4

Suppose a review of algorithms used to solve a particular problem is to be written. Is it considered plagiarism or malpractice if the algorithms to be reviewed are copied verbatim?

Do the algorithms have to be 'paraphrased' before reviewing?

  • If you paraphrase the algorithm surely it will change the function or output. Why can't you just cite the source of the algorithm? – astronat Jul 14 '17 at 12:03
  • The algorithms are quite long and I'm not sure whether it is common practice to copy them verbatim. – Zwolf Jul 14 '17 at 12:09
  • 1
    Well, the best way of finding out what the common practice is would be to read some existing review papers in your field. – astronat Jul 14 '17 at 12:11
  • The problem is that most of the review articles in my field assess standard, 'textbook' algorithms and general frameworks. – Zwolf Jul 14 '17 at 12:35
1

There are at least three distinct, though inter-related, issues here:

  • Plagiarism: As long as you properly cite your sources and you make it clear that you are reproducing the algorithms verbatim, there is no issue or concern about plagiarism. Plagiarism only applies if you are pretending that what you are presenting is your own original work.

  • Length in your own article: If the algorithms are lengthy, it might be excessively lengthy to copy them in your review. However, if it is essential that your readers see the full algorithms within your review (which is usually not necessary for merely commenting on them), then you might consider reproducing the full algorithms in an appendix rather than in the main body of the text.

  • Copyright violation: This is a sticky issue, but in general, if you copy too much from a copyrighted article (unless it is distributed as open access), then there might be a concern about copyright violation if you copy too much from any source. What "too much" means is legally murky (and it varies across legal jurisdiction), but as a very general rule of thumb, you should not copy so much that the reader decides that since you have given them the algorithm, there is no need for them to consult the original source. (I'm not a lawyer, though, so that's not legal advice.)

Based on these three issues, I suggest:

  • Be sure to properly cite all sources and make it explicitly clear whenever you copy anything verbatim.
  • Don't copy entire algorithms unless absolutely necessary. It usually isn't necessary to copy the whole thing to make your point; copy only the partial pieces necessary.
  • If you feel you have to copy an entire algorithm, make sure that you don't make your review a substitute for reading the original article (e.g., the algorithm is the only important part of the original article). If you are replacing the original, then don't copy it at all; rather refer readers to the original source, especially if it is not open access.
  • If you have to copy an entire algorithm and you feel you are justified to do so, then place it in an appendix, not in your main article text.
5

Any verbatim copying of a long block of text, including if that block of text is an algorithm, would be plagiarism if it is not attributed. However, you can avoid plagiarism by stating explicitly that your description of the algorithm is copied verbatim from another source, and citing the source. Then you cannot be accused of plagiarism, although it may still be considered poor or lazy writing if a substantial amount of your review is copied from other sources.

  • However, for me to review algorithms, they have to be present in the paper. Is there a way to include them without copying them verbatim? – Zwolf Jul 14 '17 at 12:25
  • Obviously, you can either copy the algorithm verbatim and add a citation, or write the algorithm in your own words, or not write the algorithm. Those are your options, so choose whatever works best for the kind of review you want to write. – Dan Romik Jul 14 '17 at 12:28
  • Being new to research, and with no one to guide me, I apologize for asking this here but are papers with a non-mathematical description of algorithms restricted to general or commonly used algorithms? Is this practice acceptable if I am doing it for very specific and narrow algorithms? (This is based on a quick Google Scholar search). – Zwolf Jul 14 '17 at 13:07
  • I don't understand your question at all, sorry. It might help if you explain the terms "non-mathematical description of algorithma", "general or commonly used algorithms", and "this practice". – Dan Romik Jul 14 '17 at 13:14
  • 1
    @Zwolf Just copy the algorithm verbatim and cite its source. There is nothing wrong with that if it is neccessary for your paper. – sgf Jul 14 '17 at 13:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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