(The content of this chapter is kind of basics of what I've been doing for a long time now, so my notes will be skimpy) ORMs: maybe good, maybe bad They talk about the tradeoffs of using a JSON document rather than tables, and I think it's funny that after talking about how nice it is to have one-to-many represented in a document, say that if you have a lot of references you might be better off with a table When you use an ID, your data is more normalized, lol the water under that statement is deep I'm actually surprised that they talk about the document vs relational database tradeoff more in terms of normalization & joins than they do about schemas vs schemaless, to me that's at least as important a concern. Also, they previously talked about fanout, and are kind of hinting at the connection between denormalized data and fanout, but aren't making it explicit (nvm, they do that immediately after this occurred to me) Can feel the author's pain in writing X (formerly Twitter). How many…
No comments yet. Log in to reply on the Fediverse. Comments will appear here.