> I didn't think it was that bad, free to use and everything has an ID right? My take is that the API forces the N+1 query problem on clients that start with a

> I didn't think it was that bad, free to use and everything has an ID right? My take is that the API forces the N+1 query problem on clients that start with an empty repository.