Closed thebluepotato closed 2 days ago
Revert Kluwer changes in this branch, but otherwise... @adam3smith, do we want arXiv DOIs? I really have no idea.
Oops sorry, my branches got mixed up... I'll probably end up deleting and re-doing #3161
I don't understand why we wouldn't want the DOI for arXiv? That's why they started issuing them, just like all.other preprint servers.
OK, sorry @thebluepotato, appreciate this fix but I think we're going to do the opposite.
(I see the argument for not saving redundant IDs, but choosing which IDs are relevant is ultimately the citation processor / style's job, and we should err on the side of saving as much as possible.)
After the recent changes to the arXiv translator, items imported via identifier or from arXiv search pages no longer include the arXiv issued DOI, as intended. However, when imported from the webpage, the DOI would still be set.
I believe the consensus is that the DOI should reflect the published version. We have easy access to the arXiv page anyway.