Last week, a widely-hyped Webcast by Oprah Winfrey left millions of would-be viewers with dark or stuttering video, with officials at Oprah’s Harpo Productions saying that "more than 500,000 viewers attempted to watch the event - but most could only see the first few minutes." According to Harpo, "crashed Internet servers" and "throughput demands of 242GB/s" brought the Webcast to a standstill for perhaps more people than actually were able to view it.

Aldo Cugnini
Analyst

Much of the discussion that picked up on this, claimed that "Oprah overloaded the Internet," that the network just couldn’t handle such a big load. However, technology colleague Shelly Palmer writes in his blog that the crash was almost certainly the result of a "logical error in the caching servers" bringing down the video stream. (Curiously, I experienced long email forwarding delays at about the same time from one reflector to which I subscribe. Chalk it up to coincidence.) This is not the first time something like this has happened; New York magazine and Victoria’s Secret are among the more notable casualties of similar incidents.

But will the Internet eventually slow to a crawl? Is bandwidth on the Internet keeping up with demand? Last year, Motorola predicted that by 2010, some 69% of US households would need at least 40 Mbps of bandwidth, with many needing as much as 60 Mbps. Ten years ago, engineer Jakob Nielsen proposed that a high-end user’s connection speed grows by 50% per year - a claim that fits current usage and Motorola’s prediction quite well. Apparently, service providers are already feeling the pinch. Comcast has warned broadband Internet customers across the country to limit their downloading; violators can get a 12-month suspension of service.

Government and academia are now looking for ways to develop a faster, more efficient Internet. Two such projects, Next Generation Internet, and Internet2, are now underway. In addition, the IETF has addressed the growing shortage of IP addresses by developing "Internet Protocol Version 6," or IPv6. IPv6 is expected to replace the existing IPv4, with the two coexisting for a number of years.

Many experts believe that a solution to bandwidth bottlenecks is a migration towards peer-to-peer connectivity, away from the broadcast-centric view of content distribution. Open-source advocate Tim O’Reilly is widely credited with coining the term "Web 2.0," describing, in part, the tendency for the Internet to evolve towards a social network. But before extensive peer-to-peer file sharing happens, there will have to be a major shift in the position of conservative copyright politics.

So what does this have to do with displays? Regardless of the actual cause, the perception of the frailty and on-demand limitations of both PCs and the Internet will continue to avert the shift from TVs to PCs as the dominant long-form entertainment vehicle. That means that large-format displays should continue to dominate sales for "TV" viewers, and small-format ones for PC users.

Other factors continue to stymie the PC-plus-attached-display as the home’s preferred entertainment system. I have yet to see a mainstream PC that can play DVDs as flawlessly as even a cheap DVD player. Wal-Mart has already closed its video download service, less than one year after launch. The claimed reason was a break in support from their server supplier. More likely was the fact that it simply was not profitable. Microsoft has, for more than 10 years, pushed to make the PC the home’s central entertainment device. "Microsoft TV" continues to be promoted as "the future of television." Yet, this future eludes the company, as the product is constantly morphed, year-by-year. The emergence of WiFi-enabled TVs and the SlingBox continue to suggest that the large display will always provide the "lean-back" experience that defines the home entertainment system - but don’t be surprised if a growing number of consumers drive this display, at least part-time, from a remote PC.

Reply to the author