"To me, Apple's not-yet-executed move to liberate some of the Swift 2.0 code seems a tactical stunt to win over developers who currently prefer the relatively more open nature of the Android/Linux platform."
They said why this was the case -- Swift 2.0 isn't finished yet -- but have pledged to open-source the entire thing.
It's not necessarily a bad thing to have sentinels out on the fringes, like RMS and Bradley, to remind us not to get lost. We don't have to go where they are, but we can at least navigate by them.
Having said that, there is a detail here which Bradley is omitting. Specifically, that is that Apple is not anti-copyleft, what Apple is (and what many other corporations are), is anti-GPLv3.
Apple shipped Samba for a very long time. They could have replaced it at any point, but they didn't need to. It was Samba switching to GPLv3 that pushed Apple into writing its own SMB implementation.
Now, it's possible that these things are unrelated, but I believe I am correct in stating that Apple has never shipped GPLv3 software, and anything that has switched to v3, simply gets replaced or held back at its last v2 release (see https://news.ycombinator.com/item?id=3559990 ). The most likely explanation is that they are not prepared to be bound by the patent/tivoisation clauses.
The deeper question is why these v3 terms are unacceptable. IANAL, but I have worked at places where there are pretty flexible policies around using/distributing Free Software, and it's not at all uncommon that blanket policies apply to v2 licensed software, but any and all use of v3 (or indeed AGPL) software require specific individual review by Legal to ensure that you aren't placing the company in an actionable position. That's a pretty heavy time/effort burden to place on the (ideally nimble) process of developing and shipping software, it's also a cost burden.
Personally, I love the GPLv2, I think it's a masterstroke of a license. I don't use it very much anymore, but I'm very glad that exists. I think v3 goes to far, and I think AGPL goes way too far.
I tend to think that there are strategic projects which should be under v3, to protect the wider Free Software community from being unreasonably locked out of their own future, however, it seems fairly clear to me that overall, v3 has hurt copyleft more than helped it.
>Personally, I love the GPLv2, I think it's a masterstroke of a license. I don't use it very much anymore, but I'm very glad that exists. I think v3 goes to far, and I think AGPL goes way too far.
I am really not familiar with licenses and I have a feeling the legal text won't make much sense to me either if I treid reading it by my self
tl;dr is that v3 places a much broader burden on entities releasing GPL software, wrt patents and the general digital freedoms of users.
GPLv3 = GPLv2 + you can't sue any users/distributors for violating your patents + you grant an implicit license to all your relevant patents to your users + you must protect users from patents that might be covering code you contribute
For companies with patents and lots of patents they are afraid of losing their ability to sue for patent violations. I'm unclear whether the license you grant is just for the GPLed code or just in general. And in general it's unknown how all this plays out in court, so most lawyers (and thus companies) avoid v3 like the plague.
AGPL = GPL + must distribute even if you just provide an internet accessible service.
The idea being that the GPL only applies to distribution, so if I modify some GPL software but never distribute it (because I just run it on a server I control) I never have to GPL my changes. The AGPL says you have to distribute/open source your changes even if you don't distribute if the modified software is accessible over the network to users.
In my simple mind, open source is a developer construct while free software is a user construct. Thank you for bringing a user's perspective in explaining GPL.
I see permissive Open Source licenses as giving developers all the freedom, and copyleft licenses as giving the software itself, all the freedom.
I choose this distinction only because some users are developers, and a copyleft license specifically restricts them from doing some things (i.e. closed source forks), thus restricting their absolute freedom. Instead, the software itself is promoted to the receiver of ultimate freedom, because its basic rights are enforced through all its generations/forks/integrations.
Apple have been pretty open about why they don't like the GPLv3. It's not actually the patent clauses at all (their own Apple Public Source License predates the GPLv3 by several years and includes a patent grant): https://news.ycombinator.com/item?id=8868994
Unfortunately this bis of misinformation has become part of the public consciousness.
I understand the concept when it comes to games, since a finished product is one where the developers has switched over to the next product and only occasional maintenance patches are added post-launch. It makes sense to call those finished, but a programming language that does this would likely be called dead.
One that meets a pre-defined milestone and is seen fit for public release. I guess Apple wants to get some warts out and add a few features before it releases the project to the "design by committee" open source world.
You know what else isn't finished? The vast majority of open-source software. You don't need something to be finished to open-source it. That feels like the Android model to open-source: source code dumps at convenient times. The only valid reason not to open-source when you say you will is that you don't have all the copyright to do so, and are actively rewriting the bits you can't open-source.
An open, collaborative development community in an Open Source or Free Software project, is a very good thing, but it is a separate thing.
And frankly, if there's going to be a new, modern, high quality language with support for the platforms I care about (Linux, OS X), I'm going to use it.
I wonder if RMS is going to publish a Swift version of "The Java Trap". I don't think it will meet with the same result (relicensing under the GPL).
Non-copyrightable API's is in turn is the reason we enjoy atleast some level of interoperability today, as API's and their corresponding functionality can be re-implemented across platforms despite all the 'vendor-lock in' companies like Microsoft have tried to enforce.
>If coders get to apply it to every line of copy-pasted code they "write"
This is not true, copyright applies foremost to the work as a whole, not each indivudual line, this was already proven as Oracle lost their attempt at copyright infringement on a 9 line rangecheck function, which was ruled as fair use.
Code is not code, it is a specification on how to build machine language. Does that make my code any less valuable?
> ...these specifications have never been considered copyrightable until...
That's correct. Also, nothing was copyrightable until someone started a legal campaign.
> ...copyright applies foremost to the work as a whole...
So copyright can apply to an API as a whole because an API is the same thing as code. APIs are not free. People have to design them and test them, just like when they write code. What universal law do you think grants you the right to just take someones work?
> Non-copyrightable API's is in turn is the reason we enjoy atleast some level of interoperability today...
Yes and tax loopholes are legal until they're not too. Just because you're getting something for free already doesn't mean you're somehow guaranteed entitlement to it forever.
Also sure a compiler may drop but its useless without a pile of libs.
Apple has a long proven track record of saying one thing and doing the complete opposite. I'll save my congratulations for when they actually do it.
Easy. Not say they are going to do something unless they know they can do it. Better yet: announce it when you can do it.
This isn't unique to Apple, but it's stupid to ignore reality. At the end of the day, they haven't release Swift. Banking on it now is, without a doubt, a foolish gamble.
>Apple has a long proven track record of saying one thing and doing the complete opposite.
Care to share?
And you can guarantee that Swift will be entirely open sourced? Sorry, but it doesn't change things because of reality. Apple hasn't released Swift yet as open source. So far, I'm winning.
> Care to share?
Besides FaceTime? Off the top of my head, see the releases of the iPod, iPhone, and iPad. All were released with claims that were turned completely around (only for Mac, no native apps, smaller versions are bad). This ignores other things like how multitasking is bad one year and then good the next.
Simply put, when someone from Apple says they won't ever do something, I assume they just haven't come up with a way to market it effectively.
Personally, I think Apple doesn't give a fig about software freedom. I don't think they are in favour of it, or against it, they simply want to ship their products.
For a time they were happy to use GPL software to help them do that, but then v3 happened and they have been moving away from the projects that switched to v3.
It's impossible to argue that WebKit, LLVM, Swift, etc. are intended "to kill software freedom". They are simply software projects that exist and release code under an OSI license (or will do, in the case of Swift).
If Apple/Google/others can push these projects beyond where things like GCC are, that is more the fault of the GCC project for not being nimble enough to stay on the leading edge (if it ever was there). It's also not because they are determined to kill projects like GCC, it's because they need really good compilers/renderers/etc.