Article Info

SIPlified Content Distribution

Wading through my morning reading I happened upon an Acme Packet (APKTwhite paper that did a good job explaining my apprehension about Akamai (AKAM) and how their business might be commoditized. In the long term, do SIP and managed media sessions replace the media caching model?

For decades, the experts have been telling us that content is king. Whoever controls content, they say, will profit the most from network services. And video-on-demand will be the ultimate Internet moneymaker.

But history disagrees. Interactive communication has always made more money than content distribution. In the nineteenth century, before telephones and email, newspapers accounted for 95% of U.S. postal traffic by weight, while ordinary letters generated 85% of postal revenues. Alexander Graham Bell thought that people would use his telephones to listen to distant concerts. That didn’t happen. Napster, while a huge overnight success in terms of users, never made any money. Today, data consumes half or more of public network bandwidth, but ordinary telephone calls generate over 80% of total earnings (Figure 1). In fact, the telephone industry earns as much in two weeks as the movie industry makes in a year. And while informational Web sites like struggle to survive, email remains the Internet’s killer app, with instant messaging close behind.

One thing I disagree with, though I am not certain this was the authors intent, is that content OWNERSHIP is not king. Owners and efficient/effective generators of quality content have never in history had such wide and efficient distribution. The marginal cost of an additional media customer is approaching zero. Imagine how excited Shakespeare would have been to transition from a Globe Theatre audience of 3000 to a fully distributed IPTV broadcasting system reaching billions.

I think the point is content DISTRIBUTION is not king, and that history shows facilitating interactive communication generates the most value.

By this analogy, Akamai is just in the mail delivery business. Perhaps this time is different, and since the cost of communication now approaches zero this is no longer true. I don’t have an opinion yet since predicting the future on a large scale is a risky business.

Regardless, I thought the passage was worth sharing, particularly since the reference to how Internet Video will simultaneously save chip companies, equipment companies, and Telcos/Cablecos seems increasingly absurd to me.

I’m still wading through the implications of Session Border Controllers and trying to form an opinion on what role they play in the future. If SBC technology is inexpensive enough, does the QoS it brings to media sessions drive the concept of remote caching (typically in high cost urban datacenters) to extinction, at least for all but the most popular media content? Instead of caching 10000 songs, Apple can cache their top 250 songs with Akamai but rely on SIP and (Net Neutrality supporters can leave the room now) guaranteed transit services from carriers using SBC technology.

Which approach is cheaper? Today, clearly it’s the Akamai solution. If core bandwidth gets cheap and SBC’s get cheaper, that may not always be the case. It would be very interesting to hear Akamai’s position on this, as well as those of my readers.

Full Disclosure: I am long Acme Packet, and hold both long and short positions in Akamai.


Comments are disallowed for this post.

  1. Isn’t it more likely that SBC’s are likely to be incorporated into next-generation routers (JNPR-Kagoor, etc …) further commoditizing the SBC market therefore putting a larger crimp into AKAM? I know Redback is looking to incorporate SBC’s into next-generation SmartEdge.

    Posted by BP | November 21, 2006, 11:40 AM
  2. Cisco thinks every packet should be routed and every feature should go into the router.

    I think the carriers will demand separation between the transport and service creation layers. Putting the SBC’s into the routers runs counter to this.

    Putting these features into the router reduces future flexibility for the carrier. Even if the router could do this as good as a standalone product (which it can’t) it wouldn’t happen for this reason.

    It’s a good question and a good debate.

    Posted by Andrew Schmitt | November 22, 2006, 3:34 PM
  3. If you look at Juniper’s website, you’ll see they have EOLed the Kagoor session border controllers.

    Full disclosure: I’m an employee of Acme Packet

    Posted by Albert Lew | November 23, 2006, 1:58 PM