Aereo may be long gone, but the dream of streaming affordable local television (without an antenna) lives on. And a Stanford University research project has answered the call with Puffer. We enjoy some television on-the-house and they use machine learning to improve video streaming algorithms:
We are trying to figure out how to teach a computer to design new algorithms that reduce glitches and stalls in streaming video (especially over wireless networks and those with limited capacity, such as in rural areas), improve picture quality, and predict how the capacity of an Internet connection will change over time.
The idea of this study is that about 500 people will watch TV channels here, streaming them over their Internet connections, and as they do, the Puffer website will automatically experiment with different algorithms that control the timing and quality of video sent to them, and will monitor how well the resulting computer-designed algorithms work. The more diverse the Internet connections that the 500 study participants use, the better the system will be able to learn, and the more robust the resulting computer-generated algorithms.
The TV channels are presented via a spartan but functional web player (that isn’t compatible with iPhone or iPad, sorry), with optional full-screen view, essentially encoding and re-transmitting Bay Area ABC, CBS, Fox, and PBS programming across the Internet.
I assume Netflix, YouTube, and the like similarly analyze their video presentation. But, as a non-profit, getting the technical data into the public domain would certainly serve the greater good – and possibly why there’s a legal exemption here. But no telling how long Puffer will remain available, so enjoy it while you can.
(Thanks for the tip, James L!)