2 ms·
I almost exclusively use utf8mb4_bin (for "human" text) and ascii_bin (for ID's or other "computer" text) -- they're basically the two 'extremes', and it's rare
by crazygringo 12y ago
I almost exclusively use utf8mb4_bin (for "human" text) and ascii_bin (for ID's or other "computer" text) -- they're basically the two 'extremes', and it's rare to need anything in between.
And then other collations like case-insensitive or accent-insensitive only when required, e.g. because the field is used for search.