After cutting deals with MCI and Microsoft, RealVideo's Rob Glaser is leading the web to its next level: War with TV.
Huddled with his crew in the eerie half-light of an empty auditorium in Burlingame, California, at some awful hour of the morning, Rob Glaser is running on adrenaline, plenty of it, plus maybe a quart of Diet Coke and a hundred-odd minutes of "rest," little of it sleep. "Aright," he tells his huddle. "Timeforadryrun. Withallofthedemos. Let's makesureitallworks."
Glaser has been through this drill before. His rocketship decade as one of Bill Gates's top lieutenants at Microsoft put him in the public eye plenty. So have his now almost three years at the helm of Progressive Networks, creators of RealAudio, a white-hot Internet technology that gave the once-mute Web a voice.
But this time the circumstances are a bit more challenging, the stakes a bit higher. This is his third keynote in three days (or has it been two days, or four?). The last one was in Tokyo, a dateline and however many time zones away. And this isn't the usual, half-bored trade show audience - these are his people; a cavern full of geeks paying up to US$700 each to hear the Progressive gospel over the next two days. And most to the point, what they've gathered for is no cookie-cutter corporate pep rally: it's a full-throttle induction into the ways of Progressive's pièce de résistance, RealVideo.
The stakes RealVideo raises are huge, not just for Progressive but for the Internet as a whole and maybe for the entire exploding universe that used to be called the broadcast industry. In a home-run scenario, RealVideo could transform the Net into a mass-market conduit for all manner of video content - one that could some day challenge the landed interests of the TV industry and transform Progressive into a kingmaking Internet powerhouse. To this end, everyone in the room - a couple dozen people hunched anxiously over computer monitors, sound boards, lighting gear, and speech notes - is a little on edge. "DON'T CUT THE FUCKING VIDEO BEFORE THE AUDIO," Glaser reproaches after somebody cuts the fucking video a little early. OK, maybe a lot on edge.
The auditorium doors open and waves of people - most of them guys, it seems, most of them under 40 - run/walk for the good seats. Many are Web developers, devotees of the black art of streaming media. Others come from TV and radio stations that are starting to pump their offline content onto the Net. There are entrepreneurs, hoping to create the ultimate transmedia jukebox or maybe the next MTV. Plus, there's the inevitable assortment of journalists and investors, trying to make sense of it all.
Soon enough, Glaser is at the podium. His speech is titled "The Internet as the Next Mass Medium," and he's belting it out. "We're gonna show all those people," he says, "the ones who say, 'Hey, the Internet's not gonna scale up,' or 'Gosh, there's gonna be bandwidth limitations.'" It almost sounds like a taunt. "We have the technical know-how, collectively. We have the resourcefulness and the commitment." And the crescendo: "You talk about the 15 hours a week that people spend listening to the radio, or 20 watching TV - four or five years from now we'll talk about the 10 or 15 hours a week that people spend experiencing audiovisual information over the Internet. It'll be just a standard part of the media fabric of people's lives. That's the kind of scale we're talking about here."
Glaser is intent, a blur. He machine-guns his words, pumping his arms as if jammin' on the air accordion. RealStuff flashes and dances on the big screens at the front of the room: a RealVideo newscast from Bloomberg News, synchronized to a stock ticker and an external audio track. One of three short films by Spike Lee (Spike Lee!), shot expressly for RealVideo's launch, this one featuring a young tap dancer doting over his shoes ("I like to call them weapons, you know what I'm saying?"). Clips showcasing the RealMedia Architecture, a framework for making all sorts of Net-borne media work together (and yes, allofthedemos do work). And for the finale, an animated sequence that portrays a hapless conference attendee who gets abducted by aliens. (Hey, it could happen.)
Hardened Net skeptics would find plenty to snicker about - teensy screen sizes, grainy pictures, jerky motion. But this isn't about re-creating Blockbuster Video online overnight. The aim, Glaser says, is to create "something that's good enough to criticize," borrowing researcher Alan Kay's famous quote about the 128K Mac. In other words, 1997 won't go down in history as the year interactive television burst full blown from the Internet's womb. Rather, Glaser explains, it will be for Net video what 1995 was for audio - "the year this really all starts to happen and to come together."
The analogy is a fundamentally bullish one. When RealAudio débuted in April 1995, it was an ear-wearying squall, compromised enough to squeeze through a 14.4 modem. Eyes rolled. But those who dismissed it sneered too soon. Today's RealAudio 3.0, on just a 28.8 modem, can compare with FM radio. Over a broadband connection, it can approach the fidelity of a compact disc. At the same time, a textbook-worthy marketing push has made RealAudio the Net's de facto audio standard. It's running now on 10 to 12 million desktops and thousands of servers, transmitting everything from live NBA games and ABC newscasts to the burgeoning output of garage-based guerrillas like tiny Terraflex of Burlingame, California, whose six overworked employees at last count were streaming more than 50 channels of music onto the Net round the clock - more full-time feeds than any but the largest radio holding companies.
But with RealVideo, Glaser is raising the ante. Television is our most ubiquitous, hyped, and heeded medium. It's Moby Dick to radio's school of minnows, drawing more than $35 billion in ad spending annually in the United States, plus billions more around the globe. But more to the point, TV is now on a collision course with the PC - a modest little $125 billion industry. Companies on both sides of the fence are jumping into the convergence game, bidding up the stakes to dizzying levels. There was the reported $80 million from ABC for Web-weavers Starwave, an estimated $100 million investment from CBS into SportsLine, then two ground-shaking Microsoft deals: $425 million for WebTV and $1 billion into cable giant Comcast. Yet for Glaser himself, the biggest were still to come: July's announcement that none other than his old boss Bill Gates had agreed to embed RealAudio and RealVideo deeply into Microsoft's Internet product line and to take a good-faith 10 percent stake in Progressive Networks as part of the deal. And a second deal announcement two weeks later, with MCI, creating the RealNetwork, a nationwide multicasting backbone that will help resolve the biggest problems that Internet video faces: scarce bandwidth and server overload.
Nobody - not Glaser, not Rupert Murdoch, not even Chairman Bill himself - knows exactly how the convergence game will play out. But Microsoft's adoption of RealVideo is about more than just a move closer to an Internet version of TV's NTSC standard: it's about bridging the shrinking gulf that separates computers and television. What Gates is buying into is state-of-the-art compression technology that squeezes fat, data-rich video signals through the Net's cramped capillaries by carefully removing nonessential data. As that technology improves, as processor speed increases and bandwidth widens, Net video's quality will soar. And as it does, jerky little postage-stamp video patches will blossom into full-screen, high-definition real video - skirting the cluttered airwaves and old-fashioned cable toll roads entirely.
All this could wreak havoc with the competitive dynamics of the TV industry, opening millions of living rooms to any hyperactive kid with a video camera, a computer, and a phone line - which may soon mean most kids in America. Or it could give the likes of Murdoch and Ted Turner an eight-lane highway into the heartland of the traditionally democratized, decentralized Net. Or maybe do both.
RealVideo's teensy, grainy, jerky pictures aren't going to do any of this in 1997. But when the history of post-television media gets written, at least a footnote should be reserved for Glaser and Progressive's superbly executed march on the market: Focusing first on distributing audio over the Net - a far simpler trick, technologically - then springboarding into video by converting an already critical mass of established users. Doing loss-leader content deals with audience magnets like ABC and NPR. Going engineer-to-engineer against Microsoft's own streaming audio and video product, NetShow. And finally winning (or was it re-winning?) the notoriously fickle allegiance of his old boss - instead of being bought out at a fire sale price or crushed, the usual fate of would-be Microsoft competitors.
After the speech, with sleep deprivation in full retreat, Glaser's adrenaline levels - and maybe a splash of Diet Coke - have him crackling. "This is big," he says, barely restraining those accordion arms. "So big." What he's talking about is not just the conference (although it's plenty big) nor RealVideo (although he hopes it will be, too). Rather, it's the relentless spread of the mass-market Web, the steady expansion of bandwidth, and the market-warping, at times contradictory, transformations those changes are unleashing. It's the rise of a medium that's both as personalized as a shopping list and as broad-reaching as Seinfeld. It's the grassroots invasion into Big Media and the concurrent Big Mediafication of the open Internet. It's the convergence of TV and online networks, the rise of push in a pull-based medium, and, above everything, the intoxicating fact that we're still just three years into a twenty-year phenomenon.
Glaser can go on about this at auctioneer velocities - for days. About how the coming media landscape will be transactive: "You can show a commercial, and let someone order the product or service being offered straightaway. Or you can play a song and let them order a CD." That it will be on-demand: "You can see or hear something when you want, not just when it happens to be broadcast." And, perhaps best of all, how it will have no bounds: "Content will no longer be limited to 3 networks or 36 cable channels, but rather by production economics."
Glaser sets his empty Diet Coke can down next to, well, another empty Diet Coke can, and prepares to sum up. "Now I look at all those changes," he explains, "and I say, 'This is about to have a bigger impact on electronic media than anything since we went from radio to TV. And even then, the TV industry took on so many of the structural characteristics that radio established before it: national networks, advertising interjected into the content, national programming.'
"So this is bigger. Much bigger."
A few days later, a very different group gathers amid the posh sprawl of the Inn at Spanish Bay in Pebble Beach, California. This is the Futures Summit of the National Association of Broadcasters, an élite retreat for a few dozen CEOs and other ranking barons of the airwaves. There are a couple of network types on hand. But the summit's real pillars, like those of the NAB as a whole, are the station holders - the lucky heirs of the great American spectrum handout, who each control anywhere from a fistful to a boatload of local stations.
From the looks of the summit's agenda, the NAB's future is not, well ... broadcasting. Although the crowd comes straight from the platinum-plated heartland of Old Media, the sessions and seminars will consider not spectrum law, not prime-time audience fragmentation, not the confounding popularity of news magazine shows, but... the Net. Presenting over the next three days will be a veritable chorus line of Internet somebodies. Marimba's Kim Polese is on the agenda, as are the CEOs of soon-to-be Microsoft fiefdom Web TV, ecommerce might-have-been-giant Cybercash, and one of Glaser's direct competitors - Asaf Mohr of VDOnet. Ghost of Apple (then-)Present Gil Amelio and Ghost of Apple Past John Sculley also have airtime, as has Lord of the Skies Rupert Murdoch.
This isn't the first Net-centric NAB summit - that happened way back in early 1996, when the notion that the Web and TV were converging was still downright avant garde. That might seem to indicate that the NAB has a remarkably clear crystal ball, for an old boys' club. But broadcasters have long had strong Darwinian motives to keep an eye on television's Future - they've been positively stalked by it for decades. First that Future was cable. Then it mutated into another un-broadcast form: interactive television, one of the decade's better-publicized flameouts. Then, for a brief, cushy interlude, there was no Future. And then came the Internet.
Glaser is slated to present his take on this latest Future during the conference's second day. He strides into the inn the preceding afternoon, checks into his room for a manic series of phone calls and emails ("playing a bit of catch-up from the weekend," he explains, although it's still Sunday), then heads down to a cocktail-'n'-dinner reception in a posh Spanish Bay clubhouse. Here the food is abundant, the drinks unlimited, and the dress decidedly uncasual.
The buttoned-down setting might seem unlikely for a former United Farm Workers leafleteer and stalwart on the board of left-leaning, anti-establishmentarian Mother Jones magazine. (Glaser's company isn't called Progressive Networks for nothing.) Still, he's all charm and conviviality at this Establishment to-do - and there is no artifice in it. Friends say that Glaser has always been a natural at cohering with the people around him, and that tendency is every bit as him as the restless blurs of words that he emits. But more to the point, however big the transformation he seeks to catalyze - bigger-than-cable big, bigger-than-television big - Glaser does not believe that the ancien régime tithe-earners gathered here are necessarily doomed. To the contrary. "Physical community," he explains, "doesn't stop being important, even as virtual community becomes more accessible. Local issues matter. Local content matters. Local broadcasters aggregate local content and advertising, so I could certainly see them continuing to matter a lot, even if the infrastructure that delivers their content changes."
It's an upbeat take, very much in keeping with the summit's general tone. Broadcast and the Net are friends, complements, sharers of the same Future. Right?
Well... sort of. It is true that some synergistic and profitable broadcast/Net allegiances have already been forged. At last year's summit, for instance, Granite Broadcasting president Stuart Beck got to talking with Yahoo!'s Jerry Yang, and the result was a partnership that both sides still glow about: Yahoo! designates Web sites that Granite's nine affiliate stations can point to for further information on news stories. It is also true that the broadcasters' principal bedfellows - the major networks - are positively firehosing cash into the Net. An ardent broadcast booster could probably even find a positive spin to put on Microsoft's acquisition of WebTV ("It'll get them away from their computers and put them back in front of their TVs, right?")
But much of this can also be seen in a downright somber light. Sure, WebTV might pull some Netheads back to their living room couches. But should it become a mass-market hit, it will also give millions of TV viewers a lot more to gaze at than Live at Five. But broadcast's real nightmare isn't Web-ready televisions piping in HTML flip charts. It's not Java gamelets or stock quotes or static news headlines. The nightmare is full-screen, full-motion video. Seinfeld-on-demand. Friends-on-request. Not just 24-hour, but perpetual headline news, where it's top-of-the-hour whenever you hit that button. And all of it, maybe, brought to you via RealVideo.
Some forward-looking TV industry heavies are already taking the first steps down this path: using RealVideo and the Net to burrow through to audiences that can't be reached via coaxial cables or transmission towers - at the office, say, or on the far side of the world, or in a cable market where their network doesn't have a carrier. Murdoch's Fox News, a young CNN rival with cable distribution minimal enough to make even MSNBC feel good, can now be picked up by anyone with a 28.8 modem and a downloaded copy of Progressive's free RealPlayer. So can legislative superstation C-Span. So can Trinity Broadcasting Network, "the world's largest Christian network." And KWMJ of Tulsa, Oklahoma ("family-safe programming"). Bloomberg News is using it. So is a growing list of newly global stations and national networks, from Brazil, Chile, Taiwan, Spain, France, Hong Kong, and others. It's not Friends-on-request - yet (though it could be, if an NBC affiliate somewhere decides to join the list). But Glaser is happy to provide a pointed subtext: "Anyone who underesti-mates the depth of the transformation already under way will be out of business - or, at the very least, underserving the assets they're associated with."
Diehard skeptics can take heart from those teensy, grainy, jerky pictures - just a few inches square on most monitors. But recall RealAudio's improvement trajectory, then extrapolate - and factor in geometric improvements in bandwidth, as deep-pocketed RBOCs, cable systems, even electric utilities lunge for slices of the broadband pie. How the bits arrive - cable, fiber, satellite, xDSL phone lines - doesn't affect the need for compression technologies like RealVideo. Indeed, no matter how huge the pipes get, there is never such a thing as too much bandwidth. So its the technology that becomes the compression standard, that could find itself with a bombproof niche in the Internet's substrate.
But to broadcasters, better and better Internet video looks like the ITV nightmare all over again. Only this time, matters will be far worse, because there is no reason the boundless Net will confine itself to an ITV-style menu of movies and canned programming. There will be live feeds (as there already are), be they 3-hour sporting events or 24-hour simulcasts from the other side of the globe. Reruns of popular shows will be available continuously - not just when it suits some broadcaster's whim. Even first-run shows will be accessible on demand - perhaps pay-per-view, perhaps ad-supported, but either way, no more bounded by broadcast schedules than rented movies are by theater showtimes. The NAB's worried transmitter holders have already been undermined plenty by hundred-channel cable snaking beneath our feet. How much more diminished will they be when 3,000-channel pipes - hell, an essentially unlimited number of "channels" - start sliding into our living rooms and dens? Pipes bearing signals from countless little network affiliates pumping from hundreds of little three-transmitter towns (hmmm, they all seem interchangeable, don't they?) and however many millions of garages and home studios?
None of this necessarily means the Death of Television. Indeed, Glaser believes that many of today's TV industry heavies - at least those with strong original content or coherent, well-branded themes - could ultimately thrive. "I'd rather be ESPN than the USA Network," he explains. "Networks like ESPN have deep identities - it ties together in a way that makes sense in a consumer's mind. But something like the USA Network or maybe TNT is a generalized entertainment franchise. Shelf space alone can be enough to build an audience when shelf space is tight. But shelf space per se becomes much less interesting when the shelf expands." And when choices are abundant, people tend to be drawn less by shallow, one-size-fits-all offerings. They want deep content that speaks to them directly.
So who benefits? Glaser isn't shy about predicting: "Aggregators who are all over sports, or all over news, or all over music, or kids stuff, or French impressionism, or cooking, or something that may not even have an economic model in the 36-channel world, but which has a deep and committed community - that's who will do great." At the same time, the "high-tonnage aggregators," the major national networks, may well hang onto important positions, simply because of their skills at funding original content. But lesser networks and content-poor local affiliates will be squeezed. When the dust finally settles, Glaser says, "there will still be broadcast television. There will be cable television, and there will be satellite. But the underlying consumer mechanism for delivering choice will be IP-based. There's no question about it. There is no competitor."
Four years ago, longtime Washington, DC, politico David Halperin was Glaser's charter partner in the project that eventually became Progressive Networks. They meant progressive in the vintage sense of left-leaning political activism. "It's fair to say that our original objective was social revolution," Halperin says - smuggling Fox News into CNN neighborhoods. But then, everything mutates on the Internet.
Glaser traces his interests in technology, mass media, and social revolution back to a modest neighborhood in Yonkers, New York, just two blocks north of the Bronx. His mother, a psychiatric social worker, worked with inner-city children. "For every kid that she was able to get through to, she'd see 50 she couldn't reach," Glaser recalls, "and meanwhile there'd be three or four thousand that she didn't know about." She and his father, who owned a local printing company, had no problem with extracurricular activities like leafleting for the UFW and organizing protests like 1982's massive No Nukes rally in Central Park.
Glaser was bused down to Fieldston, a private K-12 day school and arty bastion of Upper West Side New York liberalism, complete with an Ethical Culture curriculum. ("Unitarianism with even less God," says Glaser.) As a protogeek from Yonkers, he got some early training at Fieldston in what he calls "shifting gears between very different milieus of people" - his hip-effete classmates and neighborhood friends at the local bowling alley ("where they'd probably beat the shit out of you if you even used a word like milieu," he says). Glaser likes to compare himself to Woody Allen's chameleonlike antihero Leonard Zelig - an analogy old friends echo verbatim. Fieldston also gave him his first taste of the power of mass media: a "radio station" he and some friends set up that ran over the school's intercom. And it was at Fieldston that he first found happiness as a hacker, starting in fourth grade when he was sent off to play on the upper school's old DEC PDP-8 minicomputer - not exactly a fashionable activity back then. "I had a level of obliviousness that was helpful," he says. "I was so focused on what I was doing."
That focus propelled Glaser to Yale, where he served a long stint as editorial-page editor of the Yale Daily News, maintained a column called "What's Left," and ran a rather pissed-off organization called the Campaign Against Militarism & the Draft - all while putting in serious hours among the reclusive inhabitants of the CRT-lit computer labs. Cured of big-name technology by a summer job at IBM, he and three friends started a videogame company, self-consciously named Ivy Research. They went on to release two racin' snake PC games - Slynx and Viper - that were distributed nationally through Computerland. But by graduation everyone was ready to move on, Glaser holding his own personal monument to overachievement: three degrees (a bachelor's in computer science, and a bachelor's and master's in economics) in four years. He was 21.
For a while he thought seriously about working as a UFW organizer, at $10 a week. But geekdom won out, specifically a chance to move to Seattle and work for a guy who'd left Harvard eight years earlier with no degrees at all. Microsoft in 1983 was still a fine haven for a subversive techie - a feisty pip-squeak with fewer than 300 employees going up against white-shirted IBM.
In Redmond, Glaser showed an early appreciation for the levers of power. He chose marketing and management over programming, he says, "because I didn't want to create a product that would get screwed up because the messaging was wrong or because the customers weren't listened to." He managed Word, the King Kong of wordprocessing, then moved into the company's networking group. Mike Slade, now CEO of Starwave and another early Microsoft lieutenant, remembers Glaser as someone who "just stayed there all night. He would never go home. Even when the rest of us knocked off at 9 or 10 o'clock. He was always there." He also had a computer at home, Slade says, "which in those days was practically like having your own mainframe. I mean, a home computer. It was like, 'What a geek!'"
Glaser did well enough to be named Microsoft's vice president of multimedia systems - in effect, a direct report to Gates himself. He did well enough financially to be able to buy a small piece of the Seattle Mariners. But 10 years into his Redmond sprint - and having just turned 30 - he got off the treadmill. Glaser says his departure stemmed from a very personal need to "up periscope." As he explains it, "My job satisfaction had gone from 90 percent to 70 percent," adding that "I'd kind of gotten spoiled early on, when Microsoft was the David to the IBM Goliath." Some Microsoft alums say a power struggle with Gates confidante and corporate macroheavyweight Nathan Myhrold played a part. (Glaser left Microsoft "a little bit angry," one longtime colleague recalls.) But Slade puts a forward spin on Glaser's departure. "If you work at Microsoft," he says, "you want Bill to approve of you because he's so capable, because he's so difficult to please, and because he's so smart. And making it outside of Microsoft, doing something successful on your own, is the ultimate way to do that."
Officially on leave from Microsoft, Glaser upped periscope in the spring of 1993, then spent a couple of months drifting through Germany, Greece, and Egypt, thinking about what to do next. A summer course on the history of consumer culture at the University of Washington cured him of academe. He considered political activism and joined five nonprofit boards, including the Electronic Frontier Foundation and the Foundation for National Progress, which publishes Mother Jones. But Microsoft had spoiled him. "I was used to the kind of leverage you get when you enlist the power and incentive structure of the private sector," he says. "That made the incrementalism of the not-for-profit world kind of frustrating."
Glaser was left pondering a thread in his life that ran back to high school, what he calls "the nexus of media, computing, and communication" - an area that in 1993 was suddenly red hot. Glaser was particularly intrigued by the idea of harnessing new-media technology - even relatively old but underdeveloped media, like cable television - to generate and distribute "progressive" content. "People who pursue progressive agendas in the social or political sphere are often downright Luddite when it comes to technology," he says. So, instead of choosing between technology, media, and social causes, he would work in all of them.
The result was the seed of a company grandly dubbed Progressive Networks. In October 1993, Glaser was joined full time by Halperin, another Yale politico, but one who had followed the path his geekier friend forewent: not-for-profit work, law school, a staff job on Capitol Hill. Glaser's Microsoft winnings funded the effort. Pooling their connections, the pair met with media industry figures in New York, Los Angeles, San Francisco, Seattle, and Washington. Their agenda was constantly evolving - or, more bluntly, continually vague. One possibility they considered was building a "progressive" cable channel, or, perhaps just some cable programming. But what Glaser calls TV's "distribution hammerlock" was just too daunting. Meanwhile, the economics of another much-heralded new medium - CD-ROM - were proving downright dismal. Throughout it all, Glaser kept up his nonprofit work, did some consulting jobs for Microsoft, and watched his Mariners play ball.
That same summer, one of Glaser's EFF connections steered him to some interesting new software, a noncommercial Internet browser called Mosaic. Glaser downloaded a copy, launched onto the World Wide Web, and had a not uncommon reaction: "a total revelation," he says.
The 1993 Web was not much to look at, much less listen to, or "watch." A few technically savvy cliques were already using the Net to distribute audio. (Remember Internet Radio's "Geek of the Week" ?) But the download times were horrendous - a five-minute clip could require half an hour of thumb twiddling up front. The problem reminded Glaser of a demo he'd seen years before in Redmond: low-bit-rate audio distributed over an internal network. And that produced an idea: through compression technology, it might become possible to "stream" an acceptable grade of audio content over the Internet - click on a selection and have it play more or less instantly, in real time. Radio, in other words, only with the listener in control. The Internet could then be used to distribute ... who knew? Books on tape? Delta blues? Social revolution?
Glaser was sufficiently intrigued to dip back into his Microsoft stash and put three contractors to work on the problem of transferring sound across the Internet. His initial hope was that there could be "some real connective tissue" between the audio initiative and his and Halperin's would-be media project, which continued to not take off. The early demos were not flashy: Captain Picard from Star Trek: The Next Generation intoning, "Tea. Earl Grey. Hot." A baseball broadcast. The theme song from a cartoon called Animaniacs. But they pointed clearly to possibilities far larger than a politically correct cable show. And, as Halperin recalls, "Even the public-interest types seemed much more excited by audio running over the Net than by our grander social and political vision."
Building software instead of pursuing a partly formed media agenda had the added benefit, Halperin notes, of making Glaser "look less like he was Don Quixote and crazy." But Glaser himself was also convinced that the Web "was a once-in-a-generation thing" - much like the PC a decade before. And that gave even more appeal to the idea of a foot-to-the-floor, wildly-for-profit software company. "It used to seem like, how could you do software again and have the same kind of impact that you'd had at Microsoft," Glaser explains. "But once it was clear that the Net was about to change all the rules, suddenly you could have a broad impact outside the context of a big company." Not to mention, a little vindication.
Halperin went back to DC, while Glaser - still funding everything himself - staffed up his contractor force. By early 1995 he had nearly two dozen people on board and was ready for outside financing. Video was not even mentioned, at least not publicly. "There was an overwhelming impedance mismatch," Glaser says, "between the existing consumer-infrastructure 486-based PCs and 14.4 modems and the requirements of video." His most advanced demo featured an audiocast of a baseball game, synchronized to an animated diamond showing the positions of runners and the score. The primary targets of Glaser's fundraising were private individuals with a deep understanding of technology, rather than funds or institutions; the round was ultimately led by Glaser's EFF boardmate, Lotus founder Mitch Kapor.
Video remained unmentioned when Progressive launched its first product suite at the NAB's massive annual conference (not to be confused with its élite summit) in Las Vegas in April 1995. RealAudio1.0 gave away fidelity to squeeze through a 14.4 modem in real time. Music - charitably called "AM quality" in the promo material - was a torment. But more significant than sonic fidelity was the fact that RealAudio worked - it delivered voice on demand: the Net spoke. And for anyone with longer vision, it pointed the way to richer streaming media - music, multimedia, and eventually video.
Glaser's strategy with RealAudio was no secret: push the technology, bring top content partners aboard, and turbocharge the user base - from 1 million at the end of 1995 to 10 million by October 1996 and more than 20 million today. Competition was quick to follow.
Xing Technology was second to market with StreamWorks, released six months after RealAudio. DSP Group followed with TrueSpeech, then Macromedia with Shockwave audio. All were close to RealAudio technologically, in some respects even superior. But by being first, signing up more compelling content, and distributing faster and farther (including with every copy of Internet Explorer ever shipped), Progressive held onto an overwhelming market share - now more than 85 percent, by credible internal estimates. And as Microsoft's own history shows, iron market grip can trump even a serious technological edge. Remember the Mac?
The move to video was more complex, in part because the shimmering lure of TV make the stakes intrinsically higher. Pioneered by Xing in late 1995, the Internet video market was soon contested by a pack of hungry start-ups - VDOnet, Vivo, VXtreme. But most of the software simply did not provide satisfying experiences over consumer-class modems - with the dubious possible exception of the inveterately early-adopter "adult" media crowd. Internet video was a market, Glaser says, "that was continuing to not take off." It was at best a fight for mindshare - to the extent that anyone cared.
Not that it would stay that way forever. With the rapid spread of Pentium-based computers and 28.8 modems - not to mention far faster corporate connections - the impedance mismatch that Glaser had once identified was starting to close. In December 1995, while attempting to vacation in Hawaii, Glaser got some email from a two-person San Francisco company, FreeVu, which had an Internet videoconferencing tool under development. Glaser took a look, was impressed, and persuaded FreeVu's principals to sign on as Progressive employees. RealVideo's development effort had begun. Technology was only part of the problem. "You had to convince people that what you had was really better than what your competitors had," Glaser says. "But you also had to cross the threshold that got people to care about what you were doing at all. When a company like ours tells people that now is the time to care about video, and it turns out that now is not the time to care about video, we've hurt ourselves. So we looked at those two issues very carefully. And it turned out that the far greater challenge was to have something good enough to care about."
Progressive once again executed well, developing a product that key reviewers saw as superior and wooing users with high-profile content, including C-Span and Fox. RealAudio's brand-recognition obviously didn't hurt - when the first public betas of RealVideo went out in February, its older sibling was already on an estimated 15 million desktops. But the real tipping point was Progressive's ability to point RealAudio's massive user base - clients and servers alike - directly to RealVideo, making it, in effect, an instant default. As Dave Dadekian of Fox News Internet Services puts it: "Someone will come out with a slightly better product, then someone else will, and it will probably evolve constantly. But you don't want to require your users to download 40 different plug-ins just to keep up. And you don't want to be constantly changing encoding tools and servers on the back end. So it makes sense to stay with a single, widespread technology."
By the time RealVideo officially launched in February, it looked like Progressive's most serious competition would come not from Internet video pioneers but from Glaser's old friends in Redmond. Though RealAudio was still being bundled with Internet Explorer, Microsoft by then had geared up NetShow, its own streaming audio and video technology. NetShow didn't look or sound as good as RealVideo. And months after its début, it still boasted only a trivial content base - a handful of radio stations and three live video feeds: a SharkCam from Sea World, a SurfCam from a shop in Florida, and a RainCam from Redmond. What it did have was Microsoft's fearsome distributive muscle.
Glaser was not spooked, in part because of the long list of Microsoft business units Progressive was already working with. "The fact that there are 100 people at Microsoft working on streaming media," he would say, "is not the same thing as a Microsoft jihad."
Still, there were disturbing signs. Months after Progressive rolled out RealAudio's much-improved 3.0 version, Internet Explorer was still being packaged with its relatively dismal-sounding predecessor. Worse, upon downloading Explorer, some users noticed that their RealAudio 3.0 software was actually erased and replaced with the 2.0 version. Honest bungling? Or a ploy aimed at driving users over to NetShow? Soon rumors were circulating that Progressive would be dumped from Explorer altogether.
On a Thursday afternoon in June, Glaser's voicemail fielded a call. "Tell Rob Glaser," a muffled, foreign-sounding - and anonymous - voice said, "that Microsoft is about to acquire VXtreme."
VXtreme? This was no download bugaboo - this was a serious threat. VXtreme was one of Progressive's best-regarded competitors. Technologically, it had some neat tricks, including a quarter-screen display mode at high bandwidths. And unlike most of the rest of the pack, they had lined up some heavyweight content partners, including CNN Interactive and CNET. Glaser did consider the possibility that the whole thing was a Microsoft ploy to scare him into selling. But he quickly concluded: "There were plenty of far more direct and less elaborate channels they could have used to get us to the table."
Glaser phoned his old boss, Paul Maritz, Microsoft group vice president for platforms and applications, and asked if the rumor was true. Maritz confirmed it. "He told me," Glaser says, "that streaming media was becoming such a strategic area, 'Microsoft felt it had to own it.'" Maritz also said that it was their understanding that Progressive 'was not interested in being acquired.'"
That was certainly true. Glaser's stated goal was - and is - to be an independent player on the Internet software scene. He could count on strong relationships throughout Microsoft; he still caught up with Gates himself every few months. Progressive had the market momentum. And possibly the superior technology. So while Progressive and Microsoft had never discussed an equity deal, Glaser felt confident that common ground could be found. The two sides held their first meeting on a Friday, a week after the initial call. By Saturday night, there was a draft agreement. By Monday, they shook hands. On Tuesday, the lawyers signed off.
The terms of the agreement allow Microsoft to fully license Progressive's client and server technology, for undisclosed - but "very significant," as one Progressive source puts it - fees. Microsoft also agreed to bundle the latest RealAudio and RealVideo client software in future releases of Internet Explorer, and Progressive's servers with NetShow developer kits. Microsoft has the right to build upon Progressive's technology but must license back to Progressive any improvements it makes. Microsoft is also forbidden from breaking compatibility between its products and Progressive's - a lesson Glaser learned from Sun Microsystems' ongoing nightmares over Microsoft's licensing of Java. Finally, Microsoft bought a 10 percent nonvoting stake in Progressive. Neither side is commenting on the price, but a knowledgeable source puts it at $25 million.
The Microsoft deal, announced in late July, gives RealVideo a secure place in Internet Explorer, meaning a secure place in Windows - and thus, the PC desktop. To the extent there had been a question about which streaming media product would be the Internet standard, "I think we've flattened it in a very meaningful way," Glaser says.
But there is another question: Who really controls RealVideo? Progressive still owns it, but its giant partner now has access to much of the source code. Microsoft can create fancy new streaming-media products and give them away. It can run in its own direction using its old NetShow technology as a base. And it can acquire new technology from Progressive's still lively rivals. Indeed, the ink was scarcely dry on Glaser's signature when rumors began circulating that Microsoft was on the verge of acquiring VXtreme after all. Glaser was taken a little aback. "My God, what are they going to do?" he said, when asked about all the Internet video technologies suddenly in Redmond's war chest. "Glue them all together?"
The VXtreme deal did in fact go through, in August; Microsoft took the whole company, acquiring a key technology that adjusts streamed video to fit the available bandwidth. But Glaser, a team player - at least a 10-percent one, now - is philosophical: "This isn't a zero-sum struggle for some static, zero-growth market. And it isn't about Microsoft fighting for a slice of a few-hundred-million-dollar server market. It's really about television transitioning from a broadcast to an online form - and that's a hundred-billion-dollar opportunity."
He's right, of course. But a lot needs to happen before that transition can really begin - before anybody will take Internet video seriously. Above all, its image quality will need to improve - dramatically.
Glaser believes this will happen far more quickly than most expect - that by the end of the current decade, on-demand VCR-quality video will be streaming into at least some living rooms over the Net.
He cites three forces that will make that happen. The first is Moore's Law: consumer-class computers will have the processing muscle they need to draw richer and richer video streams from a given connection.
A second reason is the forward march of bandwidth. As Glaser points out, even if bandwidth-per-dollar increases at only half the rate of Moore's Law - doubling every three years, rather than every eighteen months - the two effects together "will be additive and very powerful." And the fact is that bandwidth could increase faster - a lot faster. Cable-borne networks like @Home are already bringing the Internet into homes at speeds hundreds of times faster than today's standard fare. Time Warner is rolling out its rival Road Runner service, while telephony companies are prototyping xDSL technologies that may offer equal or even higher speeds over plain old copper. And finally, Microsoft's billion-dollar investment in cable giant Comcast both validates the market and puts TV on notice that high tech's übertitan will not be sitting on the sidelines.
Layered on top of all this will be what Glaser - still a marketer at heart - believes will be the most significant force: the billion-dollar race to put Net-ready video receivers in front of the world's couch potatoes. Microsoft's WebTV is simply an opening shot - Glaser expects to see "some really, really interesting competitive dynamics" as the consumer-electronics industry, cable companies, and telcos all pile on. "If you think Steve Case was in a shoot-out over online access, wait until you get into the battle of the TV attachments," he says. And the hardware, he points out, is just a sideline: "Those devices will be wedded to a particular set of network services. So if you get your device out there, you'll have a recurring revenue of, say, 20 bucks a month from access fees, plus ad revenues, plus pay-per-view fees, plus home shopping revenue on top of that. People will look at that and say, 'Holy shit! What am I doing screwing around, trying to get $300 for a device?' It's like a cell phone - get a subscriber, you get a lock-in."
The result could be "an artificial acceleration of Moore's Law," Glaser predicts, as the awesome value of playing gatekeeper in a turborcharged, Net-based TV-and-every-other-kind-of-media industry jams the field with major players, unleashing a torrent of cross-subsidy dollars, scaling production volumes, and industrial-strength R&D. "The price per mips could then drop faster on the set-top than on the desktop. I give that a 60 percent to 70 percent chance. And if it does, wow - hold onto your hat." Even if it doesn't, he says flatly, "there is no doubt that after the turn of the century, every new TV that is sold will be an IP device."
Of course, the maximum Web=TV scenario injects a certain dramatic tension into the Net - a medium cherished by many of its pioneers as the antithesis of centralized Big Media. After all, technologies like RealVideo do ultimately promise to hand the Net's keys - or more precisely, blow open its doors - to Murdoch, Malone, and other deep-pocketed media barons, whose production and marketing budgets dwarf those of any pure-play Internet content company.
And what happens then to the once-imagined democratized legions of a billion publishers? Will garage-based Java jocks have a prayer of being seen through the all the flashing lights and connecting with an audience? Glaser believes they will - in part because what we now call TV will undergo an even more profound change: "It will be much more wide open, much more decentralized, much more personalized, with a range of choices much more like what you see on a magazine stand than in a world where John Malone or your local cable operator is the gatekeeper."
So the big guns aren't necessarily in trouble. "If Rupert Murdoch is adding value because The X-Files looks great and has great storytelling, great actors, great pacing and suspense, then Murdoch will hang onto his position," Glaser says. "But if he has his position only because he's crowding out people who can't get onto the broadcast or cable shelf, he'll be diminished. Likewise, if Suck.com is funny, irreverent, provocative, and interesting, those attributes won't be diminished by the emergence of more audiovisual content on the Web. And as Suck.com becomes bigger and more popular and better, the people who run it can decide for themselves whether to add A/V richness, or whether that's really the antithesis of what they're doing."
So there it is: the Net will let players on both ends of the spectrum find avenues to audiences that don't yet exist. Garage-based guerrilla video artists will be able to build global followings. And Rupert Murdoch will no longer be at the mercy of Ted Turner or any other pipe-owning rival. Content producers of all stripes win big. So do consumers, who will enjoy an explosion of content and an empowering shift in its accessibility.
Forms of content that either can't get carriage in the present distribution network - or literally can't exist within it - will flourish. Glaser commissioned the three Spike Lee shorts in part to prove a point: the film short, a powerful form with a vibrant past, has virtually no present, because it lacks a mass distribution channel. We no longer see shorts at the movies, we see marketing-driven previews. They don't fit TV timetables, outside of MTV. But the form resonates perfectly with the jumpy attention spans of media surfers. Not to mention the lower-than-Spielberg-class budgets of most new media producers. And on the Net, the space for them is infinite.
Some old forms will have problems adapting. One likely casualty is the notion of networks as linear, round-the-clock processions of programming. Streamed content is infinitely accessible. And that favors what might be called the "branded nonchannel" - an aggregated batch of content that can be viewed at any time, in any order. CNN and C-Span's video-enhanced Web sites are early prototypes. Another might be the ultimate Star Trek "channel" - any show, including spinoffs, available whenever you summon it. The result for any one programmer could be far more - or less - than 24 hours of transmissions a day. ESPN might stream out dozens of basketball games, all live, in a single evening. The Abyssinian Cat Fancier Network might post only a few minutes of new video per month. And both models can work economically.
Of course, all of this will be for naught should the video-fueled Net go spectacularly supernova, then catastrophically collapse under a deluge of live sports, first-run movies, and sitcom reruns. But Glaser isn't worried. "At the end of the day," he says, " it's all plumbing," and points to an array of bandwidth-optimizing technologies - multicasting, proxy caching, usage-based pricing. Progressive's RealNetwork, the new service launched with MCI in August, is designed to let more people connect to RealVideo, while at the same time reducing pressure on the backbone. Up to 50,000 streams can be carried simultaneously, enabling content producers to reach audiences that are economically significant, but too small to be reached by cable or satellite. For instance, Glaser estimates that a one-hour show can be streamed to 25,000 people - around the world - daily for just $8,000 a month. The picture is RealVideo 1.0 - far from broadcast quality. But Glaser says MCI "can ramp up the backbone" as needed, to let better pictures reach bigger audiences. ABC News, Atlantic Records, and ESPN are already signed up.
Glaser has history - however brief - on his side. The Net has so far defied repeated predictions that it will choke on its own bits. As the Super Bowl and Jurassic Park, Part 13 start sluicing through, there will be a lot more - exponentially more - to choke on. But there are worse bets than Glaser's: that the rewards should the system survive are huge enough to ensure that it will. So if there is ample room for both Murdoch and Suck on the broadband, video-injected 21st-century Internet, what kind of space will there be for Rob Glaser? In high tech's odd world of alliance-based competition, the answer will largely hang on the delicate dance he has entered with his old friends in Redmond. Microsoft has given Progressive its blessing - no small thing. But Microsoft could also crush Progressive, just as it has crushed so many first-to-market innovators and even partners. Gates can pour hundreds more engineers into streaming technology. Having digested VXtreme, he can buy up everyone else in the market, then mine their code for proprietary gems. He can build stronger, faster, more stable products than anyone else, then give them away with the purchase of a small drink and fries.
It's no secret that Microsoft is never the easiest bedfellow. "If we don't execute well, it could turn out to be a bad deal," Glaser concedes. "But face it - if we don't execute well, we would have been in just as much trouble without the deal. And if we do execute - if we keep driving our standards forward independently and remain at the forefront of the market and the technology - it'll turn out to be an amazing deal. There are pitfalls. But if you want a sure thing, you should take a job at the post office, right?" Glaser is no postal worker. And having spent most of his working life in Redmond, he is no John Sculley or Jim Manzi. He knows how the machine operates. And in the present battle, he understands how high the stakes are and where the real lines of competition lie.
Microsoft built one of the most powerful franchises in history by becoming the toll-collector on the PC desktop. Now there is another, potentially far more valuable piece of territory up for grabs - interactive Net-enabled television. Microsoft wants that, too, and has been jockeying with the broadcast industry - not to mention its consumer-electronics partners - about who will set the rules. It is a titanic struggle - one that could extend Windows World from the current 35 million US homes with PCs to the 97 million that have televisions. Not to mention the rest of the world. And if doing that means giving a few players from "the competition" seats at the table rather than shoving them onto platters, Bill Gates can certainly afford it.
And Glaser is sure to come away with more than just crumbs. Since the initial syndication, Progressive has had two additional rounds of financing, the first in October 1995 by Accel Partners, the second in December 1996, led by Goldman Sachs and Merrill Lynch. Along with slices of stock for key employees - and now Microsoft's investment - these have almost certainly left Glaser with a minority stake in his company, whose value is currently estimated at more than $250 million. For now, Microsoft's check in the bank means plenty of money in Progressive's till. But at some point an IPO is nearly certain, as the stakes - and the capital demands - of the Internet-video game rise. That could still make his position worth well north of $100 million - not quite the billionaire boys' club, but not bad for three years' work.
But while an IPO may be looming, "sellout" doesn't seem to be part of Glaser's vocabulary. Wholesale acquisition, he says, by Microsoft or anyone else, would only happen "in such an usual situation that I can't even contemplate what it would be. Progressive is where I see myself in 10 years, 20 years - basically, however far I see."
And his old field of activist dreams? Glaser says Progressive may flirt around the edges of content, but the furthest it's likely to stray from software, networks, and tools is aggregating other people's programming. The company already runs a service called Timecast, a Yahoo!-like showcase for RealMedia resources on the Web. It also funds WebActive, a Net-based clearinghouse for political activism ("Julia-Louis Dreyfus says: Become an email activist for the Environmental Defense Fund"). During the July 1 handover of Hong Kong to China, Progressive put together a special RealVideo resource site, complete with that deeply credentialed progressive Henry Kissinger gushing in a press release: "For someone like me who wants to stay current about events, these RealAudio and RealVideo broadcasts are invaluable."
But whatever Progressive ultimately becomes - Internet standard-setter, protonetwork, or another notch in Bill Gates's deeply rutted scabbard - Rob Glaser's life-after-Microsoft project has already done more to transform the Web's media environment than any technology, with the (possible) exception of Java. Comparing the Web today with its 1995 counterpart is not unlike comparing, say, the first Technicolor epics with a silent film. And as Glaser likes to point out, the real changes are still to come.
So there is plenty of time for an Act III, if Glaser wants it. And meanwhile, there's another keynote or four to prepare for auditoriums full of true believers, each wanting to see allofthedemos - and maybesomenewones.
