Windows Media Is Alive and Well
In 2008, Microsoft finally got off the “our encoder is best” bandwagon and announced support for H.264 in Silverlight at the International Broadcasting Convention in Amsterdam. For years, Windows Media and Microsoft had been the enemy of any standards body or industry group, such as the Internet Streaming Media Alliance or the MPEG Industry Forum. Now they would become friends. H.264 is today as important to Microsoft as Windows Media was a few years ago—maybe more so, since the primary responsibility for codecs was moved from the server group (Windows) to the tools group (Visual Studio, Expressions).
Just a year or so earlier, Microsoft had taken its codec to SMPTE as a proposed standard, and after a lot of puts-and-takes, Windows Media was approved as SMPTE VC-1. To help market what most people know as Windows Media, Microsoft positioned Windows Media as its implementation of the SMPTE VC-1 standard, which is sort of like getting your date pregnant then marveling at how much the baby looks like you.
Behind the scenes, AT&T was working on rolling out its U-verse service to consumers. U-verse is a private IP-based network to deliver voice, data, and video to consumers provisioned over an xDSL rather than a radio frequency-based cable system. Designing a system for national rollout takes time, and there are enormous consequences to selecting which codec to put into millions of set-top boxes and what sort of delivery infrastructure you need. Naturally, MPEG-2 was the first choice, as it is still the primary codec for conventional digital cable TV systems. But as a new IPTV service where bandwidth might be tight, AT&T selected both H.264 and Windows Media.
Really? AT&T does not talk much about Windows Media, and even the Wikipedia post says U-verse is based on H.264. But for the thousands of public, education, and government channels (yes, including your local equivalent of Wayne’s World), it’s all Windows Media, aka VC-1 delivered over a single T1 line from the TV station to AT&T’s network operating system. Actually, the T1 is a commodity business internet access line that allows AT&T to pull the Windows Media stream from the TV station at about 1.5Mbps, which gives an acceptable quality level. AT&T then distributes the stream to local subscribers via IP multicast, which gives them scale. It works quite well.
Many local TV stations cry foul, not because of quality issues, but because AT&T has segregated municipal channels to a special area of its user interface—you have to navigate there rather than finding it through random channel surfing. In other words, it’s more like web streaming than cable TV, or more like Roku. Or Boxee. Or Netflix or Hulu. The under-30 crowd won’t have a problem with this, and there are several benefits. First, in the conventional cable world, a local TV channel would only reach to the edge of the town that the cable franchise served. With U-verse, you can watch any channel within your state, so the potential audience grows. Second, because the TV station has an encoder, it typically uses the same technology to simulcast its live content on the public internet. That internet stream is usually in the same format that goes to U-verse: Windows Media. Thus, your local TV station is viewed on the web via Silverlight or Windows Media Player. Thanks to live transcoding, those same feeds are often also available via Flash Player; on iOS devices, Android devices, and Blackberries; and on Roku boxes via a live H.264 stream, even though the source originates, as it must, in Windows Media Format.
Windows Media, like MPEG-2, has an enormous installed base, and we won’t see technology updates until the existing equipment and infrastructure are fully depreciated, and then only if there is a business case to do so. In the meantime, Windows Media is surprisingly alive and well, and we see new systems going in every month.
This article was originally printed in the April/May 2012 Streaming Media magazine.
Rather than presenting new online series, Microsoft showed how advertisers could create custom experiences.