Last month, on December 29th, Akamai was granted a patent (number 7,640,303) relating to a "Method of, and system for, webcasting with just-in-time resource provisioning, automated telephone signal acquisition and streaming, and fully-automated event archival."
Reading over the details of the patent, the focus appears to be on a system that allows for the real-time provisioning of a live meeting and the automated process to be able to handle things like audio signal acquisition, encoding, monitoring, archiving, reporting and billing.
While the patent filing makes reference to prior art and systems that already do this today, the basis for the patent seems to be the claims that all of these other systems have deficiencies, which the Akamai patent supposedly overcomes. Some of these deficiencies include:
- "traditional Internet conferencing casting systems have several
deficiencies. These applications typically use databases to generate
reservation information, to initiate events, and to authenticate a host
or the attendees. The database dependency creates a potential single
point of failure because if the database is unavailable, events cannot
- "a typical application runs the event streams from a single server,
which again represents a single point of failure and limits
scalability, i.e., the number of attendees that can attend the
- "prior art conference casting systems also do not have the capability to
archive the event in an automated manner and/or to manage when
particular streams get interrupted before the event is terminated."
- "the prior art systems require advanced setup for the streams, which
dictates a blackout period between the time that an event is reserved
and the occurrence of the event itself. Thus, once an event is
scheduled, a service provider typically must provision or allocate in
advance various system and other resources. Such resources include,
without limitation, media encoders, storage, network connectivity,
streaming server ports or publishing points, and the like."
- "prior art conference casting systems do not have the capability of
reserving and then immediately executing the event, with the resulting
stream being immediately available to an audience member. In prior art
systems, stream redundancy typically requires special handling and
What seems broad to me is Akamai's assertion that all other systems reserve physical resources ahead of time. They are implying that no events based webcasting system on the market today can generate a live stream, or provision any audio resources on the fly, or what they call "just-in-time", when requested. The patent also mentions multiple times that the Akamai patent "provides significant flexibility and reduced costs as compared to prior systems", which sounds like marketing speak to me. They also say that, "In addition, the system is able to handle large numbers of simultaneous events and massive total audience sizes." Again, lots of marketing words, but no definition of what "large numbers", "massive" or how many "simultaneous" streams they are talking about.
Clearly this patent is aimed at the business side of webcasting since
typically, those are the events that have registration databases
involved and audio conferencing bridges. Most of your one-off media and
entertainment events, sports and concerts, don't involve that kind of
I'm very curious to see if Akamai does anything with this patent and if they try to go out to the market to enforce it in any way. That may not be their intention, or they may view it as something they want to try and enforce, we just don't know. Since Akamai has a company policy of not commenting on anything patent related, I didn't ask them what they plan to do with it.