We and our partners at the U.S National Cancer Institute recently had an article describing our Pathway Interaction Database accepted by Nucleic Acids Research. I’m not posting to puff that up: during the submission process, the NAR editors raised a couple of perfectly reasonable questions about preprints and unique identifiers.
We had previously put a preprint of the article in Nature Precedings. We’ve now updated that article to point to the final peer-reviewed version — and, as you’d hope, the NAR editors were happy with that. There’s a general point in that:
we encourage all authors to update their articles in Nature Precedings to point to new versions subsequently published elsewhere.
The NAR editors also raised a question about unique identifiers. Many journal publishers assign CrossRef-managed Digital Object Identifiers (DOIs) to published articles. The DOI for our NAR article, for example, is 10.1093/nar/gkn653. In Nature Precedings, we also assign unique identifiers but have chosen those managed by Handle.net to avoid any complications/confusion caused by two versions of the same article having different or the same DOI. The Handle for our Nature Precedings article, then, is 10101/npre.2008.2243.1.
I think a large fraction of the scientific research community doesn’t know about DOIs and other unique identifiers. They prefer to stick to traditional journal citations and PubMed IDs — either not knowing or not caring that PubMed IDs aren’t, strictly speaking, a stable identifier for the actual research article. DOIs are now a standard of the publishing industry, and authors are increasingly likely to need to understand what they are and how to use them.