Media
Why Joost is Toast (Toost?)
Joost, YouTube and other web-based video services will soon run up against physical limitations of the Internet, says Carl Yang, Beijing-based executive vice-president of strategy for start-up company MPi.
Yang’s company - surprise! surprise! - offers a solution to this problem, but his argument is interesting nonetheless.
Size matters
First Yang takes the vast size difference between data and video files (something I have become very aware of as Video Blogging hogs my hard drive).
Yang reckons that once 10 percent of the Internet’s population starts using video on a regular basis, 90 percent of the Internet’s traffic will be hogged by video. This traffic jam will grind Joost and YouTube-style services to a halt.
Nodes are slow (not fiber)
The Internet traffic jam will not, however, be solved by running fiber into homes. The issue, Yang says, is the relatively slow speed of traffic through the network’s nodes. In other words, it is a structural issue.
Moore’s law does not matter
You may say: “But Moore’s law will save us by doubling the speed of the processors in those nodes every 18 months.” Moore’s law is great, but Yang counters with something called Gilder’s law that states that the capacity of fiber networks triples every nine months. (Sorry, Moore.)
The result: Video Constipation
No matter whether you offer videos off of servers (YouTube) or via peer-to-peer networks (Joost) the Internet is headed for a traffic jam.
MPi’s pitch
MPi, on the other hand, says their non-IP-based system will allow Telcos and Cable companies to offer a massive population:
- High quality videos
- Two-way interaction
- In real time
Yang’s TV Vision
Yang’s vision of television moves straight to the YouTube-style model of people broadcasting to one another, but does it in real time with high quality video. (He also includes regular broadcast-style TV along with timeshift features.)
He uses the term “Switched TV” to describe a world in which any individuals on a network can broadcast television to all other individuals.
To clarify:
Stage 1: Broadcast (1950’s model of one to many)
Stage 2: Interactive (1990’s model of individuals gaining more granular control of what they receive through interactive digital TV or Tivo-like systems.)
Stage 3: Switched TV (2000’s model/YouTube of anyone broadcasting to anyone else)
The troubles facing today’s Cable and Telco companies will only worsen unless they find ways of embracing and monetizing high quality YouTube-style model of television, Yang said.
How?
Consumers using MPi’s system need a $100 set-top box and must subscribe to a Cable or Telco company with the equipment back in a central location. One limit vis-a-vis Joost, YouTube, etc: Consumers can only interact within the network, closed-garden style. A real world test of the system will be starting soon with a Gehua, a Beijing cable company.
Money?
In addition to subscriptions from consumers, video on demand, etc… some new and interesting revenue streams include adding charges for people to set up their own broadcasts (You could broadcast to all offices of your company, for example, or to all of Beijing). They also offer a way for broadcasters to charge viewers for their videos. Billing is easy since the entire system operates within the Telco or Cable company’s proprietary system.
Dangers
The weakness of MPi’s system is, of course, that Yang’s calculus might be wrong. Joost, YouTube and friends might find that people are happy enough with low quality videos. Another danger is that Joost and friends find ways to send video over the Internet in a more efficient way that does not face these limits.
Even with these two scenarios, Yang argues that Cable companies and Telcos are better off using his system to retain and gain customers.
Technorati Tags: Carl Yang, Joost, mpi, mpi (shanghai) ltd, YouTube
Discussion
Comments are disallowed for this post.
Post a comment
Newspaper Writings:
French Blogging:
Digital profiles:
Add to Favorites:
Random Faceroll
Interesting interview, but I think Mr. Yang is willingly misdirecting you when he talks about Joost and YouTube as companies that would be “replaced” by their offering. In fact, MPi’s offer sounds like a simple technology replacement for current cable infrastructures.
However, if you have more information on MPi or their product, I would appreciate if you could send it this way.
Read a more complete analysis at: http://www.webcom.it/blog/articles/2007/10/10/chinese-competition-to-iptv/
Bye,
Andrea
Thanks for your comment about MPi. They don’t actually have a website for now.
I have a powerpoint, but would like to check with the company before forwarding it.
You are right, they will be more a replacement technology for cable companies than a direct competitor for YouTube or Joost. I have no idea of their competitive environment. Sounds like you are much better informed about that.
Are there many other technologies like this?
I’m familiar with most of the technology, as well as part of the “TV over Internet” scene, but not necessarily with the “TV over IP cable company” scenario and solutions.
What do you mean with “Are there many other technologies like this”? If you mean other fiber based, non-IP solutions, then I’m afraid I don’t know of any (whether that is because they are not viable, or only because they are waiting for a breakthrough); if on the other hand you mean, solutions for a cable operator that wants to replace coax with IP, then yes, there are many.
If you want first hand experience from an early adopted European triple play operator, google for “fastweb iptv”. I’m not giving links directly because I don’t have first hand knowledge of the technology they use inside their network, but several companies claim to provide at least part of the equipment.