• by Tom Hespos , October 1, 2002

In the past few weeks, I’ve seen several articles lamenting the lack of broadband advertising options and speculating as to how much better online advertising could be if home broadband penetration was higher. But will higher broadband penetration mean that advertising will suddenly become more compelling due to fatter pipes? Not likely.

Nielsen//NetRatings pegs home broadband penetration at 21 million people in the U.S. I’ve been with RoadRunner for a few years now, and consistently connect at speeds that exceed the connection speeds that I’ve had at the office, when I worked for various agencies that used T1 or T3 lines. Last year, I gave my mom and my dad Optimum Online gift packs, so that they could have the always-on, high-speed experience that I’ve been accustomed to for the past few years. Truth be told, our online experiences are not all that dissimilar from those of 56K users. Sure, we have advantages over dial-up users: The connection is always on, and if the web servers of our favorite websites can send content to us faster, then we can get it faster. But it turns out that’s a big “if.”

It’s important to remember that home broadband only represents the last mile to the consumer. While we all dream of days when we can get full screen streaming video at high resolution with little waiting time, the truth is that the Internet’s infrastructure can’t support that at this time. When I visit a website for a video clip and click the “300K Cable/LAN” link that is appropriate for my connection speed, most times I won’t receive the video at anywhere near that speed. Is that my fault? No. It’s the fault of the content provider and general Internet congestion.

Truth be told, it’s extremely rare that I can receive streaming video at a speed that even remotely challenges the full potential of my connection. To be able to consistently deliver content at this rate would require enormous investments on the part of content providers, who would need to beef up their server farms and available bandwidth to be able to provide video and other such content at the speed, size and resolution that most consumers expect from TV. Not to mention that the Internet backbone would need some reinforcement as well. There’s even the question of whether the TCP/IP protocol, which controls the routing and switching of data packets over the Internet, is well suited to the task.

So why are we waiting for a day that obviously won’t arrive for quite some time?

Since the idea of video over the Internet became popular, we’ve all had to rely on the concept of streaming to deliver video experiences. That is, video content begins playing on a user’s computer or access device before the complete video is delivered through the Internet – content is streamed to the user over time. Most of us have been frustrated by our video players as “congestion” causes pauses or skips in the content as it plays. This isn’t due to a lack of last-mile bandwidth. The problems tend to occur upstream, as TCP/IP routes data packets through an increasingly congested Internet, or as streaming servers struggle to deal with fluctuating numbers of users in all parts of the world.

Technology providers in the streaming media space have developed some clever ways to minimize these problems associated with streaming. Streaming players like those provided by RealNetworks use a “buffer.” That is, the player gets a head start by downloading a certain percentage of the content before it actually starts playing. In most simplistic terms, a player will wait until it is reasonably sure it can stream in the remainder of the video content within the requisite amount of time before it starts playing. But sometimes, connection speed can slow down between the time a user starts downloading and finishes. This can result in a pause in the video, a refreshing of the buffer, or an adjustment in frame rate.

Another approach to solving the streaming problem is patented by Unicast. The company uses a proprietary Java applet to stream content to a user’s machine. The content is played only after it has arrived in its entirety, thus eliminating any problems with buffering or interruption of the experience. The result is a rich experience that is delivered as intended.

Most rich media advertising providers rely on streaming to one degree or another. And until the Internet infrastructure can support higher deliver speeds, they will continue to do so. Fatter last-mile pipes will help, but they are not the Holy Grail.

So why are we waiting for home broadband to achieve higher penetration to revolutionize online advertising? With the current technology and infrastructure, it will be a long time before any home user who requests video content will be able to get it without a bit of waiting, so let’s not pretend that a new era is right around the corner.

Rather, let’s take this opportunity to do the best that we can with current limitations. After all, streaming content experiences are certainly better than flat GIFs.