Your browser is not supported. Please update it.

21 June 2006

Coral raises worthy issue of persistent web based content

There is one huge difference, and several important, but smaller differences, between the way in which the Coral Consortium’s new interoperable DRM framework works, compared to Microsoft’s Windows Media DRM. The release of the new eco-system within which Coral’s previously released interoperability framework will work, came this week, keeping alive one of the few prospects for a system to protect digital entertainment without disaffecting consumers with new limitations on their ability to move content between their devices.

Both of these schemes allow a number of devices in a home domain to acquire content rights and content versions that will play without the customer being aware of DRM. This difference is that with Coral’s new Eco System A, any existing DRM can be enabled, as can legacy digital devices, which have no DRM client. It will also offer a set of web based persistent rights to content so that if consumers have a disaster like having their PC crash or stolen, all their content rights will still be intact and they can download new content copies.

There are four outcomes we can see as Coral continues its laborious but worthy approach towards its first implementations. Either a single standard, such as OMA DRM or Microsoft Window Media DRM is established as a global standard for DRM, and we all use the same code, but someone gets monopoly control of DRM. If this is Microsoft, this implies that Microsoft gains architectural control over all consumer electronic devices. At least if OMA was the standard it would be open, however OMA was designed for mobile phones, and is probably not man enough for the job in its current implementation.

Or an open source DRM emerges, as Sun Microsystems has been pushing, which would allow a single client to be written for ALL devices without ceding a DRM monopoly. It’s a nice thought, but still probably a year away.

One of these must happen if we are not to continue to live with our current splintered approach to DRM, with more and more content being riveted to a single device, such as iTunes content being pinned to the iPod, a situation that continues to drive users towards piracy.

Or Coral gets its act together in time. But it’s not just about getting it together soon. The first thing we said to Talal Shamoon at Intertrust, who’s technology sits at the heart of the this Coral effort, is that the consortium must move fast to avoid monopoly, but it must also offer reasonable royalty terms.

This last issue may not be so easy to overcome. Sun has already gone very public in stating that it will offer its DReaM for free once it is ready, and there are other open source DRMs emerging (see elsewhere in this issue). Sun would only expect Java license fees, as DreaM is implemented in Java. Sun has also publicly criticized Intertrust’s grasping attitude towards royalties, something that became clear as it requested royalties for underlying patents used in OMA DRM, along with its two parents Sony and Philips, as well as Panasonic and Contentguard.

Why can’t Intertrust, known to be a key patent holder for DRM trust mechanism techniques, license the Coral framework virtually for free, and reap all of its reward in the provision of Coral ready DRM implementations or patent licenses for the basic trust mechanism. This way Coral would win the war, and everyone would want to build a DRM that reflected their business priorities, leading to lots of business for Intertrust. Failure to correctly (and publicly) price the Coral intellectual property is likely to lead to one of the other solutions becoming more interesting for the rest of the consumer electronics world.

All delays play into the hands of Microsoft, which has recently commissioned CoreMedia to write a single bridge between OMA DRM, designed by the Open Mobile Alliance for mobile handsets, and its own Windows Media DRM. This could provide for a world where DRM on handsets or any other devices that behaved like them would be OMA enabled, but content could be swapped from your OMA domain to your PC/Windows domain through such a bridge. Microsoft is understood to have been asked for such a development by a major tier one mobile operator. It’s not the sort of idea Microsoft would just come up with.

Having said that, there is a lot in this Coral announcement of its interoperability framework and home media domain specifications
and its white paper on Eco System A.

In February the first Coral specification painted a picture of how trust can be achieved between two rival DRM architectures, specifying how the authentication of devices should work, and ways to assign key roles to each device such as license issuers, content handlers, rights mediators, rights registries and principal identity managers, but it is this new paper that puts together just how these roles will interact together to form a single home and web based ecosystem.

One key element that Faultline has been banging on about for years, is the idea of a web based rights locker. The idea is that whichever DRM is protecting the content you buy, that it will also make an entry in a centrally held rights locker so that if, in the future, a consumer loses his or her PC, or all copies of a particular piece of content, a new copy can be obtained (assuming that the service operator sets this as its policy).

We feel sure that buying a film which you know you will be able to access and refresh for life on multiple devices, purely for your own use, will lead to vastly different content pricing. We think people would pay a lot for such a service. Okay this is just the technological wherewithal to allow it, and it’s not as if Disney or someone has just endorsed the idea, but we need the technology first, before such a service can be offered.

Also as part of this announcement, two companies that we see more and more together, in Thomson and VeriSign, have both joined the Coral Consortium. Back in November 2004, they jointly agreed to create a service offering in content authentication and authorization to support DRM delivery of digital entertainment.

OK their first target was certainly a bureau for the distribution of digital films to digital cinemas, and that was announced at the end of 2005. But now they should be shaping up to take on IP film, music and games delivery to residential customers over broadband networks. The perfect pair of companies to run and manage such a Rights Locker, as well as the various trust authorities that will be required to mediate between DRM types in the Coral vision of interoperable DRM.

VeriSign already manages authentication and transaction authentication for 400,000 web sites and at the time, Thomson and VeriSign said they were developing technology to offer secure authentication interoperability between a variety of home networking devices including personal video recorders, mobile devices, computers and recordable media. It looks like they are now at least contemplating that this technology will be based on Coral’s technology.

In the new Coral digital home scheme, there is a domain anchor, most usually it will be a home router, which will determine just how far away devices are from the home, to establish them as local or remote member clients.

Member Clients that are in proximity to the Domain Anchor are considered to be in the local environment and may automatically acquire content usage rights. Other clients, deemed to be remote, will be required to provide additional authentications in order to get the same rights.

This reflects the thinking that the DVB project has within its DVB Bluebook A094 Content Protection and Copy Management (CPCM) guidelines published last November, which is being pushed for ETSI European standardization, shortly to be followed by a set of technology recommendations.

This paper put forward the idea of an Authorized Domain Manager rather similar to the Coral architecture with metadata covering all usage state information including versioning for content licenses, a content identifier, a way of creating its own licenses, a record of how the content has been used so far, an identifier for each domain and descrambling keys. It looks to us that the two systems may be being developed in parallel with common contributions from some vendors. That might mean that the usage model would not involve huge intellectual property costs.
In the diagram above we can see that Coral sees its own domain manager being on the web, and talking directly through a home network to each of the member clients.

It is clear from the paper that there is no standardized way in which this will automate copies of content between rival DRMs. Some will have to be re-downloaded, some will have a new content key issued which will work with a transcoded copy of encrypted content, and the system allows for both eventualities.

The system also even allows for existing legacy devices, which are not capable of joining in all the Ecosystem A fun. It does this by creating a proxy for them on a eco system capable device, which in turn authorizes the legacy device, creating a ‘virtual’ client, which will allow it to play content from the secure domain.

But these devices are likely to be gradually timed out of the system, and we think they are a weakness, rather like running two conditional access systems in simulcrypt. Most operators are likely to set a time limit for this capability to expire as well as setting a limit to the number of legacy devices than can be supported on any given domain.

The home domain will also allow for multiple tiers so that personally owned content can be distinguished from family owned entertainment, allowing content suppliers to offer cheaper subscriptions for individuals, or more expensive ones shared among the family.

Ecosystem A also supports the transfer of a domain account from one domain to another, which would be for when a person moves into a new house with a partner, coming from their parents’ house. Such users could take all their rights, ‘unbind’ them from their parents’ domain and then move them to the new domain. But to avoid people moving copies around, each client can only ever be a member of a single domain at any given time.

The Coral effort has, by encryption and standards schemes been relatively brisk, beginning in October 2004, just 20 months ago.

“Anyone interested in digital content distribution can now see how our technology works, and how it can be used to create interoperable domains in consumers’ homes,” said Jack Lacy, president of the Coral Consortium. “These specifications represent continued and unprecedented consensus among the Coral Consortium’s broad and growing membership. We have worked together to define interoperable content distribution solutions that are intuitive and service-rich for consumers while being robust and secure enough for content providers.”

Coral Consortium’s founding members are HP, Intertrust, Philips, Panasonic, Samsung, Sony, and Twentieth Century Fox but it also has other promoter members in IFPI, NBC Universal, LG Electronics, Cisco, Motorola, the MPSS, and conditional access firms Irdeto, Secure Media and Verimatrix. Widevine and NDS as well as AOL and Comcast.

The tempo is likely to increase now that most of the definition work is published, and implementations, as well as a logo campaign (we suggest PlayforReal instead of Microsoft’s PlayforSure) will soon get under way, but until Coral makes it clear what an implementation is going to cost in license fees and patent royalties, none of the consumer electronics companies are likely to commit.

While the scheme is complex, it doesn’t require any more processing to share rights using Coral, than it does to share rights across a ‘Microsoft only’ DRM domain, there are just more points in the process where things can go wrong. After all Coral is dealing with multiple versions of multiple DRMs, and this will mean that numerous plugfests will have to be conducted to ensure that users don’t end up with the same frustrating mess that they started with before the DRMs were allows to be interoperable, copying content that simply refuses to play on new devices.