get the video. You, Fox, control the interface.' We're not interested in shoving our brand at you."
"Real," says Pete Mountanos, former CEO of VXtreme, "is driving the content industry into Microsoft's arms. Quite candidly, Rob is acting like the old Microsoft."
But the flak isn't coming just from Microsoft and its supporters. One entertainment-industry executive calls Real's aggregation strategy "a mistake" because "if they want us to create traffic for them so their environment is compelling, at some point we'll be in competition with them."
Todd Wagner, CEOof broadcast.com, says that in his case, that's already happened. "I hate to use such an overused word," he says, "but our relationship with Real has turned into coopetition."
The most ominous peril to RealNetworks didn't emerge from Microsoft or the streaming marketplace, though. Instead it came from Germany, in the form of a compression/decompression standard, or codec, for downloadable audio cumbersomely named Moving Picture Experts Group-1 Audio Layer 3.
Downloading hadn't even been in Real's early strategic plans. The very idea of streaming was predicated on the knowledge that consumers got impatient waiting for large media files to make their way down a pipe. Real had, over time, incorporated downloading into various products - there was CD Streamer, an uploader, or ripper, of tunes from compact discs to hard drives, built by a fellow in Philadelphia using Real's software-development kit and sold on Real's Web site - but Glaser resisted calls to move formally into the downloading market.
Real's customers - especially the record labels that stream music samples to would-be buyers - hated downloadable music. Publicly, they worried that rampant piracy would cut into their revenues. Privately, they fumed over the threat to their economic hegemony if the music industry turned into a click-and-publish free-for-all. So Real stayed away. Angering Microsoft and broadcast.com was one thing; provoking the content owners was an entirely different matter.
But MP3, as the German codec was nicknamed, could not be denied. Thanks to the caliber of its compression algorithm - a CD-quality song that took up 40 megabytes in a standard WAV file was only some 4 megs in MP3 - and with the easy terms under which its developers licensed it, MP3 raged across college campuses in 1997 and 1998. Students began using it to rip CD tracks and trade them across T1 lines. MP3's popularity was soon galvanizing scores of entrepreneurs and audiophiles to manufacture encoders, players, and portals around it.
It all felt uncomfortably familiar to Glaser. "They basically did what we did with our streaming players - put out free copies for anybody who wanted one," he says. "I wouldn't say they had a conscious plan to build a system around it, but it did unleash an organic phenomenon that's created a groundswell."
In other words, somebody else had created a virtual standard and an ecosystem - one that promised to consume RealNetworks, unless it got to the dinner table first. To survive, Real needed to co-opt MP3.
Around Labor Day weekend 1998, Glaser and his senior executives met to plot their strategy. The basic idea was fairly simple: Create a "software VCR" - an old, discarded concept of Glaser's - but with audio, not video, as its foundation. It wouldn't be a grand technological leap: It could be built atop RealPlayer's core. Like RealPlayer, it would be free. Ideally, it would play every major codec, including the dread MP3, so no consumer (and no potential ecosystem member) would get frozen out. And with Real's registered base of 61 million users, the company could guarantee getting the software into the public's hands.
But how would Glaser persuade those consumers to use this amazing new software and lure the record companies to start appealing and selling to this market - the core of any virtual-standard gambit? The "eureka," he says, was to integrate the CD scanner - "to use personal-use rights as the bootstrap" - and build software that would help people upload their own CD collection to their hard drives.
"The reason we were successful with RealAudio was the content," recalls Phil Barrett, Real's senior vice president for media technologies and long one of Glaser's closest colleagues. "We had lined up ABC News and NPR for the launch, and people came and got it."
They would do the same thing again. "Where's all the music content?" Barrett continues. "It's in people's own collections." If the company could persuade audiophiles to make a "behavioral change" and think of their hard drives as music-storage bins, RealNetworks could use MP3 to piggyback onto this new marketplace.
The strategy was the riskiest gamble Real could take. Microsoft had already considered and abandoned the idea of building a CD ripper, because, as one executive in the Windows Media Technologies team says, "we didn't want to come off as promoting piracy." Real was willing to hazard that perception. The record companies would be livid, to be sure, but by the time they or SDMI could do anything about it, there would be too many RealJukeboxes to ignore. Best of all, in their default mode the new players would encode CDs in Real's own G2 format - effectively transforming untold hours of music into something only Real software could translate.
"They realized that every music lover in America already has a titanic content library," says Don Brown of Qradio, Quincy Jones' webcasting venture. "What RealNetworks figured out is that if content is king, your living room is the palace."
The labels, when they heard the plans, were none too happy. To them, it seemed as if Glaser had attacked SDMI. They were quick to point out that they could retaliate. "What would happen if we put our music out, but it only played on our software player and we put those players on each of our CDs? That's my question for Real," says Larry Kenswil, the executive responsible for ecommerce at Universal Music, the nation's top music label. "Why turn this into a war?"
Glaser's response is to deny the problem and assume that the labels will act rationally - his highest praise. "If we can demonstrate that RealJukebox causes more CDs to be bought and more digital downloads to be purchased than would be the case if they didn't work with us, then they will work with us," he says.
In the fall, Glaser and his management team presented the RealJukebox project (back then dubbed Taiko, after the Japanese kettle drums that mimic the sound of rolling thunder) to the company's board of directors. The team was prepared for the usual interrogation. "This time," says Bruce Jacobsen, who was then Real's president, "they said, 'Ship it faster.'"
As RealNetworks' stock price rose past 75 on April 1, a few programmers fashioned a wicked practical joke, writing and installing a program that scanned for a number following the company's ticker symbol, RNWK, and automatically sliced it nearly in half. Anyone searching for Real's stock price was confronted with a bloodbath. April Fool! In reality, the shares kept rising, past 120, and not long before the release of RealJukebox, Glaser mulled videotaping a "Real Rob," one of his frequent streaming videos to the staff, to tell them to focus on their tasks, not their fortunes.
The real Rob, meanwhile, was still busy taking care of politics. He revised a speech he was due to give at the Internet World conference, in which he'd challenge the record labels publicly, for the first time, to embrace MP3. He orchestrated a slew of deals: an arrangement with AOL that would bundle RealPlayer on AOL CDs and let users "instant message" Real content links to their buddies, as well as a licensing agreement to provide software to GeoCities' 4 million members to encode their own streaming content. He even managed a slight rapprochement with Microsoft, getting the RealGuide its own toolbar extension on Internet Explorer 5.0. And he got AT&T's digital-distribution subsidiary, a2b, to license its secure format for RealJukebox.
Meanwhile the ecosystem kept evolving. Liquid Audio and MP3.com filed for initial public offerings. LG Electronics of South Korea publicized its forthcoming release of walkmanlike portables with MP3 playback added to the standard fare of FM radio and cassette play. And Glaser scrambled to complete arrangements with Thomson Multimedia for its own portable device, the RCA Lyra, to be powered by RealJukebox. He did the deal with IBM that designates RealJukebox as the interface for its Madison Project, an SDMI dry run, and bought Xing, the MP3 specialist. All of which prompted one influential analyst, Rob Martin of Friedman, Billings, Ramsey & Co., to predict that Real will break into the black by the end of this year with earnings of $0.38 per share and to set a $150-per-share price target.
But then Microsoft released Windows Media Technologies 4.0 with its download ability, super-clear codec, and rights-management protection. And Casio announced a handheld PC with a Microsoft-powered stereo built in. And Sony did its deal making Redmond's Windows Media Technologies its platform of choice. And Microsoft bought $5 billion worth of AT&T shares, guaranteeing itself a place in a few million broadband set-tops. Apple came after Real too, announcing a free QuickTime streaming system. Taking a swipe at Real's reliance on server-software revenues, Apple plastered advertisements for blocks around the RealNetworks conference in San Francisco that read: "Keep our streams free. No server tax."
The day after Glaser recorded his stay-focused "Real Rob," the stock began a rapid slide that would take it down to 64 a share - before it began an equally giddy glide back up in anticipation of strong earnings announcement and news that RealJukebox had been downloaded 1 million times within 10 days of its launch.
I asked Glaser recently to project how his convergence quest will play out. "It depends upon what paradigm you're on," he answered. "If your view is that this is like the PC operating-system business, where everyone not the winner is a loser, then these coopetition dimensions are relatively nerve-wracking.
"But if, on the other hand, you think it will become more like the media business, where there's ABC and NBC and CBS and Fox, where on Monday CNN can compete like heck with MSNBC, but on Tuesday Time Warner can give carriage on its cable systems to MSNBC, then it's not winner-take-all. It's an ecosystem, where everybody jockeys for advantage but has interdependencies."
As we departed his cramped office, I paused by the aquarium and noticed that among its residents was a clown fish. Striped and placid, it is one of the few creatures that can live within the stinging tentacles of a sea anemone, which it did while I stared.
Later, I sent the RealNetworks founder an email asking him whether there's any symbolism behind his choice of aquatic companionship.
"None that I know," he wrote back. But he has clearly paid some attention to the relationship between the little critter and the deadly medusa. "While other fish stay away, I guess for fear of being eaten," Rob Glaser told me, "the clown fish swims merrily in its midst."
PLUS
