0

I noticed that Microsoft claims to apply an MIT license to their .NET core code, which appears to contain a variance:

Copyright (c) .NET Foundation and Contributors

Whereas the opensource.org MIT template reads:

Copyright <YEAR> <COPYRIGHT HOLDER>

As far as I know, after 1989, it really doesn't matter what the copyright format is (it's actually optional). I have some questions regarding this variance:

  1. What are the legal consequences of each of the two forms of copyright notice?
  2. If the year is used, does it have to be updated in each of the years the code is modified?
  3. Isn't the '(c)' redundant?
  4. Should Microsoft be claiming to use the MIT license, given the variance?

I am working on an open-source license management tool for use in DevOps work-flows. Understanding the full legal nuances of the above questions will help me design a better tool. For instance, should I separate the copyright line(s) from the actual license content for all of the OSS licenses, or is the copyright format specified by each of them, part and parcel to the license? Should I create a special category of license for the Microsoft variant? Etc.

  • 1
    It might be worth reading this question, as I think it addresses at least some of the questions you ask. – MadHatter Sep 29 at 5:45
  • @MadHatter, that was definitely helpful. In-line with what I thought I knew about copyright notices. Microsoft's legal department would probably be among the top 10 or 20 law-firms in the world, and they have some pretty good legal minds, so I wonder what it is they had in mind when they approved that notice for their .NET Core releases? – jwdonahue Sep 29 at 14:58
  • 2
    Their highly-trained legal minds know that the copyright notice is completely unnecessary, and has been in the US since the US ratified the Berne Convention. Knowing this, they will know that the exact format of it is really, really unimportant. – MadHatter Sep 29 at 15:02
  • @MadHatter, that seems a reasonable assumption to me. In addition to that, I think adding the years to the copyright notice is redundant, given the publication dates can be derived from publicly available announcements and the git repo history. I am working on an open-source tool to help manage copyright and license files and source header blocks, so I am motivated to discover if there's any more secret sauce to it, than that. As techies, we tend to assume certain levels of logic apply to pretty much everything, but when it comes to law, that seems to be a week assumption, given their origins. – jwdonahue Sep 29 at 15:13
  • Is there anything left unanswered in your question? If not, would you prefer to (a) delete it, (b) have me close it as a duplicate of the question I linked, or (c) have me write up a small answer pointing to the linked question and adding my comment about the Berne Convention? I really don't mind which, but we should get the question put to bed one way or another. – MadHatter Sep 29 at 15:34
3

The sole function of the copyright statement in modern, well-managed free software licences is as a chunk of text that must be reproduced verbatim.

This from the MIT licence:

The above copyright notice [...] shall be included in all copies or substantial portions of the Software.

This from Apache 2.0:

You must retain, in the Source form of any Derivative Works that You distribute, all copyright [...] notices from the Source form of the Work

This from 3BSD:

Redistributions of source code must retain the above copyright notice [...] Redistributions in binary form must reproduce the above copyright notice

This from the ISC licence:

Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.

I could go on, at length, but the idea should be clear. None mandates any particular form of copyright notice; they say only that if someone else attached one to a work that you are conveying, you must reproduce it.

As regards any other function of the notice,

the 1908 Berlin text of the Berne Convention forbade treaty signatories from conditioning copyright on formalities

such as the existence, position, or form of the copyright notice. The US continues as an annoying holdout, granting what it sees as Berne-compliant copyrights regardless of formalities but reserving a higher tier of protection to those who comply with s401 of the Copyright Act 1976; discussion about this, however, is not pertinent to the question at hand.

I note also that "the copyright holder is never beholden to the rules of the holder's own license grant", and that the MIT licence isn't a copyleft one. This means that even if the MIT licence did require a particular form of notice (which it doesn't) it wouldn't require that the author of the work use that form.

So in answer to your question, it makes no difference whatsoever to Microsoft's compliance with the MIT licence that they have chosen to include a copyright notice in such a form.

| improve this answer | |
  • I know all those things. Your claim that "None mandates any particular form of copyright notice", is not accurate. Most of them provide the template for the copyright notice. You don't answer one of my key questions. Why that particular variance? You seem to be too hung up on the supposed irreIevance of the copyright notice. It apparently doesn't matter to you, but I would like to know why they chose that particular form. – jwdonahue Oct 7 at 17:21
  • @jwdonahue I'm sorry, let me be clear. You and I have already discussed this, at some length. Pursuant to those discussions, you wrote your own answer, which to my mind contains a number of major inaccuracies. I in turn decided not to let that answer stand unopposed, so I wrote my own. It is not expected to please or satisfy you; it's so that the wider community can express their comparative opinions by voting, which is how the site is designed to work. Future readers who come across this question may then see more than one point of view on the subject, and can judge them accordingly. – MadHatter Oct 8 at 7:25
  • Fair enough. What exactly is inaccurate in my answer? – jwdonahue Oct 9 at 0:59
  • If I'd thought I could fix your answer line by line, I'd've commented on it instead of writing my own. I do not intend to engage further with you about it, but since you ask, let's start with "the 80's when [...] the presence and form of the copyright notice were strictly proscribed", which is about as wrong as you can get - it's never been proscribed in Berne Convention countries, merely not required, and it was required in the US until it started to drag itself into compliance with the rest of the world. – MadHatter Oct 9 at 5:36
-2

First, let me just say that I am not a lawyer, and all I can say on this topic is what I've found online, where non-lawyers tend to give out substantial quantities of bad advice, and there are few legal scholars volunteering free guidance. If anyone can post a more definitive answer to my questions, I will happily withdraw this and chose a better answer, once I've got a reasonable selection to chose from.

What follows is my interpretation of the information I have found. Hopefully, minus the noise.

Let's start with the "is Microsoft compliant with MIT license?" part of this question, since that is the title of this thread.

According to MIT License - Wikipedia, there is no one MIT license. They point out that MIT has released code under more than one permissive license, however, all of the variants listed, include the year of creation/publication, but we must keep in mind that these licenses date back to the 80's when US and other laws were substantially different, and the presence and form of the copyright notice were strictly proscribed.

The Open source Initiative only lists one MIT license which does include the year.

Microsoft's LICENSE.TXT file, for the Core repository, starts with the line "The MIT License (MIT)". It's not at all clear what variant or version they are referring to, but the parenthesized 'MIT' happens to be the SPDX short identifier for the MIT license, and since it is redundant, I assume they are referring to that SPDX, which takes us to the OSI. The fact that they do not comport with the MIT license published by the OSI, suggests what they really have is a variant of some MIT license.

Wearing my pedantic cap, I'd say the OSI and most of the various license authors have generally failed to adequately separate copyright concerns from license concerns, but I know from previous lurking on their various email lists, that many of the legal experts, claim that the copyright notice itself is integral to any OSS license, as you must first establish that; you are granting permissions based on your legal ownership of the work. Without it, there's no grantor for the grantee to be bound to contractually. It is the MIT license after all. On what basis would MIT be granting anyone a license to someone else's work?

So a statement of who is the grantor of the license, seems to me to be required for any contract to be valid. In this case, the grantor can only be the current copyright owner, hence the copyright notice. Keep in mind that the distributor of the source code in question, may not be connected/related to said owner, so including this information in the license, attached to the source, seems perfectly reasonable.

It is my opinion that Microsoft should make it clear, that what they have is a variant of the MIT license, not the MIT license itself. So the answer to my question is a qualified NO. Qualified in the sense that, I am just not a lawyer and there are no legal restrictions, I am aware of, on the use of the MIT license published by the OSI.

But Microsoft includes a copyright, it's just not of the form suggested by the OSI or used by similar licenses. So why the variance?

That brings us to "What are the legal consequences of each of the two forms of copyright notice?".

As pointed out in https://www.copyright.gov/circs/circ03.pdf (emphasis are mine):

Although notice is optional for unpublished works, foreign works, or works published on or after March 1, 1989, using a copyright notice carries the following benefits: • Notice makes potential users aware that copyright is claimed in the work. • In the case of a published work, a notice may prevent a defendant in a copyright infringement action from attempting to limit his or her liability for damages or injunctive relief based on an innocent infringement defense. Copyright Notice 4 • Notice identifies the copyright owner at the time the work was first published for parties seeking permission to use the work. • Notice identifies the year of first publication, which may be used to determine the term of copyright protection in the case of an anonymous work, a pseudonymous work, or a work made for hire. • Notice may prevent the work from becoming an orphan work by identifying the copyright owner and specifying the term of the copyright.

So two of the listed benefits, in the US at least, and probably others, mention establishing the date of publication. I seem to recall that copyright begins at the point of creation, not the point of publication, but establishing the creation date is difficult without some form of public disclosure. It seems that Microsoft is willing to forgo the benefit of a date established in the copyright notice. Perhaps they feel their publication process is adequate to the task of establishing a year of publication? Perhaps they prefer the potentially later publication year, to be the point at which their legal term of copyright begins? I can't say, but those are the only reasons I can think of at this time.

I think the answer to #2 is yes, if you wish to reset the start of the remaining term of copyright. Now this may bring us to another of Microsoft's motivations (only speculating of course). If you modify a work in a given year, and you're tracking years in the copyright notice, then you must update the notice to include the years of modification, otherwise you run the risk of not gaining the benefit of the potential extended copyright term. Now there may be other ways to establish that the work was updated, and attribute the changes to the listed copyright owner, but why complicate that process? Either update the notice in every year the work was modified, or don't post the year in the copyright notice to begin with. For a large organization like Microsoft, ensuring that a copyright notice is updated with every change, would require the build system inspect every changed file for an accurate year and either fixing the notice or stalling the build for manual intervention. The former would require tooling and build time, while the later is not worth contemplating. It's far simpler to enforce a fixed static copyright notice in every source file.

For #3, I think the answer depends on your jurisdiction. In the US, it seems redundant to me, but there may be countries where the symbol takes precedence. I believe that the copyright symbol is common to many, if not all countries, but the word "copyright" might not be so obvious in some languages. I suppose a little redundancy doesn't hurt, but I can only speculate why Microsoft chose to use both.

#4 just restated the title, and I think I've made it clear that they are using a variant. It's still not clear to me why they chose to call it "The MIT License (MIT)". Perhaps someone will enlighten us, for history's sake?

In my search for information, I stumbled across this excellent page written by another non-lawyer who seems to have done some good research, where I was reminded that, in the US, you can only sue someone for infringement if you have registered your copyright with the US copyright office. I think we can assume that Microsoft does in fact register in most, if not all jurisdictions that require it. Perhaps it is the US registration they rely on for establishing the publication date?

Apologies if I have asked more questions than answered here.

| improve this answer | |

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.