Image: NYC Media Lab/Flickr
For nearly as long as people have used the internet, they have suffered through buffering and downloading delays to watching their favorite videos. These days, bandwidth has increased at home, but under current Draconian mobile data policies, streaming content on the go is increasingly not an option, and copyright laws prevent users from buffering more than short snippets of content to take with them.
Luckily, Dr. Shivendra S. Panwar and his team of student researchers at the Polytechnic Institute of New York University have a solution: "streamloading."
Streamloading is Dr. Panwar's term for his new fusion of streaming and downloading. He hopes it will help wireless carriers get more mileage out of their bandwidth, while also helping data consumers watch more videos on the go. In the streamloading utopia, gone are the vibe-killing breaks in action when watching mobile video.This is a great technology, which promises to transform the Netflix queue from a reflexive list into a dynamic viewing aid.
The technology works by bisecting video into two layers. First is a base layer, which streams during viewing, then there would be a higher layer, which the user would pre-download from some high-bandwidth location like the home or office. While the higher layer would be useless on its own, and thus in compliance with intellectual property laws whose aim is to prevent free sharing, it would nonetheless comprise about 3/4ths of the total data.
Trinity needs the data NOW, and buffering just won't do
Because the streamed base layer would be necessary to unlock the viewing experience while still constituting only about 1/4th of the data, effective and lawful streaming on the go would require fairly low bandwidths. The low bandwidth required to stream the content would mean that data-heavy movies and TV shows would be watchable on your phone even in zones of spotty coverage. It means the death of buffering.
Perhaps more significant for carriers, content providers and viewers in the long run, the lower data requirements would mean that mobile network bandwidth could accommodate significantly higher levels of traffic at a fraction of the cost, enabling consumers to view far more video without worrying about fees.
I talked to streamloading inventor Dr. Panwar to get his take on things.
Motherboard: How did you get the idea for this?
Dr. Panwar: The law regarding streaming video doesn't allow you to buffer for more than 30 to 60 seconds at a time, so if you drive by a building and the building blocks the cell tower, you get an interruption. The legal requirements for streaming state that streaming is like listening to the radio: you only pay the price of listening to advertising as opposed to actually paying to buy the content.
There's nothing in the law to prevent you from downloading higher layers, which are useless in themselves, but which nevertheless represent the bulk of the actual data required for quality viewing. The law leaves room for viewers to preload a portion of the data when bandwidth is cheap, and then to stream a smaller portion of the data in a variety of circumstances. Unless bandwidth were totally blocked, streaming the lower data layers would not suffer from interruptions.
Why would wireless carriers be interested in this kind of technology?
I see this as a triple win scenario. Carriers are facing a bandwidth crunch. The 4G LTE systems are not keeping up with demand for data. AT&T has said publicly that they might run out of capacity this year. A crude way for them to control demand is to raise data charges, which would drive away customers. Anything delivering quality data at a lower cost is good for the carriers. That's the first win.
It's also good for streamers like Netflix and Hulu because it enlarges the number of times customers can view video. Currently, downloading a movie on a cell phone could cost ten dollars per movie. Making things cheaper means more watching of movies. Customers could watch movies on their commutes and get more high quality video at a lower cost.
What would the user experience be like?
It's the same user experience as watching a Netflix video or downloading a TV show at home, except you would predownload the higher layer beforehand and only stream the base layer on the go. You would only pay a data charge on the base layer at about 1/4 the cost.
What are the wireless carriers saying? When do you see this becoming available?
I'm a professor and I've come up with this idea. Ideally, I'd like to license the idea and have carriers implement it, because then I don't have to do too much. Potential partners all say yeah, this is interesting, but then there's a whole chain of negotiations, from the carriers to the streaming providers. I have to get all parties willing to invest and license.