If (in the confines of a “MOVIE”) you stop thinking about taxonomies you have a “section” (movie, actor, director) with subitems (films, names, names).
content/
content/movies/
content/movies/die-hard/
content/actors/
content/actors/alan-rickman/
content/actors/bruce-willis/
content/directors/
content/directors/john-mctiernan/
each of these folders (page bundles) has an index.md with info about it’s subject.
Then let’s cross reference:
die-hard/index.md has a frontmatter:
actors:
- bruce-willis
- alan-rickman
director:
- john-mctiernan
---
that does not need to be a taxonomy, because you can access it via .Params.Actors.Bruce-Willis
and such.
actors/bruce-willis/index.md` has the following frontmatter:
---
movies:
- die hard
- die hard 2
- die hard with vengeance
---
Then you group your “items” (movies, actors, directors) in sections and check for "in section actors which item has the directory john-mctiernan
" and return that, which will return all actors that worked with mctiernan.
That works. You can define permalinks per section (movies, directors, actors) and interlink.
The problem is, to create this abstract structure BEFORE you start collecting the data.
I think the main problem in this whole abstract thought palace is, that a section is basically a taxonomy and only by the way you are utilizing it and configuring it you define what it becomes.
I would create sections for the “main” “categories” of your structure (probably movies, actors, directors, writers, aso.) and then add a level of taxonomy to it (like tv, movie, series, documentary as taxonomy to movies).
The sad thing is, that we will always end up with IMDB as role model. But that might be a good thing too.