Welcome to the discussion about this lecture. Here you can ask questions or post feedback about this specific lecture.
Here is my simplified idea:
-
Use token to pay content creators
-
Stakeholders of governance are: creators, advertisers and users. They decide fair use policy and what is allowed.
-
Creators to decide how the advertising revenue is distributed and how it is presented to users
-
Automate (like in a bidding process) the sale of advertising.
-
Centralize storage of most videos but leave option to store some videos on the blockchain (for example to be used as proof. This feature will be chargeable)
Videos are memory intensive, so they probably should remain on a centralized server or on each content producers computer, which is accessible via the internet. Advertisers need proof of the consumers watching the videos. So, the blockchain can record the honest watching record by users. A payment smart contact can decide a payment scheme based on the watching record. So, the watching data and payment based on watching data is protected on the block chain, transparent and all that good BC stuff. Payment can be made via a crypto currency that is donated by consumers and paid by advertiser. Perhaps consumers and advertisers pay a crypto currency subscription fee and payments are made from these fees that are tracked in the BC ledger.
I still want everything decentralized so i would partner with STORJ or SIA to make sure that we donāt have to delete videos. That seemed to be the main issue if its processing power the Golem , if its ram EOS . my point is use other platforms strengths to run yours.
Hello, in my opinion, the part to put on the BC is the revenue distribution part. In other words, how many views, and how much advertising per video, and how it is distributed. That needs to be put on a BC. Those elements are the parts where we donāt want to trust a 3 rd party, and we donāt want an intermediary. That is what is opaque and needs to become transparent. The filing of the videos are better done in a centralised fashion. What people reproach to Youtube is that they take too much money, so that part needs to be decentralised, trustless, and verifiable.
Essentially, most succesful YouTubers are business people who are monetizing their content, even if that wasnt their original intention when joining the platform. The problem with a centralized platform like YouTube is that they(YouTube) can, at will, remove or reduce exposure to someones content thereby adversely effecting their business.
The solution, in my mind, is to decentralize only the metadata which would include a link relating to the video content ,by placing it on a blockchain. The video content itself could be held on a centralized server which could be hired or owned by the video creator themselves.
This is, of course, a compromise on a fully decentralized video platform but could serve as a better option than a fully centralized platform at this point in time.
Why put a āgeneral purposeā video sharing platform, on the blockchain? Why put anything on a blockchain? An earlier text reading indicated there are three key criteria that drive making a database a blockchain:
-
Uniformly ordered: Every block contains a record of all the most recent transactions and a reference to the block that came immediately before it. This results in every transaction being able to be traced back to its origin. This would be useful for video content creators to assign attribution and get rewarded for providing content with tokens.
-
Redundantly maintained (by multiple computers) to guarantee the consistency or nonrepudiation of the data: Each node (a computer containing the ledger) in a blockchain has a copy of the transaction ledger. There is no ācentralā owner. All nodes see the same data at the same time and all nodes have equal rights to the data. This perhaps may not be suitable for all video content creators who may prefer access to the data to be controled by a token. This would allow tracking and ranking of consumed content at a more granular level.
-
Validated using cryptography: Cryptography is the science of protecting information by transforming it into a secure format. This technology allows blockchains to verify users and blocks. For a āgeneral purposeā video sharing platform, content creators would sign their work, and content consumers would sign transactions. The utility of that depends on the community.
So perhaps a āgeneral purposeā video sharing platform, has no inherent requirement for putting any data on a blockchain database. The need for putting such data on a blockchain is heightened when perhaps factors involving other content data considerations, such as immutability, censorship resistance, non-fungability, and identity. Or ensuring fair and equitable models of value distribution. Then, depending on the purpose of the platform, perhaps parts of the platform which are not typically thought cost efficient or performant in a general platform solution may be worth the investment.
Perhaps it depends on what audience or community your video platform serves.
Many people have proposed good ideas here. Even though I donāt necessarily want to defend youtube, I have to say running all their servers and warehousing all the videos, must have a cost for them. Acquiring content from creators also has a cost. So if we just distribute all the revenue directly to the content creators, that is not quite fair. We need to take into account youtube cost structure.
I would say that amount of videos and the capacity would be a significant problems for blockchain, therefore this should stay centralized (centralized servers). However, the quality of content, subscribers, likes, channels etc. could/should be recorded by blockchain. For example, content creators should be getting paid via blockchain as soon as a new subscriber is added. Also blockchain should help to distinguish between watched video or just āclickedā video, that is to count a view if at least e.g. 50% of a video was properly watched. This should help reward and encourage content creators to provide not just some content but to increase quality of videos.
I want
Dezentralized
Controle of my own data
Seamless sharing
Good UX
Flexible functions of who can read personal data
Searchable in a search engine
advertising possibilitys
Open Sourse
Problems
Controle of content in different channels
Decreasing of free Speech
Different regulations of what type of places that are allowed to visit on the web
Space of streaming media
Less ability to track sensitive data like IP, without losing ability for advertising.
Ordinary users must store their files at an ISP
Internet today are based on different search engines. Google, Bing, Amazon, Alibaba, Youtube, Twitter, Pinterest, Facebookā¦
The last years for example Google also use API from Facebook, twitter, Pinterest. Results of different types of social channels are chowing up on Google on the serp more and more.
I would like to dive into how toā¦
- Build a new type of searche engine with an API that can be integrated with other engines
- A sort of application with templates so you can create your own place similar like WP templates but with integrated social functions.
- You put sensitive personal data in the blockchain. Only you can administrate your own data.
- Investigate how to take advantage of new technic like 5G and 6G
- In order to save space. Is there new ways to zip streaming media and unzip on the client in real time, what if there are other technical solutions than normaly are used? New solutions for pictures and video formats so they can be stored in a blockchain.
- Integrated advertising system that, in first place give miners an income for living.
There are many differet ways but we all want to own and controle our own content and the right for free thinking and speaking.
Aside from monetization and followers I think itās useful to put the video description/title/comments or at least give it as an option to users to do so. Some content creators might fear that the platform will delete their videos so they might include a link to it uploaded somewhere else in the video description and still get their subscribers to watch it.
You donāt even need your own database you can use Youtube, Vevo, or any other platform to query the video.
Hope I didnāt get too verbose. . .
Problems associated with blockchain videos
- Storing files in the block chain is expensive/inefficient
- End users want transparent monetization for views, likes and shares
Solution
The hybrid network will include conventional file storage (cdn) and delivery, along with a
block-chain-based backend to facilitate membership, video submission/management and monetization (payments)
Rationale
- Storing files in conventional way is cheap/easy/perpetual.
- Using the blockchain for membership/accounts will ensure authenticity in the users.
- Using the blockchain for membership will inextricably link a video to its uploader/author.
- Using the blockchain to manage the monetization policy will let users pick a policy that best meets their needs.
- Perhaps a policy will let an idle video expire/archive/delete after x months.
- Perhaps the policy will reward likes more than shares.
- Perhaps the user has in-video advertising and only cares about views.
- Perhaps the content is time-sensitive, so monetization only matters the first 2 weeks itās available.
- Using the blockchain to manage the shares/likes/views will inextricably link who liked what. Further, it will handle the monetization in a rigid mathematical way.
Use Case - Create account
- User visits website and logs in or creates an account/key. Not much to really say here.
Use Case - Add New video
- User log in
- User clicks the āUploadā button and is prompted to select a video file, pick categories, thumbnail, etc.
- User picks a monetization policy for this video.
- Block Chain Peer nodes process the request and add the file reference (and settings) in the ledger. The file is uploaded to the CDN.
- The overall application is updated to make the video visible.
Use Case - Random user āViewsā or clicks āLikeā or 'Shareā
- User log in
- Watches a video - this triggers a smart contract to update the view counter and process any monetiziation payouts.
- Clicks āLikeā or āShareā - This triggers a smart contract to update the āLikeā or āShareā counter and process any monetization payouts
I think my big learning after watching filipās solution was the slowness of the blockchain and how that might alter a solution. I forgot about that (as it pertains to user management, in particula). I also didnāt really understand the essence of social media and monetization and censorship. I built monetization around likes and shares, when it should have been around followers. In my solution, I have a lot of it in the blockchain and that might be ill-advised if I want a fast user-experience. A ālikeā shouldnāt take 20 seconds.
Hmmm, I am not an avid content creator myself, but from a business perspective, both as a content creator, content host, and sponsors (advertisers), I think the money trail has to be considered for any blockchain solution to help.
So I would first try to clearly identify the problem.
Lets say, Iām a content provider. I want to ensure after I have built a tremendous following, and I produce some content that is deemed āquestionableā by a central authority that does not want the liability of being associated with providing the content, I want to ensure, my content is preserved, and my ad revenue continues. The problem is that a third party can decide to pull the plug at any time. Whereas I want to ensure I can keep building on what I have started forever.
The solution I would propose is to have content providers associated with a key, which would tie the creators and their content. The content could still be stored on a centralized server, or any server that was adequate, letās say SIA had some break through, and was able to be as fast as Google for example.
Now you could set up certain metadata, collect certain data about your followers, that they could control, and use that data to provide targeted advertising, all through smart contract that would facilitate, consumer preferences, advertiser preferences as to audience, and enable a system for payment based on whatever agreement.
So to be somewhat clear, the parts that would be:
On the blockchain, or would use the blockchain:
- Setting up agreement parameters with advertising or other revenue generating , either by selling data, or watching ads. As well as with the storage provider of the file.
- Content users or consumers could customize their preferences for one or many different ways for different content providers. Eg: General use, and a special allowance for āWhatever you wantā
- Agreement and identity verification of content creation, which would hopefully satisfy any reservations that central authorities would have about liability for displaying such content
Off the blockchain:
- The actual hard copies of the data files (video content),
- server
Exercise:
First of all, I would like to describe my own idea of what a content platform could be, in terms of its structure. For this exercise I imagine a platform that has elements such as:
ā¢ Videos
ā¢ Users information
ā¢ Comments
ā¢ Like&Dislikeās
ā¢ Followers
ā¢ Views
ā¢ Payments
I classify these elements in two groups. Each group is a database as follows :
Information, a Private SQL DB with sensible information like: User_ID, Followers, Payments.
Content, a Public NoSQL DB: Videos, Video_Views, Comments and Like&Dislikeās.
The platform has an API which connect both databases with a Blockchain in order to store all transactions depending on the user activity. Each transaction has its specific value and payments will be split between the platform owner and users.
Activity demands trust so it is necessary to have a blockchain to store this part of the structure. That involves transactions, like upload a video, its interaction between followers and general public and payments.
The other information is too heavy to manage into a blockchain so it can be stored in a CRUD database.
Create a public Blockchain that is open source
On Blockchain
Profile Data
Likes, Comments, views and a hash of all content posted are held on the Blockchain using private keys or nemonic passphrases to prove that the user is who they say they are.
Not Blockchain
The video content
Can have centralized storage
Torrent Files
Direct Download
Even your friend with a USB stick
Because the Hash of the original is stored publicly and securely redundant across multiple systems. This could be how we confirm the authenticity of every file ever made. People could refuse to open anything that the hash is not on the blockchain. All front end versions could then just match the hash of the file with the hash on the blockchain. If I view it or make comments etc while running the blockchain front end that is all stored in blockchain. This could also create profiles that show trust worthiness of the source.
I am not sure the best mean to monetize this some type of crypto would have to be added to allow the consensus mechanisms to stay secure.
Let us divide the youtube processes and consider each separately;
Channel
All text data inside the channel which is open to the public such as channel name, about page, video name, description etc.
Execution: A permissionless blockchain with a public ledger.
Private Information
Personal information such as Name, birthday, e-mail address, IBAN Nr. etc.
Execution:
1- A permissionless blockchain with a private ledger, with zero knowledge proof to hide the information from the public and still work as a permissionless blockchain.
2- The monetization mechanism, payment processing intervals etc. will be coded inside the separate smart contracts which will be executed fulfilling the timeframe and view count and other parameters set within the contract.
3- Profit generation will happen through Advertisement process which will be organized on smart contracts that will consist of the algorithms, advertisement provider information, content, payment rules and intervals.
Video, Image(Thumbnails) Content
Basically, the content that requires the 99% of the storage
Execution:
1- The content itself is stored in the decentralized databases with a reduced number of nodes. Letās say 10 server locations will contain the copy of all contents.
2- We need to generate an algorithm that can read the video file and render it into a hash. It could be a process where the certain timestamp(image) color matrix could be converted into a number which then will be hashed and stored on the blockchain. This will allow us to check if the content in any of the servers has been maliciously changed. This way a certain level of decentralization will be achieved.
3- The Database will contain the CRUD functions but these functions will be either controlled by the owner of the content with its private key or with >50% consensus. A central authority will be eliminated. Consensus will decide to remove the disturbing content (Hate speech etc.)
Right, this is an important aspect not always mentioned. But storage costs are not that high these days, so YouTube probably takes a disproportionate share of the revenues for itself. That is where the opportunity exists, for other alternative modes of revenue distribution to be considered.
Unzipping video content at the client is not always a good idea, as some clients have very limited computing power, significantly affecting UX.