Personaly, I'm fan of Cypher. We just published a blog post about different graph query languages few days ago.

https://memgraph.com/blog/graph-database-query-languages-you...

Did you know you can create polls here on HN instead of linking to Twitter?

https://news.ycombinator.com/newpoll

This is great. I didn't know about this feature. Is there some docs on pools? Can they be included in regular post?

When I first saw GraphQL, I was like "This is some overengineered shit that is going to be the next cargo cult"

Then I got actual exposure to it, and I think it's amazing in certain scenarios. You can easily make CRUD apps that allow a front-end to pick which database rows you want to update, and you create a metadata file that describes the database and applies permissions.

You get so many things for free, and the result is you have far less code to write and fewer bugs, especially around RBAC.

That being said, like many technologies, there will be people that try to use it for things it shouldn't be used for.

GraphQL or not... cargo cult explains a lot of what happens in tech.

Basically, "Google does X so my smaller company should do X too."

That's just VERY often not the case.

What do you think is the reason for this? Is it maybe something along the line "If company X uses it, and they are big, this standard / technology will for sure stay here some time"?

I think it's more like "Company X is big and we have to be like company X if we want to be big too"

A large part of it is "FANG does x y,z and I want to work at FANG. So I must so x,y,z at smaller company."

Compared to REST APIs, I like how it eliminates the coarse-grained vs. fine-grained decision for designing queries in APIs. You can just spec out the model and let the clients decide what they need.