I've been down this road multiple times with Arango DB, and Orient DB and IDK how many others. Doing too much of everything usually means not good at everything. I will stick to doing one job good, and battle tested tools.
Indeed. At this point if I'm building an app, I don't want surprises in the technology side. If I hit a problem, I'm more likely to find a solution had I used Postgres than had I used the database du jour, SurrealDB in this case. I'd much rather go through composing building blocks together (Apollo GraphQL with Postgres) than depending on one tool to do them all well, because if one part sucks, at least I can swap it out.
Keep your technology boring and make your app interesting, as they say [0]. Given only three innovation tokens, I don't want to burn all my tokens on something the end user won't even see.
Keep your technology boring and make your app interesting, as they say [0]. Given only three innovation tokens, I don't want to burn all my tokens on something the end user won't even see.
[0] https://mcfunley.com/choose-boring-technology