14

Many entries in the DOI catalogue contain incorrect information. Is there any central way to report such entries (to the DOI team or the publisher*)?

Example:

When importing a work into a reference management software by its DOI this results in:

title={This Is An Example},
journal={Big Journal},
author={Author, A.},
year={2000}

But correct would be:

title={This Is An Example},
journal={Big Journal},
author={Author, A. & Bauthor, B.},
year={2000}

* I am aware of the possibility to write each publisher an email. But I am looking for a faster solution

7
  • 2
    I don't understand the question. You may want to elaborate exactly what you mean. Commented Feb 12, 2023 at 14:26
  • @user2705196 does the example help?
    – user1
    Commented Feb 12, 2023 at 15:48
  • Importing where? How? Commented Feb 12, 2023 at 22:02
  • 1
    I'm still confused. What is a doi catalogue? Commented Feb 12, 2023 at 23:03
  • 1
    You seem to be fundamentally confused about what a DOI is and what services commercial databases for the scientific literature provide. But it looks like one of the answers figured out where you're coming from and is trying to clarify! Commented Feb 13, 2023 at 13:59

2 Answers 2

17

I can understand your frustration at trying to create an accurate bibliography, only to find that the data you believed would be correct actually turns out not to be! To answer your question, however, requires an understanding of how, and where, your problem might have arisen.

The first thing to understand that there is no DOI catalogue. As Wikipedia explains, the Digital Object Identifier (DOI) is "a persistent identifier or handle used to uniquely identify various objects". It's primary purpose is to provide a persistent link to a digital "object" such as a journal article, despite the fact that the actual location of the journal article on a publisher's website might change. One consequence of this is that it becomes the responsibility of a publisher to inform the maintainers of the DOI when a link to which a DOI resolves (for example, when one browses to https://doi.org/nn.ddddd/thisisthedoilabelgivenbythepublisher) is changed.

Moreover, when a publisher first publishes an article and notifies the DOI maintainers, the primary thing that they are doing is telling the DOI maintainers the Uniform Resource Locator (URL) to which the DOI should resolve.

The question that underpins your problem is not really about a non-existent "DOI catalogue" (which at best might be conceived of as being the database that links a DOI to a URL) but instead, a question about how, and from where, your reference management software (which I'll call RMS for conveniece) is acquiring the metadata about an article, and how errors might be introduced in the process.

There are several possibilities, so much more likely than others. Here I simply list some of the possibilities with different possible variations:

  • Scenario 1: When you enter a DOI into your RMS it uses https://doi.org to resolve the DOI to a publisher's URL but the resolution is incorrect and the reference management software ends up on a page for the wrong article. In my experience, this is just about the least likely of the various possible errors. Nonetheless, if it occurs, it is almost certainly the publisher who has made an error by incorrectly notifying the DOI maintainers of the correct URL and so it is the publisher whom you would have to contact to correct your problem.
  • Scenario 2. When you enter a DOI into your RMS it uses https://doi.org to correctly resolve the DOI to a publisher's URL where the journal article appears. Several things can now go wrong.
    • 2a: The publisher has published incorrect metadata on the page (e.g., omitting an author's name or incorrectly stating the page numbers of the article). In this case, you would have to contact the publisher.
    • 2b: The publisher has published the correct metadata and you try to import the bibliographic information directly from the correct web-page but your RMS does not scrape the information from the page correctly. In this case, you would need to contact the maintainers of your RMS.
    • 2c: You choose not to have your RMS scrape information directly from the web-page but instead choose to export the bibliographic information from the page using one or other available export formats such as *.ris . The export process might either go correctly, and export the correct data, or fail and export incorrect data. The only way of finding out which has occurred would be to examine the exported metadata file using a text editor. If the exported data is incorrect, contact the publisher. If the exported data is correct, then the error will have occurred at the final stage of importing the metadata file into your RMS. The problem should be reported to your RMS producer.

There are, of course, yet more possibilities, but these are the most likely ones.

4

Assuming the metadata retrieval method you're using uses CrossRef behind the scenes you can contact them. This is the case, for example, when adding references by DOI in JabRef and Zotero. (If the method doesn't use CrossRef's services I suppose you can still verify that the problem is present in their metadata before contacting them.) According to the FAQ:

I’ve found a problem within your metadata - how do I get you to fix it?

While we aren’t able to correct the metadata provided by our members, report any metadata issues to our support staff and we’ll contact the responsible member and ask them to make corrections.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .