• Putting the Broadband Inauguration's Flameout Into Perspective

    There has been no shortage of stories in the last few days about the travails many people experienced when trying to watch President Obama's inauguration via broadband. While things worked flawlessly for many, for far too many others it was a frustrating and unfulfilling experience. As I wrote on Tuesday, the first "Broadband Inauguration" was a milestone opportunity for this new medium. Instead, it was a confusing flameout.

    While I'm disappointed, I can't say I'm terribly surprised. Even as I was writing Tuesday's post, I found myself wondering if the Internet was really up to the task of handling this colossal and highly compressed live event. What I experienced personally that day, and have heard and read since, all underscore the massive inconsistency in users' experiences.

    For example, Hulu worked fine for me. But when I tried to watch on CNN.com I couldn't even get the Flash plug-in to download (and btw CNN, talk about an inopportune time for a download!). I had no luck at NYTimes.com either. One person I spoke to this week said he couldn't get a stream at any of the major news sites and ended up watching at MLB.com of all places. Conversely, others reported no problems at all. No doubt you have your own particular stories.

    So here's an attempt at putting all of this into perspective: no communications or transportation system is ever built to serve extraordinary peak demand. Instead they are built to serve typical demand plus an increment for periodic bursts. We don't have 15 lane highways so there's no congestion on Thanksgiving Day, while the other 364 days of the year 90% of the space is unused. Likewise, on inauguration day, wireless carriers were urging attendees to refrain from using their handsets for fear of overloading their networks. I can even remember back to the '80s when on heavy call days like Mother's Day, Ma Bell's gold-plated network would sometimes stymie me with an "all circuits busy" message.

    And these are just a few examples. The reason things are this way is purely financial. It simply doesn't make economic sense to invest in so much extra capacity that's unused most of the time. No venture capital or institutional investor would tolerate capex budgets not supported by realistic use cases. The result is that on surge days like on Inauguration Day incremental available capacity is quickly swamped and many users expecting a flawless typical experience are disappointed.

    If that's the sobering reality, then here's some good news. Tuesday's massive broadband interest, coupled with other heavily viewed live events, will likely spur further investment in all links in the Internet/broadband delivery chain. History shows us this is true. We may not have 15 lane highways today, but we often do have 4-5 lanes instead of old dirt paths because of cars' growing popularity 50-60 years ago. And we have 5, 10 and even 50 mbps broadband service now instead of pokey old dialup because Internet usage has soared in the last 10 years, demonstrating users' widespread willingness to pay and prompting huge broadband ISP investments.

    For all of broadband's progress, it is still a relatively nascent medium. "Rome wasn't built in a day" as my father used to admonish me in my moments of adolescent impatience. On Tuesday, broadband's limitations became obvious. That was unfortunate, but I'm betting that next time around will be better.

    What do you think? Post a comment now.