re: SQL 101: All about SQL JOINs VIEW POST


I'm tossing up as to whether I explain how a B-tree works or just stick to a high-level explanation. That will dictate the strategy I take with gifs, images and which direction I take it.

I also enjoy the library metaphor:

"Databases are like libraries.

Tables are like books stored in a library.

Rows are like pages of a book and to make getting to the page you need quicker, you need to create an index. Flipping through a textbook page by page looking for that one page you need is going to take time, the same way querying millions of rows in an unindexed table is going be time-consuming and tedious."

I think a high-level B-tree explanation would be interesting — but perhaps you could warn that this needs some CS fundamentals (trees)?

Your approach to the library metaphor is interesting — I actually thought of it one level up, i.e. "a table is like a library" because I was thinking about a book as a row in a table. For example, a book has many properties (like genre, publish date, content, title, authors…) that we can relate to columns and relationships. Also, I'd be cautious not to cause confusion because "page" also have a specific meaning in the realm of databases (grouping of rows stored on disk). :-)

Anyway, excited to read this next article!

This is why I’ve held off tackling indexes so far. I’d like to pitch it to beginners but feel like I’m going to get too far into CS topics I wasn’t intending to explain.

Back to the draft :)

code of conduct - report abuse