I'm trying to think through the normalisation, particularly because a number of people play a number of different roles, for instance say an actor might direct an episode, a person who's normally a writer has a cameo role in one show and so on.
If I were just doing an "actors_play_roles" kind of normalisation, based on the assumption that actor John Smith always plays the character Jake Jones, I could just have tables like this:
which says that Actor 88 played role 99.ACTORS: id | Name ---------------- 88 | John Smith ROLES: id | Name ---------------- 99 | Jake Jones ACTORS_ROLES: actor_id | role_id ------------------ 88 | 99
But what if he plays Jake Jones in one episode, and then Mycroft Jones, Jake's Evil Twin, in another?
What if he plays both parts in one ep?
Do I just need one big table of "people in general"? And then separate table of directors_of_episodes and actors_in_episodes and so on?
That seems over-complex, though I guess it's still normalised.
Or do I want to somehow do a three-column table like:
STUFF_PEOPLE_DO episode person function 1 | 88 | 23
where 23 is "actor", so we know that in episode 1, person 88 was an actor? And put which part he played somewhere else?
Any suggestions gratefully received.
($_='kkvvttuubbooppuuiiffssqqffssmmiibbddllffss') =~y~b-v~a-z~s; print
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |