Codec Wars vs Serving The Common Good

There has been for many years a subtle conflict ongoing in telecom space. Various vendors have created digital encoding techniques (codecs) that target common network issues. Since various network realities exist so too do various approaches to the problems faced. So a range of codecs exist in the marketplace.  Typically a high-quality solution comes with an associated cost, reflecting the very fact that the solution has merit.

The poster-child for this is the G.729a codec. Over time this patented codec has become the industry standard low-bitrate codec for voice applications. Who can argue. It works well. It squeezes reasonable voice quality down to under 30 kbps and it’s compute overhead is acceptable on available hardware.

The fact that it allows you to cram more calls into the same amount of bandwidth gives it real value in the eyes of network operators. As a result they elected to live with a licensing scheme that involved paying per channel in use. Often the cost of the codec is bundled in gateway hardware. Even so, the creators of the technology have been paid for its use for years.

Even telecom bad-boys Skype use G.729a when passing calls to/from the PSTN.

Barring future revision of our patent laws the patent on G.729a will eventually expire and the codec will become available for use without license. If you want a really good, entertaining read on patent and copyright issues I suggest Spider Robinson’s 1983 short story Melancholy Elephants.

Why does all of this come to mind? Well, consider the amount of action on the codec front in recent times. Last summer Polycom released its Siren7 and Siren14 codecs under a royalty free license. Earlier this past week Skype announced a similar arrangement for their new SILK codec. Interesting parallel, no?

polycom_ip650_256All three of these codecs support wideband telephony at reduced bitrates, making them all desirable in one fashion or another. Of course, Polycom has hardware support for their codecs in their conference systems, but oddly enough not in their more common SoundPoint IP phones.

Nothing supports SILK yet except the Skype 4.0 for Windows client.

Here’s an idea…why don’t these two companies cross-license each others codecs as a courtesy to the industry? That opens the door to every Polycom device being a Skype end-point. It also allows Skype end-points to engage in conferencing with existing Polycom systems. It’s a win-win, right?

ipevo_s010_ga_c03Is there demand for such a move? Absolutely! In fact, Ipevo is trumpeting new two-stack desk phone that combines SIP and Skype functionality. The trouble is that, however nice that phone might be, it’s not likely to be enterprise class hardware. It probably won’t get much traction beyond home users.

Imagine an office where the outstanding Polycom hard phones could make and receive calls via SIP and Skype.  We’d have best in class hard phones acting in concert with a best in class soft phone in the Skype client.

Update: March 8, 11am Wirevolution’s Michael Stanford agrees noting that

“Polycom has been leading the wideband codec charge on deskphones, and it already co-brands a phone with Skype. It would make sense for Polycom to add SILK to its entire line of IP phones.”

  • I like where you are going with this, but cross licensing the codec doesn’t mean cross platform compatibility. If I had the SILK codec on my phone today, it still doesn’t provide me any ‘ins’ to the Skype network.

    Having said that, I do think this move signals a trend for Skype to open up a little – and I don’t think you can argue they are at least starting to talk (internally) about peering with other VoIP networks. It certainly would be a lot easier for them in the long run if all hardware/gateways supported SILK, so perhaps this move is a precursor to a larger strategy.

  • We already know that Audio Codes is working with Skype from statement made by Alan Percy a few months ago. We also know that Digium’s Skype For Asterisk project is coming along.

    What interests me with respect to Skype is end-point compatibility. There are many good SIP hard phones around, but really no comparable Skype hardware. Most Skype hardware is in the consumer space. Availability of business class phones that could interoperate with Skype would be a game changer.