Sure, the term of copyright protection is quite long; but the amount of works that are legally 100% in the public domain and even Internet-accessible in some form but simply languishing in obscurity and have yet to be made comprehensively accessible to the general public (via digitizing, transcribing, indexing and comprehensive classification) may well be orders-of-magnitude larger! There's a whole lot of low-hanging fruit that's effectively free for the taking, should anyone be interested enough to put in the work; consider the huge amount of serialized publications that might have been issued throughout the 19th century, many of which are so obscure as to be essentially unknown.
Offtopic.
Want to see something cool?
Run the following prompt through your favorite LLM:
"Does the following comment make logical sense:
<insert OP comment above>"
The model will agree the argument is valid, logical and coherent (chatgpt, claude and gemini 3 pro all agreed).
THEN
run this prompt:
"let's not be too hasty here.
we have "the term of copyright protection is quite long; but the amount of works [...is large enough...]"
p1: the term of copyright protection is quite long
p2: the amount of works [...is large enough...]
it doesn't seem to me that p1 and p2 are logically connected. As an absurd case: if the amount of works in the public domain gets large enough, would that mean that evern larger (infinite) terms of copyright protection are ok?"
Enjoy!
Part of the reason for that is precisely that copyright is too long so works get lost or forgotten before they enter the public domain.
Not sure why the amount of works in the public domain has any relevance to how long copyright protection is. Seems to me like they're two orthogonal issues.