Here is my idea.
The storage of content(videos) should be centralized. The monetization and engagement interms of followers should be decentralized. The censorship of comments and content should also be decentralized.
What do you think? Share your opinion and let’s connect.
Excercise: build your own business blockchain (based on youtube system):
Definition of the problem:
Youtube hosts videos that are uploaded by individuals who want to show something to the world.
Individuals that post the videos can gain popularity wich will result in followers. also the individuals that post videos get revenue from youtube when the video is viewed many times.
The problem is that youtube is a central authority that can do whatever it wants to the visibility of the video’s posted and to the video’s itself:
what do we want to solve to get a better product?
how can we solve it?
authentication (make sure that a real person logs in safely): make use of a blockchain technology that specializes in KYC/AML (for example eKYC built on hyperledger sawtooth, maybe there are much better solutions but it gives you an idea).
In this way: the data stays secure / can be used to communicate with financial institutions (needed to receive the monetary rewards from the platform)
log every valid user on the blockchain so nobody can remove (ban) a user from the platform.
authenticity of video: store a hash made from the original video on a blockchain to prove it’s authenticity and link it to the authenticated person
because we do KYC/AML we know that all the logged in persons are real persons and the only thing we need to know is what video’s they look at and how long they watched it. This can also be stored on a blockchain so nobody can argue/change those statistics (and end users stay owner of their own data and this data is highly secured)
make sure the rewards are payed: when you have a smart contract that evaluates the watched time per video (and we allways know who posted the video), the smart contract perfectly knows how much the content creator must be payed.
prevent censoring of video by not showing it in the search results: you can log the popularity of a video and store it on the blockchain. Popularity can be expressed by end users for example through likes (only 1 like allowed per video per user). So when someone searches for a video, the most popular video should allways show on top. If not, end users should be able to see this.
prevent video from deletion by youtube: we could put the video on the blockchain but that would be a very bad idea because of the space it will take and how slow it will be. So the video storage must stay centralized. But since every video’s authenticity exists on the blockchain we can easily see if a video is being deleted so if the platform would delete it, it will not be accepted by the community and people will stop using the platform.
I think these general implementations could make a better platform. Off course you can go in much more details and discuss even more features (for example what to do with ads) but these are in my view the most important parts that have to be solved. A video platform can never be fully decentralized if storage space stays expensive. When there is a moment in time when we have unlimited storage space for allmost no money, then a bitcoin-like network could be created where everything is decentralized and where every user is a node in the network to secure it.
Well It’s complicated. But if I’m thinking from the client perspective I wouldn’t want to pay YouTube that much for their service. It would be hard but I would build a business , set up deals with ads and marketers and then set up smart contracts to pay per view on the videos. That way content owners could keep the majority of the money.
Also I would want to make sure I’m paid fairly so I wouldn’t want anybody controlling the money.
The videos don’t have to be on a block chain because it would probably make things slower. That could work as a normal database.
Also you would think that you don’t wanna censor peoples content and that’s why people would want a decentralized video company but people would then be able to put sexual, offensive, violent videos and nobody would be able to stop them. So you would have to be very careful with allowing any content.
I would start with making the payment part Decentralized and I’m not sure if we would need a token .
Each viewer automatically duplicates/seeds/has a copy of the blockchain of the videos he/she likes. Kinda like torrent seeding. Those would be true likes and would get deleted when the video gets deleted by the viewer who liked it.
A like would emit a coin/token while unliked=deleted video would burn that token.
This would prevent something like Luke’s letter (if it would be a video) from the future to be just deleted like Reddit did…
To prevent existence of consortiums that uploads ‘1000 hours of silence’ videos and like among each other (EOS haha) structure something like ERC-1155 would have to be used. Where the depth of your token pack tree would have a reputation angle to it.
More lesser quality videos liked would oversupply the market with tokens/coins and increase the individual data storage needed. So people would keep/like/store only content that they really deem worthy. So hopefully this would reach a sort of an equilibrium at some point. Kinda like difficulty in Bitcoin…
This wave of content would follow with the society’s interests as the older videos would get deleted.
The forces who ‘change’ history at the moment would approve of this platform… As the new ideas/concepts/‘truths’ would automatically blend in the wave and the original truths would disappear with the generations. Like Pascal coin…
‘Sick’ side of the society would be a problem as the recent video from New Zealand would most likely be among the top. At the same time when sickness is identified organisations which would try tackling it would appear. Kinda like Free Software Foundation in the IT world…
Actually each video essentially would have an individual sub-token and chain so that Luke’s letter would win over the consortium of ‘1000 hours of silence’ videos. So if you have liked 9 videos, you would have a copy of them/their chain on your device. And a mix of private and privacy videos/chains could be utilised like in Optherium.
Plus some propellant tokens to generate residual income/rewarding for keeping/seeding your videos and their chains online.
Also video content creators would end up delivering the information in a lot more dense and efficient manner to reduce the file size as viewers would prefer shorter videos to store/like. Again a balance would end up being created. Still, more tokens to the longer videos and less to the shorter to prevent the short/‘frequency’ video creation by consortium liking bots… Longer would not be efficient for them because of the storage cost. Unless they create something like that small Rar within Rar file which weights petabytes when attempting to extract it…
No more post edits of SpaceX live streams like they do on YouTube :)))
It would be a nice gig for me for the next bull run…
The best exercise so far! I actually enjoyed this.
That was everything on blockchain example.
For a mixture of blockchain and centralised solution - the videos could be stored on ‘nodes’ and only likes being on the chain.
at first, I would made an research about storing data in this decentralized market. I think that storing data using some of these storing services is more effective than centralized way. if its not possible or so effective as I thought, then store all videos in centralized way. every other information and transations put into the blockchain.
Here is my thought process:
What goes on Block chain:
Identity management
Monetization smart contracts
source data location (like AWS S3 bucket location)
thumbnails
date and time uploaded
digital signature
video length
Video headers
Followers
What goes on Cloud or central database
Actual video
Centralize storage of content (especially due to file size), subscribers, likes, etc.
Decentralize metadata related to the content (content hash, ownership, date/time stamp, etc.)
Smart contract(s) can be used to track views and related advertising payments
Well, this is exactly my honest thoughts would have been too regarding the subject.
But since Filip wanted to channel our thoughts, I would like to say:
Critical pain point:
The key asset for Content Creators (CC) are their followers and being able to reach them. There’s a critical pain-point should the video-sharing platform become unavailable for CC and/or their followers. Why might this happen - CC may have their account suspended, access restrictions may be placed on platform from a specific country meaning these CCs or followers residing there can no longer access the platform or CC channel. The CC then has no way of contacting their followers to put alternative arrangements in place.
It seems to me that video content itself is of secondary importance, as CC should have their own copies of content and so if necessary content could be upload to another video service. Also video content itself is not currently suitable for keeping on blockchain (due to its size and performance required to deliver content).
Solution:
Keep each CC follower’s details on a blockchain, separate from the centralised channel that holds the video content. In addition the CC could write to the blockchain which platforms they are using currently. The CC would need to encourage their followers to also register as a follower on the blockchain, e.g. via a link in the channel description (as I assume it would not initially be possible to get all platforms such as YouTube to directly integrate with such as system).
If a CC creator looses access to a particular platform, by the CC notifying the blockchain they could inform their followers and suggest they continue to follow him/her through a specified alternative platform. Specifically, perhaps the blockchain holds each followers email address (encrypted) and a smart contract emails the CC’s followers automatically with the details of the alternative video platform.
Likewise followers could flag up to the CC (by updating the blockchain) if they are banned or blocked from the video sharing platform - the CC could then see if large segments of their followers are no longer able to follow them (e.g. as a country has blocked access to a specific platform for its citizens). This would make the CC aware, who could then make alternative arrangements so these blocked followers can access their content once more, e.g. by putting content onto a secondary platform that isn’t restricted for these followers. The CC would then update the blockchain with the new platform details and the blockchain smart contract would notify the CC’s followers.
There is already lot said, difficult to come with some news, so lets take a complete exotic new path. Why don’t we offer a complete decentralized video hardware box on something like a raspberry pi with 2 TB harddisk. Cost for the creator is let’s say 350 euro, which offers a complete setup to upload video-content and as soon it’s connected to a network peers will find each other, channels will be opened. User can make use of a mainURL where all the videoclips comes together or user can create his own domain (non stoppable domain for example). On the node there is Bitcoin, lightning and IPFS running. Channels will be opened between the peers and the amount of upload and download will be recorded between the nodes. Creators can choose to activate advertisements which users can get rid of when they pay an amount of sathoshis. By using Bitcoin with lightning users and advertisers will be rewarded per second. A RSK layer for the smartcontracts between advertisers, creators and watchers. Probably a lot to shoot on, but maybe a total new direction as inspiration 
Discuss a business model to compete with Youtube.
It’s an interesting topic and I have a couple idea.
To compete with Youtube, we can deploy cutting edge strategy where the the uploader or the content creator need to have responsible for what they doing. They need to stake amount of token (to prove their good will) to upload a video to the platform. And the viewer can choose to support the video content they like by paying the tokens to the content creator.
In youtube case, the thing going wrong is not only the content but also it’s the suggestion system. The content can easily manipulate the suggestion system and the viewer don’t have the option to opt out what they don’t want to see immediately because content creators pay lots of money to advertise their content. In the new platform, when the content creators want to advertise their content, they need to stake an amount of token to cover the numbers of people they want to deliver the advertisement. The larger number of people they want to reach, the bigger amount of token they have to stake and once the viewers show their dislike (veto) the content uploader will loose their stake.
The data that consumes large space shall not be put on the blockchain like the video itself. The videos shall be in a centralised database. The users’ personal details, his subscribers, his views, advertisements linked to his account shall be kept in the blockchain.
Video sharing proposed arrangement
Concerned parties
Items involved
Design structure:
I personally think decentralization would solve most of the problems, that are currently the main talking points around Youtube blocking/deleting video’s. Advances in technology happens at a fast pace, so I think that data storage are the main issue that hampers the use for blockchains at the moment, but will not be a everlasting problem.
My solution to this for now would rather come back to permissions to access content maybe through a smart contract. On sign up, give the user all the responsibility, have a list of all major contentious topics and age inappropriate content that the user have to activate to get access to that specific topic, but also when watching something that you do not approve of be able to ban content or a specific channel from your list of accessible video’s, replacing the Report Button with a Ban Button. Add tipping from the users to the content creator. In the back-end, also give advertisers the ability to ban their ads from topics or creators that they don’t want to be associated with a Ban Button and the ability to choose general and specific creators where they prefer to advertise. This type of approach can be greatly enhanced upon, but this is according to me is a viable type of approach where human beings are the variable - let them make their choices.
Is YouTube even a blockchain candidate? pure
no. Hyperledger has some possibilities though.
the bigest issues wit YT imho are. views, payments, copyright and censorship.
can views can be faked? if so put em on the chain. (will this cause performance issues?)
censorship is an issue. I’d would like to see some sort of governance protocol tied to that. videos/channels just disappear arbitrarily.
videos are very memory intensive so I wouldn’t put them on the chain.
views are tied to payment so we need a ledger for that. (chain it)
really ANYTHING dealing with payment should be chained this increases trust. More trust more use. Particularly by content creators.
copyright could be tokenized and therefore easily monetized transferred/tracked ectera
the resulting YTBC api could communicate with ascap / BMI ectera.
Distribution of royalties could be streamlined aswell and made easily verifiable via the blockchain ledger.
However purchasing “likes” and “views” still remains an issue
Nice try Filip & Ivan… trying to make us do all the dirty work for your video platform 
On blockchain
-Payments to content creators for views through advertising (similar to BAT token)
-Also integrate a ‘tip’ section where micropayments can be donated, even fractions of cents
-Hashes of original videos. Although someone can edit content after they steal it, it can still be proven that someone had the video originally if it came down to it.
-Option for voluntary identity verification of users that can boost their reputation or authenticity ranking
-Constructive criticism system or token whereby you can give back to someone that commented on something or helped you out
-Payments to music artists per play if you use their music
Centralized storage / servers
-Storage of all content
EDIT: typo
Thought process
1st stage: immediate thoughts
On blockchain:
identity of owner, user and payments, rules and regulations of the platform, information about the users behaviour. Most of this information should be secured and private though.
Why? I want traceable and secure payments. But: transactions need to be immediate.
Probably not on blockchain:
-relationships between the content creators and followers, between users/followers etc.
Centralized: content as we don’t want it to stay. Alternatively other solution but certainly no deleting.
-> not sure if all is feasible that way
2st stage: general issues and problems to take into account:
3rd stage: Summary
after having read many of the previous answers, I saw a few interesting ones but I believe, because we are trying to design form an already existing platform, we are very biased. A lot more is possible than we probably believe possible now.
Difficult question but good to go through the thought process. It would definitely be interesting to see more real life examples from close up when companies implement blockchain technology.
Develop a video sharing platform.
A- As a first step, what would be my main product?
1-The platform for sharing content.
2-The security and veracity of the users.
3-Free communication by not allowing Shadowbanning.
4-The monetary incentive (token) to the users of the platform to share real and / or valuable content, as well as interact in a respectful way on the platform.
B- Of the previous points, which do you really need decentralization?
1- For the part of video storage being decentralized is not a primary point since they can be stored and accessed more agilely in a centralized point.
2- The subject of the users I feel if it must be decretrlized by the issue of validation of the veracity of the information and also in this way the publications, modifications and deletion of content that make them is registered.
3- The Shadowbanning goes hand in hand with the previous point so it is not decentralized content is blocked but if it is penalized if it violates any of the rules of the platform.
4- The incentives must be decentralized to be able to make use of the token and that this is of value for the users since with it it could have access to premium features when paying them or to be able to give strength to the post of the video and that it appears in the first wall results. So you can also give tips to the best content creators.
C- When analyzing it we can see how several points can take advantage of decentralization, but we must think about which of all would be the first one to carry out in order to trace the project in an orderly manner. It is not only decentralize you also have to analyze Hyperledger help? Can I do it on an existing blockchain? (Ethereum, Cardano or EOS) Should I create the blockchain from scratch? What would be the consensus to use? (PoS, DpoS, PoC, etc.) Looking for public or private decentralization?
Pain points:
Censorship: content will be public and not restricted/stored to a single database/country/owner
Copyright: can transverse the blockchain to verify who is the real owners of the content
Payments/donation: can use payment systems or crypto tokens to incentivise content creators. E.g. BAT or DLive
Follower/audience level subscriptions: can create different permissioned channels to present different content to audience based on their level or following maturity/seniority
Ownership: if blockchain is public, the content creator has full control of their product (no “central” owner of the database). Node or server providers storing the content could be incentivised for doing so (by being paid from the content creators)
Video platforms like YouTube have complex configurations that I’m guessing I wouldn’t even guess exist. So, in short, I would say to keep video storage centralized for speed and storage capabilities. Put the most crucial information for the content maker related to subscriber base and monetization on a blockchain. The other less crucial data could be centralized for efficiency.