10

When MongoDB submitted the Server Side Public License to the OSI for approval, the OSI has publicly announced that they do not consider it an open-source license.

Has the FSF (or, if relevant, a prominent member) ever commented on the SSPL in a similar fashion? As of writing, Wikipedia states that it is not "FSF approved", but despite my best efforts at searching, I can't find any comment from the FSF on the SSPL, not even in their list of licenses.

3 Answers 3

11

I dig some digging, and found RMS confirming in Oct 2021 that "we", presumably meaning the FSF, had still not got around to considering the SSPL, saying that it wasn't a priority for them, and that he didn't expect they'd find it to be a free licence as and when they did get around to it.

That was about a year and a half ago, and a full three years after the release of the SSPL. Given how little discussion I've seen about the SSPL in the last 18 months, compared to the furore that surrounded its publication, I can't imagine that the FSF would have seen any good reason to bump it up the stack since then.

That said, I freely accept that absence of evidence is not evidence of absence, and will be fascinated to see if anyone can dig up fresher data.

2
  • 1
    Indeed, if the FSF has never officially commented on the SSPL, I don't think it would be possible to definitely prove it. I'll accept this as the best source I can reasonably expect. Thanks for the answer!
    – Invizio
    Apr 1, 2023 at 15:43
  • @Invizio welcome to OS.SE, and many thanks for the prompt acceptance. Hope you stay around and continue to write interesting questions - and answers - for some time to come!
    – MadHatter
    Apr 1, 2023 at 19:45
2

As your question acknowledges, the SSPL does not appear on the FSF's license list. Given how comprehensive this list otherwise appears to be, it is my interpretation that the SSPL's absence indicates that the FSF has not formally evaluated it. If they had, they would have put it on the list. The list contains many licenses which the FSF describes as non-free or otherwise problematic; licenses are not excluded from the list simply because the FSF disapproves of them. It is rather implausible to me that they would exclude the SSPL from this list, yet simultaneously take an official position on the license in a different forum.

Having said that, it is not clear to me how often they update this list. A date in the footer indicates March 4, 2023, but it looks automatically generated, and it might simply indicate that e.g. someone refactored the HTML or CSS without changing the contents of the list. It is possible that the FSF has gotten out of the business of evaluating licenses altogether (or at least, evaluating licenses that everyone already knows to be non-free, anyway).

-1

I ran into this license recently and wondered the same thing. It's still true as of 2023-12-05 that the FSF license commentaries have no entry for this. Here's my own (non-legally-informed) analysis:

Section 13 holds the most notable change. Here's the addition that sent me running for the hills:

all programs that you use to make the Program or modified version available as a service, including, without limitation, management software, user interfaces, application program interfaces, automation software, monitoring software, backup software, storage software and hosting software, all such that a user could run an instance of the service using the Service Source Code you make available.

This is a very strong position, vastly stronger than AGPLv3 gets. I wouldn't feel comfortable running this sort of software (as least in a manner that counts as public serving) on anything but a dedicated bare-metal hardware instance. From my (again, not at all legally-trained) perspective, this prohibits running on e.g. AWS because that involves automation that is not FOSS.

The MongoDB SSPL FAQ calls Section 13 out explicitly:

What specifically is the difference between the GPL and the SSPL?

The only substantive modification is section 13, which makes clear the condition to offering MongoDB as a service. A company that offers a publicly available MongoDB as a service must release the software it uses to offer such service under the terms of the SSPL, including the management software, user interfaces, application program interfaces, automation software, monitoring software, backup software, storage software and hosting software, all such that a user could run an instance of the service using the source code made available.

This is the portion that the OSI objected to in their rejection of SSPF as an open source license and why Stallman stated the FSF has no incentive to comment (FSF-approved Free Software licenses are, afaik, a strict subset of OSI-approved Open Source licenses).

5
  • 1
    Note this GPL FAQ entry: gnu.org/licenses/gpl-faq.html#ModifyGPL Dec 6, 2023 at 8:00
  • 1
    Thanks for the clarification, @BartvanIngenSchenau! I was wondering why the FSF hadn't acted on the copyright violation. I've removed that part of my answer (context: here's the original answer, see also the diff).
    – Adam Katz
    Dec 6, 2023 at 16:32
  • 1
    With respect, the question doesn't ask if the SSPL is free or not, it asks if the FSF has ever made a ruling on it. Saying why you think it's non-free, for all that I agree with your reasoning, doesn't answer the question.
    – MadHatter
    Dec 6, 2023 at 17:05
  • @MadHatter – I was just pointing out that (1) it's not Free Software, (2), the OSI rejects it as such, and (3) the FSF won't likely review it because they consider the OSI statement to be sufficient. This was meant to supplement your answer, not replace it. I even gave your answer a +1.
    – Adam Katz
    Dec 6, 2023 at 17:35
  • 2
    That's very kind of you, and you are of course free to write anything you want in your answer, but it seems to me that (1) answers something not asked in the question, (2) is stipulated in the question, and (3) is speculation, unless you can reference it. Your last sentence is genuinely interesting, and new information, but it's unreferenced, so I don't know if it's true. If you could dig out a Stallman (or official FSF) quote saying that it's the OSI's rejection that means the FSF won't bother to review it, I'd be really interested to see it, and would definitely upvote that.
    – MadHatter
    Dec 6, 2023 at 18:07

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.