Interesting discussion about pain point of centralized vs decentralized, If no size fits all approach then that means we can slowly implements better solutions that both world can bring to the table. Especially areas that need to use public or private blockchain such as reduction of paper works but more on secure and efficient communication of all parties involved.It can solve any inconsistency of information travel around the globe in terms of trade as well.There are specific use case for a supply chain to be effective if used hyperledger or private blockchain or semi decentralised blockchain such as VeChain that can track the origin of the products and info stored to authenticate the products and check by validators and users to make better decision about the items.
As has been highlighted previously a full on Blockchain based video sharing solution is challenging. The space required for each video is also a challenge for a decentralized blockchain which will lead to videos being taken down after a time. Also, it was clear that content creators are highly reliant on centralized entities for income/following, and the bans have been rampant with more censorship.
As such, I am proposing to decentralize the below parts:
- The followers/audience will be kept as UTXOs for each user where instead of funds you have users and your portfolio fetches your users whenever queried from the DL. (I’m not even sure that is possible but I’m just throwing ideas
) - The identity management of who can access what channels in the Blockchain with what responsibilities so your channel or your email cannot be banned or removed.
- The payments to the content creators to be paid through smart contracts established clearly and immutable so no centralized platform has control of their livelihoods. It will be linked to a bank API such as the examples we saw with CITI previously
- The approval of the content and overall governance is decentralized in a similar way that people approve transactions in BTC to resolve the censorship issues.
What needs to remain centralized is:
The videos/interactions need to be stored in a database after being approved by the Blockchain and can be fetched by query whenever needed.
The video content must be stored centrally with the best value for money.
The creators and advertisers would use a public blockchain such as Ethereum with smart contracts and tokens. The tokens would have to be bought by both Advertisers and Creators via different tiers of subscription, ie. Bronze, Silver and Gold based on levels of use. Advertisers for example would buy a Gold package to advertise on the largest viewed/most popular subscribers content, and Creators would qualify for Gold with a combination of views and subscription to Gold which they would be incentivised to do with more views.
The advertisers and creators would also use a private blockchain with a database demonstrating all revenue, income and expenditure etc… And would have consensus to decide and vote upon things such as which video platform to use, how much each level of advertiser and creator gets paid, subscription level would determine the level of authority and responsibility for key decisions.
Since storing the content on the blockchain takes a lot of storage, that should probably then be kept off-chain on a centralized server until a second-layer blockchain solution emerges that can solve the centralization problem.
Blockchain-based system should be kept in the part where the value of the content is measured. There should be some kind of voting system based on decentralization to decide what is good content and what is not. Here is where a utility token/stablecoin would come in. Content creators would be rewarded this way, and maybe users would be incentivized to consume content they like by receiving rewards.
My 2 cents.
Video over internet at the moment is not practical to put on b/c due to the large file size and speed required. However there are some people that are working on different aspects of this problem. I see a mish mash of all these as a possible solution.
Theta for example (backed by samsung) are leading the field in video streaming. Video hosters are paid in theta by streamers and advertisers. This is a centralised operation but it may be possible to use cloud storage in conjunction with the devices/computers of a large and strong community, to fileshare videos among nodes. Blockchain could then be employed to handle minting of rewards for hosting, hodling tokens, contracts between users, advertisers, content creators, viewing records etc.
Pain points
Trust to the platform
Payment
Censorship
Accounts getting hacked
Identity
Possible solutions
with limiting the amount being put on the Blockchain in mind
Establish private ledgers for every channel where the tx consists of stats and followers. The ledger to be contained so that possibility for channel owner to interact with followers and prove stats exists no matter what happens to the platform itself.
Payment would be a side chain in symbiosis with this ledger. There would be a possibility to implement a dc token system on the platform, for payments and interactions. Still believe that payment in preferred currency should be possible, which of course comes with the need for trust that someone will act on the “alarm” of the payment ledger and pay you in other currency.
Identity management through digital signature to verify ones ID.
Even tho the censoring part of YT can be used for personal/political agendas/winnings and can cause a lot of problems for creators getting banned “by mistake” or because of some bot/algorithm, the necessity for human correction based on the case’s specific nuance might be needed.
Since YT being as big (or bigger than) of a search engine as google is, it would be impractical to put all the data on the BC because of its slow querying speed.
Hi Filip,
So I’m creating my idea from the perspective of an existing video service business such as youtube.
I have 2 areas I would potentially address:
Channel Identity Management / Content Management / Censorship Management
I think one of the current pain points for Youtube is likely the censorship of channels, removing scam channels and identifying hacked channels on their platform and the General Identity & Content management perspective.
Often when it comes to Youtube identifying content creators and managing, sharing and verifying their identities it can be very difficult. It can at times be almost impossible for Youtube to differentiate between a genuine channel creating and sharing legitimate and non harmful content, and a channel creating harmful or misleading content or scam content.
Currently Crypto scam giveaways we’ve seen on Youtube. These are Live streams sometimes on hacked channels, naming public figures figure such as Vitalik Buterin or Robert Kiyosaki in the thumbnail or the live stream. Youtube is completely unable to remove these Scam giveaways quickly and effectively without Censoring keywords such as Cryptocurrency, Bitcoin or live streams of this context.
For Youtube this provides a genuine dilemma. This shear cost and inefficiency of manually identifying and removing these channels or protecting the hacked channels owner to begin with makes the process completely out of the question and instead Youtube opts to censor key words and Live streams affecting genuine content creators in the process.
This is a problem, Not only does Youtube lose money from revenue and traffic these content creators bring to the site, but also these particular breed of content creators fight back requesting for their content to be put back up and in doing so makes the process of removing these scam channels and live streams even more difficult. Some Youtubers such as Ivan on Tech no longer live stream at all on the site and have opted to host elsewhere. This is my opinion should motivate Youtube to look to a blockchain solution to manage this ever increasing issue.
So my proposal is that video hosting platforms such as Youtube find a way to integrate a blockchain solution to work alongside the existing centralised infrastructure, tackling the issues around the censorship of harmful content without affecting genuine creators and their content and to prevent channels being hacked. If they are able to utilise identity management and certificates similar to how the Hyperledger blokchain functions then they could potentially regulate and govern the content of each channel more effectively and on an individual to individual basis.
The use of a consensus and validation of identity along with traversing content combined with the use of smart contracts (chaincode) to automate the processes, I believe Youtube can effectively tackle this issue and optimise their platform to be fully inclusive and also effectively censor specifically harmful content only.
Hacked channels will almost immediately be flagged meaning that it would be difficult for such a thing to happen. With the presence of ledgers across nodes hacked channels on the centralised infrastructure could potentially be reversed or completely prevented in the future. Harmful content will not be present long before being taken down and content creators such as Ivan on Tech and Datadash among others will no long need to fear the removal of genuine content and for Youtube this could mean the retention of many of these profitable content creators.
Communication of data between Youtube servers and advertisers and 3rd parties
While I have focused mostly on the issue of identity management and censorship, I do also believe there could be a potential pain point for Youtube when it comes to sharing data about channel content and their viewers data to advertisers and 3rd parties.
Many of the advertisements are personalised based on the channel content as well as the viewers interests. Therefore efficient communication of information and data between advertisers and Youtube is essential for the optimisation of this process and for maximum profitability.
Blockchain could automate the sharing of this data and also the strategic placement of adverts for advertisers. Intermediaries could potentially be removed and manual processes equally. While I have focused much of my thought onto the first issue, I also believe a deeper look into this area may proof beneficial to video sharing platforms such as Youtube.
Thanks for reading.
- Governance structure should be decentralized controlled by the viewership. What videos should be banned/blocked. (In the BlockChain)
- The advertising revenue and payment for content creators should be decentralized (In a smart contract)
- The media itself should be centralized, but distributed due to propagation delay/lagging.
An ideal video sharing platform should be a hybrid between centralized and decentralized elements. Due to the massive requirement of storage for the content, that part should be stored in centralized databases. Voting mechanisms such as likes or blocking content should be part of the blockchain with a trustless consensus algorithm. Due to the massive market impact of such a solution, a native token could be created to incentivize creators and viewers. Advertizers could required to purchase these tokens for uploading their content. I would also insure the integrity of the videos through storing each video’s hash on a blockchain. Any hash that is not identical to the one on chain could cause a reupload of the content from a safe server outside the system.
- Contributors get rewarded by content and reward goes up by viewers
- Ads spend could be contributed to contributors and to viewers (watch the ad get reward)
- Engagement reward to channels and viewers that engage
- Video storage on traditional db but censorship removes reward (penalty for censoring) or maybe videos are on decentralized network but distributed by bits only which are shared by some but not all nodes
Having a centralized and decentralized section where if they see that some of the content does not meet their centralized requirements then they can place a rating on it and send it to the decentralized part of the system
I would try to keep data like followers, comments, views decentralized so that this data could not be manipulated from a central entity. The videos itself I would store centralized, to get the right performance to the users…
In my mindmap I tried to represent my idea of a self-sustainable video platform on blockchain.
For speed and scalability reasons just a part is in blockchain. The miners mantain the node but at least Block One also runs the logic. Based on the user profile, it matches the ads with the video requests and pass the match to the app on the user’s device. The content creators can choose to publish on their own servers or on 3rd parties: the token reword (“tk”, 10tk per ad) varies consequentially. Obviously on the bc is stored the premium’s wallet for tipping and incoming transactions if ads-viewing is opted).
I hope my sketchup is understandable, I’m here for clarifications.
Thru the Blockchain
Monetization. Based on number of views & sponsors each content creator has. Users can also get involved in the pmt of the content creator
YouTube Algorithm: helps content creators understand how to reach a wider audience. Some channels post great content but get left behind due to the algorithm. BC would make this transparent to users as to how to get more views
- Proof that users enjoyed the video is watching in full. Create a trustworthy source for users such as the avg completion of this video is 78%. Could feature into monetization aspect as well as just benefitting users for recommended videos.
Centralized
- Video storage
Hi,
I propose creating using a public blockchain to handle user identity of the video platform, monetization for the content creator through tokenized payments awarded from other users, and a fee system for processing transactions of these payments and counts of video id that was served.
The serving of the video would live off chain, but would be linked to the blockchain by sending back to the blockchain only the information regarding the server (video miner), amount of videos served (maybe in file size) and video ids that where served. This way, you have an incentive for video file storing, serving and maintenance in a slightly more decentralized manner and would link the video miner/server to the blockchain with revenue for handling the streaming and validation of what the network has streamed.
If someone one wanted to censor video material it would be harder as more video miners would step in to serve the video that might have been questioned by another miner for its content. Content creators would be more protected by this.
Video Sharing Platform
On the Blockchain
-
Native cryptocurrency
-
Special native objects (things that can somehow be found on the site by completing tasks and are scarce, these can be used to incentivize initial adoption by users and creators)
-
Rules and Regulations (amendable only through voting of users and developers who hold the native token)
-
User profiles with wallets
-
Content creator profiles (excluding videos)
-
Advertiser profiles
-
Miner Profiles
Off the Blockchain
-
Content creator videos
-
Comments
-
Advertisements
Explanation
In order for a hybrid blockchain approach to work for a video sharing service we will have to take several factors into account. The first is addressed above with what should and should not be put on the blockchain.
The next thing to address is how to avoid censorship when you can, since this is the whole point of creating this type of service in the first place. This is tricky because the videos and comments are off the blockchain (mostly due to size limitations), but you must also be able to remove (for legal reasons) any videos or comments that break the law for fear of facing lawsuits sometime down the road. For this reason the rules and regulations should be embedded on the blockchain, and those rules should be governed by the developers of content and the token holders (users of the platform). They have an incentive to not vote for anything that could hurt the value of the token they hold. In other words, while censorship is a concern, they would also likely not want to vote against their own financial interest by allowing the company to take on lawsuits that would hurt the value of their token.
The next is how to incentivize creators and users to move over from popular platforms like Youtube. This can be done through the introduction of a cryptocurrency which the producers of content would be paid out in (as well as paid in tips from users). The users of the platform could also purchase the token, earn it through tips on comments, or also earn it for things like “cakedays” (this is a Reddit term denoting that you have been on the platform for a certain amount of time), and for viewing a certain amount of videos or for viewing a certain amount of time in terms of video content.
Another way to incentivize adoption would be through the ability of early users to find special items that give them special permissions on the platform. These items would have a limited supply and only be found a few times in each way (in order to avoid having it spread online that “this is the way you find it”). Perhaps they would provide a lifetime free subscription to the services premium package (like Youtube has a subscription service), or they could guarantee that a supply of the native token be deposited into the users’ wallet every month indefinitely (just some ideas). They could also be sold at auction by the owner if the owner wished, or if they really wanted to they could gift it, or they could pass it down after they died through a smart contract.
The native token would of course be provided liquidity by advertisers, as well as by users who decided they wanted to exchange their fiat on the platform for the native crypto. Miners would also be able to join the fray through a proof of stake algorithm, meaning that people holding the tokens could earn more tokens by locking them up in order to approve blocks (which could consist of batched new users, crypto transactions, trophy transfers, and other things on the platform).
I had this idea which was basically a blockchain platform the keeps record of events as they unfold and stores them them blocks, so that our history cannot be altered.
This would include speeches from persons of influence and politicians as well and monitoring press/news/article releases and time-stamping them, so that if there are any re-writes/updates to an article (let’s say on the Daily Mail website) they would again be recaptured and timestamped in a future block.
It came out of the frustration that dependent on where you have been brought up, history always seems to contradict another history. The image we have been sold can be so inaccurate.
Imo, the task/exam cannot be solved, as there are too many unknown parameters.
We have to get a complete ecosystem together with producers, viewers, miners, producers, viewers, advertisers.
So the alignment and the focus of the project very much determines how things are implemented.
For consideration:
- If I want a censorship resistant (underground) platform, I need to have the videos decentralized (but at this time needs compromises. (large distributed storage, big processing power, synched on blockchain) and should not have personal data stored of the creator
- If I have a lifestream platform, those requirements are even hire (blockchain-tech)
- if Yvan wants so setup his own channel, he may use a more centralized approach and he will certainly collect a lot of user data (btw: in the academy, I only have the option to accept all cookies)
- if I build a youtube-contender, i will need a moderation functionality to accept a video to the platform (even though it is dezentralized)
Centralized related
a good, fast user experience is important, at that time quite some arguments for a centralized approach
Blockchain related:
- validation of financial transactions for producers, viewers and advertisers makes sense
- the full content distribution network could be a next step, once technology proceeds
- incentives makes sense, so the platform attract miners and can grow its bandwidth, storage etc
- privacy is, where blockchain is perfect (user data)
I would have the Video streaming service setup as follows:
Centralised:
- servers where video data will be stored centrally for quick access, ease of storage and lower costs
- Frontend UI
-advertisement
Decentralised:
- monetisation/validation scheme will be recorded on the blockchain (user voting, views, subscriptions, uploader info).
- Governance model should also be decentralised so that a single authority cannot make decisions. Incentivised voting (stakeholders/stakers are rewarded to tell the truth) Network voting for large changes to the platform
-user deposits and tips
Multi-Tiered approach of centralised reporting system which feeds into another governance model of decentralised voting where actual users are rewarded to determine truth of infringement on T&C.
Opinion:
All media content should be stored on a centralised server in order to ensure efficiency and high storage space availability. Your previous example of D-Tube storing everything on the blockchain does not sound like the perfect solution due to lower speed and storage capabilities.
Website hosts should not be able to take down or ban any content but instead have a decentralised structure where people can vote if a video/user should be removed from the platform.
All other user data could be centralised as long as monetization is kept separate and does not depend on the platform for pay outs. Instead a model like Andreas Antonopolous (excuse my spelling) once proposed, where content creators are paid directly from the customers.
Eg. 1 min watched is 1 cent paid to the content creator directly from the users wallets. With smart contracts surely this would be possible to implement.
The people watching the videos could possibly make their money back by leaving comments. 1 upvote comment = 1 whatever coin into my wallet?