First thing you'll probably want to do is check if the work is public domain (https://guides.library.cornell.edu/copyright/publicdomain) or not.
Second thing is a four-factors analysis like Lena suggested. Speaking as someone who just spent approximately two years as the interim Copyright person, I always took a very generous approach to what is or isn't a fair use.
Other countries, ie., Canada, have similar laws: https://copyright.ubc.ca/public-domain/ <https://copyright.ubc.ca/public-domain/#>.
So a suggested workflow might be something like:
1. what kind of material is it and was it published here? (different works have different public domain periods)
2. does public domain apply to this work (if yes = approve)
3. does it meet the criteria of a fair use (aka the four factor test---you'd need to crudely suss this out with questions in the submission form)?
4. send to a human to review if it fails at step 3?
I don't like the "it uses more than X% of the work" test. It's mostly arbitrary and leads to gaming the system. We used a 25% test even. Only if I couldn't find a way to say "this is a fair use because X, Y, and Z" would I look at the amount.
not legal advice/not a lawyer/please don't sue me Elsevier
On 2021-12-02 at 15:24 (-05:00), Lena Bohman wrote:
> So, this may not be the answer you are looking for, but instead of looking
> at 10%, you could use the 4 factors to determine fair use:
> On Thu, Dec 2, 2021 at 12:27 PM Julien Tremblay McLellan <[log in to unmask]>
> > Hi everyone,
> > Does anyone have an idea how I can figure out if a scan request is
> > compliant when patrons request chapters from a book.
> > We are on Alma but I don't have API access. Looking to hack together a
> > script to discover if chapters X and Y are less then 10% of the book.
> > Any ideas sometimes Amazon and google have the book's table of contents.
> > Are there any other tools ? Books range from 1900s to present.
> > I have page count in the 300 a marc21 tag.
> > All ideas are welcome I'm a bit stuck.