This is the only reasonable path for regulation in my opinion. The networks can't be 'broken up' due to the natural networks effects. But forcing the companies to adopt open protocols will foster innovation and force them to behave better.
I love this idea and believe it the future. However, one of the main benefits of centralization is performance. How can we hope to provide the same UX ( real-time complex queries ) if the data is dispersed geographically? Are there ideas on how to overcome this fundamental limitation?
There is no ‘GraphQL only DB’. GraphQL makes no assumption about the data sources, that’s up to the individual resolver functions. It can be a db, local file, redis, apis, etc. The apollo client doesn’t know what backend you are using, you simply point it to a GraphQL endpoint. That separation of concerns is very important.
Hey Vincent, thanks I’m glad you found it useful. GraphQL has nothing to do with what database tech you use, it’s a tool for crafting APIs ( similar to swagger ), and you can use it with any database. If you do not need the powerful graph features of ArangoDB, I would recommend using a more managed solution like prisma. If you are not very technical…
Are you familiar with https://dgraph.io ? I think the main problem with distributed graph dbs is the query performance problem. ArangoDB has a feature called ‘Smart Graphs’ that tries to solve this, but it needs to know information about your domain to perform the optimization.