I’ve read a fair bit of your whitepaper as well as did some searching through the forum and didn’t see mentions of DRM, Rights Management and Copyright Protection. I was wondering if you could comment briefly on how you see Livepeer integrating (or not) with those concepts.
For elaboration, when I refer to those “concepts” I’m basically trying to get at the problems surrounding the protection of creator/publishers content and prevention of fraudulent creator claims and ensuring that payments or incentives go to the actual creator and not a malicious entity that happens to be scraping video/audio frames from their instance of an LPMS.
Yes very good questions. As an open platform we welcome anyone building existing DRM solutions into Livepeer Media Server itself to satisfy the needs of copyright protected content in centralized use cases.
When it comes to using the Livepeer network itself to transcode and distribute video, there are solutions for doing key exchange and encrypting/decrypting the video for specified users who pay for the content…however at the current moment the transcoder is still a special role, and as you point out, they will certainly have access to the raw frame data within the media server when doing the encoding. As they are elected, and have a lot to lose, reputation-wise and therefore economically, by acting in a way that isn’t in the best interest of the network. If they leak your content, then they could lose their valuable job.
One of our community members does have an idea for how to use DRM to implement a “proof of play” at the interface level, and he’ll be writing up a proposal in the coming weeks. In the meantime, if you have ideas for incentives to facilitate the roles necessary to support DRM in a decentralized way, please share them. Whether they support challenge/response, or encryption of the stream at the edges, or key generation/exchange services, it would be interesting to think about how to layer in the incentives. It’s not at the top of our roadmap for release our current release milestones, but it will be important to allow Livepeer to satisfy all use cases in the future.
Thanks Doug. You touched on an idea that I think I’ve now seen on the Lbry docs as well. You stated,
As they are elected, and have a lot to lose, reputation-wise and therefore economically, by acting in a way that isn’t in the best interest of the network. If they leak your content, then they could lose their valuable job.
The idea underpinning this seems to be one of – if it’s more profitable to contribute legitimate activity, then it discourages illegitimate activity. Or why disrupt and gain less, when I can help and gain more.
If it is impossible to keep drugs out of prisons, it will never be possible to enforce copyright via analogous tactics on the infinitely less-controlled internet. Instead, focus on enticement. While legal compliance is paramount, concentrate as much as possible on making a system that relies more on giving people no excuse to do the wrong thing.
They also make some interesting points about the durability of the blockchain revealing a record of illegitimate activity (like double publish) and also the idea that content owners still have to protect their works by seeking out violators with available tools and persecuting theft appropriately.
I’m putting together a research package for a commercial entity and this copyright issue is a big concern for them as they work with large studios. The issue is one that they will likely use as a claimed competitive advantage over decentralized content publishing systems like Livepeer. However, If the incentive system and ledger arguments work as characterized it could prove to nullify that claimed advantage. That being said it will take time for fears to subside and the model to prove itself out.
Could you please elaborate on the notion of “proof of play”? What is being proved?
This is related to the economics of who pays for bandwidth. If the broadcaster wants to pay the cost of bandwidth for delivery to an unlimited number of viewers (the way CDNs traditionally work today), then this presents some challenges in a decentralized network. Anyone providing bandwidth could spin up a number of nodes (think clickfarm) to request the content and inflate the viewership numbers, thereby directing money from the broadcaster into their own pocket.
This wouldn’t be a problem if the viewer pays for bandwidth…but this breaks traditional conventions of user experience.
What one of our community members was suggesting via “proof-of-play” was something along the lines of using DRM like solutions to essentially create unique copies of the content per viewer, and to receive some sort of confirmation that the video actually was played in a player for a human viewer. I’m not certain what the details of this proposal were or how it would work exactly, but solving the problem of letting the broadcaster cover the costs for verified viewers in a decentralized context is important for many use cases.
DRM (Digital Restrictions Management) are Non-Free. Making provision for DRM in Livepeer would mean that it would be classified as non-Free software by the FSF (Free Software Foundation).
Free Software should be supported, as it is a firm foundation for digital platforms. Communications are less trustworthy if parts of the code cannot be audited.
I hope that there is no support for DRM in Livepeer.