IPUrbia

For those out there who haven’t heard of Joost, it’s been receiving a lot of positive spin in the video blogging community. A product of the founders of Skype, Joost seeks to do to online video what Skype has done for Internet telephony.

Andy Abramson posted a less than positive review of the concept, blaming a shoddy backbone infrastructure in the US for the infeasibility of the project. I’m less than sold on the project, largely because I don’t the the barriers to a viable video over IP market are technical. There are dozens of Internet video solutions out there that, in my mind, deliver a reasonable experience to a PC (although not HD suitable for that new plasma) in real-time over an average broadband connection. I see the barriers to market entry entirely revolving around content–the producers of the “desired” content (ie ABC, NBC, etc…) are very particular about their distribution channels. I think that ad revenue is somewhat of a red-herring in this argument–in this case, it is more about keeping control of the distribution. I think the real issue with the recent Viacom/YouTube spat is simply that Viacom is crapping their pants that people visiting YouTube for recent Daily show clips will stay for some user-generated programming. Every time I visit watch online, user-generated video, I am amazed at the quality improvements, and it is to the point that better online content is more enjoyable than the poorer offerings of the network. However, there is very little good “serial” content–a good clip here, a good clip there, but, Coke and Mentos is not (I hope) a weekly viewing for many people. The lack of predictably good “weekly serial” content makes any sort of online station programming difficult and really favors the YouTube model.

Still, my puzzlement is not about the lackluster reception of the concept, but Andy Abramson’s reasoning, blaming shoddy network infrastructure in the US, foreseeing gridlock. Hogwash, I say. Sure, there isn’t oodles of excess capacity, but that’s economic efficiency–simply put, there’s not the demand. I get into this argument regularly at conferences with people. At one point in a past life (managing technical operations for a small ISP) I calculated out the average cost of bandwidth per DSL customer (1.5-3.0Mb/s pipes) to run about $0.60 with even small amounts of scale. Similarly, I often do consulting for major cable MSO’s and am always surprised by how small their aggregate cable modem traffic actually is. Once the demand is there, the supply follows. Right now, there really isn’t much of a demand–hence, low supply.