Most of the schemas I have worked with used “id”. I agree it’s the default. But aliasing was inconsistent and made it hard to figure out which table id is bei

Most of the schemas I have worked with used “id”. I agree it’s the default. But aliasing was inconsistent and made it hard to figure out which table id is being referred to later in the code. I know it is a matter of discipline but any tool that encourages consistency helps. A database schema is definitely one of those tools. One unrelated idea is to include the entity in the id itself. I have never done this but I’d imagine it would help with things like logging / observability. It would not play nice with indices though.