I am working on a webmail client. I want to store the remote mailboxes information in a Postgres database. As a general rule of thumb, most IMAP servers will have a similar default folder structure like so:
- Sent Messages
- Junk Mail
As I add users to my system, this information will be repeated across most users. So here's my question:
Considering these are small-ish strings, am I better off to just store all the mailbox information in one row (ie: mailbox name, attributes, delimiter, parentId, hasChildren) or should I store the mailbox names in a separate table and join on a foreign key?
Right now I'm thinking the former as the cost of storing the extra information is probably less than the cost of throwing more computing power at the database to maintain the same performance. But I am a novice when it comes to database design and I want my application to scale well.