OpenAlex Exports Names too long

brierjonOU
I'm running into multiple sync errors due to too long of names. Some from the American Chemical Society and other publications.

Example The creator name “Pedro C1 - Department of Physics, Department of E…” in one of your items is too long to sync. Shorten the field and sync again. Associated with https://doi.org/10.1021/jp409833b

I exported records affiliated with our institution from OpenAlex and many are not stripping the affiliation correctly and includes it in the the author field. The pattern often being capital letter followed by a number then space and dash.

Example: Last, First C1 - Department of Physics, Department of Electricaland Electronics Engineering, UNAM—National Nanotechnology ResearchCenter and Institute of Materials Science and Nanotechnology, Bilkent University, Ankara 06800, Turkey C1 - LUMINOUS! Center of Excellence for Semiconductor Lightingand Display, School of Electrical and Electronics Engineering, Divisionof Physics and Applied Physics, School of Physical and MathematicalSciences, Nanyang Technological University, Singapore 639798, Singapore

I can share the export records if this detail is insufficient.
  • adamsmith
    which format did you export from OpenAlex?
  • brierjonOU
    I exported as RIS.
  • adamsmith
    FWIW, Zotero imports both from the Save to Zotero button and openAlex JSON (as obtained from the API), those are richer formats than RIS.

    You can open one of the RIS files and copy its contents here, but this looks like something went wrong in how you saved those files (or maybe on the openAlex end).
    The C1 is the affiliation and should be on a new line in the RIS, and if it is Zotero ignores it. I tested with a sample RIS generated by openAlex and that worked as expected.
  • brierjonOU
    The issue was with only less than 10 publications in over 40-80k imported via the exported RIS generated by OpenAlex. It was opened with Zotero to import.

    There was no issue indicated until sync is attempted and reached that content. Which from a UX is kind of frustrating when you're trying to sync thousands and sync keeps stalling (at least that is the perception) until each item is addressed.

    Thanks for the tip on the json, this was just a one off test, but the UX could be improved to flag length issues prior to encountering when syncing.
  • adamsmith
    The original rationale for not just shortening them on import (or a similar solution) as I understand it has been that authors that are too long to sync typically signal a larger issue with data import. I agree that, especially as Zotero is getting used more and more for systematic reviews and other large-scale bibliographic projects, this isn't great and having better ways to handle this would be good (there are also regular threads from people doing large imports from OVID databases which export broken RIS).

    I'm not sure how devs are thinking about this -- this isn't easy to handle well in terms of UX, but it's exceedingly frustrating when it occurs. In your case, when you only have 10 broken entries that you can just fix, perhaps less so than when you end up with several hundred.
  • adamsmith
    But if you only get this for a small number of items, it might be worthwhile looking at the RIS because that suggests that the problem is faulty export from OpenAlex, and they've been very responsive to bug reports in my experience.
Sign In or Register to comment.