On sites like Stack Overflow and Wikipedia (both use CC BY-SA 3.0), users can collaboratively create/edit a work. If I distribute such a work, I have to attribute all contributing users.

But what if I only want to distribute a part of the work?

Do I have to attribute

  • only the user(s) that created/edited the part in question, or
  • all users that edited the whole work, even if they had nothing to do with the specific part?


  1. Bob creates an answer to a question on Stack Overflow:

    You can solve it by adding a . to the foo.

  2. Alice edits Bob‘s answer by adding a sentence:

    You can solve it by adding a . to the foo. Adding dots is fun.

  3. Carl edits Bob’s answer by changing the first, but not the second, sentence:

    You can only solve it by adding a . to the foo. Adding dots is fun.

I want to use/distribute the second sentence ("Adding dots is fun.").

Whom do I have to attribute?

  • Alice (because it’s only about her contribution?)
  • Bob and Alice (because Alice’s sentence only exists in the context of Bob‘s answer?)
  • Bob and Alice and Carl (because every contributor is relevant, not only the first author and the direct contributor?)
  • 1
    In your specific case, I think the issue might be muddled, because such small changes might not be copyrightable. Are we to ignore this aspect, and assume that the changes are substantial enough to be copyrightable? (E.g. whole paragraphs instead of sentences.)
    – svick
    Jun 28, 2015 at 16:57
  • 1
    @svick: Yes, it’s just so short for the sake of the example; could as well be a paragraph or even more text, or an image, etc. (However, even if it would be such a short sentence, assuming that it’s too trivial to be copyrighted can be risky, no? So I’d better be safe and use it in accordance to the license, even it’s not necessary.)
    – unor
    Jun 28, 2015 at 17:45

2 Answers 2


When multiple people collaborate on a product, they produce a combined work, where each new revision is a derivative work of the previous revision.

In your example, Bob owns the copyright to the first phrase. Alice owns the copyright to her change to the work "Adding dots is fun", and the combined line is a combined work, the copyright owned by both (and thus should be attributed to both)

When you only want to use part of the combined work, it is generally understood to be part of the shared copyright. It is really difficult to make sure that the part you are using is only the free-standing addition, and not a part of the combined work.

To explain what I mean by this, take, for example L.H.O.O.Q. This is a combined work of the Mona Lisa, the title, and the drawn upon mustache and beard.

It would be very difficult to argue that the drawing of the mustache in isolation can be seen separate from the combined work. If I for example took the combined work, and removed all parts that are not the mustache, beard and title by coloring all other parts white entirely. It would be difficult to argue that this is not a derivative of the combined work in its entirely rather than a re-use of just the parts Duchamp created.

(the Mona Lisa, of course, is in the public domain, but that's besides the point here)

I am not familiar with any case law where part of a combined work is reduced back to the part, and is more in the scope of law stackexchange than it is for this one.

In practice, attribution is cheap, and legal morasses are swampy. Just attribute the entire combined work, and don't stray into the morass.

  • @FreeRadical Most appropriation art defended in court was defended with a fair use defense. This does make it a derivative work, but one that doesn't break copyright law. Duchamps ready mades make non-copyrightable materials into artforms by placing them in a different context, but LHOOQ is not a ready made, and is very much a copyrightable work, just one for which the copyright has long expired. I don't agree with your interpretation.
    – Martijn
    Jul 31, 2015 at 9:06
  • 1
    I find this conversation fascinating, but urge the participants to be careful to distinguish between an artistic judgement that a work is derivative, and a copyright judgement. Both are legitimate judgements, using the same language, but the outcomes may not be the same.
    – MadHatter
    Jul 31, 2015 at 9:08
  • @FreeRadical let's (later, I don't have much time ATM) continue this discussion in chat. I'd have to look up one or two things as well.
    – Martijn
    Jul 31, 2015 at 9:24
  • quick note: docs.law.gwu.edu/facweb/claw/Lhooq0.htm
    – Martijn
    Jul 31, 2015 at 10:26
  • Good reference. I concede. Jul 31, 2015 at 12:17

In the very specific case described in the question, Alice owns the copyright on the phrase: "Adding dots is fun." The most correct attribution, if you re-use only this part, is to just credit Alice. Assuming that this is a wiki, and anybody can add content without consulting other authors, attributing Bob and Carl would be morally wrong. You cannot be sure how they feel about this. They may think adding dots are boring, and will embarrassed if they're publicly associated with adding dots being fun.

I should also add that in the case of conventional multi-authored works (e.g. books and academic papers with more than one author), all co-authors has agreed to have the final work published carried their byline, so that would be different from the wiki-situation.

Also it should be noted that for many multi-authored works, no revision history is available to outsiders, so when you use parts of the work, you have no of way of knowing who did what- In that case, your only option is to attribute all co-authors.

However, while not said explicitly, this question seem to be about use cases where the revision history is available to outsiders. This is typically the case with free software projects where the SCM-system is publicly accessible (e.g. projects hosted at GitHub), and free culture projects such as Wikis, where the editing history publicly available.

First, let us look at SE. While not really a Wiki, it allows user edits, and it makes the editing history publicly available.

There is some guidance about attributing questions ans answers pulled from StackExchange in this blogpost by SE co-founder Jeff Atwood: Attribution required, where the most relevant line is this:

Show the author names for every question and answer

This, presumably, refer to the publicly visible author names. Every non-wiki post on SE has a single, clearly identified, author. There should be no need to dig into the revision history on SE, as user editing should only be trivial copy-editing (spelling corrections, language improvements, etc.) that would be below the threshold of creativity. Trivial copy-editing need not be attributed. Unlike a wiki, creative changes that actually alters the meaning or message of a posting, is generally not allowed on SE.

If we move on to real wikis, multiple edits by different contributors to a body of text tend to be much more entangled than in the Bob, Alice, Carl string of edits in the OP's question. In practice, it may not even be possible to sort out who did what. Such entangelements are sometimes known as massively collaborative projects or massively co-authored projects and depending on the number of co-authors, it becomes decreasingly practical to attribute all individual co-authors by name.

This problem is addressed in an article by (then) vice president of the Creative Commons, Mike Linkswayer, and includes a summary of a survey conducted by the English and German Wikipedias regarding how contributors feel about what constitutes appropriate credit for using Wikipedia content. These are, in order of preference:

  1. Link to the article must be given.
  2. Collective credit (e.g. Wikipedia community).
  3. Link to the version history must be given.
  4. For online use: link. For other uses: full list of authors.
  5. Full list of authors must always be copied.
  6. No credit is needed.

Mike does not conclude, but notes that the method ranked top in this survey is "Link to the article", and that the the CC attribution license has since ver. 2.0 recognized the use of a link as a legitimate way to provide attribution.

In CC BY-SA 4.0 it is no longer specified how to attribute, instead this is delegated to the Licensor:

identification of the creator(s) of the Licensed Material and any others designated to receive attribution, in any reasonable manner requested by the Licensor (including by pseudonym if designated); (my emphasis)

The current Wikipedia guidelines for reusing contents made available under CC BY-SA says:

To re-distribute a text page in any form, provide credit to the authors either by including a) a hyperlink (where possible) or URL to the page or pages you are re-using, b) a hyperlink (where possible) or URL to an alternative, stable online copy which is freely accessible, which conforms with the license, and which provides credit to the authors in a manner equivalent to the credit given on this website, or c) a list of all authors. (Any list of authors may be filtered to exclude very small or irrelevant contributions.)

So three alternate methods of attribution is suggested, and two of them involves using a hyperlink, not a long list of authors.

The Wikipedia Example [attribution] notice (reproduced below) also contains no list of co-authors, just a link to the article.

This article uses material from the Wikipedia article
<a href="http://en.wikipedia.org/wiki/Metasyntactic_variable">"Metasyntactic_variable"</a>,
which is released under the
<a href="http://creativecommons.org/licenses/by-sa/3.0/">Creative Commons
Attribution-Share-Alike License 3.0</a>.

In the cases where the license is not a CC-license, or when the Licensor does not provide any guidance about how to attribute, you may need to review the specific license the material is made available under, and your use case for reusing the materials, in order to decide how to best fulfil the attribution requirements.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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