You are quoting yourself.... not the license.

To avoid a couple of extra roundtrips...
I really doubt that interpretation of the license. And I have never heard of anyone else actually interpreting it that way, and the license actually goes out of its way to avoid using the concept of "user".
It talks about user _products_ and the like and you're most definitely not "transferring the right of possession and use" of any product by manipulating a touchscreen in a museum.

I don't know if the source of this FUD is either Jolla or if this is your interpretation. If it's Jolla, it's really shameful.

Please stop this nonsense!
I linked to the GPLv3 to read, specifically terms #2, #3 and #6, plus this paragraph in the preamble:

Some devices are designed to deny users access to install or run modified versions of the software inside them, although the manufacturer can do so. This is fundamentally incompatible with the aim of protecting users' freedom to change the software. The systematic pattern of such abuse occurs in the area of products for individuals to use, which is precisely where it is most unacceptable. Therefore, we have designed this version of the GPL to prohibit the practice for those products. If such problems arise substantially in other domains, we stand ready to extend this provision to those domains in future versions of the GPL, as needed to protect the freedom of users.

This is no "interpretation", it is what the license states and what the FSF deliberately intends (and clearly states so)!
And it is the reason why Google, Jolla and many others avoid *GPLv3 code like hell.

This is nitpicking, but yes, while as museum definitely "transfers the right to use" its explanatory, fixed touchscreens to visitors, it does not "transfer the right of possession" for these.
But as soon as the visitors are provided with tablets for that purpose (had that in a Dutch museum) the "right of possession" for these is transferred to their users (note, that "owning" a product is not necessary!), as with centrally administered computers, MDMed tablets and phones etc. in companies, government offices and many other use cases (i.e., most non-private ones).
The latter are the use cases Jolla, Google have to respect, if they want big licensees for their software stacks.

Sorry, who is spreading "alternative facts" here?
Please, try to disprove my statements by quoting the GPLv3 or shut up.

P.S. / edit: For me the *GLPv3 family of licenses are "unfree" licenses, which should never have been OSI-approved, because they factually restrict the absolute "Freedom to use" by ultimately stating, "one may use *GPLv3 software to create lock-down mechanisms, but then must provide users with means to circumvent them", which renders lock-down mechanisms useless in the first place.

Do note: I am not questioning that you need to provide the "encryption keys" if you transfer ownership.

I am questioning that you have to provide the "encryption keys" if you don't transfer ownership. i.e. the situation where a museum lets you use a touchscreen fixed device, or it lends you a tablet.

That is as absurd as claiming that if let you use a device of mine with GPL3+ software, temporarily or permanently, I would have to provide the source code of it. It's just absurd. No legal system is going to parse the license that way. There is even a FAQ item about it:


What I am asking is precisely for evidence of any legal system of any country where the license was actually parsed in the way you described. An interpretation that
1. Contradicts the intention of the license, as evidenced by the FAQ
2. Contradicts _everyone else's_ reading of the license, to the best of my knowledge (I have never seen the license ever interpreted like that anywhere other than in this thread).
3. Is just a plain absurdity and would need for everyone to rethink the way they use software completely.

i.e. what is this "Dutch museum" story?

There are companies shipping products with GPLv3. Just not Jolla. Google ships products with GPLv3 software just fine.

Please quote from the license!
I cannot find any of the terms "ownership", "owner", to "own" or "encryption keys" in the GPLv3.
So what are you talking about?

As you quoted, the point is "transferring the right of possession and use" in legalese.
To "own" (as a legal term) is not a requirement.

Which?
If you really know any (and are not just trolling), please provide a pointer to Google software for the aforementioned target audience ("big licensees") with GPLv3 components.
I cannot see any in AOSP, ChromeOS, Chromium etc.!

Please stop FUDding and substanciate your statements with sources.

Side note:
Many of the FSF's explanations around the GPLv3 state some things (their details vary), which are definitely not in the GPLv3.
Some believe this is done deliberately.

You are making the insane interpretation. Not even Jolla!
I am just asking for a source of your bizarre reading and I am yet to see it! Even the "Dutch museum" story would do.

ChromeOS ships damn Bash! The original motivation for the entire Jolla GPL3 rant was about this package itself!!!!

And you still claim that this is not just plain FUD???

Okey, waste of time...

Merged with next posting

That is not true!
You have to enable "developer mode", then a bash is being installed, which you cannot if your device is locked, ...
... exactly my point: GPLv3-free for the "big licensees" target audience.
Thanks for pointing to this good example!
Plus it is nicely depicting the conclusion / "solution" Jolla has recently come to, too.

No, this thread is about Qt "stuck" at v5.6 in SFOS, as its title clearly says.
But exploring the GPLv3 issues Jolla has and sees definitely plays a role here.
Though bash is irrelevant in this context.

True, my dear troll.
I was still hoping for anything substancial from your side.

To get back to the original topic of this thread, "Qt "stuck" at v5.6 in SFOS" and in this very context "Jolla's relation to the *GPLv3 license family", they have achieved a little step:

As denoted in the IRC community meeting log of 2021-02-25, Jolla ceased to avoid GPLv3 software in general.
Instead they started distributing GPLv3 software as optional RPMs for SailfishOS, i.e. what formerly was (until SailfishOS 2.2.1) in the mer-tools repository. Now (since SailfishOS 4) only the RPMs for the basic SailfishOS (i.e., which may be installed by default) are still kept free of GPLv3 software.
For example, this change in strategy allowed Jolla to update the historic, last GPLv2 version of bash (v3.2.57) to a recent bash v5, but only as an optional RPM. Side note: At the same time a recent busybox version of ash is installed by default (which carries a "permissive" / "weak-copyleft" license), along with a symlink which calls ash in its (limited) bash compatibility mode.

Well, to put this into context, that strategy change addresses half of point 3 of the identified items Jolla could be asked for.
And while this is nice and viable for small, interchangeable tools (as a UNIX shell or UNIX command line utilities), applying this scheme seems to be technically hard to infeasible to big, local infrastructure components, as ... Qt with all of its components.

IMO the major takeaways from this license strategy change are:

• Jolla has shown for the the first time (in 7 years), that they are not only "thinking about addressing license issues" (as they stated so many times), but are also able to further develop their license strategy (just a little, this time) and to execute that in practice.
  Even though this is a small step taken now, I think "execution over blah" cannot be overevaluated.
• For the first time Jolla is willing and able to clearly communicate an aspect of their license strategy (i.e., aforementioned change) publicly, though only when asked and with a tiny audience (IRC community meeting).
• Jolla sounded much more confident in recent IRC community meetings to update Qt beyond 5.6 "soon" (well, we know that Jolla slang means "some time").
  Sadly they provided no indication how (WRT licenses); I guess we will know from the header files, when it happens, but not before that.

This discussion on GPLv3 between Javispedro and Olf is very interesting to a noob like me. (even though it escalated a bit at the end )
I would not say that one won over the other but I tend (with my zero knowledge and so without weighing anything in the discussion) to be convinced more by Olf ;-)
And I believe even tiny Jolla paid quite some bulk to some lawyers over the years to try and find the best way to deal with it. I have difficulties to believe that they avoid GPLv3 for nothing.

Regarding Qt 5.6, I have one question: If Jolla pays the commercial license to the Qt fundation, that means I guess they can avoid GPLv3 for the Qt pack. Do you think they will go that path?
Because I do not think to have Qt5.X above 5.6 can be optional...as done for bash

1. Not only that is not true, but considering that literally chromeos-base already depends on bash, I find that hard to believe. So, source?
2. Even if that were true, it would mean that Google _is shipping GPL3 software_! So much for "Google avoiding GPL3 software".
3. That Google was forced to offer "developer mode" as a consequence of shipping GPL3 (this I don't dispute).. how is this a bad thing?

https://together.jolla.com/question/...embrace-gplv3/


Aaaaaand we start the namecalling.
Summary what I see from my side:

1. A claim (that "GPLv3 licensed software [is] unsuitable for devices which are not user-controlled, e.g. a "presentation point" (in a museum etc.)" which I find hard to believe and I asked for sources, only to be shown that it didn't came from Jolla, but from you (my mistake).

2. When I inquiry into the claim, I get that "this is nitpicking; I meant tablets lended by museums", and a vague reference to a story about a "Dutch museum"

3. This is directly contradicting the GPLv3 FAQ by the own FSF, so I quote the relevant FAQ item, while asking for further details on the "Dutch museum" story. I specifically mention that I am interested on the claim about the "museum devices".

4. Again, I don't get an answer.
Rather you start subjecting me to an higher standard, e.g. now if I use a word like "encryption keys" apparently I waste a roundtrip in the discussion since you will simply stop the conversation to ask me to use clearly-defined words. Do you realize how this sounds to others? What if every time you used a phrase like "cryptographic mechanisms" (which you have) I would stop the conversation and ask you to please limit yourself to license terms, wasting another post?

5. Despite claiming that you are not trying to FUD, you try to FUD your way out at literally _every_ _single_ _opportunity_.
Every message of yours has a totally gratuitous remark about how you think the FSF is more dangerous than people think or why the FSF is lying about the meaning of the licenses. If this is not FUD, I don't know what it is. Examples:

"For me the *GLPv3 family of licenses are "unfree" licenses, which should never have been OSI-approved," -- Guess what. They are approved.

"Many of the FSF's explanations around the GPLv3 state some things (their details vary), which are definitely not in the GPLv3.
Some believe this is done deliberately. " -- Here claiming the FSF literally lies _deliberately_ about the GPLv3. A claim that apparently _I_ am supposed to prove is not true (so I have to prove a negative! how nice!) . Also note the prevalence of weasel words like "Some people say" , which is a textbook example of FUD.

"And it is the reason why Google, Jolla and many others avoid *GPLv3 code like hell." -- Despite the fact that Google is literally shipping GPLv3 software in products _right now_,
and I cannot think of any company avoiding the GPLv3 other than Jolla and maybe Apple.

"And specifically for the future of the (L)GPLv3 Qt releases: The KDE community is committed to handle that somehow (trying to convince the Qt company to alter their plans for the GPLv3 releases or to "soft-fork" Qt), " -- the KDE community is NOT threatening to use the FreeQT clause because of the GPLv3, which they actually _requested_ back in the Trolltech days (and literally applauded when it was announced). The reason is because Qt is delaying open source dumps for up to 12 months in some cases, which is not related to the GPLv3 at all.

6. I try to ignore these off-topic baits (as best as I can, but I admit I am not very good at not biting them ), but fail. I claim that Google is shipping GPLv3 software.

7. Again I don't get an answer about the "Dutch museum story".
Rather, I get namecalled, and then you claim that Google only ships GPLv3 software in "developer mode". Not only this is nonsense for reasons mentioned above, it also shows a failure to understand the point of the GPLv3. Plus the fact that Google ships GPLv3 software is already enough to dispel your FUD.

8. Then, in an additional post, you even claim that "Jolla is now going to ship GPLv3", as if it was some type of miracle.

First, this contradicts any argument regarding Jolla "not being able to" ship GPLv3 software.

Second, what is so hard about it? I have right now a Remarkable Tablet in front of me. This was built by some KDE guys. It uses GPL+3 software like Bash or Qt 5.15. (So much for the FUD about KDE disliking GPLv3!) The only thing they have to do is to give me the root password, which they do!

Look, even Microsoft is shipping GPLv3 bash these days. There is just no point to the discussion. It is particularly offensive that you even try to bring in KDE (which happily distributes GPLv3 software too) in your poor attempts at FUD. Cut it .

I am only interested in the story about the strange legal interpretation (the "Dutch museum thing").