From other discussion in Academia SE and other source, I got an impression that redrawing is to avoid Copyright infringement. link1 link2 link3 not from SE link4 link5

I cannot comment there to ask doubt because lack of reputation.

My post here is about redrawing schematic diagram/schematic circuit/3D visualization, not (re-)plotting data.

My question:

  1. Do I still need permission from the original paper/author if I redraw schematic diagram/schematic circuit/3D visualization?
  2. Why rotating 3D graph is considered OK ? link4. I am not sure about "OK" here, is OK=no need for permission?
  3. The author in link1 found a better drawing than his/her own that describes same idea. in the comment section, he/she was planning to use his own drawing if the author could not get permission or did not want to pay licence. From here, I got an impression that redrawing is a work around if we do not want to deal with permission/licence. But, is it legit?
  4. How about citation? Do I need to cite after redrawing?


  1. "fair use" does not apply here.
  2. asking legal permission from publisher is not free.

I am writing a technical report for a research project, a project before master thesis.

  • 1
    In general, there is no "default" legit -- you can cite another paper by copying text/image from there, and if the copyright holder brings you to the court, it's up to a judge to decide whether it was proper citation or infringement. Is asking the orginal author for a permission not an option (just to be on the safe side)? – rg_software Jan 26 at 22:46
  • copy/paste without permission is against the law. I checked it online, the permission is not free. – Codelearner777 Jan 26 at 22:58
  • 3
    As was asked in the first comment, ask the original author for permission. Remember, plagiarism is also taking ideas... so just moving a couple of components won’t cut it... – Solar Mike Jan 26 at 23:26
  • @Codelearner777 Citing someone is also "copy&paste". In most books they say explicitly that "no part of this book can be reproduced", and yet we cite them, i.e. reproduce. My point is that I personally believe that copying a picture might constitute legitimate citation, but there is no "default" rule -- they bring you to the court, and then the judge will decide whether it was appropriate. Not the best perspective, of course. In your case I would simply refer the reader to the image in a particular paper without providing the image (given that it is just a tech report). – rg_software Jan 26 at 23:37

Question 1 Since you have asked us to assume that fair use does not apply (a big assumption!), your question hinges on "copyrightability." In the United States, one requirement for copyright protection is that the material have "at least a modicum of creativity." (That comes from Feist, a Supreme Court case about the uncopyrightability of the white pages of a phone book.) The standard line is that facts aren't copyrightable, only expression. Relatedly, the scènes à faire doctrine eliminates copyright protection for tropes and conventions, and the merger doctrine eliminates copyright for very simple expressions of facts. A selection or arrangement of uncopyrightable components is copyrightable, so long as there is creativity in that selection or arrangement.

Here's an example from the Compendium of U.S. Copyright Office Practices (2017):

The Office may register a work comprised of rocks that are selected, coordinated, arranged, and fixed in such a way as to result in a sculptural work. Likewise, the Office may register a photograph of a rock, a drawing of a hand-tool, or a written expression of an idea. However, the Office cannot register a mere “compilation of ideas,” a mere “selection and arrangement of hand-tools,” or a mere “compilation of rocks,” because ideas, handtools, and rocks do not constitute copyrightable subject matter under Section 102(a) of the Copyright Act.

If you use only uncopyrightable components of the original and do not copy a creative arrangement of those components, then you do not need permission (nor do you need to rely on fair use). This is simply something copyright permits, without permission from the rightsholder in the original work.

I've based my answer on U.S. law, which is my expertise, but this concept of copyrightability has equivalents in every copyright regime. The details, including the threshold of copyrightability, do vary significantly. For a slightly more international perspective, you might want to check out Copyright for Librarians, Module 3: The Scope of Copyright Law.

So, the answer to question 1 is, "it depends." Sometimes yes, sometimes no. (A lawyer's answer, I know. Sorry!)

Question 2 As for question 2, rotating a 3D graph, I disagree with the assessment that that's "OK," but perhaps I'm misunderstanding. Certainly reproducing a 2D graph with a 90 degree transformation would implicate the reproduction right. Perhaps what is meant with the 3D graph is such a rotation as to require creation of new material (e.g., showing only the side that wasn't visible).

Question 3 Certainly it is "legit" if it doesn't involve copying copyrightable material.

Question 4 Whether to cite the original depends on what your version looks like, as well as the norms within your discipline. In the U.S., citation is not a matter of copyright law. In countries with stronger moral rights regimes (virtually all other countries), attribution can be a matter of copyright law, but if your new figure only uses uncopyrightable components, that still shouldn't be an issue.

Expertise: I am a U.S. lawyer who works in an academic library teaching faculty, staff, and students about copyright law.


(Not an expert on this topic, just wrestled with the same questions)

I think what you ask is largely covered in the accepted answer to this question you linked. In particular,

The new plot [schematic in your case] needs to do one of two things:

  • present materially different content relative to the old image
  • present the same material in a different context.


  • If this is a schematic of a fundamental concept (e.g., Gauss's Law) and you redraw it in a way that is substantively different (e.g., not identical with trivial recoloring, etc.) then you are fine.

  • If the schematic is of proprietary technology (e.g., an Intel E6590 chip) and you redraw it in essentially the same way with the same context, then it is copyright infringement.

  • In practice, you are likely somewhere between these extremes, but the specifics matter a lot -- it is hard to define what is "reasonable" and what is not; it comes down to what a judge decides.

As for citing, that really has nothing to do with copyright infringement. In the first case, you wouldn't need to cite since you are presenting fundamental information in a new way. In the second case, citing would not be enough (saying thank you as you steal someone's IP does not help). After you are clear on the copyright, you can decide if you have (legally) used enough content that citation is justified.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.