I have to agree that just renaming Publications to Edition Groups would be no more helpful in drawing a distinction between Editions and Publications. In fact, I recently filed a JIRA issue for documentation to be added to the user guide to clarify what the data model really is.
So more to the point, @LordSputnik, what defines “same piece of literature” for the purpose of an Edition Group/Publication, especially with respect to what might actually be a distinct Work? For example:
- Is my English translation of Don Quixote a distinct Work from the original Spanish text?
- Would an illustrated edition of Dickens’s A Christmas Carol be a distinct Edition Group, or is that also just another Work?
- If it’s an Edition Group, what if that illustrated edition was translated into Russian as well?
- A critical edition of Shakespeare’s Hamlet seems like it would probably a separate Edition Group of the Work Hamlet, but what happens when they make a revised text of that critical edition? Would I then just add each new manifestation (hardcover, paperback, eBook) of that revised text as a new Edition next to the original ones?
- What if a text is published by the same publisher without any modifications but with a new Introduction? Is that a new Edition Group, or just a new Edition?
- What do I do about the various printings of works like the Fellowship of the Ring like the 2012 reissue by Mariner Books, the 1988 hardcover by Houghton Mifflin Harcourt or the 1965 mass market paperback by Ace? Presumably the text is exactly the same, so are they all just different Editions? Or are they grouped by publisher into different Edition Groups, because they have substantially different cover art and publishers (perhaps even publication countries)?
In offering this list of questions, I don’t mean to be glib. These aren’t exactly easy questions to answer, and various librarians have, as far as I can tell, spent years working on their own representation models attempting to answer these questions, of which perhaps FRBR is the best known. But even these models have ambiguity that comes down to “house style”. But, depending on what the goals are, it may suggest that deeper model fixes and new relationship types are needed.