|
|
Subscribe / Log in / New account

Mono Relicensed MIT

At the Mono Project blog, Miguel de Icaza announced that the Mono runtime has been relicensed, moving from a dual-license slate (with LGPLv2 and proprietary optiona) to the MIT license. The Mono compiler and class libraries were already under the MIT license and will remain so. "Moving the Mono runtime to the MIT license removes barriers to the adoption of C# and .NET in a large number of scenarios, embedded applications, including embedding Mono as a scripting engine in game engines or other applications." De Icaza notes that Xamarin (which was recently acquired by Microsoft) had developed several proprietary Mono modules in recent years; these will also now be released under the MIT license.


to post comments

Mono Relicensed MIT

Posted Apr 1, 2016 4:00 UTC (Fri) by rsidd (subscriber, #2582) [Link] (163 responses)

So, first Microsoft acquires Xamarin, and then all this stuff is released under the MIT license. This on top of the Ubuntu-for-Windows news, MS's contributions to the Linux kernel, etc. Satya Nadella, who are you and what have you done with Microsoft?

Mono Relicensed MIT

Posted Apr 1, 2016 9:49 UTC (Fri) by mpr22 (subscriber, #60784) [Link]

I have seen it suggested that some of the groundwork for moving in this direction was actually set up during the reign of the chair-thrower.

Mono Relicensed MIT

Posted Apr 1, 2016 13:49 UTC (Fri) by tshow (subscriber, #6411) [Link] (1 responses)

Microsoft is on the same trajectory as IBM, about 20 years behind.

Likewise, Apple is on the same trajectory as IBM and Microsoft, and is about 40 years behind IBM and 20 years behind Microsoft.

There was a time when IBM was the proverbial 900 pound gorilla. They were rich, they were successful, their whim dictated the success or failure of other companies. They were the Great Satan of Computing, as it were. "Nobody ever got fired for buying IBM" was a phrase that IBM sales people would use during sales, with the subtext of "because if you don't buy IBM, we're going over your head and if *anything* goes wrong we'll be there to tell your boss who's fault it is".

Eventually (around 1995...), Microsoft managed to take the crown from IBM. It took IBM years to reconcile themselves to it; you can see it in things like OS/2; there was clearly a faction at IBM that thought they could fight their way back to dominance, but Microsoft took over the 900lb gorilla role and eventually IBM (mostly) reconciled with that. IBM is now a services company.

You're seeing the same thing now with Microsoft and Apple. Microsoft is having to realize that no, it can't dominate the way it did back in the 1990s and early 2000s. Interestingly, we're also seeing the kind of software quality problems from Apple that we saw from Microsoft when they were at the height of their powers. Things seem to start slipping when a company gets big enough that it's trying to hold down multiple fronts at once.

But I think Microsoft is in the process of becoming a cloud services company, and they're gradually realizing that as a services company they have to play nice with everyone. So what we're watching is the process of them going from a "crush everyone" strategy to a "we can make anything play nice with anything, whaddaya need?" strategy, with the old guard muttering quietly in the background that back in THEIR day, we wouldn't have put up with these shenanigans, we'd have shown these little whippersnappers what's what, what they need is a taste of the bayonet...

In a few more years the young'uns around here will wonder why anyone ever had a problem with Microsoft, much the way people do now with IBM. By that point, I presume we'll be watching this whole process again with Apple as the Great Satan starting to come to terms with an upstart unseating them.

Mono Relicensed MIT

Posted Apr 5, 2016 0:37 UTC (Tue) by jospoortvliet (guest, #33164) [Link]

My crystal ball is out of service but I bet if it worked it'd show you right ;-)

Yeah, I've been around for a bit but am post-evil-IBM and smack in the middle of Evil Microsoft, having participated in debates at local computer/business clubs in NL arguing with a MS sales person over a decade ago... But I'm quite open to a changing MS, in no small part because I don't believe companies have a soul* ;-)

* or anything else for that matter

Mono Relicensed MIT

Posted Apr 1, 2016 16:09 UTC (Fri) by kena (subscriber, #2735) [Link] (1 responses)

Amen. "This is not your father's Microsoft." Or, really, any of ours, since I imagine most of us remember the bad-old let's-break-DR DOS-because-we-can, etc. days.

Mono Relicensed MIT

Posted Apr 1, 2016 16:17 UTC (Fri) by ejr (subscriber, #51652) [Link]

"DOS ain't done 'till Lotus don't run." They hate it when I mention that phrase.

Mono Relicensed MIT

Posted Apr 1, 2016 18:48 UTC (Fri) by orev (guest, #50902) [Link] (2 responses)

I guess you haven't been paying attention to how they're rolling out Windows 10. It truly is "resistance is futile".

Mono Relicensed MIT

Posted Apr 1, 2016 22:00 UTC (Fri) by jhoblitt (subscriber, #77733) [Link]

Vogon Constructor Fleets don't turn on a dime.

Mono Relicensed MIT

Posted Apr 1, 2016 22:05 UTC (Fri) by dfsmith (guest, #20302) [Link]

"He gives the kids free candy, because he knows full well |
That today's young innocent faces, will be tomorrow's clientèle,"

Mono Relicensed MIT

Posted Apr 2, 2016 3:59 UTC (Sat) by ssmith32 (subscriber, #72404) [Link] (151 responses)

Depending on your point of view, moving from LGPL to MIT is not making things more free. Instead, it enables open core, and locked down, proprietary enhancements. Nothing new here.

Mono Relicensed MIT

Posted Apr 2, 2016 4:22 UTC (Sat) by Cyberax (✭ supporter ✭, #52523) [Link] (149 responses)

And also allows developers to create applications for iOS, providing applications for hundreds of millions of users.

Mono Relicensed MIT

Posted Apr 2, 2016 8:26 UTC (Sat) by ssmith32 (subscriber, #72404) [Link] (1 responses)

As vlc demonstrates, it makes it much easier to distribute, but the apps can be made regardless (I think, corrections welcome). I do think of it as a less free license, but I also do respect Miguel's contributions and decisions over the years. And am happy he found a route that he is comfortable with for mono and mobile and all the other things.

Mono Relicensed MIT

Posted Apr 2, 2016 8:30 UTC (Sat) by Cyberax (✭ supporter ✭, #52523) [Link]

Publishing LGPL-containing code in the AppStore is technically not allowed. That's why Xamarin made a lot of income selling proprietary licenses for Mono to allow app developers to use C# in iOS.

Mono Relicensed MIT

Posted Apr 3, 2016 0:33 UTC (Sun) by donbarry (guest, #10485) [Link] (146 responses)

Therefore making the jail for hundreds of millions more comfortable and making it less likely they would fight to leave.

Mono Relicensed MIT

Posted Apr 3, 2016 0:44 UTC (Sun) by Cyberax (✭ supporter ✭, #52523) [Link] (145 responses)

Or perhaps making the world's greatest collaborative project greater, thus bringing improvements to the life of hundreds of millions of people.

Mono Relicensed MIT

Posted Apr 3, 2016 12:05 UTC (Sun) by DOT (subscriber, #58786) [Link] (2 responses)

You think the iOS app ecosystem is the world's greatest collaborative project?

Mono Relicensed MIT

Posted Apr 3, 2016 17:34 UTC (Sun) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

Not yet, but it's on the way there.

You don't like that it's based mostly on proprietary stuff and not on Free(tm)(r)(c) Software? Then perhaps you should re-evaluate your assumptions about the merits of it and the strategy used to promote it.

Mono Relicensed MIT

Posted Apr 5, 2016 0:41 UTC (Tue) by jospoortvliet (guest, #33164) [Link]

Well, it's not collaborative much if each piece can only be worked on by a single person... That they're all in one store doesn't change that. It is no different from the Windows ecosystem which is still larger I'd bet.

The largest actually *collaborative* IT project is the Linux Kernel. Granted, that's a single project. If you want multiple collaborative projects, you're probably looking at the Linux distro's.

The iOS ecosystem is a platform, a large one, sure. Collaborative I wouldn't call it, there are other platforms far more fitting that term.

Mono Relicensed MIT

Posted Apr 3, 2016 23:44 UTC (Sun) by linuxrocks123 (subscriber, #34648) [Link] (134 responses)

> Or perhaps making the world's greatest collaborative project greater, thus bringing improvements to the life of hundreds of millions of people.

"World's greatest collaborative project?" First, it's not a "collaborative project", it's a store where people buy stuff. Typically frivolous, overpriced crap. Second:

https://en.wikipedia.org/wiki/Hoover_Dam
https://en.wikipedia.org/wiki/Roman_roads
https://en.wikipedia.org/wiki/Cathedral
https://en.wikipedia.org/wiki/Great_Wall_of_China
https://en.wikipedia.org/wiki/Channel_Tunnel
https://en.wikipedia.org/wiki/Panama_Canal
https://en.wikipedia.org/wiki/Golden_Gate_Bridge
https://en.wikipedia.org/wiki/Taj_Mahal
https://en.wikipedia.org/wiki/First_Transcontinental_Rail...
https://en.wikipedia.org/wiki/Three_Gorges_Dam
https://en.wikipedia.org/wiki/Empire_State_Building

...the word "perspective" must not exist in your vocabulary.

*mic drop*

Mono Relicensed MIT

Posted Apr 4, 2016 4:31 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (133 responses)

> "World's greatest collaborative project?" First, it's not a "collaborative project", it's a store where people buy stuff. Typically frivolous, overpriced crap.
Well, and Taj Mahal is just a shrine serving no useful functions at all. So?

You might not like my perspective, however it still leaves a problem with Free Software growing less relevant by the day.

And I say it's a good thing.

Mono Relicensed MIT

Posted Apr 4, 2016 8:54 UTC (Mon) by DOT (subscriber, #58786) [Link] (129 responses)

I stumbled over your usage of "great" because I thought you shared our values for software user freedom, even though you don't like said freedom to be forced. But your comment leads me to believe you don't value this freedom at all.

Mono Relicensed MIT

Posted Apr 4, 2016 9:02 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (128 responses)

I absolutely value freedom - I don't want patents on software, DRM or overly restrictive copyright (like copyright on APIs). Yet I don't like the FSF's usurpation of the "Software Freedom" term.

In short, FSF's opinion is that progress in the IT world shouldn't happen unless it's all Free Software. See: emacs vs. LLVM-based autocomplete.

I have a completely opposite opinion. Anything that can ease the progress should be celebrated and promoted. So far liberal licenses seem to be doing this in an exemplary manner.

Mono Relicensed MIT

Posted Apr 4, 2016 9:15 UTC (Mon) by DOT (subscriber, #58786) [Link] (125 responses)

I share your opinion that it can be strategically helpful in certain cases to use BSD-like licensing to encourage freedom-respecting software, but how does that translate to "iOS app ecosystem is great" and "it's a good thing that free software is less relevant"? The iOS ecosystem is a prime example of user subjugation: you can only install apps approved by a secret Apple committee, and there are many examples where this limited developers in their free expression, or users in the usage of their device.

Mono Relicensed MIT

Posted Apr 4, 2016 9:30 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (124 responses)

iOS is not "great" in the sense that it's perfect. It's "great" in its impact on quite literally hundreds of millions of people (total iDevices sold: 1 billion). If your ideology prevents software from reaching that many people - there's probably something wrong with it.

"User subjugation" - are we in Godwin's territory yet? Nobody is subjugating iOS users, absolute majority of them bought iDevices because they solve their problems.

Mono Relicensed MIT

Posted Apr 4, 2016 10:35 UTC (Mon) by DOT (subscriber, #58786) [Link] (19 responses)

I don't believe a free software ideology prevents reaching that many people. Android started out as a very open platform, which overtook iOS' market share a few years ago. It's sad that Google doesn't release more of its software under a free license, and in fact increasingly closed the system, but that was not at all necessary for the initial growth of the system.

Like Google but more so, Apple is subjugating its users. The fact that users voluntarily (mostly unknowingly) set themselves up to be subjugated doesn't change that fact. This is slavery in a microcosm. Compare this with labour laws: they limit what employers can require of employees, because otherwise people would voluntarily enter into actual slavery because it solves their problems (security, sustenance).

Mono Relicensed MIT

Posted Apr 4, 2016 12:00 UTC (Mon) by ghane (guest, #1805) [Link] (18 responses)

> Like Google but more so, Apple is subjugating its users. The fact that users voluntarily (mostly unknowingly) set themselves up to be subjugated doesn't change that fact.

I cannot comment on the "unknowingly" part, as that would involve speaking with users; but if it is voluntary, how do we correct this? Short of legislating, and enforcing, that users must only use our approved "Free Software", what do you suggest?

The figure of 1 billion iDevices sold has been mentioned. Calling them ignorant slaves (my words) is not likely to be productive to the cause, unless the cause is to irritate them..

Note that this cuts both ways, Corbet has somehow convinced me, perhaps unknowingly, to part with $70 every year for nothing more than early access to a web blog! A blog that he gets Cyberax and you to write for, for free! Who will save me from his subjugation?

Mono Relicensed MIT

Posted Apr 4, 2016 12:15 UTC (Mon) by DOT (subscriber, #58786) [Link] (17 responses)

Calling them slaves is unlikely to produce results, I agree. ;) I debated with myself whether I should have avoided the term "slavery" in my last post, for the same reason. Education is a very important step though, but I'm not sure how to reach iOS users. They tend to be emotionally attached to the Apple brand, so that makes it very difficult to avoid entrenching them in a tribal mindset.

I agree that theoretically Corbet is our overlord here, but a seven-day embargo is quite a far cry from 70 years after publication, and our esteemed editor is very conservative when it comes to deleting our posts. I tend to think of paying for LWN as more of a donation with benefits. There have been some calls for the LWN source code to be opened up though.

Mono Relicensed MIT

Posted Apr 4, 2016 15:49 UTC (Mon) by bronson (subscriber, #4806) [Link] (16 responses)

Yes, but that education is needed on both sides. How can the iOS-using hordes reach the free software community about what they need in a phone or desktop environment? Because we don't seem to be anywhere close. Android even provided a road map.

You up for some learning as well as teaching?

Mono Relicensed MIT

Posted Apr 4, 2016 15:55 UTC (Mon) by DOT (subscriber, #58786) [Link] (15 responses)

Ubuntu seems to be going in the right direction. Freedesktop is doing the plumbing (xdg-app) to let the rest follow Ubuntu. I'm not sure the iOS-using hordes are interested in teaching 'us', since they don't even know we exist.

Mono Relicensed MIT

Posted Apr 4, 2016 21:30 UTC (Mon) by bronson (subscriber, #4806) [Link] (14 responses)

Oh, I know scads of iPhone-toting web developers who are very familiar with the GPL. They certainly know that free software exists and what it means, yet they choose to reject it. Why do you suppose that is?

Mono Relicensed MIT

Posted Apr 4, 2016 23:21 UTC (Mon) by DOT (subscriber, #58786) [Link] (13 responses)

I bet they don't reject all the software they download from NPM. They know that free software doesn't cost money to use, but do they really know what it means in terms of safeguarding freedom? Have they internalized that their phone is essentially spying on them? Can they even prove it, since all the software is closed source? What could they even do about it, if they wanted to?

Sadly, the current phone market makes it next to impossible to maintain your freedom. The only somewhat practical free solution is to buy an Android phone, hack it, and put CyanogenMod on it, while resisting the urge to install the Google Apps and instead getting everything from F-droid. And even then you don't get free drivers (and thus no timely updates), since hardware companies routinely violate the GPL.

Mono Relicensed MIT

Posted Apr 4, 2016 23:57 UTC (Mon) by bronson (subscriber, #4806) [Link] (12 responses)

Of course they know. They're actually very smart. They just have less tolerance for putting up with user-unfriendly BS.

"Trade a little privacy for less time learning how this arbitrary nerd stuff works? And avoid bricking my phone? Sure, sign me up! Here's my credit card number and map coordinates."

Mono Relicensed MIT

Posted Apr 5, 2016 1:03 UTC (Tue) by jospoortvliet (guest, #33164) [Link] (3 responses)

... and suddenly NPM breaks their software and they wake up, eyes glassy, realizing they were pretty ignorant all that time.

To doze off again until things break even worse the next time. Boiling frog syndrome I'd say.

Mono Relicensed MIT

Posted Apr 5, 2016 2:44 UTC (Tue) by bronson (subscriber, #4806) [Link] (2 responses)

Huh?? NPM is full of open source, including the GPL. You're saying developers should be scared of it? If so, I strongly disagree.

If you're saying that NPM is poor and unpublish is grossly idiotic, then I agree. But that's a different thread.

Mono Relicensed MIT

Posted Apr 5, 2016 15:08 UTC (Tue) by jospoortvliet (guest, #33164) [Link] (1 responses)

NPM is ran by a company in sole control over the ecosystem, that's the problem... The GPL and AGPL are great licenses but the platform you distribute them on should be under that license too, clearly ;-)

Mono Relicensed MIT

Posted Apr 5, 2016 16:51 UTC (Tue) by bronson (subscriber, #4806) [Link]

Agree 100%. The current situation is venture-backed crazy pills. It's the Node community's biggest liability right now.

Mono Relicensed MIT

Posted Apr 5, 2016 8:01 UTC (Tue) by DOT (subscriber, #58786) [Link] (4 responses)

This is what Apple and Google are counting on. They have created a system where it is much easier to just give up your privacy out of despair and just 'take' it. It is an abusive relationship, and the victims are vehemently defending their abusers.

Mono Relicensed MIT

Posted Apr 5, 2016 10:48 UTC (Tue) by nye (subscriber, #51576) [Link] (3 responses)

>It is an abusive relationship, and the victims are vehemently defending their abusers.

Surely you must at least be willing to entertain the idea that if several billion people disagree with you, they might not *all* be imbeciles suffering from severe Stockholm Syndrome?

Most people do not consider the things that you hate so much to *be* abusive, at all.

I *want* my phone to track where I am, and to track where other people are. This allows it, for example, to give me weather reports and directions, locally relevant searches - and locally relevant *voice recognition*. Voice recognition that *actually works* because it has enough information about the surrounding area to make *intelligent* guesses about what you might have said, like the name of a nearby road or shop with questionable pronunciation. Tracking multiple people, in aggregate, allows it to give predictions about traffic patterns that are *actually useful* in working out when you might get to your destination, not on average, but *today*.

(I won't bother to harp on about everything else, but suffice it to say that location tracking is just a single example. I'm pretty much living in the amazing sci-fi future of my childhood, like at least once a week something makes me stop and think "this is the future - I'm living in the future now".)

Nobody is giving anything up out of 'despair'; we're using useful tools that intrinsically work by processing certain information, so of course we want them to have that information. Most 'enhanced privacy' things that are popular with a certain subset of FOSS advocates look like deliberate antifeatures to me, and apparently to most of the world if their behaviour patterns are any indication.

Mono Relicensed MIT

Posted Apr 5, 2016 11:18 UTC (Tue) by DOT (subscriber, #58786) [Link] (2 responses)

Location tracking is useful, yes, but indefinite storage of that information goes way beyond usefulness into creepiness. Do you think people realize that Google stores every single movement in their entire life? Obviously, we're not telling each other new information here, so maybe we should just agree to disagree and leave it at that.

Mono Relicensed MIT

Posted Apr 5, 2016 12:13 UTC (Tue) by raven667 (subscriber, #5198) [Link] (1 responses)

The real problem is the lack of required oversight to prove that nothing unethical is being done with the data, currently there are few laws about this and everything is locked up behind proprietary walls and trade secrets with unverifiable privacy reports as the only mechanism to validate what is happening.

Mono Relicensed MIT

Posted Apr 5, 2016 15:12 UTC (Tue) by jospoortvliet (guest, #33164) [Link]

Exactly. I'm with Nye in that I want these functions - but I should have some certainty as to how it's used. Maybe it should be on my own private server, or a server I choose at least. Or maybe there should be decent protection offered by the law. Whatever - not the wild west it is right now.

Because it is guaranteed to go wrong at some point. Next up - a company running your pacemaker (which needs a regular check-in to the cloud to function) goes bankrupt and shuts off the server...

Mono Relicensed MIT

Posted Apr 5, 2016 15:09 UTC (Tue) by jospoortvliet (guest, #33164) [Link] (2 responses)

And then this happens and the hipsters realize that they made the wrong choice... Forgetting it the next day of course.

Mono Relicensed MIT

Posted Apr 5, 2016 16:47 UTC (Tue) by nye (subscriber, #51576) [Link] (1 responses)

>And then this happens and the hipsters realize that they made the wrong choice... Forgetting it the next day of course

This seems to be a *much* more pertinent complaint about cloud services. Not based on ideology, because *everyone* glazes over questions of ideology when they don't share it, and convincing people to share your ideology is *hard*, but based on practicality.

I know I live my life assuming that any and all Google services will shut down tomorrow, or sooner, because history has shown that to be the safe bet. This is actually a constant nagging worry because I do currently pay for Gmail in particular, and I don't see any serious competitors, Free or otherwise. Some day I need to put some serious effort into contributing to Mailpile, but the realist in me says I will probably never find the time, because I won't really need it until it's too late :-(.

Mono Relicensed MIT

Posted Apr 14, 2016 3:10 UTC (Thu) by linuxrocks123 (subscriber, #34648) [Link]

Shameless plug:

I recently wrote an email application and it's awesome: https://github.com/linuxrocks123/MailTask

It's probably not for everyone, but, for me, it's awesome, and even if no one else ever uses it ever, I'll be glad I wrote it. I'm trying to get other people to use it, too, though; hence it's FLOSS. It probably needs more docs but feel free to post issues on GitHub if you run into setup or usage trouble.

Mailpile is webmail. Webmail is (imo) brain-damaged nonsense.

Mono Relicensed MIT

Posted Apr 4, 2016 11:09 UTC (Mon) by krake (guest, #55996) [Link] (101 responses)

> If your ideology prevents software from reaching that many people - there's probably something wrong with it.

Exactly! Apple's ideology prevents FOSS software from reaching their customers. We all agree that there is something wrong with that.

Mono Relicensed MIT

Posted Apr 4, 2016 17:12 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (100 responses)

> Exactly! Apple's ideology prevents FOSS software from reaching their customers.
Apple is perfectly OK with other open source licenses. It's GPL that is causing problems.

Mono Relicensed MIT

Posted Apr 4, 2016 17:27 UTC (Mon) by krake (guest, #55996) [Link] (78 responses)

Are you sure?
I was under the impression that the app store policies where the ones imposing restrictions.

Which part of the GPL prohibits distribution in binary form or distribution by download?
Not likely distribution via the iOS app store specifically as the GPL predates iOS, right?

Mono Relicensed MIT

Posted Apr 4, 2016 17:53 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (77 responses)

Ok, re-wording. It's only GPL that would force Apple to commit to conditions that are not acceptable for them just by distributing the software from their market.

Mono Relicensed MIT

Posted Apr 4, 2016 18:13 UTC (Mon) by krake (guest, #55996) [Link] (10 responses)

I don't see how the GPL would force Apple into anything but for the moment lets go with that narrative.

That would mean if someone would create an app store that would require availability of the source it would be all proprietary licenses' fault that they don't?

What if the license requires to distriminate based on ethniticy, gender or faith and the store restricts distribution to a subset of these?
Also the fault of the ideology of the app authors for not wanting to discriminate against their fellow humans, yes?

Mono Relicensed MIT

Posted Apr 4, 2016 18:59 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (9 responses)

> That would mean if someone would create an app store that would require availability of the source it would be all proprietary licenses' fault that they don't?
Not a "fault" per se, but yes.

> What if the license requires to distriminate based on ethniticy, gender or faith and the store restricts distribution to a subset of these?
> Also the fault of the ideology of the app authors for not wanting to discriminate against their fellow humans, yes?
Oh, wow. I haven't seen such a masterful strawman in some time. May I also do it?

Suppose that the FSF in the GPLv4 suddenly decides to require people to kill small children in order to be allowed to distribute the software. Why do you want to kill small children?

Mono Relicensed MIT

Posted Apr 4, 2016 19:27 UTC (Mon) by krake (guest, #55996) [Link] (8 responses)

> Not a "fault" per se, but yes

So the choice of the authors is bad by default?

> Oh, wow. I haven't seen such a masterful strawman in some time. May I also do it?

I don't know what a strawman is but I take the masterful as a complement.

> Suppose that the FSF in the GPLv4 suddenly decides to require people to kill small children in order to be allowed to distribute the software.

Well, the equivalent to the sitation would be that the GPLv4 require that you don't kill small children but the app store would, which in your model would be bad on the part of the license because the restrictions imposed by the authors are bad and the restrictions imposed by the store are good.

I don't think you would be ok with small children being killed, but maybe even in your model there are things a store can not impose and still be good?

Mono Relicensed MIT

Posted Apr 4, 2016 20:32 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (7 responses)

> So the choice of the authors is bad by default?
If it includes 'GPL' in it, then quite probably yes.

Note, that I don't argue that authors are not free to make bad choices.

Mono Relicensed MIT

Posted Apr 5, 2016 7:33 UTC (Tue) by krake (guest, #55996) [Link] (6 responses)

> If it includes 'GPL' in it, then quite probably yes.

That at least gives it more perspective.
A license in favor of the user is bad, licenses or store restrictions against the user are good or at least not as bad.

I guess that is a valid philosphy, but others will still believe that caring for the users is also valid.

Mono Relicensed MIT

Posted Apr 5, 2016 7:37 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (5 responses)

How do you care for non-existent users (GPL does not work on iPhone)?

Why not release your code as Apache2 or BSD and help real-life users instead?

Mono Relicensed MIT

Posted Apr 5, 2016 8:03 UTC (Tue) by krake (guest, #55996) [Link] (4 responses)

I am not buying into the urban legend that an author can not both license their software as GPL and distribute it via the iOS app store.

But if I were, the best option would probably be to license the software as GPL in general, but upload the app with some simple proprietary license that allows the user to treat the software as licensed under GPL on their choice.

Then the app's license is proprietary as far as Apple is concerned and GPL as far as any user who wants that is concerned.

Mono Relicensed MIT

Posted Apr 5, 2016 8:12 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (2 responses)

> I am not buying into the urban legend that an author can not both license their software as GPL and distribute it via the iOS app store.
It's spelled out quite explicitly in the Apple developer agreement.

> But if I were, the best option would probably be to license the software as GPL in general, but upload the app with some simple proprietary license that allows the user to treat the software as licensed under GPL on their choice.
Even better. A user will be able to get a GPL-ed source but won't be able to distribute it or even use it, except by paying Apple and putting their own devices in developer mode.

Mono Relicensed MIT

Posted Apr 5, 2016 8:55 UTC (Tue) by krake (guest, #55996) [Link] (1 responses)

> It's spelled out quite explicitly in the Apple developer agreement.

Wow, really? Can you provide a link? All I could find yesterday were some articles that claimed sime restrictions imposed by the app store would somehow be a problem, but nothing as concrete.

At least that should put all questions at rest which side is at fault, such an explicit rule makes it clear that it is the app store.

> A user will be able to get a GPL-ed source but won't be able to distribute it or even use it, except by paying Apple and putting their own devices in developer mode.

I am afraid I don't understand where you are seeing any problem there.
Anyone with access to a GPL-ed source can distribute it, e.g. putting it on a digital media and given that to someone, or uploading it to a server.

For using source you need to either be a developer or have easy enough to follow instructions, but that is a property of source code, not of the source code's license.

Also not entirely sure what you mean with the last part but my guess is you are referring that deploying a modified version onto an iOS device would require Apple's development tools and a device that allows out-of-store installation.
In that case I also don't see how using e.g. a BSD licensed source would not also require that. Again a property of having access to a programs source, independent of the source's license.

Mono Relicensed MIT

Posted Apr 5, 2016 15:14 UTC (Tue) by jospoortvliet (guest, #33164) [Link]

We solved it this way.

Mono Relicensed MIT

Posted Apr 5, 2016 15:13 UTC (Tue) by jospoortvliet (guest, #33164) [Link]

That last thing is what we do with the ownCloud app for iOS... Works.

Mono Relicensed MIT

Posted Apr 4, 2016 19:50 UTC (Mon) by pizza (subscriber, #46) [Link] (64 responses)

Let's be honest here -- Apple has a raging hatred for all things copyleft that predates the iOS app store's creation. They wilfully chose terms and conditions that exclude all copyleft stuff from their platforms when there was no technical reason to do so.

That choice, of course, is their right. Just as it is my right to not give a proverbial bowel movement about Apple's platforms and instead support those who better suit my principles.

Mono Relicensed MIT

Posted Apr 4, 2016 20:23 UTC (Mon) by Wol (subscriber, #4433) [Link] (29 responses)

> Let's be honest here -- Apple has a raging hatred for all things copyleft that predates the iOS app store's creation. They wilfully chose terms and conditions that exclude all copyleft stuff from their platforms when there was no technical reason to do so.

Actually, I think Apple has a major legal problem on their hands with the app store and the GPL. Namely, that *Apple* can be held liable for J. Random Developer's actions. And rant all you like, it's actually not acceptable under any "civilized" legal code for one person to be held liable for another person's actions.

Let's say I develop an app, and use a load of GPL code. I sell the binary on Apple's app store. As per the terms of the GPL, who is liable for providing the source? APPLE!!! (Or at least, that is a perfectly reasonable argument.) But do they even have the source?! THAT is why distributing GPL apps through a store like that is tricky.

I don't blame Apple, Google, et al for not wanting to get involved in that can of worms. And if you think that's being OTT, just look at the way Cisco et al are being sued for selling routers where they didn't even realise GPL software was being used!

Cheers,
Wol

Mono Relicensed MIT

Posted Apr 4, 2016 20:41 UTC (Mon) by pizza (subscriber, #46) [Link]

> Actually, I think Apple has a major legal problem on their hands with the app store and the GPL. Namely, that *Apple* can be held liable for J. Random Developer's actions.

Given Apple's already not-inconsiderable app review process, and the very voluminous T&C one has to accept in order to publish to the app store (last I checked it was over 50 printed pages) I find it hard to believe that Apple's seriously concerned about their direct liability from copyleft as opposed to any other type of copyright infringment.

> I don't blame Apple, Google, et al for not wanting to get involved in that can of worms.

Oh, I don't necessarily blame Apple or Google for wanting to ignore that can of worms, but since you brought it up, Google at least provides a supported mechanism for alternative app "stores", including those which seem to have no problems complying with copyleft source requirements. Apple, on the other hand, actively works against anyone trying to use non-blessed-by-Apple software on their platforms.

> And if you think that's being OTT, just look at the way Cisco et al are being sued for selling routers where they didn't even realise GPL software was being used!

As for Cisco, if they put their name on a box, it behoves them to know what's in that box. If Cisco *still* doesn't have the internal processes in place to keep track of the licenses of the software they incorportate into their products -- more than a decade after they first ran into that problem -- they only have themselves to blame.

Mono Relicensed MIT

Posted Apr 4, 2016 22:27 UTC (Mon) by ssmith32 (subscriber, #72404) [Link] (1 responses)

Whaa??? People are held responsible for other people's actions all the time. In fact, no civilized society does *not* do this. Bosses are fired because they fail to monitor their employees actions. Companies are held responsible for creating hostile work environments, and the actions they can precipitate. The whole idea of any functioning modern society is that we are at least partially responsible for each others actions.

"I am not my brothers keeper" hasn't passed as a viable excuse since the excuse was first written down.

How *much* we are responsible is often debated. But the idea that we have some sort of responsibilty to each other is what knits society together.

Mono Relicensed MIT

Posted Apr 5, 2016 9:05 UTC (Tue) by Wol (subscriber, #4433) [Link]

> How *much* we are responsible is often debated. But the idea that we have some sort of responsibilty to each other is what knits society together.

Fair enough. But "best efforts" and all that. If you shoot someone and I get held liable that's called "miscarriage of justice".

Equally, if someone else knocks a pot of paint over, I see it and ignore it, then you slip and end up in hospital, I'm on the hook for negligence and that's fair - I knew about it and I should have done something about it.

That's the thing with the app store - if Apple knows people are breaking the rules, and does nothing about it, they are liable for the own inaction. That's why Cisco are in trouble on this point. At least, by trying to keep out of this situation, both Apple and Google at least have a defence of "best effort" if things go wrong. Cisco don't have that...

Cheers,
Wol

Mono Relicensed MIT

Posted Apr 4, 2016 23:09 UTC (Mon) by lsl (subscriber, #86508) [Link] (23 responses)

> Let's say I develop an app, and use a load of GPL code. I sell the binary on Apple's app store. As per the terms of the GPL, who is liable for providing the source? APPLE!!!

How is that any different from just taking a random proprietary app, change the name on it and upload it to the app store? This happens all the time. Those apps just get deleted as soon as they're noticed or someone complains.

Mono Relicensed MIT

Posted Apr 4, 2016 23:33 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (22 responses)

Sure. It's exactly the same situation.

If Apple were to continue distributing a stolen app then they would be legally liable for up to $150000 per _copy_ distributed. Just like with the GPL.

Mono Relicensed MIT

Posted Apr 7, 2016 13:48 UTC (Thu) by aggelos (subscriber, #41752) [Link] (19 responses)

If Apple were to continue distributing a stolen app then they would be legally liable for up to $150000 per _copy_ distributed. Just like with the GPL.

Eh, you say that as if apple didn't explicitly select to be in that position re: GPL'd programs.

Also, you fail to provide convincing reasons to sympathize with apple when their terms of service take care to restrict the ability of users to copy a _free software application_ to their own devices or to use it for commercial purposes.

On top of the unfounded sympathy for apple's restrictive (some would say exploitative - towards both users and developers) terms, your position carefully avoids sympathizing with the people who made sure to allow the users to arbitrarily run, modify and distribute their code, only to have a mediator impose arbitrary restrictions.

Your narrative above pretends that there's some fear of "OMG so many $$" being awarded as damages for copyright infringement, as if there were some intense GPL litigation going on, whereas there's scarcely any. In fact, for the two cases of someone distributing a piece of copyleft software in the app store that I'm aware of, (GNU Go and an early port of VLC) both were simply removed from the app store and, legally speaking, the matter ended there. So the text quoted above reads as fear-mongering as well.

All in all, I'm afraid this narrative is in need of a leg transplant.

Mono Relicensed MIT

Posted Apr 7, 2016 19:47 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link] (18 responses)

> Your narrative above pretends that there's some fear of "OMG so many $$" being awarded as damages for copyright infringement, as if there were some intense GPL litigation going on, whereas there's scarcely any.
Correct. However, companies think about maximum potential damages. On one hand there's a possibility of multi-million damages and on the other hand the loss of GPL applications.

Coincidentally, is there any significant amount of non-trivial GPL-ed applications that might be interesting to regular tablet/phone users?

Mono Relicensed MIT

Posted Apr 7, 2016 23:16 UTC (Thu) by DOT (subscriber, #58786) [Link] (17 responses)

Ah, you're still at it I see. Yes. Take a look at what's available from F-Droid. Especially OsmAnd, IMHO the killer app for Android. Free maps and navigation that works online, *offline* and without sending your every move to some company. It's GPL-3.

Mono Relicensed MIT

Posted Apr 8, 2016 1:32 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (16 responses)

> Especially OsmAnd, IMHO the killer app for Android.
Like Sygic? Or a plethora of other similar applications?

> It's GPL-3.
Of course it is. It would eventually allow authors to charge for a non-free version with more features, while eliminating all the competition.

Proof: https://github.com/osmandapp/Osmand/blob/master/LICENSE
> * Pull-requests and translations:
> - All pull requests are accepted under MIT License (most honorable contributors are mentioned in AUTHORS list)

Mono Relicensed MIT

Posted Apr 8, 2016 1:46 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (15 responses)

http://osmand.net/features?id=contour-lines-plugin - that didn't take long.

GPLv3 preventing abuse of users and non-free plugins?

Yeah, sure.

Mono Relicensed MIT

Posted Apr 8, 2016 7:03 UTC (Fri) by DOT (subscriber, #58786) [Link] (2 responses)

Wow, so now you concede that permissive licensing doesn't protect the user! I guess any argument is fine as long as it can be contorted into an attack on the GPL.

Mono Relicensed MIT

Posted Apr 8, 2016 7:07 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

So you're conceding that your star example of a "GPL killer app" for Android is actually living a secret life as a MIT-licensed base for shareware? And GPL is used in the worst possible way - to allow copyright holders to exploit other contributors without giving them the same benefits?

Mono Relicensed MIT

Posted Apr 8, 2016 7:23 UTC (Fri) by DOT (subscriber, #58786) [Link]

The default PR license is MIT, no doubt to be able to distribute a proprietary version on iOS. But it is not the only license in the project. Note also that the iOS version is severely limited because it misses out on a lot of code. You were saying?

Mono Relicensed MIT

Posted Apr 8, 2016 7:47 UTC (Fri) by DOT (subscriber, #58786) [Link] (11 responses)

By the way, this is released as GPL-3: https://f-droid.org/repository/browse/?fdid=net.osmand.sr...

Mono Relicensed MIT

Posted Apr 8, 2016 8:06 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (10 responses)

Yes, it is indeed released as GPLv3. You can be sure of that, after all it's built by fDroid from source!

And let me quote the relevant part of it (out of the whole 60 lines of code):

>Intent intent = new Intent(Intent.ACTION_VIEW, Uri.parse("market://search?q=pname:" +
> OSMAND_COMPONENT_PLUS));
>try {
> stopService(intent);
> startActivity(intent);
>} catch (ActivityNotFoundException e) {}

The parking and address plugins are similar in content.

Mono Relicensed MIT

Posted Apr 8, 2016 8:25 UTC (Fri) by DOT (subscriber, #58786) [Link] (9 responses)

I haven't investigated the code. What is your argument?

Mono Relicensed MIT

Posted Apr 8, 2016 8:31 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (8 responses)

These plugins in the fDroid repository are just references to paid closed source plugins on Google Market. There is no actual functionality in the "GPL" version of these plugins.

But be sure, download the source code and check for yourself.

To recap, OsmAnd is just another example of corporate shareware with paid versions providing additional functionality, that uses GPL to prevent other companies from providing competing offers.

Mono Relicensed MIT

Posted Apr 8, 2016 8:36 UTC (Fri) by DOT (subscriber, #58786) [Link]

I'll take your word for it. I guess a fork is in order, if someone is willing to take the time. I don't quite see how a proprietary program is better than an open core program, considering the large open core.

Mono Relicensed MIT

Posted Apr 8, 2016 19:29 UTC (Fri) by mathstuf (subscriber, #69389) [Link] (6 responses)

Umm, I have the Contour plugin from F-Droid and it works just fine…

Mono Relicensed MIT

Posted Apr 8, 2016 19:36 UTC (Fri) by mathstuf (subscriber, #69389) [Link] (5 responses)

OK, more misleading content from you (or you're not really grokking that code). That code seems to be looking for OsmAnd itself and opening *it* if you click the plugin from the application drawer (or whatever), the code you reference is to pop up an alert that you need OsmAnd for the plugin to actually work.

Mono Relicensed MIT

Posted Apr 8, 2016 20:05 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (4 responses)

This is the code of the plugin:
https://github.com/osmandapp/Osmand/tree/master/OsmAnd/sr...

It depends on hillshade file, which is not present in the default OsmAnd package. I just downloaded it and checked, again.

Mono Relicensed MIT

Posted Apr 8, 2016 20:13 UTC (Fri) by mathstuf (subscriber, #69389) [Link] (3 responses)

Is there some import that isn't satisfied or are you referring to something like this:

log.info("Indexing hillshade file " + filename);

Hillshade files are available just like any other map: you download the file for the region(s) you're interested in.

Mono Relicensed MIT

Posted Apr 8, 2016 20:16 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (2 responses)

I don't see it in the UI of the basic version. Will try again...

Mono Relicensed MIT

Posted Apr 8, 2016 20:38 UTC (Fri) by mathstuf (subscriber, #69389) [Link] (1 responses)

The only difference between the "basic" and the "plus" edition is that the "plus" one is a donation. It's the same code (AFAICT); F-Droid builds the "plus" edition. Maybe they disable the plugins in the "basic" version, but then it's also convenience-ware in that you could build your own "plus" edition, but if you donate some money to them (and Google…), you can use the plugins. Personally, I go through F-Droid. Nothing in the GPL says you have to provide binaries…

In any case, you may need to enable the plugin for contours in the plugin menu to be able to enable the hillshade or contour layers.

Mono Relicensed MIT

Posted Apr 9, 2016 20:22 UTC (Sat) by DOT (subscriber, #58786) [Link]

Yeah, these plugins also seem to work on my phone without paying for it through the Play store. OsmAnd seems to be GPL entirely, and my assumption about their MIT Pull Request policy seems to be correct: it is to please Apple.

Mono Relicensed MIT

Posted Apr 7, 2016 13:58 UTC (Thu) by aggelos (subscriber, #41752) [Link] (1 responses)

If Apple were to continue distributing a stolen app then they would be legally liable for up to $150000 per _copy_ distributed. Just like with the GPL.

Eh, mispaste, the first paragraphs of the previous reply should have been:

I'm wondering if there's some other online forum where you publish polemics on how proprietary software is EVIL because if apple were to distribute it illegally, they would be liable for $150k per COPY distributed :-)

This makes zero sense on so many levels. (a) this $150k number originates from copyright law, the license has got nothing to do with it (b) your narrative isolates and only talks about the license you want to bash (c) your narrative axiomatically assumes that people should _not want_ Apple to be bound by copyright law (copyright law which they're making heavy use of for their profit, one could add).

Mono Relicensed MIT

Posted Apr 7, 2016 17:56 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link]

> (a) this $150k number originates from copyright law, the license has got nothing to do with it
Correct.

> (b) your narrative isolates and only talks about the license you want to bash
Incorrect.

> (c) your narrative axiomatically assumes that people should _not want_ Apple to be bound by copyright law (copyright law which they're making heavy use of for their profit, one could add).
Incorrect. I have never advocated violating licenses, I'm advocating avoiding certain licenses for new projects.

Mono Relicensed MIT

Posted Apr 7, 2016 12:37 UTC (Thu) by paulj (subscriber, #341) [Link] (1 responses)

Your argument applies equally to RedHat, Oracle, Debian, Android, Microsoft, etc. Further, it applies far more generally than just GPL or even free software.

All they have to do is some level of reasonable due diligence to ensure they're acting in good faith in distributing some software according to the applicable terms. Nothing GPL specific about that! And if some submission says it's using GPL software, then they could have a process to ensure the source is also uploaded and made available.

*Any* distributor of software - proprietary or FOSS - should do some basic checks to ensure licences are honoured. If they do that and also act reasonably when any issues are notified to them, I'd suggest any problems with wilful licence violations will fall on the upstream.

Mono Relicensed MIT

Posted Apr 7, 2016 12:56 UTC (Thu) by paulj (subscriber, #341) [Link]

s/any/the worst/

Mono Relicensed MIT

Posted Apr 4, 2016 20:30 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (33 responses)

> Let's be honest here -- Apple has a raging hatred for all things copyleft that predates the iOS app store's creation.
Apple was A-OK with GPL up to GPLv3.

> They wilfully chose terms and conditions that exclude all copyleft stuff from their platforms when there was no technical reason to do so.
Apple distributes applications from the market and so they're on hook for GPL's patent and source code clauses.

Mono Relicensed MIT

Posted Apr 5, 2016 0:50 UTC (Tue) by jospoortvliet (guest, #33164) [Link] (4 responses)

They have expensive lawyers. I'm sure they can argue they are just providing a platform and the actual producer/seller is the one doing the distribution. Heck, what's next, you demand the source code from the DHL dude delivering your new router? That argument leads to madness. It isn't hard to define who sold you the app is and it isn't Apple.

Mono Relicensed MIT

Posted Apr 5, 2016 1:40 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (3 responses)

> They have expensive lawyers.
And they probably told Apple not to risk it.

> I'm sure they can argue they are just providing a platform and the actual producer/seller is the one doing the distribution. Heck, what's next, you demand the source code from the DHL dude delivering your new router?
DHL in particular is shielded by the common carrier laws. They are not legally responsible for the contents of parcels. You _might_ be able to go after the seller of the router (be it a brick-and-mortar store or an online retailer).

However this is irrelevant in this case. There's an absolute trigger for copyright protection - it's the creation of a new copy. The law is quite specific on that point and Apple does create new copies of applications when you download them.

Please note, that even simple stuff like loading data from disk into RAM might count as a "distribution": http://digital-law-online.info/lpdi1.0/treatise20.html

Mono Relicensed MIT

Posted Apr 5, 2016 2:06 UTC (Tue) by pizza (subscriber, #46) [Link] (2 responses)

> The law is quite specific on that point and Apple does create new copies of applications when you download them.

So I guess it's a good thing that the 50 or so page T&C everyone agrees to in order to publish stuff on the iOS App store includes a clause granting permission to Apple to do just that.

(And, I might add, so does every OSI-compliant software license..)

Mono Relicensed MIT

Posted Apr 5, 2016 2:11 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

The problem is that it will trigger _other_ GPL clauses like no-TiVo-isation clause or patent clause.

There's also an issue with the "system library" clause which might not apply to Apple, since they actually _own_ the underlying system.

Again, you might argue that they don't apply, but this is just that - an argument, untested by courts. And the stakes are very high.

Mono Relicensed MIT

Posted Apr 10, 2016 7:49 UTC (Sun) by jospoortvliet (guest, #33164) [Link]

I don't think the stakes are high at all. Generally, violating the GPL is extremely cheap - worst case is usually that you are forced to stop violating. Point me to a GPL violations case where the perpetrator had to pay millions... You can't just keep claiming Apple does this for a legal reason while there simply isn't one - loads of companies and communities distribute GPL software without any problems, from Red Hat to the many mirrors of Debian (which, incidentally, do what Apple would do: DISTRIBUTE. Yet no cases of them being sued or getting in trouble instead of the project they provide the mirror for).

I've read all your posts here and, again, there isn't a single, rational, convincing argument against the GPL. Perhaps accept that you won't convince others and stop argueing unless you have a good argument.

Mono Relicensed MIT

Posted Apr 6, 2016 14:36 UTC (Wed) by pboddie (guest, #50784) [Link] (11 responses)

Apple was A-OK with GPL up to GPLv3.

Citation needed. Particularly in light of the well-known GCC Objective-C incident.

Mono Relicensed MIT

Posted Apr 6, 2016 17:08 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (10 responses)

That was in 95 or so. And they decided to contribute to it. After that Apple contributed quite a lot to GCC, Samba and other projects.

All of this has stopped after GPLv3 switch.

Mono Relicensed MIT

Posted Apr 6, 2016 19:09 UTC (Wed) by mathstuf (subscriber, #69389) [Link] (9 responses)

I think you're getting things backwards. Apple contributed to GCC because there were no other compilers to use at the time and the GPL forced them to release the source if they wanted anyone else to actually use the language. If it weren't GPL, I'm sure Apple would have kept it secret. If something like clang had existed in the 90's, the GCC with the GPL wouldn't have even been on the drawing board. The GPL allowed the community to benefit from Apple's contributions which it otherwise is highly unlikely to have done. I think you keep forgetting that the GPL's main goal is to empower users, not developers.

Mono Relicensed MIT

Posted Apr 6, 2016 20:06 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (8 responses)

> If it weren't GPL, I'm sure Apple would have kept it secret.
Like, you know, clang? That they developed fully from scratch?

> I think you keep forgetting that the GPL's main goal is to empower users, not developers.
By keeping them off iStore. I know.

Mono Relicensed MIT

Posted Apr 6, 2016 20:11 UTC (Wed) by mathstuf (subscriber, #69389) [Link] (7 responses)

They have patches which aren't upstream. Xcode's clang version scheme does not even follow upstream (and is usually based off of some random mid-release cycle revision).

Yeah, it's totally the GPL's fault that the ToS Apple drafted years after the GPL was created is incompatible with the Apple ToS.

Mono Relicensed MIT

Posted Apr 6, 2016 20:32 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (6 responses)

> They have patches which aren't upstream.
They are pretty trivial.

And they open sourced the whole thing, even though they could have kept it closed. Think about it.

Mono Relicensed MIT

Posted Apr 6, 2016 20:43 UTC (Wed) by mathstuf (subscriber, #69389) [Link] (3 responses)

Are they? Do you have a link to them?

And I really don't understand how a handful of applications being FOSS really absolves Apple of all the crap they sling at FOSS everywhere else.

Mono Relicensed MIT

Posted Apr 6, 2016 21:12 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (2 responses)

> Are they? Do you have a link to them?
They usually have some internal changes for iOS builds that take some time to be cleaned up and pushed to the mainline clang which has a very linear development model (they still use SVN).

But hey, that handful of patches totally make all of the contributions to clang and WebKit completely irrelevant.

> And I really don't understand how a handful of applications being FOSS really absolves Apple of all the crap they sling at FOSS everywhere else.
A 'handful'? Really? If you want to do mud-slinging then why not call all FSF products something like "insignificant trinkets" or "obsolete obsessions"?

Mono Relicensed MIT

Posted Apr 10, 2016 7:54 UTC (Sun) by jospoortvliet (guest, #33164) [Link] (1 responses)

WebKit is lgpl so they have to make it public - we wouldn't have that entire ecosystem without the license.

They opened CLANG for the same reason MS here opens Mono: to fight competition and strengthen its position and ecosystem. Same reason Google made Android open source. And once the ecosystem has crushed the competition, they close it off and strengthen their control, just like Google is doing with android now. Not impossible with the GPL, for sure, but far easier with a more 'business-friendly' license.

Mono Relicensed MIT

Posted Apr 10, 2016 20:46 UTC (Sun) by Cyberax (✭ supporter ✭, #52523) [Link]

> WebKit is lgpl so they have to make it public - we wouldn't have that entire ecosystem without the license.
And now WebKit is mostly Apache2. The few remaining LGPL pieces are not used by Safari anymore.

Yet it's still alive.

Mono Relicensed MIT

Posted Apr 8, 2016 0:09 UTC (Fri) by lsl (subscriber, #86508) [Link] (1 responses)

>> They have patches which aren't upstream.
> They are pretty trivial.

You mean as trivial as their ARM64 compiler backend that they kept closed and only released when the merging of an other, independently-developed, competing backend into upstream LLVM was imminent?

Mono Relicensed MIT

Posted Apr 8, 2016 1:39 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link]

Yes, like this. They promised to do it once it is production quality and ready for release. And did just that:

> A number of you have asked about the 64-bit CPU in the iPhone 5s, and what that means for LLVM. The iPhone 5s is based on the ARMv8 / Aarch64 instruction set, but the clang compiler in Xcode 5 is based on a custom LLVM Aarch64 backend, not the one currently on llvm.org. Apple is committed to contributing its Aarch64 backend to the community (merging it "the right way" with the existing backend), but it was a significant amount of work, and will take at least several months to work out all the details. I'll keep you posted.

Mono Relicensed MIT

Posted Apr 7, 2016 13:50 UTC (Thu) by aggelos (subscriber, #41752) [Link] (15 responses)

Apple distributes applications from the market and so they're on hook for GPL's patent and source code clauses.

Sheesh, the FUD campaign keeps on going. What the GPLv3 (since you speak of a patent clause, you're refering to GPLv3) is saying (Section 10) is that, if you're conveying a piece of software and then go around and sue the people you've conveyed it to for patent infringement, in relation to their use, further distribution etc of this piece of software, then you lose the license to distribute the software. At no time do you lose any right you have to enforce your software patents.

Now, the sole inference I can make from the above is that you consider enforcement of software patents to be something that one should not hinder. Perhaps I'm misinterpreting your claim though; if so, please clarify.

Section 10 is the 'minimal patent defense' clause of the GPLv3 ("the authors of this program do not want to help you to both distribute a piece of software and sue the recipients for using it; the only thing they can/will do is revoke your license if you do so").

Section 11 is the actual patent clause and it says that if you're actually a copyright holder (i.e. a developer/contributor to a piece of software) who has specifically authorized use of their code under the GPLv3, then again you cannot go around suing people for patent infringement when they use your code. This, of course, does not apply at all to any redistribution of code through the app store.

Which of the two clauses above seems unreasonable?

With regard to source code, (a) this seems straightforward to enforce when people submit an application to the app store (b) the GPLv3 makes it even easier than the GPLv2 to cure (Section 8). Do you have a specific complaint there? If developers submit proprietary applications which infringe on others' copyright, how does apple deal with that and why is that solution not applicable to e.g. GPL'd software?

Mono Relicensed MIT

Posted Apr 7, 2016 18:23 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link] (14 responses)

> Sheesh, the FUD campaign keeps on going. What the GPLv3 (since you speak of a patent clause, you're refering to GPLv3) is saying (Section 10) is that, if you're conveying a piece of software and then go around and sue the people you've conveyed it to for patent infringement, in relation to their use, further distribution etc of this piece of software, then you lose the license to distribute the software. At no time do you lose any right you have to enforce your software patents.
Sheesh. I wonder how GPL defenders know almost nothing about the license they try to defend. Did you even read it?

No. GPL doesn't work like this.

Suppose that Apple has a patent license for showing animated background image with small kittens. Then they allow a 3rd party GPLv3 application which shows animated kittens.

Whoops. Clause 11 kicks in:
> If you convey a covered work, knowingly relying on a patent license, and the Corresponding Source of the work is not available for anyone to copy, free of charge and under the terms of this License, through a publicly available network server or other readily accessible means, then you must either

Apple now has to grant the transferrable patent license to all recipients. And before you object, "knowingly" bar is very low.

Mono Relicensed MIT

Posted Apr 7, 2016 19:34 UTC (Thu) by aggelos (subscriber, #41752) [Link] (13 responses)

Sheesh, the FUD campaign keeps on going. What the GPLv3 (since you speak of a patent clause, you're refering to GPLv3) is saying (Section 10) is that, if you're conveying a piece of software and then go around and sue the people you've conveyed it to for patent infringement, in relation to their use, further distribution etc of this piece of software, then you lose the license to distribute the software. At no time do you lose any right you have to enforce your software patents.

Sheesh. I wonder how GPL defenders know almost nothing about the license they try to defend. Did you even read it?

No. GPL doesn't work like this.

Umm, I believe I've read it, and the part you quote below does not apply to what I typed in above; you do not lose any right you have to enforce your software patents, merely by distributing. Again, AFAIK that's not even possible via copyright law.

Below, you shift the discussion to talking about software patent _licenses_, which is a different thing (and, needless to say, you don't lose those either by distributing something that's covered by the GPLv3).

Suppose that Apple has a patent license for showing animated background image with small kittens. Then they allow a 3rd party GPLv3 application which shows animated kittens. Whoops. Clause 11 kicks in:
If you convey a covered work, knowingly relying on a patent license, and the Corresponding Source of the work is not available for anyone to copy, free of charge and under the terms of this License, through a publicly available network server or other readily accessible means, then you must either
Apple now has to grant the transferrable patent license to all recipients. And before you object, "knowingly" bar is very low.

Oh finally something concrete to talk about.

It is absolutely the case that apple is not allowed to distribute code which is under the GPL, when (a) that code might be infringing on patents (b) they /know/ that the code is infringing on specific patents and (c) they know that they, themselves (but not their customers and downstream recipients), are shielded against a patent infringement lawsuit as a result of a separate license.

Say apple is notified of there being software patents that the code infringes on. If apple doesn't have a license to these patents, they have to stop distributing anyway.

If apple does have a license to these patents, then they cannot be extracting value from GPL'd software by, yes, knowingly, exposing others to danger which they themselves are immune from.

They are not _required_ to grant the a patent license to all downstream recipients. Indeed they can't be, as they are not the patent holder and the patent licenses they have might very well not be transferable (probably the case).

What the GPL says in Section 11 (which you've quoted completely without context, even cutting in the middle of a sentence, to make a misleading point) is that you lose the right to distribute the software if this situation persists. If, despite the fact that there is a patent holder who might sue your customers, you wish to keep distributing it to them (and should we pretend there's nothing wrong with that?) the only way you can do that under the GPL is to either share in the legal risk by arranging no longer to be covered by any patent licensing agreement (as clearly defined in Sec. 11) or by negotiating a transferable patent license with the patent holder and transferring it your downstream recipients.

In neither case are you forced to do anything with a patent license (that would again be legal fiction - you're not the patent holder, you cannot legally be forced into providing others any rights WRT the patent). In the specific case that you've brought forward, if you don't want to share in the danger with your customers or convince the patent holder to extend the license, then you simply stop distributing the software and are done. Your precious patent licenses (again, those are patent _licenses_, not _patents_ that you hold) are untouched, nor do I see how they could ever be touched.

Of course, I imagine the authors of the GPLv3 (I never payed much attention to the discussions back when it was being drafted) did not so much word this to prevent a company or individual from exposing their downstream to "external" patent danger, as to incentivize entities to negotiate any patent deals to the best interests of the community around a piece of code that this entity is also taking advantage of. FWIW, to the degree that the GPL is mostly a tool to enable collaboration and to keep a somewhat level playing field for the collaborators, I would say this is pretty much in the spirit of all previous GPL versions.

Oh, regarding the low bar of "knowingly". The next sentence to the one you half-quoted is

“Knowingly relying” means you have actual knowledge that, but for the patent license, your conveying the covered work in a country, or your recipient's use of the covered work in a country, would infringe one or more identifiable patents in that country that you have reason to believe are valid.

If this is a low bar, please explain how.

Mono Relicensed MIT

Posted Apr 7, 2016 20:08 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link] (12 responses)

> It is absolutely the case that apple is not allowed to distribute code which is under the GPL, when (a) that code might be infringing on patents (b) they /know/ that the code is infringing on specific patents and (c) they know that they, themselves (but not their customers and downstream recipients), are shielded against a patent infringement lawsuit as a result of a separate license.
Yes. Or if they own the patent in questions.

> FWIW, to the degree that the GPL is mostly a tool to enable collaboration and to keep a somewhat level playing field for the collaborators, I would say this is pretty much in the spirit of all previous GPL versions.
Nope. GPL is a vehicle for "software freedom" promotion. Promoting collaboration is decidedly a secondary goal.

Apache or BSD licenses are much more conducive to collaboration.

> If this is a low bar, please explain how.
You need one employee to send an email about violations and that might be enough for the "knowingly" bar to trigger.

Mono Relicensed MIT

Posted Apr 7, 2016 20:33 UTC (Thu) by mjg59 (subscriber, #23239) [Link] (8 responses)

> You need one employee to send an email about violations and that might be enough for the "knowingly" bar to trigger.

At which point Apple can stop distributing that specific work?

Mono Relicensed MIT

Posted Apr 7, 2016 20:44 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link] (7 responses)

Also pull the app from the store. And quite likely from owners' devices.

And if this email goes unnoticed or ignored, then whoops.

Mono Relicensed MIT

Posted Apr 7, 2016 20:57 UTC (Thu) by mjg59 (subscriber, #23239) [Link] (6 responses)

> Also pull the app from the store.

Well, yes, that's how they're distributing it.

> And quite likely from owners' devices.

Why?

> And if this email goes unnoticed or ignored, then whoops.

Sure, just like if you ignore a notice of copyright violation, then whoops. There are still no circumstances in which Apple could be forced to do anything with their patents.

Mono Relicensed MIT

Posted Apr 8, 2016 2:16 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (5 responses)

> Why?
Restoration of an iDevice from a cloud backup might well be considered "distribution".

> Sure, just like if you ignore a notice of copyright violation, then whoops. There are still no circumstances in which Apple could be forced to do anything with their patents.
There's a legally prescribed way to give a notice and all companies have processes to take them into consideration.

In this case it's enough for one employee to send an internal email and for their boss to ignore it (maliciously or just because of a regular human forgetfulness).

And this has actually been explained to me by a company lawyer.

Mono Relicensed MIT

Posted Apr 8, 2016 5:46 UTC (Fri) by mjg59 (subscriber, #23239) [Link] (4 responses)

> Restoration of an iDevice from a cloud backup might well be considered "distribution".

If so, restoration of any copyright infringing material would also leave Apple with liability?

> There's a legally prescribed way to give a notice and all companies have processes to take them into consideration.

Not in every country that Apple do business in, but that's besides the point. In the US, 512(c) shields you from liability only in the case where you're unaware that you're distributing infringing material. If this is a genuine concern:

> In this case it's enough for one employee to send an internal email and for their boss to ignore it (maliciously or just because of a regular human forgetfulness).

then the same argument can be made around straight copyright infringement.

Mono Relicensed MIT

Posted Apr 8, 2016 6:15 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (3 responses)

> If so, restoration of any copyright infringing material would also leave Apple with liability?
Correct.

BTW, GPL-ed software _is_ copyrighted material.

> then the same argument can be made around straight copyright infringement.
Not really. The situation is different - in one case Apple needs to ignore an external note of an infringement and in the other case just an internal memo is enough.

Mono Relicensed MIT

Posted Apr 8, 2016 6:48 UTC (Fri) by mjg59 (subscriber, #23239) [Link] (2 responses)

> Correct.

I send you an email. I attach the binary of a GPLed work. You back it up to iCloud. You restore your backup. Do Apple have an obligation to provide you with the source code on request? If not, then it's clearly not distribution under the GPL.

> BTW, GPL-ed software _is_ copyrighted material.

Well, yes.

> in one case Apple needs to ignore an external note of an infringement and in the other case just an internal memo is enough.

Like I said, that's not true. A service provider is only shielded from liability if it "does not have actual knowledge that the material or an activity using the material on the system or network is infringing" (512(c)(1)(A)(i)). The takedown notice procedure is the defined mechanism for an external party to inform a provider of infringement, but if the provider becomes aware in some other way (such as someone inside the organisation noticing) then the shield is removed. Note that "actual knowledge" is the exact language used in GPLv3 - there's no reason to believe that different standards apply. If someone sending an email to their manager is enough to trigger the GPLv3 language, it's enough to trigger the DMCA language.

Mono Relicensed MIT

Posted Apr 8, 2016 6:58 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

> I send you an email. I attach the binary of a GPLed work. You back it up to iCloud. You restore your backup. Do Apple have an obligation to provide you with the source code on request? If not, then it's clearly not distribution under the GPL.
Details matter. In this case iCloud acts as a dumb storage of the stuff you upload into it. Apple is not liable in this case.

However in case of iOS backups, it downloads new copies of binaries from iStore during the restoration process.

But yeah, that could be fixed fairly easily.

> but if the provider becomes aware in some other way (such as someone inside the organisation noticing) then the shield is removed. Note that "actual knowledge" is the exact language used in GPLv3 - there's no reason to believe that different standards apply. If someone sending an email to their manager is enough to trigger the GPLv3 language, it's enough to trigger the DMCA language.
Yes, OK. However, you're still missing the point.

If Apple distributes GPLv3 software then they're bound by GPL. They can't impose additional restrictions on the license. And the GPL says that Apple is on hook for patent clause.

And the only way to avoid it is to not distribute GPL-ed code. And they have requirements for authors that shield Apple from this scenario.

Mono Relicensed MIT

Posted Apr 8, 2016 7:10 UTC (Fri) by mjg59 (subscriber, #23239) [Link]

> However in case of iOS backups, it downloads new copies of binaries from iStore during the restoration process.

Well sure - if Apple remove something from the store, a consequence of that would be that you wouldn't be able to obtain replacement copies from the store. Isn't that already true if an app is removed for copyright infringement?

> If Apple distributes GPLv3 software then they're bound by GPL. They can't impose additional restrictions on the license. And the GPL says that Apple is on hook for patent clause.

No. Apple have obligations if they have actual knowledge that they are relying on a patent license that they hold in order to be able to distribute the work. Once they have that knowledge, they can do three things:

1) Cease distribution
2) Ensure that they can provide a license to downstream recipients
3) Do nothing and incur copyright liability

But that applies equally well to *any* work that Apple distribute - if Apple have actual knowledge that a work they are distributing infringing someone's copyright, they can do three things:

1) Cease distribution
2) Ensure that they have a license to distribute the work
3) Do nothing and incur copyright liability

Realistically, Apple are not going to choose (2) in either scenario. And outside that, the potential liability to Apple is identical in both scenarios. If Apple's concern is that an internal failure might result in them ending up in case (3) rather than case (1) then that applies equally to both.

Mono Relicensed MIT

Posted Apr 7, 2016 21:24 UTC (Thu) by aggelos (subscriber, #41752) [Link] (2 responses)

It is absolutely the case that apple is not allowed to distribute code which is under the GPL, when (a) that code might be infringing on patents (b) they /know/ that the code is infringing on specific patents and (c) they know that they, themselves (but not their customers and downstream recipients), are shielded against a patent infringement lawsuit as a result of a separate license.
Yes. Or if they own the patent in questions.

Let me quote the part _immediately_ before the sentence which you half-quoted from Section 11, which defines "patent license":

In the following three paragraphs, a “patent license” is any express agreement or commitment, however denominated, not to enforce a patent (such as an express permission to practice a patent or covenant not to sue for patent infringement). To “grant” such a patent license to a party means to make such an agreement or commitment not to enforce a patent against the party.

I hope you'll agree with me that "have a patent license" clearly does not include the case of you being the patent holder. If you're a patent holder, but not a contributor (i.e. a copyright holder), who only conveys the software in question, then you do not automatically grant a patent license to the downstream recipients. Section 11 simply does not apply to entities which are conveyors of, but not contributors to, a piece of software. This is intentional and is meant precisely to make sure that conveyors will not grant patent licenses accidentally.

FWIW, to the degree that the GPL is mostly a tool to enable collaboration and to keep a somewhat level playing field for the collaborators, I would say this is pretty much in the spirit of all previous GPL versions.
Nope. GPL is a vehicle for "software freedom" promotion. Promoting collaboration is decidedly a secondary goal.

I /was/ making a secondary claim though :-) [edit: oh now I see the "mostly" there, I just misspoke. Focus should be on "to the degree that"]

Apache or BSD licenses are much more conducive to collaboration.

Interesting point! Now all you have to do is back that up.

If this is a low bar, please explain how.
You need one employee to send an email about violations and that might be enough for the "knowingly" bar to trigger.

I should hope that explicit communication within the company counts. What should one wait for, a majority of the company employees reading about it in the weekly newsletter? Please explain where the line should be drawn (and justify the line). This is not really unexplored legal territory here. Also, what mjg59 says.

Mono Relicensed MIT

Posted Apr 8, 2016 2:59 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

> I hope you'll agree with me that "have a patent license" clearly does not include the case of you being the patent holder.
Actually, it might in case of subsidiaries or other complicated structures.

> Interesting point! Now all you have to do is back that up.
I believe that to be successful software projects need to have a reliable and significant income stream. Without funding a project can really succeed only in niche areas (like academia) or/and as a hobbyist toy. GPL (the _real_ GPL, not dually-licensed corporate shareware) greatly reduces the number of available ways to do it.

And this actually greatly bother me (me being a card-carrying liberal pinko commie socialist).

I make a prediction that in the coming years we'll see GPL-only projects being even further sidelined.

> Please explain where the line should be drawn (and justify the line).
It's OK where it is. Just avoid the GPL.

Mono Relicensed MIT

Posted Apr 8, 2016 14:53 UTC (Fri) by aggelos (subscriber, #41752) [Link]

I hope you'll agree with me that "have a patent license" clearly does not include the case of you being the patent holder.
Actually, it might in case of subsidiaries or other complicated structures.

Let's not forget to mention how you (a) quote the GPL out of context to arrive at a fear-mongering conclusion (b) keep shifting your original claim to increasingly intricate cases when your previous claims are shown to be unsupportable.

The typical scenario where you do not actually hold your own patents is that you've transferred them to a shell company which can sue for patent infringement while not being vulnerable to countersuits (since it doesn't actually produce anything). Is that the scenario you're talking about? If so, are we expected to sympathize with software patent aggressors here? If not, please provide the concrete scenario, how it applies to apple and why one would want to enable it.

I'm afraid my weekly quota for posting online is almost out, so I'll have to postpone replying to the rest of your points (especially since they're about a separate discussion).

Mono Relicensed MIT

Posted Apr 7, 2016 12:29 UTC (Thu) by paulj (subscriber, #341) [Link]

So your argument is that authors using copyright to impose conditions to keep the source of some work of theirs available is bad, because it prevents Apple from closing such code and other derived code?

Your argument therefore is that authors of code shouldn't be able to place conditions on the use of their code? You're against copyrights?

Mono Relicensed MIT

Posted Apr 6, 2016 11:24 UTC (Wed) by Del- (guest, #72641) [Link] (20 responses)

> Apple is perfectly OK with other open source licenses. It's GPL that is causing problems.

The information content of your posts in this thread is unusually low. Are you trying to say that Apple is not OK with copy-left in general? It certainly sounds like you are making that argument, but then again, your signal-to-noise ratio is horrible in this thread.

You are against DRM, but still promote appstore as the greatest collaborative project (as if there is anything collaborative about it, but hey, let us redefine the whole English language to fit the discussion of the day).

Because one billion devices runs an OS, anybody not distributing their software on it must be doing it wrong. Do you even believe that yourself?

Mono Relicensed MIT

Posted Apr 6, 2016 17:24 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (19 responses)

> The information content of your posts in this thread is unusually low. Are you trying to say that Apple is not OK with copy-left in general?
Not the copyleft (MPL is fine) but with the [L]GPL in particular. GPLv3 has extremely dangerous patent clauses and GPLv2 might force Apple to disclose their proprietary source.

> You are against DRM, but still promote appstore as the greatest collaborative project
It's simple - I want stuff that benefits actual real-life breathing users now. iStore does this. GPL doesn't.

Mono Relicensed MIT

Posted Apr 6, 2016 21:35 UTC (Wed) by hummassa (guest, #307) [Link] (18 responses)

Cyberax, you are an intelligent person, but you (for some reason) have been repeating one piece of FUD that I find intolerable:

> GPLv2 might force Apple to disclose their proprietary source.

Let me explain to you: any source code that Apple could be "forced to disclose" would not be, in any capacity, "proprietary" as in "sole (intellectual) property of Apple" but it would be

A Derivative work (let's call it "D") of some Original work (let's call it "O") that was not a creation of Apple, but of third parties. Those third parties had the sole right to create derivative works, but they Licensed it to Apple (and to the General Public) under some conditions stated on the text of the General Public License (yeah, you grasp where I'm going here).

So, when Apple makes a derivative of a original work using the license granted by O's copyright holders under the terms of the GPL, the differences between D and O can never be "proprietary". It's not their work. D wouldn't exist if O didn't exist first.

Mono Relicensed MIT

Posted Apr 6, 2016 22:17 UTC (Wed) by Cyberax (✭ supporter ✭, #52523) [Link] (17 responses)

> > GPLv2 might force Apple to disclose their proprietary source.
> Let me explain to you: any source code that Apple could be "forced to disclose" would not be
Let me repeat, Apple may be forced to disclose their proprietary code if they allow 3rd party applications into their App Store.

You find that intolerable? So does Apple.

Mono Relicensed MIT

Posted Apr 6, 2016 23:33 UTC (Wed) by anselm (subscriber, #2796) [Link] (1 responses)

Let me repeat, Apple may be forced to disclose their proprietary code if they allow 3rd party applications into their App Store.

You mean, like Google is forced to disclose their proprietary code because they allow GPL applications to be sold through the Google Play Store? Yeah right.

Mono Relicensed MIT

Posted Apr 7, 2016 1:07 UTC (Thu) by rsidd (subscriber, #2582) [Link]

You mean, like Google is forced to disclose their proprietary code because they allow GPL applications to be sold through the Google Play Store? Yeah right.
The relevant Android code (that most play store apps link to) is not proprietary. It is Apache-licensed. Some of the APIs connect to proprietary stuff like Google Play Services, but I believe the "system library" exception would apply there. (But then it would to Apple's iOS too.) Plus in Android's case there are two levels of linking -- you link to the open-source and Apache-licensed AOSP, which has APIs to connect to a proprietary Google Play Services; I doubt any court would claim that the proprietary stuff is thereby "infected".

Mono Relicensed MIT

Posted Apr 7, 2016 12:55 UTC (Thu) by paulj (subscriber, #341) [Link]

Cyberax, that's just not true. No matter how many times you repeat it, it stays not true.

Mono Relicensed MIT

Posted Apr 7, 2016 13:51 UTC (Thu) by aggelos (subscriber, #41752) [Link] (13 responses)

Let me repeat, Apple may be forced to disclose their proprietary code if they allow 3rd party applications into their App Store.

The app store predominantly distributes 3rd-party applications. I'm assuming you misspoke here and meant to say GPLvX, for some value of X.

First, you persist in repeating the false claim that people would be forced to GPL stuff. Copyright does not work that way. All the GPL can do, and is trying to do, is to prevent people from creating proprietary derivatives of the GPL code, i.e. close it up. If you think that's unfair, let's talk about that, but you've been well aware for a while now that your claim above is false, only you keep repeating it as part of this FUD campaign.

Second, which components, specifically, do you claim would end up constituting a derivative of GPL code if apple were to amend their ToS and subsequently distribute GPL'd projects? Further, which specific clauses do you base this claim on?

Mono Relicensed MIT

Posted Apr 7, 2016 17:42 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link] (12 responses)

> First, you persist in repeating the false claim that people would be forced to GPL stuff.
A simple question: how do I release a software package linked with a GPL-ed library and which doesn't include its source code?

After all, nobody is forced to GPL anything, right?

Mono Relicensed MIT

Posted Apr 7, 2016 17:52 UTC (Thu) by rsidd (subscriber, #2582) [Link]

A simple answer: by not using that GPL'd library. Nobody's forcing you to, right?

On Linux very few system libraries are GPL'd (readline is one of the few exceptions). On Android/iOS no system libraries are GPL'd.

Mono Relicensed MIT

Posted Apr 7, 2016 18:03 UTC (Thu) by rsidd (subscriber, #2582) [Link]

I think what you're asking w.r.t. iOS/AppStore is the reverse of what you say -- how does Apple release an OS to which you're linking a GPL'd app that you seek to release through the app store?

Answer: the "system library" exception to the GPL applies.

Mono Relicensed MIT

Posted Apr 7, 2016 21:30 UTC (Thu) by aggelos (subscriber, #41752) [Link] (9 responses)

First, you persist in repeating the false claim that people would be forced to GPL stuff.
A simple question: how do I release a software package linked with a GPL-ed library and which doesn't include its source code?

Umm, to me this reads like "I want to violate the GPL and include copyleft code into a proprietary product". Is that what you mean? If not, please provide a specific scenario to clarify (and see also rssid's reply re: the system libraries).

After all, nobody is forced to GPL anything, right?

Not by the GPL or any copyright license. A license is not a contract; all it can do is dictate the terms under which you are allowed to use/modify/distribute/etc the software. If you infringe on the license, you may be liable for damages as prescribed by copyright law. Now if you're feeling helplessly drawn to integrating GPL code... you're still not forced. You're given an _incentive_ to produce free software (specifically, software under a GPL-compatible, i.e. necessarily free, license), so as to be able to make use of the existing code base. If that doesn't appeal to you, and as has been pointed out a few dozen times by now, by quite a few people, you're free to not create derivatives.

Mono Relicensed MIT

Posted Apr 8, 2016 2:09 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (8 responses)

> Umm, to me this reads like "I want to violate the GPL and include copyleft code into a proprietary product". Is that what you mean? If not, please provide a specific scenario to clarify (and see also rssid's reply re: the system libraries).
To me this means that GPL forces one to comply with its terms if one wants to use it in their products. In particular, it forces opening of the source code. The rest is just details.

> If you infringe on the license, you may be liable for damages as prescribed by copyright law.
And this is just useless lawyering.

It's kinda like saying that you're totally free to go and kill people, but if you do this you might be captured by law enforcement and put into jail for life. But nothing stops you from doing it!

Mono Relicensed MIT

Posted Apr 8, 2016 8:49 UTC (Fri) by paulj (subscriber, #341) [Link] (6 responses)

Rubbish, rubbish, rubbish. You are deliberately ignoring the other option: Don't use the code whose licence has conditions that don't suit you.

That isn't even GPL or FOSS specific. Proprietary software could have disagreeable licence conditions too, not least the price.

If you disagree with that, then what you are disagreeing is the whole concept of authors having a say in how their code may be copied via copyright and/or you disagree with derivative works.

From reading your collective comments here on licensing make it clear that your problem is that you (and a few entities you favour) may not be able to do with other people's code exactly as you wish. Either that, or you simply don't have a coherent view of what rights you think a cod author should and should not have over that code, that doesn't reference your interests at least.

I ask again: Am *I* allowed to do whatever I want with *any* of _your_ code?

Mono Relicensed MIT

Posted Apr 8, 2016 20:07 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (5 responses)

> Rubbish, rubbish, rubbish. You are deliberately ignoring the other option: Don't use the code whose licence has conditions that don't suit you.
What part of: "GPL forces you to disclose your source code if you distribute it as a part of your product" is incorrect?

> If you disagree with that, then what you are disagreeing is the whole concept of authors having a say in how their code may be copied via copyright and/or you disagree with derivative works.
No, I'm not. You're just misrepresenting the truth by cowering behind technicalities.

Mono Relicensed MIT

Posted Apr 8, 2016 20:14 UTC (Fri) by mathstuf (subscriber, #69389) [Link] (3 responses)

> What part of: "GPL forces you to disclose your source code if you distribute it as a part of your product" is incorrect?

The part where the GPL isn't applicable to code that you don't ship in the first place?

> No, I'm not. You're just misrepresenting the truth by cowering behind technicalities.

Um…wow. You're saying people shouldn't be free to license their code under the GPL because you (and Apple) don't like it.

Mono Relicensed MIT

Posted Apr 8, 2016 20:18 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (2 responses)

> > What part of: "GPL forces you to disclose your source code *if* *you* *distribute* *it* as a part of your product" is incorrect?
> The part where the GPL isn't applicable to code that you don't ship in the first place?
Highlighted.

> Um…wow. You're saying people shouldn't be free to license their code under the GPL because you (and Apple) don't like it.
Uhm, sorry I was not clear. I meant that I absolute agree that authors should be able to use whatever license they want for their code. It's just that GPL is almost never a _good_ choice.

Mono Relicensed MIT

Posted Apr 8, 2016 20:43 UTC (Fri) by mathstuf (subscriber, #69389) [Link]

Well, yeah, if you ship code deriving from GPL code, you have to ship your source as well. In aggregate, only the GPL bit needs source (though I'm not as familiar with GPLv3 here, care to fully quote the relevant section(s)?). The alternative to not shipping code for your code is to not have the GPL code in there in the first place.

> Uhm, sorry I was not clear. I meant that I absolute agree that authors should be able to use whatever license they want for their code. It's just that GPL is almost never a _good_ choice.

In your opinion. Others have different weighting functions.

Mono Relicensed MIT

Posted Apr 8, 2016 20:55 UTC (Fri) by paulj (subscriber, #341) [Link]

"It's just that GPL is almost never a _good_ choice for Cyberax."

Fixed it for you.

Mono Relicensed MIT

Posted Apr 8, 2016 20:52 UTC (Fri) by paulj (subscriber, #341) [Link]

In a very strict reading of just that part of your comment, it's correct. However, each of your comments is part of a body that paints the fuller picture of your opinion. To walk back up 2 nodes in this thread tree to your original question:

"how do I release a software package linked with a GPL-ed library and which doesn't include its source code?"

That is the context for that statement in the child comment of "GPL forces one to comply with its terms if one wants to use it in their products.", along with the wider context of your comments which make it clear you have a major issue with the fact that some people might write code and make it generally available under conditions that don't suit you.

Why don't you just move on? Go find a BSD/MIT/Apache version, or a proprietary for-€ version, or else write it yourself, why not do that?

Why is it you find it so hard to accept that other people have made decisions about the conditions on their code that aren't "Let Cyberax (and other entities he favours) do whatever they want with the code, for no consideration in return"? If you don't have a problem with the notion of copyright, then what's the issue? Do you similarly go on long campaigns on forums of proprietary software? Or is just GPL?

Tell me, what are all the licenses that code you have copyright over is under? In particular, do you ever write proprietary software?

Mono Relicensed MIT

Posted Apr 9, 2016 7:10 UTC (Sat) by anselm (subscriber, #2796) [Link]

To me this means that GPL forces one to comply with its terms if one wants to use it in their products.

It's a copyright license. What do you expect it to do?

If you don't like the terms of the GPL, don't base your code on GPL'ed software. If you don't like the terms of Microsoft's proprietary SDK license, don't base your code on Microsoft's proprietary SDK. Same difference. Your choice.

Mono Relicensed MIT

Posted Apr 4, 2016 14:06 UTC (Mon) by nhasan (guest, #1699) [Link] (1 responses)

He is just an Apple fan trying to reconcile his iLove with Free Software community but being a complete ass in the process. App Store...a great collaborative project? Wow.

Mono Relicensed MIT

Posted Apr 4, 2016 17:11 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link]

> He is just an Apple fan trying to reconcile his iLove with Free Software community but being a complete ass in the process.
Actually, no. I do use a Mac desktop, but my phone is Android and I don't have other Apple products.

Next personal attack, please.

Mono Relicensed MIT

Posted Apr 7, 2016 12:24 UTC (Thu) by paulj (subscriber, #341) [Link] (1 responses)

From other comments you've made, I suspect that you have a rather self-centred definition of "freedom". You want freedom to do with code - anyone's code - as you wish. Literally "you", and you alone. Cause you don't want others to have the same freedom with your own code.

Mono Relicensed MIT

Posted Apr 7, 2016 17:37 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link]

Nope. I want _everyone_ to be able to use all the available code without restrictions. I personally don't have problems with GPL or anything, it's just not relevant for me or my job.

> Cause you don't want others to have the same freedom with your own code.
What code I or my company releases is under permissive licenses.

Mono Relicensed MIT

Posted Apr 4, 2016 9:11 UTC (Mon) by anselm (subscriber, #2796) [Link] (2 responses)

Well, and Taj Mahal is just a shrine serving no useful functions at all. So?

People are prepared to travel long distances at some inconvenience to look at the Taj Mahal. It is a major contributor to the economy in Agra today (nearly four centuries after it was built) because otherwise tourists would have much less of a compelling reason to go there. That may not have been Shah Jahan's original plan but it does make a difference now.

On the other hand, it is highly unlikely that the iOS app store will have any influence on the economy anywhere, 400 years from now – never mind attracting people from far away to come look at it. And as far as I'm concerned it is a “collaborative project” only in the sense that Amazon and eBay are “collaborative projects”, IOW not that much – it relies on lots of people giving it stuff to sell but these people don't really collaborate with one another on improving the iOS app store or “ecosystem” (beyond making it more interesting by adding their stuff to it, anyway).

Mono Relicensed MIT

Posted Apr 4, 2016 9:20 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (1 responses)

> On the other hand, it is highly unlikely that the iOS app store will have any influence on the economy anywhere, 400 years from now – never mind attracting people from far away to come look at it.
I can see people coming to Cupertino to look at the giant statue of Steve Jobs that will be erected some time soon, probably.

Mono Relicensed MIT

Posted Apr 4, 2016 20:25 UTC (Mon) by Wol (subscriber, #4433) [Link]

One name ... Ozymandias

Cheers,
Wol

Mono Relicensed MIT

Posted Apr 4, 2016 8:48 UTC (Mon) by niner (subscriber, #26151) [Link]

By your reasoning the world's greatest collaborative project is probably Wallmart or Amazon though neither Apple's App Store, nor Wallmart nor Amazon qualifies for the common definition of the words "project" or "collaborative".

Mono Relicensed MIT

Posted Apr 4, 2016 11:11 UTC (Mon) by krake (guest, #55996) [Link]

> Or perhaps making the world's greatest collaborative project greater, thus bringing improvements to the life of hundreds of millions of people.

You count Mono as the "world's greatest collaborative project"?

Mono Relicensed MIT

Posted Apr 5, 2016 14:00 UTC (Tue) by aggelos (subscriber, #41752) [Link] (4 responses)

Or perhaps making the world's greatest collaborative project greater, thus bringing improvements to the life of hundreds of millions of people.

Right. It is of course important to deconstruct the above, well, claim. However, it seems to me it's even more important to discuss some meta considerations.

This is not to do with any qualities of the person I'm replying to, but with a larger pattern I notice in these posts which, to me, appears to be the primary means by which they're trying to sway people closer to their position. And IMHO this pattern needs to be talked about far more than the individual posts.

Off the top of my head, some claims I remember coming from the same poster in the last few weeks are:

[Apple's app store] is the world's greatest collaborative project

The GPL is cancer, infective, it propagates to everything it touches and of course is viral.

[GPL software] isn't actually free in any sense.

Let me put aside the discussion on whether or not there is any kind of valid point behind those posts for a bit. From my PoV, what I see is a series of posts specifically phrased to agitate, making what the poster can be reasonably sure will appear (to a large percentage of the readers) to be preposterous claims, with absolute conviction and no supporting argumentation (definitely not at the time of the claim).

FWIW, I see two aspects here which are problematic. One of course is the setting of the agenda. Constant discussion on the utility or undesirability of (various versions of) the GPL in the comments on LWN, necessarily shifts the perception of the license in this community from uncontroversial or, at least, a reasonable licensing choice (perhaps my perception is skewed, but I think that was the social norm here, say, 5 years ago) to "eh, permissive seems to be the safe choice here, nobody ever got called out for going with permissive". Regardless of what one thinks about social norms marketing, I hope we all agree that perception matters.

Now, this agenda setting would be OK (if someone honestly thinks there's something hugely problematic about X, they should feel comfortable in speaking up and directing attention to X), but for two things. (a) So far, the strategy for maintaining interest in this (judging by the narrative seen so far) non-topic is not to present convincing arguments, but to manipulate people into engaging, i.e. the strategy employed by trollumnists and linkbaiters. (b) propaganda-style repetition of the same old narrative, pretending it wasn't just torn down (in the eyes of other participants at least; not addressing their arguments is a means of keeping the discussion going, instead of moving it forward).

There's a second aspect in which the (IMHO consciously) absurdly exaggerated nature of those claims is problematic. I find it hard to believe that this practice is intended to convince; AFAICT, it only serves to stretch the spectrum of expressed opinions so far in the desired direction, so as to unduly skew any 'compromise' position in favor of the poster's position. I mean, any casual, disinterested, or just mildly interested reader will very likely not take the trouble of following the arguments (or lack thereof). Even if they come out with the feeling that one side is clearly more convincing and the other side is 'at most 10% right', a 10% shift towards an artificially exaggerated position might still be a lot.

Which is not to say there might not be an element of venting here too (we all need to rant, occasionally). This is understandable, but I would humbly suggest that venting is better done among friends and can be toxic to discussion forums. I dare say we've all been there of course.

Mono Relicensed MIT

Posted Apr 5, 2016 17:48 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (3 responses)

Me? I really dislike GPL now. These opinions are somewhat exaggerated, but I do get annoyed by GPL love-fests each time I see a discussion about "software freedom".

Mono Relicensed MIT

Posted Apr 5, 2016 21:54 UTC (Tue) by lsl (subscriber, #86508) [Link]

> Me? I really dislike GPL now.

How come? Over the last few weeks your posts really come across as though you were beaten up by someone utilizing a printed-out GPL hardcopy. That's not what actually happened, is it?

Mono Relicensed MIT

Posted Apr 6, 2016 4:39 UTC (Wed) by rsidd (subscriber, #2582) [Link] (1 responses)

These dozens of comments all being sparked by my comment that "all this stuff is released under the MIT license" after the Microsoft acquisition of Xamarin... and I primarily meant the proprietary extensions, not the LGPL stuff. Lesson to me to be clearer when posting? Or would this thread have happened anyway?

Mono Relicensed MIT

Posted Apr 6, 2016 12:29 UTC (Wed) by aggelos (subscriber, #41752) [Link]

Or would this thread have happened anyway?

Umm, this FUD campaign has been going on for a while now, yah. Seems jra at least was tired of it by October 2015.

Unfortunately, with FUD campaigns there's no winning ('winning' here would be an actual discussion). Don't play and there'll be dozens of deceptive or misleading comments standing uncontested at a central discussion forum. Try to engage the originator in a discussion and they'll put up a pretense of debating while repeating their narrative and ignoring any inconvenient points you make. Hence my 'meta' post above.

Mono Relicensed MIT

Posted Apr 3, 2016 15:56 UTC (Sun) by rsidd (subscriber, #2582) [Link]

Many previously proprietary components were relicensed under MIT too. That is what I was referring to.

Mono Relicensed MIT

Posted Apr 3, 2016 13:36 UTC (Sun) by dakas (guest, #88146) [Link] (2 responses)

Well, LGPLv2 is becoming less and less interoperable, and upgrading to LGPLv3 would have implied a patent license.

The main criticism against Mono, long before getting acquired by Microsoft, has always been the lackadaisical stance towards being patent-encumbered.

Moving to MIT license does nothing to change that. Rather it cements the "no patent grant implied" state that always had been the main argument against relying on Mono.

This merely clarifies that Microsoft is not going to rely on copyright for monetizing Mono, but since Microsoft did not rely on copyright for monetizing C# either, maintaining the previous revenue model would have been contradictory under common corporate ownership.

In other words: this is not really newsworthy. It's just common sense license cleanup after hauling Mono into the corporate mothership, consolidating the patent trap strategy of C# and Mono so that you can choose the bait to swallow without bothering about copyright.

Mono Relicensed MIT

Posted Apr 3, 2016 15:43 UTC (Sun) by rahulsundaram (subscriber, #21946) [Link] (1 responses)

I am assuming you weren't aware of

https://github.com/mono/mono/blob/master/PATENTS.TXT

It is far more important for MS to get more adoption of .NET compared to the patent licensing revenue.

Mono Relicensed MIT

Posted Apr 3, 2016 15:52 UTC (Sun) by pizza (subscriber, #46) [Link]

It's worth mentioning that the scope of that patent grant is limited to the core language, CLR, and only a subset of the runtime libraries (including stuff that's considered "standard" if you're writing C# stuff under Visual Studio on Windows)

Mono Relicensed MIT

Posted Apr 1, 2016 22:40 UTC (Fri) by ChingDucXiabong (guest, #107997) [Link] (3 responses)

Miguel de Icaza is a Microsoft whore, who likely damaged the GNOME project, in order to keep the Linux desktop at 1% market share. Do you think a $50 billion corporation like Microsoft pays whores like Miguel to do what they want? Of course they do, and $50 billion buys a lot of whores, bad GNOME code, and open source licenses. Windoze 10 is EVIL NSA spying software.

Mono Relicensed MIT

Posted Apr 2, 2016 6:49 UTC (Sat) by lkundrak (subscriber, #43452) [Link]

This kind of comments is not welcome here.

Please

Posted Apr 2, 2016 8:28 UTC (Sat) by corbet (editor, #1) [Link]

This kind of attack is really not welcome here; do not do it again.

Mono Relicensed MIT

Posted Apr 2, 2016 9:13 UTC (Sat) by jrigg (guest, #30848) [Link]

Seems to be a pattern here (see comment from guest #107866):

http://lwn.net/Articles/681116/#Comments


Copyright © 2016, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds