VideoFS (VFS)

• 

  Purpose of video fs

Inspiration

Recently i noticed saving videos offline on youtube in good quality (720P and 1080P) now needs premium. While i understand companies need to make money and youtube is not wrong. But It got me thinking how companies change their policies and agremments overtime without considering their user base and then the content creators are also locked into the platform and there is no easy choice for both. There is also no easy way where creators could choose which type of monetisation stratagey ( pay per miniute etc ) they want. Uploading same video on many platform's is also cumbersome. There is also all forms of imperfect ai and powerful entities censoring a certian type of content or content owner, maybe sometimes even threatening them with lawsuits etc.

Looking at all this one might say orgs nowdays have soo much power over what people can do on their platforms but it doesn't have to be this way. If creators could leave platforms ,users would follow suit and people may have more say in the end.

To address all these issues i made VideoFS

Its a decentralised storage for pointers that point to videos uploaded on IPFS which allows user to grant access to their content to any platform they want and till they want. Making it seamless to integrate with platforms they choose join.

What it does

It allows user to upload content once and share it on many places( different platforms like youtube ) with a click or two. It allows user's profile (could be anonymous but linked to some web3 identity) to be attached to their channel and content thanks to IPFS Cids It allows users to revoke access of their content on platforms if they are not happy with platform in any way possible Allows users to get paid in web3 native ways from get go Allow users to have a choice over which platforms monetisation stratgey they prefer, their policies, t&ncs etc etc. Allows users to issue tokens natively tied to their account and channel to their loyal viewers and in best paragmatic ways. No need to wait for platforms to implment smart contract support etc

How we built it

We first found a way to attach profile of users to their channel and videos (all decentralised and verifiable) and provide them freedom to be anonymous if they want but at same time assoicated to some form of web3 identity like hedera accountID.

Then we built a video uploading and processing functionality (transcoding, thumbails etc) that uploads the videos to IPFS and also links user's profile and channel identity with that video

Then we integrated Hedera topics to make sure the pointers to these videos and identities is never ever censored and stays decoupled from api gateway for persistence purposes and real ownership.

We made a platform layer (incomplete) which will allow users to provide access to view and manage their content from 3rd party platforms like say youtube ( should the user want other awesome features from these awesome platforms in future like advanced AI processed video etc) . The final change will always need user's permission before modification unless they allow for more permissions for a vendor in api gateway

Challenges we ran into

Most of the days were spent researching about other projects and what meaningful use cases could be built over them and that lead to lack of sight till the aderline kicked in and at that point we only had 7 days left.. This time constraint lead me to sprint for 7 days since im alone in my team. And this was one of the biggest challange for me which does show its mark in some places

Next we had issues getting a async architecture built for video uploads and processing (that to with resume support) but i wasn't going to be satisfied with mediocre implmentation and it took quite some time

We had hard time coming up with ideas and interfaces to implment this whole paradim of IPFS based identities and signatures and what not and connecting them to hedera account keys and topics

Accomplishments that we're proud of

Even with this huge time constraint im very proud that i was able to pull through this to at least 90% of my expectation.

New Ways to attach and verify ownership of content based on signatures and IPFS objects.

I choose rigirous implmentation for certain things which i was doing first time like resumable server and video processing etc and got really frustated, having to go back to pen and pair repeteadly but in the end it turned out to be really well done and async in nature.

I like how well the whole application works now and is soo much maintainable apart from some shady interfaces here and there. Also the whole project is in typescript and highly rigrious NestJS for server which has made a huge project a breeze for me for most part.

What im not proud of is not being able to make the platform layer more demonstrable and the show real power of it.

What we learned

About IPFS and Web3.Storage

About hedera smart contracts

API gateways like this (VideoFS) might be a confluence between web2 and web3 in one of the most scalable ways possible

To not sprint in a marathon but save energy and make progress consitently

What's next for VideoFS (VFS)

More finished and complete platform/vendor layer.

Add encryption and key managment to prevent user's content from being watched without permission by unauthorized entities over internet

More ways to get users paid without much hassle from platforms

More types of Smart Contracts for users to choose from for tokens etc

Fix Rough edges and make it industry capable.

Looking for more people interested in such type of ideas

Look for possible entities interested in funding such idea further

Built With

• ipfs
• nestjs
• typescript
• vue