Are data that can be obtained from the text of a report considered to be in the public domain?

Question

I have been working with some data from a technical report that is freely available online as a PDF document from the website of the institution that published the report. I've cut and pasted the data from tables in the report for my research (and written a program to process it as the data are stored in an odd order in the PDF tables). I'd quite like to use this data as a part of a machine learning benchmark (with proper attribution, of course). Would the data be considered as being in the public domain and available for use for academic purposes, and made available in a more machine readable format?

I should add, I will of course be asking permission to distribute the data, I was just interested in the expectations surrounding this; is it a courtesy (as it is in the public domain) or is it a substantial favour that is being asked (or somewhere in the middle)?

Answer 1

Your question asks about "public domain" but these are really the wrong words to use, as a) it's complicated but unlikely that these data are public domain, and b) it might not matter much for what you are asking to do.

If you just want to use the data, I don't know of any law that would make this illegal, even if the data could be considered a copyrightable database (that is, data are not copyrightable, but collections of data can be). Academic convention is of course to cite, and if the primary data in your paper are from some other source you'll end up going quite a bit beyond just a simple citation and likely have to describe specifically where and why you got the data from that source.

If you want to also distribute the data, rather than just your analysis of the data, that's where you could possibly get into trouble if they were indeed deemed a copyrightable collection (which, of course, they might not), were not associated with a license that allowed you to distribute in the manner you would like to distribute, and it would be safest to just get permission. However, you probably do not need to post the data at all if they are freely available, just point others to where they can be found.

This is a good reminder that if you do produce and distribute data you would like others to be able to use to remember that you should include a license with those data to avoid any ambiguity as to whether people can use the data; if there is no license given, it must be assumed that the authors reserve all rights allowed to them. If your preference is that others use your work, let them know by using a license! Note also that even for data that are released as public domain or with a similarly permissive license, academic standards of citation still apply; licenses give you legal rights only, not ethical ones.

Answer 2

Taking your question literally: In the United States, data is not subject to copyright. Some sources state that only creative works can be in the "public domain." If you use that interpretation, then data is not in the public domain.

Example references: https://sco.library.emory.edu/research-data-management/publishing/copyright-data.html https://guides.uflib.ufl.edu/copyright/datamanagement

Answer 3

Without a license or a clear statement that it is in the public domain you must assume that the creators hold all rights. In most jurisdictions these days copyright is automatic. It need not be asserted, though it is helpful to do so.

You should ask the creators for permission (a license) for any use beyond fair use. Using "all" of it is likely not fair use.

But no, you can't assume that the things you find online are public domain.

You may be able to "use" the data for another analysis, but probably not able, ethically, to republish it in any form without permission.

---

Note that I've assumed that some level of "creativity" was involved in producing the data.