Here is an interesting case: A company provides a free (MPL-licensed) library (called JointJS), and furthermore a number of proprietary extensions (called rappid).

On their web server they ship a file that combines code from both parts of the library at http://jointjs.com/js/joint.all.min.js (gist for historical reference), as it is quite common for JavaScript libraries. But note that this file contains only a copyright or license statement for the free JointJS part (and a few other free licenses for third-party tools). So the structure of the file is:

/*! JointJS v0.9.3 - JavaScript diagramming library  2015-06-26 

This Source Code Form is subject to the terms of the Mozilla Public
License, v. 2.0. If a copy of the MPL was not distributed with this
file, You can obtain one at http://mozilla.org/MPL/2.0/.
[Similar license header for first third-party library]
[first third party library]
[Similar license header for second third-party library]
[second third party library]
[JointJS code (likely intended to be released under the MPL]
[rappid code (when separately distributed, not released under the MPL]

So assuming I were pointed to that file only (as you are now), there is no way of knowing that parts of that code are not intended to be licensed under the MPL license as stated. And given that fact that the actual copyright owner did that, does this imply that all parts of that file can now be used under the terms of the JointJS part itself?

  • 3
    An interesting approach would be to contact the company and ask them the same question - or invite them to join here, as the jointjs site already links to jointjs discussions on stackoverflow. – Michael Schumacher Oct 2 '15 at 13:15

The license explicitly claims to apply to the source code. A concatenation of files is not source code by the usual definition: it is not the original file that would normally be modified, but a mechanically-constructed derivative that just happens to be somewhat human-readable. Thus the text you cite does not in fact apply to the file at all.

You could make an argument that the inclusion of the license text is misleading, and that you believed that this file was the source code. After all, how were you to know that the file was mechanically generated? (Or even granting that it was at some point, the result might then have been misguided manually.) The onus is generally on the party that writes the contract to make it clear, so that ambiguities benefit the other party. That would only fly if you'd never read this thread, of course.


You're overthinking it.

The license preamble clearly states that JointJS is licensed under MPL.

It does not say that anything else is licensed under MPL therefore the other stuff is "copyright all rights reserved" as per standard copyright law.

The answer will hinge on weather or not a developer could reasonably think the code is part of JointJS.

Most developers would see that comment at the top, then search for JointJS and download the latest version that hasn't got any (potentially buggy) modifications. At this point it would become clear the other stuff at the end of the file is not part of JointJS.

So no, it is not legally significant. Copyright law doesn't care about engineering details of code distribution ad the code is not open source.

It could be confusing certainly, but as has been proven many times if you're confused that is not a valid legal defence.

This is no different to an OS X computer having some BSD licensed components - it does not make the whole of OS X BSD licenced.

  • “It could be confusing certainly, but as has been proven many times if you're confused that is not a valid legal defence.” too bad :-) – Joachim Breitner Oct 6 '15 at 21:25

Quoting Wikipedia:

U.S. law no longer requires the use of a copyright notice, although placing it on a work does confer certain benefits to the copyright holder. Prior law did, however, require a notice, and the use of a notice is still relevant to the copyright status of older works.

The MIT license places no restrictions on the license of derived works. So, if someone starts with a blob of MIT-license javascript, and adds in a heap of proprietary code, they can claim copyright on the derived result, and offer it under restrictive terms. So, you really can't conclude anything from the information you have.

  • 1
    I am aware that the MIT license is not copyleft. But is it unreasonable to assume that all code in the file is distributed under the MIT license, as the proprietary components are not in any way distinguishable from the free parts? – Joachim Breitner Oct 3 '15 at 22:29
  • No. It's not reasonable to make assumptions about copyright law. – bmargulies Oct 5 '15 at 18:19
  • 1
    So how would I distinguish this release file from one where all content is covered by the MIT license? Or, put differently: If someone gives me a file, with content he owns, and it has a MIT license header, under what circumstances can I rely on that header? – Joachim Breitner Oct 5 '15 at 21:30
  • You didn't say that the file had a header that has the usual format for a notice for the entire file, or, at least, I didn't read you to mean that. In that case, you could reasonably act upon it. – bmargulies Oct 5 '15 at 23:58
  • I made the question (hopefully) more precise. I also noticed only now that it is the MPL, not the MIT license, but I hope that this does not affect the analysis. Does this affect your answer? – Joachim Breitner Oct 6 '15 at 8:58

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.