I work on a popular Google Docs alternative product in the market.
I gave some thoughts around this idea of experimenting a standard for collaborative rich text, that any client can implement. The problem though is that people want easy collaboration, more than the freedom to bring their own clients.
To simply put how can we deal with assigning people (for @mentions, comments, document ownership and content locking for a group of people) in such file formats? SaaS universe has a concept of a userbase with unique ids. So when a person is assigned/mentioned the product knows who is relevant and what to do. This implies we need a universal userbase standard (which is already hugely complicated) and is adopted by the SaaS that your target users belong to.
This is one huge roadblock that I don't see any practical solution for.
Decentralized Identifiers are aimed to solve this, basically having interoperable identifiers you can use across different applications but still referring to the same user.
Initial concept for DIDs were made for blockchains and other decentralized projects (I think, someone please correct me if I'm wrong) but useful for federation or for centralized projects that want to offer flexible data migration too.
Rich text is expressed with markup. The rest can be done with protocol.
Just like clients would have to implement markup parsing and rich text rendering, they would need to implement protocol endpoints like notifications,etc.
I gave some thoughts around this idea of experimenting a standard for collaborative rich text, that any client can implement. The problem though is that people want easy collaboration, more than the freedom to bring their own clients.
To simply put how can we deal with assigning people (for @mentions, comments, document ownership and content locking for a group of people) in such file formats? SaaS universe has a concept of a userbase with unique ids. So when a person is assigned/mentioned the product knows who is relevant and what to do. This implies we need a universal userbase standard (which is already hugely complicated) and is adopted by the SaaS that your target users belong to.
This is one huge roadblock that I don't see any practical solution for.