There are always one or two vendors suffering from FOMO (fear of missing out) following a Faultline discussion story. In this case, it was our coverage (in both written and podcast formats) on the ramifications of Synamedia’s acquisition of ContentArmor that caught the attention of UK-based anti-piracy outfit FriendMTS, abundantly aware of being left out of the watermarking conversation.
FOMO is nothing new for FriendMTS. The company has a substantially smaller marketing budget than many of its bigger digital security rivals, as well as a thinner product portfolio, fewer customers, and a failed history of cracking the US market. What it does claim to have, however, is the technology behind the world’s most widely deployed watermarking system.
Our eyebrows raised almost comically, until the names Sky and the Premier League were thrown at us by a representative from FriendMTS, who asked not to be quoted. This ongoing work with the Comcast-owned operator and the UK soccer franchise is ultimately protecting what is widely considered the globe’s most valuable content (the NFL will try to claim this crown, but in terms of pure value of pirated content, that is firmly with the Premier League, we are told).
This is all good and well, particularly as we are informed hardly anyone steals Premier League content from Sky anymore because FriendMTS finds it all. However, part of the reason for FriendMTS’ absence from security discussions is its lack of a holistic approach compared to the big guns of Nagra and Synamedia, preferring to ply its trade purely in anti-piracy and monitoring. It doesn’t dabble in delivery security either, where it claims most of these larger rivals are focused.
Re-streaming piracy is the greatest concern for FriendMTS in the OTT era. Essentially, this is anything with an HDMI connector, where pirates can siphon off content simply using a $5 HDMI splitter, then re-encode content and upload it to the internet to a pirate wholesale ecosystem – and away they go. Desktop players are still a target for pirates, but the rise in CTV viewing has correlated with a rise in CTV piracy.
Therefore, FriendMTS sees itself competing more against companies who want the real DRM business, finding itself partnering with the likes of Intertrust rather than competing with ContentArmor and NexGuard in watermarking, and their parent companies in a broader anti-piracy sense.
That said, FriendMTS says it holds some patents related to watermarking, within its proprietary ASiD (Advanced Subscriber ID) technology, and is not dependent on any third-party patents, unlike others in the security scene have been over the years.
There are several methods of watermarking. ContentArmor is known for its bitstream-based approach, and A/B watermarking is a common method, while FriendMTS’ favorite approach is client-composited watermarking, where it generates an imperceptible (not invisible as others would say) watermark. Client-composited has a fast extraction time (because the payload of the watermark is throughout the video, so there is less messing about to find the start of code), and lower costs based on no increases in delivery infrastructure or storages costs. This method is also multi-CDN ready as it can be embedded on the client device, and can adapt to the behavior of pirates when they think they may have found a way around it. It can be deployed to broadcast platforms too.
The advantages of Client-Composited watermarking are best visualized when outlining the downsides of the common A/B variant watermarking technique.
A/B watermarking is optimized for on-demand content, has a slow extraction process, can be easily broken, is inflexible to pirate behavior, comes with significant costs (increased delivery infrastructure and storage), is designed for streaming media only, and is not multi-CDN ready. We are aware of A/B being adapted to multi-CDN environments, but this supposedly increases complexity and why bother when it can be easily broken anyway.
While bitstream modification is a strong technology, we are informed this method doesn’t work so well for live content due to heavy processing requirements. That said, ContentArmor has developed a technique for reducing storage and CDN bandwidth requirements by removing the need to duplicate video streams – improving cache performance and speeding up watermark embedding. There will always be trade-offs and mud-slinging.
Zooming in closer, ASiD Embedded is the existing client-generated system for subscriber-level watermarking on set tops. ASiD OTT is a client-composited, server-generated system for internet streaming devices. ASiD OTT Edge-switched is a server-generated, server-composited A/B variant watermark, suited for devices where you have less control of the client side.
Although all three are suitable for live and VoD, FriendMTS warns there may be some added latency when using ASiD OTT Edge-switched for live content, plus users may experience a longer time to extract watermarks due to the temporal sequencing of A/B watermarking.
While the 4th generation product presentation was a long way from blowing us away, ASiD customers will get important incremental updates including optimizations to enable variable watermark parameters for different content types (UHD, HD, SD, linear, on-demand) in a way that is CPU and memory efficient. Client library hardening has also been tweaked and now supports multiple encryption modes.
Watermarking could be considered a surgical scalpel in the anti-piracy arsenal, to accurately pinpoint sources via smartcard ID or OTT account, while something like real-time delivery server blocking is another technique employed by FriendMTS which is a more sledgehammer-type approach, but equally as successful.
Looking at the business model, we were surprised to learn that, in most cases, FriendMTS operates a payment structure where customers don’t pay any fees if the technology doesn’t do its desired job, which could either scream of stupidity or confidence, whichever way you interpret it.
We believe FriendMTS has been held back from reaching its potential by founder Jonathan Friend, who has been blasted in the past by sources speaking to Faultline for his incompetency in the CEO position, while clearly a pioneering mind in the content protection space. This was finally nipped in the bud in June this year, as Friend stepped back into the CPO position, allowing him to focus on product development, while Derek Chang has been appointed the company’s new CEO.
Now at over 100 people and hiring like crazy, a shake-up of the oft-criticized FriendMTS hierarchy with fresh blood in the CEO driving seat could shape up to be one of the most significant decisions in the firm’s history.