Those Widevine patents that have been in the news of late, new patents issued in the US, similar ones in Korea, and the August filing of a patent infringement suit against arch-rival Verimatrix, have all been sitting heavily on the collective Faultline mind of late. So when Verimatrix called up to give their take on the burgeoning market for DRM which hits three screens, we naturally preferred to ask about that lawsuit.
Verimatrix kindly rolled out its CEO Tom Munro to field the questions, and you could tell, he was pretty miffed about the whole thing. Our take back in August was the these two IPTV DRM companies were fighting the wrong enemy. Right now the big Conditional Access vendors Nagravision, NDS, Irdeto and set top makers Motorola and Scientific Atlanta, which also offer CA, are the real targets, because between them they hold the gate keys to around 300 million pay TV homes, not to mention Microsoft, which uses very similar technology to both of them, and which has 18 or so major IPTV deployments using its own DRM. Start ups like these two, still drawing down on VC funding, have no time to go around suing each other, we reasoned.
‘It’s the same old story,’ said Munro, ‘if they can’t win in the marketplace, win in the courthouse. What Widevine seems to be claiming is that any stream of content must be either entirely encrypted or not encrypted at all, and that it solely invented the idea of partial encryption. That denies the entire past work on MPEG2 Transport Streams, the DVB encryption process and every version of DRM that enables trick play (DVR functions).’
It turns out that the discovery process or this case is some way off. First there will be a Markman hearing, one to decide if these patents are really valid, probably some time during 2009.
‘We are asking for an ‘ex parte’ re-examination of these patents, since the prior art is several inches thick,’ said Munro, ‘but there’s not much else we can say at present,’ but he did refer us to the libraries within 3GPP encryption as yet another a for-instance.
The truth is that every single DRM offering that has attacked IPTV has realized that either the MPEG2 transport stream packet headers have to be in the clear, or that I-Frames need to be in the clear, or that something needs to be in the clear so that anything which happens to the video, such as storing it, re-organizing its packet order, synchronizing it with sound or fast forwarding on a DVR, can be achieved without decoding it. Each of the suppliers do this in slightly different ways, but it has certainly been used for longer than the existence of Widevine as a company, so its ‘partial encryption’ patent is unlikely to prove valid in the long run, unless it’s just talking about a highly specific approach to this process.
What is important here is that it is exceptionally difficult to attack the installed base of a DRM or Conditional Access supplier, since their customers rarely ever change a system once it is installed. Suppliers need to get in on the ground floor and stay there. It is possible to swap out DRMs, and since around 1994 the DVB Project has had a Simultcrypt process to prevent permanent CA lock in, allowing two CA systems to work alongside one another. This gives an operator a matter of months, if not years, to clear out a set of smart cards or set tops that have been either compromised or which don’t have the feature set it wants, while still delivering TV services through both the old system and the new. CableCard in the US is also a way of shifting the CA away from the set top so that it might be replaced without replacing the entire set top.
But as yet there has been no significant win by a new breed IPTV specialist DRM player, among the older worlds of cable and satellite pay TV. The closest thing to that we have seen is Irdeto losing out to Widevine at Surewest, which wanted to get rid of its expensive smart card system.
So these IPTV suppliers have to live on the promise that there will be 50 million to 100 million IPTV homes by 2012 or so, knowing full well that this limits them to a small piece of this action, and puts a cap on how big they can become, especially when someone like Microsoft comes along and grabs huge chunks of this market for its own DRM.
There are two ways of dealing with this, and Verimatrix and Widevine show the two approaches off nicely ‘ though neither have yet succeeded. The first is to move the science along a little, patent the R&D and ask for royalties when the established conditional access giants begin to use your patents. But of course you have to establish the right patents. The other way is to add features to your product set that attack the specific problems that exist in smart card based CA markets, primarily huge cost issues once any part of the system is compromised.
About 18 months ago Verimatrix tried a little of each, when it launched its VCAS 2.0 product line, adding patents that authenticate the broadband line a set top is attached to, watermarking, its own analog copy protection and a new concept called Quantum authentication, which uses the customer behavior to form part of the decryption keys. It was a major offensive outside of the IPTV markets while those markets seemed to have stalled, but despite its ingenuity, that largely failed. Instead it has now found a resurgence in its traditional customer base with companies like KPN, Belgacom, and some new customers it doesn’t want announced, which put it back in the IPTV mainstream.
The Widevine move over patents and legal actions seems pretty much a similar search outside of core business for revenue, and one that is unlikely to be successful, or even to resolve itself in time to have an effect on how this market develops.
Right now everyone is talking about a sub-market ‘ the ability for a single head end to encrypt for the IPTV, PC, Web and handset market ‘ hoping that an entirely new front on the handset will mean a head start over the more conservative traditional conditional access suppliers ‘ but this also takes them up against the powerful lobbies in the 3GPP standards body, and handset vendors like Nokia, as well as OMA DRM specialists, and once again head to head with Microsoft. The head end that can do this has yet to be invented, and that’s the key part of this, and perhaps the reason behind companies like Ericsson buying out Tandberg. The DRM part of this move will be resolved, most likely, by handset makers such as Nokia and that has already put its money behind Microsoft on its PlayReady DRM.
That leaves two other ways that consolidation can happen here. Either one of these two leaders, Widevine or Verimatrix, begin buying out the smaller players ‘ Latens, Secure Media, etc.. or even merge with each other, or one or two major companies such as Cisco, Motorola, Ericsson and Alcatel-Lucent, decide that they need in-house IP based DRM capability to complete their feature set.
Either way patent litigation is unlikely, we feel, to play any real part in shaping the future of IP based encryption, quite simply because it won’t complete in time, and it specially won’t if the patents don’t survive intense scrutiny.