Last January Miramax Films, the movie distributor that brought us such modern classics as Shakespeare in Love and Good Will Hunting, became the first bona fide Hollywood player to make the leap into the world of broadband entertainment when it made its 1999 movie Guinevere available as a copy-protected download over the Internet. Web surfers were invited to pay $3.49 on their credit cards for the privilege of skipping the trip to the video store.
Fuzzy bootleg copies of other movies, such as The Matrix and Blade Runner, abound on the Web, along with infinite numbers of amateur videos. But Guinevere is the first Hollywood movie to be offered online in a legal, non-pirated way. It’s an Internet milestone-even if Miramax has chosen one of its less successful films with which to make history. So I decided to give this brave new world of broadband entertainment a try.
A day and a half later I was still trying. Much of that time was taken up borrowing a Windows-run computer, since Miramax’s Web distribution partner SightSound Technologies has yet to support Macintosh; then I discovered I needed two plug-ins before the movie would download itself onto my computer. Using my digital subscriber line (DSL) service, the download itself took one hour and 14 minutes; but add time for locating the required plug-ins, downloading and installing the software, rebooting when asked, and then going back online after the movie finally downloaded to give my credit-card information and pay, and the process easily took two hours. That’s 15 minutes longer than the movie. In two hours I would have been watching not the titles but the credits, if only I’d gone the low-tech route and rented from the local video store.
What did I get for the trouble? A grainy film on a computer monitor. The version of Guinevere available over the Internet is only about one-tenth the file size of the version available on DVD; as a result, the picture quality is much lower. There is also the problem of milieu. Watching a film-especially one that advertises itself as “a May-September romance with an edge”-while leaning forward staring at a computer isn’t exactly a bring-out-the-popcorn kind of experience. Fifteen minutes into the film I gave up and took my dog for a walk.
One of the most pervasive visions of the broadband future is “video on demand,” the ability to order up movies and other programming and have them appear promptly on your screen, ready for viewing. Miramax is forging ahead, planning to offer a dozen movie titles online for download by the end of this year. Other Hollywood studios are following suit, citing the lucrative opportunities created by the fact that, for the first time, millions of consumers are gaining broadband access to the Internet.
But in the hour or so it took to download Guinevere, I had plenty of time to wonder, Is this it? Is this the best that the creative minds of Hollywood can come up with for the broadband Web?
Certainly, broadband Internet connections can transmit large amounts of data at speeds inconceivable only a few years ago. These connections, using either a digital subscriber line service running over the pair of copper wires from the telephone company or a cable-modem service delivered over cable-TV wires, are now in over five million U.S. households. They greatly outperform conventional dial-up modems, increasing the amount of data per second that consumers can reliably receive over a network and into their PCs by up to 50 times.
Still, it isn’t nearly enough. At least not for some of the grander visions offered by Holly-wood studios and other entertainment companies looking to exploit the bandwidth explosion. Not only will the Internet of the foreseeable future give you a fuzzy version of Guinevere after nearly a two-hour wait, but many other promised broadband applications-choose-your-own-replay sports sites, say, or interactive 3-D video games-are also proving to be far more technologically difficult than many had expected.
This doesn’t necessarily spell doom for broadband entertainment. But it does mean the future will look quite different from what some Internet pundits predicted even a year ago. Forget about downloading Hollywood movies or real-time video of the soccer championship being played in Brazil. What will work are Internet sites that take into account the limitations of even a broadband Web and nevertheless manage to offer customers something new.
“What consumers want from broadband is still something of a question mark,” says Ford Cavallari, an executive with Adventis, a Boston-based management consulting firm. “But it’s going to have nothing to do with two-hour movies, no matter what kind of fire-power you see coming out of Los Angeles.”
Despite broadband’s limitations, people are signing up in record numbers. It took cable television more than 30 years to reach 15 million customers; broadband Internet services in the United States alone may hit that mark by 2005, about 10 years after the technology was first introduced commercially. Why is broadband catching on so quickly?
In a word: Napster.
“The single most important reason people have signed up for broadband is they want to download music files faster,” says Cavallari. “Next there will be some sort of equivalent to Napster for still pictures, where people can share images or display what they’ve created with their digital cameras.” Indeed, broadband provides a real benefit to anyone trying to download large digital files: a five-minute song, for example, will take less than a minute to download on a broadband line, compared with 10 minutes or more on a dial-up line.
But outside the music- and baby picture-swapping community, some of the most popular applications on the Web can still run afoul of bandwidth shortages. The problem is not so much how many bits of data per second are theoretically able to reach your desk, but rather the maddening variability with which those bits arrive. Broadband users experience wild swings in performance, from near-instantaneous delivery of large files to slower-than-dial-up speeds. Sites that succeed on the broadband Web take these limits into consideration.
Broadband line speeds can be affected by a whole host of factors, from how far the customer’s computer is from the central office providing DSL service (farther equals slower) to, in the case of cable modems, how many other people are trying to download files over the same service. Cable-modem service shares bandwidth among many users; the more traffic there is, the less performance any single user gets. This variability, combined with an Internet protocol that wasn’t designed to make bits arrive on time and in order, often produces video, even at broadband speeds, with that characteristic stop-and-go quality.
Not surprisingly, stop-and-go performance is particularly hard on interactive games. Because packets of information from different computers may travel the Internet over different routes and thus arrive at their destination out of sequence, it’s impossible to predict precisely when a keystroke or mouse click will have the desired effect online. “Say you want to develop an online basketball game,” says Mark Blecher, vice president of marketing and sales for Electronic Arts, a maker of computer games. “One player shoots, the other player attempts to block. The winner will be determined more by what the network does than anything the players do.”
The stop-and-go problem stems from the way that video traveling over the Internet arrives at our machines. Most video available over the Internet is in a streaming format; that is, only a portion of the video file is downloaded into your computer’s electronic memory at any given time. After you view it, it is discarded to make room for the next bit of video. Streaming means you don’t need to use up disk space on your PC to view a video, and that you don’t need to wait long to start watching. But quality takes a hit whenever Internet routers decide to send packets in routes other than a direct line from the server to your computer. One alternative is to download a video file to play later on your desktop processor. That’s what I did to watch Guinevere, and as I found out, this method has plenty of its own problems. Although you avoid network hiccups, you still must contend with the file’s size.
The most effective sites avoid both of these problems with content you can access easily and look at quickly. In other words, they appeal to the “lean-forward” market-computer monitors rather than La-Z-Boys. Take for example artist J.otto Seibold’s Web site at www.jotto.com; in many ways the site is a perfect example of what works in the broadband Internet world. The site uses animation rather than video, so it works well within the bandwidth constraints of cable modems and digital subscriber lines: it’s far easier to stream the simple drawings of an animated short feature than it is to make video streams work, since so much less information needs to be transmitted. The site is interactive in humorous ways, appealing to the lean-forward crowd-your mouse can make a dog run around chasing birds, or make an animated character get on a bicycle and race after an ice-cream truck.
Videos that respect the Web’s limitations are also flourishing on sites that aggregate the work of many artists, such as AtomFilms (www.atomfilms.com) and iFilm (www.ifilm.com). These Web sites are meant as a platform to give exposure to unknown or emerging directors, and films on these sites are short, usually less than 15 minutes long. Directors on these sites often experiment with images that take advantage of the jerkiness of online video: characters, for example, might run in slow motion.
Take the film “405” on the iFilm Web site. Created on home computers by Bruce Branit and Jeremy Hunt, friends who in their day jobs create effects for television shows, the film depicts a jumbo jet landing on top of a sports utility vehicle in the middle of a Los Angeles freeway. The film “405” manages to tell a story in less than three minutes, and uses effects-like blurring the edges of the plane to denote speed-that translate effectively into the broadband medium.
Show Me the Money
Besides the technological obstacles to making video over the Internet a part of America’s viewing habits, there are also economic obstacles. In truth, no one has figured out how to make broadband entertainment pay. One reason is inherent in the medium itself. While TV broadcasters incur a large up-front cost to deliver their signals everywhere, once that investment is made, the cost per viewer to deliver a program becomes ever smaller as the audience grows. The cost of broadcasting a television program is the same, whether it is watched by one person or 60 million.
Not so with the Internet. The cost of delivering a video stream to an individual over the Internet doesn’t decrease as the audience grows. Content providers must pay per-stream licensing costs to RealNetworks or Microsoft or Apple Computer, so that customers can view video in RealPlayer or Windows Media Player or QuickTime. Internet service providers also charge on a per-stream basis to carry the traffic. And expensive technology may be required to manage simultaneous requests for the same material if a site becomes wildly popular. So costs expand proportionally with the audience-and the provider never gets the benefit of scale economies.
“The early model for broadband has been TV,” says Joe Laszlo, analyst for Jupiter Research, a Web advisory company. “But broadband is exactly the opposite of the broadcast world. Your costs go up as your audience grows.” According to Jupiter, the cost of providing video streams to a thousand viewers is nearly twice the amount advertisers are willing to pay to reach those viewers. Unless some fundamental shift occurs in consumer demand, so that people will pay for broadband entertainment, or in advertising prices, so that entertainment sites can charge more than they do now, broadband Internet sites will never be profitable.
This gloomy economic reality is reshaping how broadband content providers view the Web. “People are now turning to a subscriber model instead of relying on advertising,” says Laszlo. “There’s more potential there. But with the quality of service so unpredictable, it’s difficult to say whether sites will be able to charge enough to cover their production costs.”
As a result, many of the companies offering original broadband content-Icebox, Pseudo Programs, Digital Entertainment Network and others-failed within the last year or so. What’s left in many cases are Web sites that use streaming video to promote their own products. Movie trailers. Survivor updates. World Wrestling Federation interviews. To name just a few. Sites whose purpose is to promote successful programming in other media, such as TV.
“What you see developing online is the use of three- to four-minute promotional clips,” says Daryl Schoolar, industry analyst for Cahners In-Stat Group in Scottsdale, AZ. “Until the cost to deliver streaming media comes down, that’s primarily what you are going to get.”
Even many serious broadband content developers view the Internet as just a stopping ground, using it as a test market for eventual theater showings or television syndication. Heavy.com is a mix of alternative music and crassly humorous video clips aimed at a young, urban, mostly male audience. Its “Behind the Music That Sucks” animation sequences-a takeoff on VH1’s Behind the Music docu-mentary series-skewer everyone from Britney Spears to Eminem. While the site gets a million visitors each month, cofounder Simon Assad sees it primarily as a way to promote new programs, build an audience and eventually graduate to television.
“We’re expecting the Web site to be profitable, but we’re not depending on it,” says Assad. “Instead, we’re using the Internet to promote our television programming. I don’t need a broadcast license, I don’t need a distribution network, yet I’m broadcasting television right now over the Internet.”
Up Close and Personal
What does all this mean for the future of the broadband Web? For one thing, the services that will survive and flourish are likely to be those that don’t need massive audiences to be profitable. According to Jupiter, while 15 million U.S. households will have broadband Internet access by 2005, these households will still be far outnumbered by customers using dial-up modems to access the Internet. “Two-thirds of us will still be using dial-up in 2005,” says Laszlo.
The broadband Internet is, however, great for person-to-person file exchanges. You’ll likely see more ways to share information that takes advantage of broadband’s faster speeds-Napster-like sites that allow sharing of personal photo-graphs and videos, perhaps even allowing for on-site editing. Sportscapsule is one such service: it allows high school coaches and parents to upload their videos of local sporting events, and to edit them with appropriate prerecorded comments from TV football commentator John Madden. “For video to make sense on the Internet, it needs to be personal,” says Lawrence Rowen, vice president of marketing and sales at Sportscapsule. “Television can take care of the mass market. We’re servicing a market that may only include 30 people.”
It may not be the type of application that Hollywood moguls drool over, but you are also likely to see an evolution of personal communications on the broadband Web, from textual e-mail and instant text messaging to picture mail and video-enabled instant messaging. Successful broadband applications are almost certain to evolve toward this person-to-person sharing over the Internet rather than the downloading of mass-audience movies that are available elsewhere.
While a few years ago the Web seemed poised to threaten cable tele-vision, the technical and economic limi-tations of the medium are now much clearer, and the threat is correspondingly reduced. Though Miramax and other Hollywood studios might be exceptions, even large entertainment companies are developing more realistic expectations about broadband use. For example, in July 2000 Blockbuster announced an exclusive twenty-year deal with broadband service provider Enron Broadband Services to deliver movies over the Internet; last March, Blockbuster killed the deal.
“It’s finally sinking in that streaming video will never be as good as broadcast television,” says Cavallari. “But there will be many things for which it will be good enough.”
Good enough, and perhaps even better for the populist kind of communication and entertainment that makes up the best of broadband today. The limitations of the broadband Web have created a fertile ground for innovative forms of entertainment to flourish, whether they be three-minute animations that are too raw to find an audience in broadcast TV, or three-minute highlights from last week’s high school football game that can be shared with a proud grandmom a thousand miles away. And in the end, it may be exactly the limitations of broadband that will allow the Web to evolve in unpredictable and interesting ways we’ve only just begun to appreciate.
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
Inside Charm Industrial’s big bet on corn stalks for carbon removal
The startup used plant matter and bio-oil to sequester thousands of tons of carbon. The question now is how reliable, scalable, and economical this approach will prove.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
How Charm Industrial hopes to use crops to cut steel emissions
The startup believes its bio-oil, once converted into syngas, could help clean up the dirtiest industrial sector.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.