The discrepancy is based on Amazon’s own inconsistency, but it seems like it could be easily detected: those books will always have a word in the title like 無料読本 or 試読版. I guess there aren’t that many possibilities. Plus: they are always kindle only, while the rest would usually have a printed version, and they are much shorter.
In terms of handling on Natively, I can see two options:
short term: store as standalone book. The reason to make it standalone is that the difficulty can vary wildly between the sample and the full book (that’s the case for my example above, as I have read both and the crazy vocabulary comes in after the cut). Optionally, those books could be put in a category of their own (“samplers” or something similar) since, for light novels at least, they often merge a sample of the novel and a sample of the manga adaptation.
long term: Add extra links between the book being sampled and the sampler, such as an extra entry on the info page of each linking to each other. There could be an option for the community to add those links (so that they don’t have to be automatically generated)
For the particular sampler you note, is that fully contained in the first main book? Is it separate?
If it’s fully contained in the first book, I think we should remove the separate entry and simply list the main book with a link to the free sampler, noting that it’s only partial portion of the novel.
I think I wasn’t clearly thinking what a sampler was… which is simply a slightly longer free sample of the first book. I was bucketing in with side stories / supplemental content, but this isn’t really supplemental content, just a free sample.
People can just record as having read a certain amount of pages of a book and then marked stopped… which I think is a more accurate portrayal here as they didn’t read the full book.
So I think I was wrong, let’s remove the standalone