Regarding I, l and L, what you're seeing is the product of a careless person, or a stumbling OCR (Optical Character Recognition) system. Or both.

I remember reading a blog about the issue in typography with examples from signage around the world.

Anyway, with human error it can come from a typo, bad cut and paste job, stupidity, cognitive impairment, etc. From an OCR perspective it's likely someone may have scanned material such as an album cover and then tried to automate the recovery of track details from that scan. Those details get submitted to places like FreeDB and then get repopulated onto tracks you might rip from your own collection.

This issue isn't specific to Spanish.