Efficient management of preprints.
When I organizing the papers, I found many preprints coexists with conference or journal version. Both the preprint and the formal versions are of the same work, and shown in the Duplicate Items. It would be better if we can merge the preprints and the formal version in a unified item rather than showing separately in the list.
You will likely have entered some or all of those as individual items at different times. At best currently you can then use Zotero's Related tab to link them manually, if you recognize that they are basically the same work (although you might also choose to delete the earlier, less complete versions). The problem with that is that you later have to click on the Related tab of such an item to see if there are any related items there (there is no 'count' to alert you in the Related tab title). Also, the Related tab may of course be used for other things, like other papers cited by the item (that you already have in your library).
I agree with @anders221 that there ought to be a better way of dealing with this. Maybe a "composite" or "parent" item type, under which all of the above could be filed. Perhaps a different kind of 'merge', that keeps all of the individual items (also kind of like a sub-collection). It really needs to be something in the main library pane, not just in the right pane (like Related).
A similar problem exists for multi-chapter books, which would often all best appear under their book parent, not as separate items. Also, the ways of dealing with Supplemental material published with a paper are not always ideal (eg its existence is not registered in Zotero's paper metadata, it doesn't usually get recognized for download by the Connector, and it is not always obvious when stored under an item). There are probably other examples that need this type of grouped storage too.
A composite or parent item type would perhaps deal with all of these. But that would have to allow citation of any one of the included items.
"I'm not sure, there are credible situations where someone may want to reference both publications."
But you may find publications that cite an pre-version and you want to have access to that version in your library.
Or for your argument the change between pre-released version and current version is important to point out.
And as @tim820 pointed out, the problem is of general nature for different scenarios. Think about "living" journals that change articles over time. For e.g. a historian like me, these changes are important to trace.
- versioned arXiv pre-prints
- versioned books (edition 1/2/3)
- versioned datasets
- ...