Software Freedom Conservancy: Distributing Linux With ZFS Is Illegal (phoronix.com)

LichtSpektren writes: Phoronix reports that Bradley M. Kuhn and Karen M. Sandler at the Software Freedom Conservancy (SFS) have posted a blog post today arguing that Canonical's plan to distribute Ubuntu 16.04 LTS "Xenial Xerus" with support for the ZFS file system violates the Linux kernel's GPLv2 license.

On February 18, Dustin Kirkland at Canonical wrote on his blog: "We at Canonical have conducted a legal review, including discussion with the industry's leading software freedom legal counsel, of the licenses that apply to the Linux kernel and to ZFS. And in doing so, we have concluded that we are acting within the rights granted and in compliance with their terms of both of those licenses...The CDDL cannot apply to the Linux kernel because zfs.ko is a self-contained file system module — the kernel itself is quite obviously not a derivative work of this new file system. And zfs.ko, as a self-contained file system module, is clearly not a derivative work of the Linux kernel but rather quite obviously a derivative work of OpenZFS and OpenSolaris. Equivalent exceptions have existed for many years, for various other stand alone, self-contained, non-GPL kernel modules. Our conclusion is good for Ubuntu users, good for Linux, and good for all of free and open source software."

The SFS's blog post of today states: "We are sympathetic to Canonical's frustration in this desire to easily support more features for their users. However, as set out below, we have concluded that their distribution of zfs.ko violates the GPL."

Comments

No winners here.

[Anonymous Coward]

I'm not sure what anyone would stand to gain by arguing that Ubuntu is violating the GPLv2 license. It seems kind of ridiculous.

Re:No winners here.

[Austerity Empowers]

As usual, it has less to do with the specific example as it has to do with precedents that may not be desirable. It seems like if you release these two things "separately", then nothing is wrong. However by including this other binary with this problematic license as part of a single distribution you are "apparently" breaking the terms of the GPLv2 which requires the distribution be under GPLv2.

Hairs can be split about what a "distrubution" is. I can add ZFS to my own system and not be wrong. Why cannot a script add ZFS to my system for me during install? When does it become a "distribution", given that most of us don't install from optical media anymore, and frequently download bits and pieces as we need them for our system anyway. I'm trying to see the evil here that this narrowly avoids, but I don't yet...provided the terms of the various pieces of software are still met on their own.

Re:No winners here.

[nycsubway]

That's the distinction of a distribution vs a software project. The distribution is heterogeneous: loaded with tons of open-source projects of different licenses. Maybe acknowledge this during the install process by having the user agree to the multiple licenses that are included in the distro.

Re:No winners here.

[Anonymous Coward]

Why not just make it illegal to use all software released under GPL v.2? That would make it easier for everybody and people could move to a sane license.

Re:No winners here.

[Cramer]

Translation: *whaaa*We don't like ZFS*whaaa*

The only possible grounds for complaint is that it's shipped with the base kernel (linux-image-4.4.0-7-generic) I've not unpacked the source package to see if they ship it with zfs already shoe-spooned into the tree. (but, there is an update-zfs.sh script in the debian specific parts) In fact, most of the modules are "GPL", not "GPL v2"; and yes, the zfs stuff are the only non-GPL modules in the package.

HOWEVER: Adding a file to an archive of your own making (i.e. a dot.deb, dot.rpm, dot.pkg, dot.zip, dot.tgz, ...) has ZERO entanglement with the GPL. If that were true, ISO's and disk images would be "illegal".

Re:No winners here.

[Anonymous Coward]

When you compile Linux+ZFS into one binary you must decide on a license for this binary.
This license can be neither GPL2 nor CDDL.

It's different if the user gets Linux+ZFS as two different source packages with
their respective licenses. Then the user can compile them into a binary
which he does not distribute, so there is no need for a license of that binary.
(This is the way Debian goes.)

Re:No winners here.

[buchner.johannes]

I'm not sure what anyone would stand to gain by arguing that Ubuntu is violating the GPLv2 license.

Btrfs developers. ZFS on Linux makes alternatives less important and fewer people/companies invest time/money into creating that software.

Then again, as TFA says: Oracle, why don't you just relicense the bloody thing.

Re:No winners here.

[Grishnakh]

No one is combining Linux + ZFS into a single binary. Ubuntu has already said, over and over, that ZFS will be a separate kernel module (zfs.ko).

This is no different from how Nvidia drivers are distributed, and Nvidia's drivers are completely proprietary. They've been distributing them this way for well over a decade now.

Re:No winners here.

[Anonymous Coward]

Uhhh... No., nVidia drivers get around it by being compiled by the user - and thus not distributed as a binary.

Re:No winners here.

[edtice1559]

This doesn't "get around" anything. Having the user compile them is just another means of distributing a binary. The user compiles the NVidia drivers because it's the easiest way to ensure that they are compatible with the running kernel. The NVidia driver is a hybrid model. The user compiles part of it and part of it is a blob. The user-compiled part is GPL. The interesting stuff is in the blob.

Re:No winners here.

[AvitarX]

I think you can run a script, as long as it compiles from source and says the license.

You can't distribute as a binary compiled against the kernel.

Re:No winners here.

[mysidia]

I'm not sure what anyone would stand to gain by arguing that Ubuntu is violating the GPLv2 license. It seems kind of ridiculous.

I think the argument against it is a political one, not a legal question.
Ubuntu having reviewed it by lawyers, I would feel comfortable that what they are doing is on the up-and-up.

The 'Software Freedom Conservancy (SFS)' is an activist organization.

They have an ideological stance against that which is non-GPL, even if what is being done is legal, they don't like what Ubuntu is doing, so they have their lawyers come up with a biased perspective.

Re:No winners here.

[someone1234]

Nope, haven't compiled an nvidia driver for 1-2 years. Getting the compiled binary for quite some time.

Re:No winners here.

[Anonymous Coward]

Well, sure. The nvidia kernel module is open source though and under a compatible license. Specifically to allow this. The userspace library that talks to it, though, is very very closed.

Re:No winners here.

[lkcl]

I'm not sure what anyone would stand to gain by arguing that Ubuntu is violating the GPLv2 license. It seems kind of ridiculous.

large companies - whether they are software libre companies like redhat or canonical, or proprietary companies like HP or Samsung - are supposed to be the "shining example" of what laws individuals are expected to follow. when those large companies are also making their money out of software libre, it is even *more* important that they set a good example.

otherwise what happens is that smaller companies - and chinese and other asian companies across the world - go "oh look: canonical's lost its rights to distribute the linux kernel due to a Copyright violation, and nobody did anything, therefore we don't have to worry about it either. let's continue to blatantly break the law, dishonour the spirit of software libre and make as much money as we can by spongeing off of other people's expertise".

for example: i'm one of the people who is actively involved behind the scenes in convincing allwinner to comply with the GPL. how the fuck am i going to do that - how the fuck am i going to convince allwinner that they've done something wrong by distributing source that does not come with the proper GPL headers - when even fuckwit companies like Canonical blatantly break the law?

if you have a good answer to that, please let everybody know.

Re:No winners here.

[I4ko]

And it is useless as you can't set your partitions to ZFS unless there is a complete development environment in the installer so it drops the kernel module at the installation time, so your system can boot. Oh, wait, it can't do that, since if you want to format your partitions as ZFS the installer kernel needs to be able to write to them, and that won't work, unless you can load the module dynamically in the installer kernel right after it is compiled. It is bad enough that Ubuntu mangled the installation options so badly that to tell that I want my partitions as XFS inside LVM I need to jump though 10 dropdowns and windows, and now this. IANAL, but I do agree with Ubuntu's interpretation and not with Conservancy's interpretation.

Re:No winners here.

[EndlessNameless]

The article explains why it is not ridiculous.

Everybody in the free software community benefits from strong GPL enforcement. E.g., when Microsoft reused GPL code in one of their tools, they were forced to release their source when it was discovered.

The ZFS license forbids releasing it under GPL. The Ubuntu binary distribution must be all GPL in order to satisfy the GPL's requirements.

If Ubuntu starts picking which parts of the license it follows, you can bet everyone from minor devs to major corporations will start doing the same thing. Proprietary shops will absorb open source contributions in whatever predatory manner they can.

The way to prevent it is simple. Enforce 100% of the requirements in the open source license 100% of the time, and set legal precedent whenever possible to establish those obligations as legally binding.

Re:No winners here.

[Znork]

Making ZFS incompatible with linux was the whole point of putting it under the CDDL from the start.

That said, having run ZFS since pretty much the start on Solaris servers here, it has to be the most overhyped piece of software ever released. Initially it was pretty much unusable for things like database loads, it was unstable as hell and had serious memory usage issues. These days, the glaring problems are largely fixed, but in an enterprise environment most of its features are of limited use as most of the storage will be on centralized SAN/NAS arrays anyway.

The whole discussion is one of those that gives me flashbacks to the 90's, same as when some database guy specifies that they want their volumes on this many striped spindles...

Re:No winners here.

[Anonymous Coward]

Who gains is irrelevant if they are violating.
If you let them violate, then you also have grounds for other companies that you don't like as much, to violate too

Re:No winners here.

[epine]

Having the user compile them is just another means of distributing a binary.

You're the reason the rest of us have to constantly write IANAL as if that's the only thing that makes it permissible to so much as clear your throat, when all that really needs to be clear is that we're capable of distinguishing A from B to any necessary degree.

"Having people cook their own pseudoephedrine is just another means of distributing methamphetamine."

Actually, no. What you actually distribute, and the end result of what people are perhaps likely to do with it are two entirely different things. This is not Hair Splitting 101 with a laser scalpel. This is retiring your flexible plastic picnic knife. There exist, of course, various ways to spin language to link these things together, if that is indeed desired. But then we should be analyzing that extra language, not wishing it into existence by waving our hands and muttering "kids there days".

Digression: Lawyers might make better guesses about the "necessary degree", but in practice lawyers are about as useful as political pundits empanelled on ISIS Today in speculating about the future will of the court. Maybe we should just all write CTLNKS instead (concerning the law, nobody knows shit). The main difference between critical thought at large and your lawyer is mainly that your lawyer can get into heaps of trouble by giving you a failing grade of shit. Pretty much the only thing I really want from intelligent agents of xmas future is to see the profession of law refactored, with all of their muddles and uncertainties hopelessly exposed to eternal ridicule, until we actually find in necessary to draft legislation to a sane standard (including an automated regression test suite, an unintended consequence verification gauntlet, and some Valgrind fuzz testing).

Back to the matter at hand, I completely disagree that this analysis should start with the semantics of static/dynamic linkage.

The real question should be this: if zfs.ko is erased from your system, does what remains function 99.99% the same way, minus only those features that zfs.ko provides?

With static linkage you can't do this, and thus you can't ask this question, either.

The key issue here is separability. What natural line exists between dynamic linkage and a ZeroMQ socket layer? And what about Joyent's "double hull" virtualization where ZFS is native to the underlying OS which provides a Linux API to the client OS? "Linking" is just indirection with more established history. And we all know you can solve every problem in computer science with another layer of indirection, except for too many layers of indirection.

I think my "separability" question above is far more likely to be legally productive, but then IANAL YMMV & CTLNKS.

Re:No winners here.

[Anonymous Coward]

This doesn't "get around" anything. Having the user compile them is just another means of distributing a binary.

No, the binary is being created by the user, it is not distributed. Source code is a program or library in human-readable form that can be fed into a compiler which will produce the binary, which is the machine readable code interpretation of the source code. So distributing source code is not "another means of distributing a binary", they are different things which is why the GPL makes an explicit distinction between distributing source code and binary.

Re:No winners here.

[exomondo]

Getting the user to create the non-free binary that you cannot distribute yourself does indeed undermine software freedom. If the non-free binary weren't a problem wrt software freedom then the vendor could just distribute it and make the free bits available in source form as restrictive OSS licenses mandate but the license does not allow this method when the result is a combination of free and non-free bits so they get around it by having the user connect the free and non-free bits.

In both cases the actual result is the same: user has access to the free bits in source form with the non-free bits in binary form and has a resultant non-free binary combination of both bits.

Re:No winners here.

[exomondo]

So what does anybody gain by saying you can combine the free bits and non-free bits into a non-free binary but you cannot give that to anybody else even if you give them the free bits in source form the same as you recieved them? You both have the same resultant non-free binary and the same source code and non-free component(s) from which it was produced so what is gained by making each user combine them themselves rather than just doing it once and distributing the result along with everything else?

Re:No winners here.

[Tough Love]

No one is combining Linux + ZFS into a single binary. Ubuntu has already said, over and over, that ZFS will be a separate kernel module (zfs.ko)

Nonsense. Loading a module ("linking it") into the kernel makes a single binary, that is unambiguous.

Re:No winners here.

[Timex]

Nonsense. Loading a module ("linking it") into the kernel makes a single binary, that is unambiguous.

Bull.

Assuming one is not actively using ZFS and removed all traces of the library and its tools, will the system work the same as before?

If so (and this is readily apparent), then the library is separate. If it is under a separate license and this is understood by the user/admin at the time of install, who cares? I don't.

I'm pretty sure that if you look at the several other file systems that have been supported by Linux over the years, more than one have had different license terms.

Re:No winners here.

[dbIII]

There is a part compiled on installation that is just a wrapper to make sure the binary hooks in properly to whatever kernel you have - a translation layer really. Some distros do that for you for their standard kernels.

Re:No winners here.

[dbIII]

It also potentially dilutes the authority of the FSF and some leading members of that like to imply they own linux - even going as far as putting "gnu" in front of the name of something that is not a gnu or FSF project.
Having another non-GPL thing in the mix, after X, apache, nvidia and so on annoys them.
It would be a non-story without such agitation.

Re:No winners here.

[dbIII]

it has to be the most overhyped piece of software ever released

I thought too until I starting using it.

most of the storage will be on centralized SAN/NAS arrays

Many of which run - guess what?

Re:No winners here.

[shutdown -p now]

FSF has been claiming that (with DLLs and Java .class files), but it was never actually tested in court.

Re:No winners here.

[Tough Love]

Nonsense. Loading a module ("linking it") into the kernel makes a single binary, that is unambiguous.

Bull.

Assuming one is not actively using ZFS and removed all traces of the library and its tools, will the system work the same as before?

If so (and this is readily apparent), then the library is separate. If it is under a separate license and this is understood by the user/admin at the time of install, who cares? I don't.

I'm pretty sure that if you look at the several other file systems that have been supported by Linux over the years, more than one have had different license terms.

Wow, I'm just floored by the lack of understanding I see here. Look, a Linux kernel module is just a chunk of memory saved on disk along with some information needed to fix up the bytes that need to change in order to run it at some particular place in memory. To load it, Linux just copies the contents of the module from and does the necessary fixups. There is no question whatsoever that the result of that is a single binary. That's just basic computer science. Very very basic. Very very very basic.

As for whether you care or not, that's entirely up to you. But if you plan to make a career of distributing software in violation of license agreements, good luck to you, you'll need it.

And as for whether all the other filesystems distributed as part of Linux comply with the GPL, yes, every single one of them does. Either by being licensed explicitly under the GPL, or licensed under a GPL-compatible license.

Re: No winners here.

[Entrope]

The FSF usually talks about software freedom as the ability to modify the source code for the software you use and share it with others, with no real restrictions except that you cannot forbid others from doing the same. In that sense, how is "user compiles a binary to combine with another binary, but incompatible licenses mean the combined binary cannot be distributed" a problem for free software?

The user can modify the source code for Linux or for zfs.ko. The user can share those modifications in source form. The user cannot change the licence for either piece of software, and cannot distribute the combined binary, but the person who gave them the software did not, and could not, do those things either. I'm not seeing a software freedom problem here. (And I'm setting aside for now any licensing issues for the user-compiles-it approach: they are at most incidental to the question of software freedoms.)

Re: No winners here.

[Entrope]

So what if loading a kernel module makes a single binary? Nobody is distributing the combined binary.

Re: No winners here.

[exomondo]

Sorry yes that first line was unclear, I see no difference between these things. What I meant was that it is the view of the SFLC that this undermines software freedom, they obviously see it as a problem despite the fact that there doesn't seem to be a difference between the two scenarios outside of who actually compiles it.

Re:No winners here.

[arglebargle_xiv]

That was my feeling too. Microsoft is giving the Linux community a huge boost with their consumer-hostile Windows 10, and instead of taking advantage of it they're busy shooting each other in the foot over irrelevancies so trivial that you can't even explain them to a normal person (it came up in a discussion last night, one non-geek eventually summed it up as "this is why we think computer people are all a bit crazy").

Re: No winners here.

[MemeRot]

Nothing is gained, but copyright law is not about maximizing public benefit.

Re:No winners here.

[Kjella]

I'm not sure if it's relevant but from what I understand it's a little bit different, nVidia has a small GPL-licensed patch to the kernel which essentially gives them raw kernel memory access (read/write/allocate/free) and then puppeteers the kernel from user space. So the nVidia proprietary code doesn't live in the same memory space like a kernel module would, they just have a lot of very close inter-process communication.

Re: No winners here.

[exomondo]

But copyright law is about protecting the interests of at least one party involved - even if you don't agree with it or with their point of view - so what is the interest of the concerned party in this case?

Re:No winners here.

[Grishnakh]

Nope, not at all.

The Nvidia driver (nvidia.ko) lives in kernel space, and links to the kernel. It also links to a small GPL-licensed "shim" driver, because the Nvidia driver is a proprietary blob (and likely very close to their Windows driver architecturally), and the kernel changes frequently. So that they don't have to maintain a different proprietary blob version for every single kernel version out there (which would be a daunting task, given all the kernel versions and all the different subversions and variants made by all the distros), they have a single blob, and then the shim gets compiled for the kernel on the target system, and links the two.

Inter-process communication would be a disaster for something very high-performance like a 3D video driver.

Re:No winners here.

[Anonymous Coward]

These days, the glaring problems are largely fixed, but in an enterprise environment most of its features are of limited use as most of the storage will be on centralized SAN/NAS arrays anyway.

That's bull, there's things like ridiculously simple quotas, snapshots, advanced ACLs that all benefit any environment, and ARC is more efficient than MRU alone.

Boot environments. Snap your root filesystem, update it offline, boot to it, boot back to the previous environment, copy to a different server for troubleshooting. Who doesn't want that tool in their bag?

By "most features" you're just counting what, RAID, compression, deduplication?

Re:No winners here.

[s4m7]

This is no different from how Nvidia drivers are distributed, and Nvidia's drivers are completely proprietary.

Almost completely incorrect.

nvidia proprietary drivers (at least in Ubuntu) come as a DKMS package. The GPL source portion which interacts with kernel ABI is compiled at installation time, or on the installation of a new kernel. This in turn calls on the BLOB which is plainly not at all open source. This has been a sticking point for some time, as it is clearly nvidia getting around the rules in a sneaky way. Linus is not a fan.

Relevant to the current situation, because in 15.10, ZFS is available as a DKMS package. it compiles on the user system at installation and that is a difference from distributing a precompiiled binary that is dynamically linked against the specific binary signature of the distributed kernel binaries.

Re:No winners here.

[Grishnakh]

But if you plan to make a career of distributing software in violation of license agreements, good luck to you, you'll need it.

Nvidia has been distributing their proprietary drivers this way for well over a decade now.

You obviously have no idea what you're talking about. Computer science is not relevant here, only law, and the law is that people who have standing are the only ones who get to sue over license agreements. No one has bothered with Nvidia yet, nor AMD.

Re:No winners here.

[Tough Love]

Nvidia has been distributing their proprietary drivers this way for well over a decade now.

And this is what they got for that.

Law or no law, facts are facts. I hope you don't try to defend further your indefensible position that facts are irrelevant when technological matters become legal matters.

Re: No winners here.

[Tough Love]

So what if loading a kernel module makes a single binary? Nobody is distributing the combined binary.

Somebody up above said "no one is combining Linux + ZFS into a single binary", which is just stupid because somebody is obviously doing exactly that, as I pointed out. I did not comment on the actual issue, but I will now. The issue, if you've being paying attention, is that Canonical is distributing a derived work of the Linux kernel that they do not have the right to distribute. Don't argue with me about that, I'm just the messenger. If you have your own opinions about what constitutes a derived work or what rights Canonical actually has in this matter, then RTFA.

Then maybe go read this page. Here's an excerpt: "U.S.D.J. Scheindlin found Westinghouse’s infringement to be willful and therefore awarded treble statutory damages of $90,000. The court also entered a permanent injunction prohibiting distribution of HDTV products with the BusyBox software and further ordered all infringing HDTVs to be forfeited to the plaintiff"

So far, the GPL has stood up in court every time it has been tested, which by now is quite a few times all around the world, including the good old US of A. Most violators aren't dumb enough to take it to court, they either get into compliance sensibly or settle early.

Re:No winners here.

[Tough Love]

You seem to be suffering from a case of zero research.

Re:No winners here.

[dgatwood]

Wow, I'm just floored by the lack of understanding I see here. Look, a Linux kernel module is just a chunk of memory saved on disk along with some information needed to fix up the bytes that need to change in order to run it at some particular place in memory. To load it, Linux just copies the contents of the module from and does the necessary fixups. There is no question whatsoever that the result of that is a single binary. That's just basic computer science. Very very basic. Very very very basic.

Technically, yes, but legally speaking, that's not actually the case. When we talk about a single binary, we're talking about a single binary in non-ephemeral form. At least in the U.S., copyright law[1] grants a very broad exemption for ephemeral copies of software that exist solely in memory, and solely for the purpose of executing code. Under U.S. law, executing a program is not legally considered making a copy, so that ephemeral copy is not a copy, which means that the single binary does not exist, for legal purposes.

As a result, merely linking a kernel module does not cause it to be a derivative work of the kernel. More to the point, a kernel module is not a derivative work unless it contains some significant portion of kernel source code above and beyond any data types and function names contained in the headers. Galoob v. Nintendo made this pretty clear.

For further reading, I would suggest this article from the software pluralism site on University of Washington School of Law's website. After you read it, I think you'll come to the same conclusion—that a filesystem that exists on other platforms cannot possibly, by any stretch of the imagination, be considered a derivative work of the Linux kernel, and thus cannot be tainted by the kernel's GPL license in any way, shape, or form. As a result, the Software Freedom Conservancy's conclusions are likely in error.

[1] Title 17, Chapter 1, Section 117a(1)

Re:No winners here.

[TheRaven64]

Inter-process communication would be a disaster for something very high-performance like a 3D video driver

For the nVidia driver (for cards from the last 5 or so years), that also includes communication between userspace and the kernel. Nothing in the kernel driver is on the critical path for performance. The kernel driver is responsible for mapping device rings into userspace and setting up memory mappings. Everything else goes directly from userspace to the hardware. The userspace driver writes data to shared buffers and sends commands into the memory-mapped ring buffer, then pokes a couple of memory mapped device registers to tell the card to read them commands. For Tesla, the ring buffer design was completely batshit insane, but apparently it's improved since then.

Re:No winners here.

[Anonymous Coward]

No one is combining Linux + ZFS into a single binary. Ubuntu has already said, over and over, that ZFS will be a separate kernel module (zfs.ko).

That would be allowed by the LGPL, but we are discussing the GPL.

When you have two variants of a license, with the difference being that one allows dynamic linking, it is pretty clear that the intent is that the other does not.

This is no different from how Nvidia drivers are distributed

The NVidia drivers get around this by being the Windows driver[1] compiled for Linux. It can't be loaded by the Linux kernel. In between sits a translation layer, which is GPL+exception for linking with the NVidia driver, with the NVidia driver license also allowing for linking with the translation layer. NVidia can do this because they own the copyright for the driver.

Ubuntu does not own ZFS, and cannot change the license. ZFS is under the CDDL license which does not allow linking with GPL software, and Linux is under the GPL which does not allow linking with CDDL. Thus a translation layer would be required to be under (GPL and not CDDL) and (CDDL and not GPL). Which is clearly impossible.

This could be changed by Oracle changing the ZFS license, but they have no interest in that.

TL;DR: nVidia is willing to jump through hoops to get the driver legal. With ZFS, both licenses say no.

[1] It used to contain the string "HKEY_LOCAL_MACHINE", but I'd expect they'd got that part optimized away nowadays.

Re:No winners here.

[BadDreamer]

The concept of "single binary" is misleading. Computer science doesn't even include formal definitions for that concept, so it is not at all something basic. And frankly, I don't even understand what you mean by that in the context of distribution. I suspect that neither do you.

What this is about is distribution of various executable files. At the core is the question if one executable (zfs.ko) is derived from or to be considered a part of another executable (the kernel) *during distribution*. What happens at runtime is immaterial.

Re: No winners here.

[Entrope]

I'm not responsible for what someone else said in this thread, so don't expect me to defend random other people.

Which derivative work of the kernel do you think Canonical is distributing without authorization? I'm not going to spend an hour knocking down theories where you might respond "but that's not what I meant!"

The cases you linked to are cases where a GPLed work was distributed in direct violation of the GPL (e.g. without source, or purporting to only be usable for "personal, non-commercial" use). That is easily distinguished from what you are suggesting here.

Re:No winners here.

[segedunum]

That's nVidia's attitude rather than how their drivers work. I can't see what you're trying to argue here.

Re:No winners here.

[Anonymous Coward]

It also potentially dilutes the authority of the FSF and some leading members of that like to imply they own linux - even going as far as putting "gnu" in front of the name of something that is not a gnu or FSF project.

You didn't understand a word of what the FSF has been saying. That slash was never supposed to mean "GNU Linux". It was always to be understood as "GNU + Linux". As in "GCC, etc + Linux".

It used to seem pointless (except from a marketing point of view), but nowadays you have at least Android/Linux, Busybox/Linux and GNU/Linux, in addition to GNU/Solaris, GNU/FreeBSD, etc.

Linux has a much larger market share than GNU/Linux, and logically, "Linux" alone should be understood as either all of them or the largest one - which is by far Android/Linux.

Re: No winners here.

[Anonymous Coward]

The interest of Sun (creators of ZFS, now owned by Oracle) was to sell Solaris, which was at the time being out-competed by Linux.

The interest of the software freedom conservancy is to get Oracle to change the license of ZFS to make it legal for Linux use, AND to show that the GPL is a well-defended license, rather than a license nobody cares about people breaking.

Re: No winners here.

[Anonymous Coward]

That's funny, I thought it was.

Re:No winners here.

[edtice1559]

It is very important to understand that distributing source and having the user compile it is another form of distributing a binary. Attempts have been made in the past to violate the GPL but creating tools to compile the source, delete it, and then claim that this is allowed because no binary is being distributed. A lot of effort went into crafting the GPL in order to ensure certain freedoms. If it could be easily circumvented, those freedoms would be lost. Sometimes compliance is tedious (such as in this case) for commercial entities but it's a small price to pay for what we get. None of this affects individual users since we aren't commercial distributors.

Re: No winners here.

[edtice1559]

The issue here is not the *user* doing at. As you've said that's fine. The issue is that Canonical is doing it and distributing it. Some argue that they shouldn't be doing that. The argument being made is that Canonical can't distribute their binaries. Most of us think that they *can* distribute the binaries as we consider their Linux distribution to be a collection of individual works not a work that is a derivate of all of them. I can combine GPL with non-GPL and do certain things with it as you have said. GPLv3 makes it even clearer that I could, for internal use at a company, do this as long as I keep it within the company. But in this case Canonical is releasing a product. I think that they are in the clear but I still think there's value in understanding the argument against it.

Re:No winners here.

[Znork]

Quotas we barely ever use anymore, to the extent they exist they tend to be integrated into applications if there's a point to them. Wasting employee time is extremely expensive compared to disk. Most systems support snapshots on multiple levels already, from OS/LVM and virtualization layers down to the SAN/NAS. ACL's, after 30 years, managing probably about 10k unix systems, I have run across a handful of situations where it would have been useful, and exactly zero cases where the cost benefit ratio would have made it economically viable. Most modern filesystems support them, but for applications that need that access granularity, the functionality tends to end up in application or database layers. Minor cache improvements pale in comparison to simply throwing the entire performance demanding application on flash-only, FusionIO or NVME disks.

Boot environments like that have been done in various ways for as long as I can remember, where the earliest were basically the diskless NFS based clients where you could simply copy the filesystem and run the update on that. After that, any simple disk mirror could be split off and cloned for a snapshot system to work with. Thankfully, such functionality is approaching irrelevance as well, as application design is growing up enough to actually build redundancy into the application layer so you can take any number of servers offline at any time. Snapping a root filesystem isn't exactly necessary when the application is built to live on a server instance that will disappear and be replaced with a fresh disposable instantiated image on next reboot...

And yes, most features mean pretty much raid, compression, caching, deduplication, snapshots, etc.

It's not that it's a bad filesystem, it's a quite good one. But the problems it solves are becoming legacy issues of less relevance in an industry where the discussion is shifting toward whether there will be an OS as we know it underpinning the application infrastructure at all.

Re:No winners here.

[Anonymous Coward]

It is very important to understand that distributing source and having the user compile it is another form of distributing a binary.

I don't know how much more wrong you can get. I'm baffled.

Distributing source is in no way equivalent to distributing a compiled binary, either legally or technically.

Re:No winners here.

[david_thornley]

The FSF does not own Linux, and does not pretend to. They like calling it "Gnu/Linux" because Linux fits so nicely into the Gnu project, and because the Gnu software used is essential to making the system useful. (Other userlands exist, of course, but the FSF isn't going to use "Gnu" to refer to any of them.) The Linux kernel is useless by itself, and is useful when the Gnu stuff is added on, so there's reason to consider both Gnu and Linux to be the vital parts of the system.

The FSF also feels very strongly about the GPL licenses and their enforcement. They have no problems with anyone adding stuff with GPL-compatible licenses to the code, and they took specific action in GPLv3 to make it compatible with the Apache license. They have problems with people violating the GPL, since that weakens the GPL in general.

Re:No winners here.

[Tough Love]

a kernel module is not a derivative work unless it contains some significant portion of kernel source code above and beyond any data types and function names contained in the headers. Galoob v. Nintendo made this pretty clear

Whaaat??? Galoob v. Nintendo did nothing of the sort, the court found that a "product that allowed users to alter codes transmitted between video gaming console and game cartridge did not infringe console manufacturer’s exclusive right, under federal copyright law, to create derivative works". Your interpretation is wildly creative.

A central point that will be argued is whether a work that relies substantially on the interfaces of the Linux kernel is a derivative work. Though I haven't researched it deeply, my impression is, that's a pretty solid yes.

I would suggest this article from the software pluralism site on University of Washington School of Law's website. After you read it, I think you'll come to the same conclusion—that a filesystem that exists on other platforms cannot possibly, by any stretch of the imagination, be considered a derivative work of the Linux kernel, and thus cannot be tainted by the kernel's GPL license in any way, shape, or form. As a result, the Software Freedom Conservancy's conclusions are likely in error.

The arguments there did not persuade me, and neither did your presumptive conclusion. You're clearly aware of the importance of the derivation question, but imho, your interpretation is incorrect and likely to fail. It's enough for only part of the work to be derivative to constitute a violation.

Your link provides little in the way of case law and a lot in the way of speculation. You might consider doing a little more research yourself, perhaps focusing on the GPL paragraph 5 observation that "nothing else grants you permission to modify or distribute the Program or its derivative works". Canonical proposes to distribute the Zfs binary and Linux binary together. Their argument that the GPL v2 grants them the right to do this is very leaky indeed.

In any case, settling the question will be a source of endless entertainment. After all this work, we deserve some entertainment, don't we?

[1] Title 17, Chapter 1, Section 117a(1)

Eek, a straw man that conflates copying with derivation. Let's just forget about that.

Re:No winners here.

[dublin]

Pretty much everything the FSF claims has never been tested in court. That's half the problem...

Re:No winners here.

[shutdown -p now]

FSF GPL FAQ:

"Linking a GPL covered work statically or dynamically with other modules is making a combined work based on the GPL covered work. Thus, the terms and conditions of the GNU General Public License cover the whole combination. ...

A consequence is that if you choose to use GPL'd Perl modules or Java classes in your program, you must release the program in a GPL-compatible way, regardless of the license used in the Perl or Java interpreter that the combined Perl or Java program will run on. ...

If the program dynamically links plug-ins, and they make function calls to each other and share data structures, we believe they form a single program, which must be treated as an extension of both the main program and the plug-ins. This means you must license the plug-in under the GPL or a GPL-compatible free software license and distribute it with source code in a GPL-compliant way."

Re: No winners here.

[Entrope]

The problem is that the SFC's position seems intentionally murky. They claim that Canonical's (planned) distribution of Linux with ZFS is a single work, and imply -- but as far as I could tell, do not specifically say or support -- that the combination is more than a "mere aggregation" (which is perfectly okay under the GPLv2). The problem for SFC is that the Ubuntu ZFS functionality is essentially the same whether zfs.ko comes in the same ISO as the Linux kernel or not: the installer could easily download zfs.ko, load it from another medium, or compile it from source, and the behavior would be the same. They don't provide a clear explanation for why it goes beyond "mere aggregation", and I don't think they can.

Re:No winners here.

[dgatwood]

Whaaat??? Galoob v. Nintendo did nothing of the sort, the court found that a "product that allowed users to alter codes transmitted between video gaming console and game cartridge did not infringe console manufacturer’s exclusive right, under federal copyright law, to create derivative works". Your interpretation is wildly creative.

Because it held that the manufacturer didn't create the derivative work. One reason for this was that the manufacturer did not use any of the other manufacturer's code or other creative work in their own work. The other reason was that the customer did the actual creation by the act of running the cartridge through the Game Genie. Those exact same two arguments preclude kernel modules from being considered derivative works unless they borrow Linux kernel code.

Your link provides little in the way of case law and a lot in the way of speculation. You might consider doing a little more research yourself, perhaps focusing on the GPL paragraph 5 observation that "nothing else grants you permission to modify or distribute the Program or its derivative works". Canonical proposes to distribute the Zfs binary and Linux binary together. Their argument that the GPL v2 grants them the right to do this is very leaky indeed.

The GPL also says that mere aggregation does not trigger those provisions. Canonical proposes to distribute the ZFS binary and the Linux binary separately as a loadable module and a kernel binary. That falls squarely into the "mere aggregation" category.

[1] Title 17, Chapter 1, Section 117a(1) [cornell.edu]

Eek, a straw man that conflates copying with derivation. Let's just forget about that.

That's not a straw man. Prosecution for creating a derivative work cannot happen unless the work is in "concrete or permanent form". That bit of copyright law means that the mere act of loading a piece of software does not legally create a copy in concrete or permanent form. Therefore, the result cannot be an infringing derivative work. My opinion in the matter is further supported by parts of the opinion in Galoob and was later affirmed by the 9th circuit in Micro Star v. FormGen Inc.. So it is very unlikely that a loadable kernel module could ever be considered a derivative work unless it incorporates actual code borrowed from the kernel.

Further, this particular LKM is merely a port to the Linux kernel interfaces of an existing filesystem that is available on multiple platforms. That means it is in effect a standalone work whose connection to the Linux kernel is purely supportive. Per Micro Star v. FormGen Inc., that also means that it cannot be considered a derivative work.

So before the Linux ZFS port could be considered a derivative work of the Linux kernel, you'd pretty much have to throw away all the existing case law in this area and start over. I'm not saying it can't happen, but it's about as likely as Obama winning reelection for a third term as POTUS.

Re: No winners here.

[cthulhu11]

Maybe that's why the filesystem complement is so pitiful. No bundled compressing filesystem? In 2016? Seriously? Part of the beauty of ZFS is using it for a boot volume, Live Upgrade etc. Which are much harder when it isn't bundled. Let's get real here: in the end what it's about is Stallman's mental illness.

Re:No winners here.

[dbIII]

If you've come in late suggest you follow it from when the suggestion was made to call it LiGnuX and you'll get a clearer idea of what was behind the suggested renaming. Don't take it from me, use google and track down those FSF newsletters.

Re:No winners here.

[Anonymous Coward]

The FSF does not own Linux, and does not pretend to. They like calling it "Gnu/Linux" because Linux fits so nicely into the Gnu project, and because the Gnu software used is essential to making the system useful.

Actually they like calling it "GNU/Linux" because their position is that GNU is the operating system, Linux is just the kernel.

That's true, of course, but it ignores the fact that the kernel, while a small part of the overall OS, is the most important part by far.

Re: No winners here.

[exomondo]

The interest of the software freedom conservancy is to get Oracle to change the license of ZFS to make it legal for Linux use

It already is legal for Linux use, anybody can use ZFS with Linux perfectly legally.

AND to show that the GPL is a well-defended license, rather than a license nobody cares about people breaking.

In this case they are talking about a completely pointless clause, the end result is the same the only difference is you force the user to put the free and non-free bits together, this is nothting more than the GPL making it harder for the user with absolutely zero benefit whatsoever.

Re:No winners here.

[Anonymous Coward]

And this is what they got for that.

An obscene gesture, wow.

Law or no law, facts are facts. I hope you don't try to defend further your indefensible position that facts are irrelevant when technological matters become legal matters.

It isnt about being a single binary, it is about being a derived work or an aggregate work. In this case the argument is that it constitutes a derived work, now that may be true but even if it is there is nothing to stop the two parts being distributed separately and then combined by the user nor is there anything you or anybody else can do to stop the user from doing that. So this is just a question of how it is distributed, which seems rather pointless given the end result is the same whether you put the two pieces together and give it to the user or whether you give the two pieces to the user and make them put them together.

Re:No winners here.

[Anonymous Coward]

As usual, it has less to do with the specific example as it has to do with precedents that may not be desirable.

It only becomes a precedent if somebody sues. Until then, it is just another opinion.

I doubt any copyright holders on the GPL side of things here is silly enough to try setting a precedent here. The other player is of course Oracle, but I haven't heard anyone claiming that *their* licence is being violated. I doubt it matters if the legal opinion was given by an expert or a muppet, or if the whole things is a bluff. Canonical put their balls on the table and dared people to cut them off.

Re:No winners here.

[Anonymous Coward]

Not quite right.

Distributing source is perfectly fine. But once compiled the zfs.ko is a derived work of the linux kernel according to the Software Freedom Conservancy
  and therefore the GPL applies to it too.

Re:No winners here.

[Anonymous Coward]

A central point that will be argued is whether a work that relies substantially on the interfaces of the Linux kernel is a derivative work. Though I haven't researched it deeply, my impression is, that's a pretty solid yes.

That would mean that pretty much any software that operates directly with the kernel would constitute a derivative work which is exactly what the preamble in the kernel COPYING file serves to prevent.

NOTE! This copyright does *not* cover user programs that use kernel services by normal system calls - this is merely considered normal use of the kernel, and does *not* fall under the heading of "derived work".

It is very clear.

Re:No winners here.

[Anonymous Coward]

Like ESRI?

Re:No winners here.

[Anonymous Coward]

They're clear about what they want to cover, but I don't see any technical difference between a user program and a plugin module when it comes to kernel services.

Hell, most programs since the 1960s use relocatable code - which was kind of novel back then - and require linking during the load phase (thus the term "linker/loader").

Just because FSF chooses to single out "normal system calls" from modular interfaces doesn't mean they're right.

bunch of hot air - use FreeBSD then?

[ecloud]

I happen to like using ZFS.

But fine, distributing drivers with binary blobs is OK, while this little license incompatibility between two open-source projects is a big deal. Whatever, dudes.

Re: bunch of hot air - use FreeBSD then?

[Billly Gates]

Thanks to Oracle's ruling on Java a clean room implementation of ZFS is still owned by Oracle as the API and keywords are owned.

No distro outside Solaris is safe and could open your employer to liability from an Oracle audit

Re: bunch of hot air - use FreeBSD then?

[mysidia]

clean room implementation of ZFS is still owned by Oracle as the API and keywords are owned.

You can use new keywords. Search and replace if you want.

It will still be a filesystem if you use different keywords in the implementation..... it's not like Java, where all the keywords in the code have special significance, because they are meant to be invoked directly from other programs.

Re: bunch of hot air - use FreeBSD then?

[brambus]

No distro outside Solaris is safe and could open your employer to liability from an Oracle audit

I see this mistake repeated over and over in every licensing discussion on Linux and ZFS. The CDDL is not the license crying foul here, so Oracle has no standing to sue. It's the GPL that is trying to infect the CDDL'd code and the CDDL won't let it, hence the incompatibility. IOW, if anybody wanted to file suit, it'd have to come from the Linux side, not the Oracle side. I'll let you imagine the media fallout over anybody who would attempt to sue Canonical for including a piece of open source software in their distro that said zealot doesn't like (remember, the CDDL is the freer of the two).

Re: bunch of hot air - use FreeBSD then?

[Anonymous Coward]

If no distro outside of Solaris is safe, how do the BSD's get away with it (FreeBSD and NetBSD have ZFS kernel modules)?

Software Freedom?

[KermodeBear]

So, Software Freedom means that you can't do what you want with your software? Is this one of those Richard Stalin - I mean, Stallman - groups or something?

Re:Software Freedom?

[slashping]

There's no such thing as unlimited freedom. You have to pick the freedoms you care about, and then forbid those that would violate it.

Re:Software Freedom?

[gstoddart]

You're free to be free as long as your freedom matches what you've been told you're allowed to do to be free.

Some software freedom limits what you can do. Other software freedom doesn't give a crap.

Re:Software Freedom?

[Anonymous Coward]

You can do whatever you want to do with your own software. Its when you distribute that software and hand it out you must abide by the copyrights.

Re:Software Freedom?

[edtice1559]

Software freedom means that you can't create derivative works and then distribute them with a license that restricts other people's freedom. It's the same reason that my freedom to swing my fist ends at your nose. You can do anything you want with the software and combine it with proprietary software even. You just can't *distribute* the changes.

Re:Software Freedom?

[Marginal Coward]

Let's not give Mr. Stallman too much credit, here. I believe George Orwell thought of it first:

“Doublethink means the power of holding two contradictory beliefs in one's mind simultaneously, and accepting both of them.”

For example, it's a pretty short trip from Orwell's "Freedom is Slavery" to Stallman's "Freedom means not using proprietary software." For many more such gems, search for the word "freedom" on his Wikiquote page.

Re:Software Freedom?

[Anonymous Coward]

It's the same reason that my freedom to swing my fist ends at your nose.

Even if you only swung close to my nose, I could probably still have to arrested for assault.

Re:Software Freedom?

[Marginal Coward]

Which reminds me: here's something from "Life of Brian" that stuck in my mind when I saw it the other day:

[the members of "The People's Front of Judea" are sitting in the amphitheatre; Stan has just announced that he wants to be a woman and wants to be called "Loretta," and is explaining why]

        Stan: I want to have babies.
        Reg: You want to have babies?!?!
        Stan: It's every man's right to have babies if he wants them.
        Reg: But ... you can't HAVE babies!
        Stan: Don't you oppress me!
        Reg: I'm not oppressing you, Stan. You haven't got a womb! Where's the foetus gonna gestate? You gonna keep it in a box?

So, I don't want to use proprietary software, I just want the right to use proprietary software, even if I can't use proprietary software. Waitaminute...I actually do that all the time. So it looks like the "Reg" in my own story hasn't actually succeeded in oppressing me...

Re:Software Freedom?

[drew_kime]

So, Software Freedom means that you can't do what you want with your software? Is this one of those Richard Stalin - I mean, Stallman - groups or something?

If you didn't write it, it's not your software. You can acquire the right to use it in exchange for your agreement to the terms of the license. The freedom part comes in where the terms of the license say that you get to modify and distribute it without anyone's permission. Again, in exchange for your agreement to the terms of the license.

Re:Software Freedom?

[GLMDesigns]

So you don't have freedom of speech? (Don't bring up the canard about "yelling fire in crowded movie theater.")

Canard because first it's "falsely" yelling fire in crowded movie theater. And second there is a difference between political speech and slander. Freedom of speech never included that you can slander anyone you want and be free from consequences. There isn't an internal contradiction in the previous statement because one must qualify what speech means.

Now If what you mean by "unlimited freedom" is that "freedom" means whatever pops into someone head then you're correct. Freedom of Religion means, in essence, freedom of conscience; that you can have quiet enjoyment of your beliefs. Quiet enjoyment does not include freedom to kill disbelievers for "insulting" the religion.

Re:Software Freedom?

[oh_my_080980980]

You douche bag. They're not distributing the changes. There aren't any fucking changes. All Canonical is doing is providing the ability for people to use the ZFS file system. They are not creating works off of ZFS and they are not making kernel changes for ZFS. Nothing is derivative. For fucks sake let it go morons. We users want to use the software. We don't fucking care to view the source code or recompile and jack off! Some do, that's fine but most do not and if you have a fucking problem take it up with Oracle. Canonical is giving the user the option of using ZFS. The user does not. No harm no follow.

For fuck's sake the fucking point about the GPL is that software is not included in a fundamental way they prevents you from using it. You now like a proprietary codec for trans-coding video where you have to pay a fee or what not. Something that prevents you from using the software. Including ZFS does not prevent you from using Linux. You can use other file systems. Sure it *might* be risky to use ZFS *if* Oracle does *something* but that's your risk and should be pointed out to you up front. But won't stop you fro using Linux.

Fucking idiots!

Re:Software Freedom?

[BarbaraHudson]

The only part of their blog post you should adhere to is "Do Not Rely On This Document As Legal Advice". Everything else is political posturing.

The Software Freedom Conservancy doesn't have standing, and if they tried, Oracle would probably kill them good!

Remember, morons like these want to include scripts that have NO linking at any stage as being "linked" when they call each other.

Dynamically loading and executing code in a separate module is not linking, even though we unfortunately refer to it as "dynamic linking." The code in a module is separate from the caller, written by others, and since either can be distributed without the other, they are obviously not "linked" - especially since you can change whether any module is loaded or not at run time if you so choose to include the necessary code. Including them in the same distro is no different than including gplv2 and bsd code in the same distro.

The kernel and zfs are are distributed in the same way that a can of beans and and apple are a "combined distribution" when you leave the grocery store. They're still separate entities, and a worm in the apple does not affect the beans in any way whatsoever - not even legally. To say that they need to be distributed separately is total bullsh*t.

By the same token, since the two are separate, even on an ISO of a distro (oh look - they are indeed separate files), they do not form either a derivative or combined work. The two works are still independent of each other.

The Software Freedom Conservancy is doing a great job of showing how freetards think.

Re:Software Freedom?

[slashping]

So you don't have freedom of speech?

I didn't say that. I didn't even mention speech. I said freedoms are limited, and the freedom of speech is no exception. There are limits to the things you can say.

Re:Software Freedom?

[armanox]

Who is making changes to anything?

Re:Software Freedom?

[TsuruchiBrian]

You can have freedom of speech, but that is mutually exclusive with the freedom to silence people. By granting freedom of speech, you are implicitly removing the freedom to silence people, and vice versa.

Re:Software Freedom?

[bluefoxlucid]

Basically, Richard Stallman's line is that the MIT license restricts your freedom as a software developer and a software user, and the GPL protects it. It's backwards double-think.

This is an example of protectionist politics: the SFC is afraid of the precedent by which a work of non-GPL status constructed in such a way would somehow not be forced to release under GPL. Then someone might find a way around gettext by linking against a stub library with the same ABI and then distributing with the actual gettext in a repo (the weak argument for GPL provisions about dynamic linking is ridiculous, and only survives by continuing to play on the ignorance of judges; when the judges realize you're laughing at their stupidity, they will be *very* unhappy with you).

Re:Software Freedom?

[TsuruchiBrian]

It's not like Stallman makes no sense. Imagine 2 societies. Society A in which you are free to enter into any contract want, and society B which is exactly like Society A except that contracts that involve slavery are not legal. Which society has more freedom?

Society A provides the additional freedom to to sell oneself into slavery, so society A is more free.

Society B lacks the freedom to to sell oneself into slavery, so society B is more free.

Which of these statements is true? It's just a pointless semantic debate that depends on your personal definition of a "free society". Is agreeing to use commercial software similar selling yourself into slavery? Not really, but I think it's still an appropriate analogy even if the magnitude of the consequences are not comparable.

Re:Software Freedom?

[Anonymous Coward]

You are free to do with the software as you see fit, as long as you do not distribute it.
That means you can alter the software and use it for commercial purposes or whatever, as long as you don't distribute it doesn't matter.
Once you start distributing you have to follow the rest of the license, which states that all the sources must be available and whatever else it is that makes ZFS incompatible.
The freedom in this case refers to the users freedom to obtain the sources etc.
Not the programmers freedom to do what he wants.

Re:Software Freedom?

[squiggleslash]

Yes, but just because you didn't say it doesn't mean the GP can't have a rant at you as if you mentioned speech and set up an entire straw man arguing with you about fires in theaters!

Re:Software Freedom?

[Anonymous Coward]

You seem to take issue with the idea of copyright in general. Copy "left" is an attempt to use the infrastructure created to support copyrights against itself. You should rail against proprietary software loudest, then "open" but not free licenses, etc., before finally coming down against GPL (leaving only MIT/BSD-ish licenses without your scorn).

Also, it's disingenuous to not include slander in speech: (a) it's not clear at the time whether or not it's slander, and (b) one might for the same reason not include "what would otherwise be speech, but is not because it violates the GPL." It's like excluding squares from rectangles.

Re:Software Freedom?

[squiggleslash]

Basically, Richard Stallman's line is that the MIT license restricts your freedom as a software developer and a software user, and the GPL protects it. It's backwards double-think.

It would be if he'd ever said anything of the sort, which he hasn't. His only objections to licenses like the MIT license is they open the potential for someone to produce a closed fork of your work, he has absolutely no moral objection to it, and the statement he thinks it restricts your freedom is 100% false.

Is he Stallman, or Strawman? Because half the stuff posted here about what he supposedly believes seems to fit the latter.

Re:Software Freedom?

[Marginal Coward]

If I may adapt the quote above from Mr. Orwell, I think you just said "Freedom is Licensing Restrictions."

I don't think licensing restrictions are necessary bad - let's just not adopt doublethink along the way. Then again, I fully respect your freedom to doublethink as much as you want. ;-)

Re:Software Freedom?

[Grishnakh]

Actually, you don't need to make up crazy analogies about worms and apples and beans here. Just look at how the Nvidia proprietary kernel modules are shipped, and how they've been shipped for over a decade now. Nvidia's driver also links to the kernel as a dynamically-loaded module (nvidia.ko). Is anyone complaining? Outside of a few extremists, no.

This situation with ZFS is no different than the situation with Nvidia.

Re:Software Freedom?

[GLMDesigns]

Freedom of speech is a limit on government. Government cannot say "you cannot insult the flag | president | religion | x"

Now Twitter or Facebook or Slashdot can remove people from their service for whatever reason. If we run a blog and find a contributor to be abusive or simply annoying we can remove him from our service and this is not a freedom of speech issue.

Now if I slander you. The government cannot prosecute me for saying what I said but YOU can say that my slander caused you financial harm and try to recoup said loses. Slander is knowingly (AFAIK its "knowingly") telling falsehoods about another and causing financial harm due to said falsehoods.

Example you're a programmer: and I say that you put backdoors into everything you write; insert malware and then blackmail clients to remove the malware. As a result of my lies you lost business. Therefore you can sue me for slander. Of course the flip-side to that is that I could be telling the truth and you could still sue me for slander - but that's another story.

Re:Software Freedom?

[Anonymous Coward]

Slander is a civil tort, not a criminal issue, so freedom of speech doesn't apply. You might have more traction if say criminal copyright laws violate freedom of speech (as opposed to civil enforcement).

Re:Software Freedom?

[GLMDesigns]

The OP made a blanket statement about freedom(s). Therefore mentioning an individual freedom is logical statement.

OP said that "All X are Y."

I countered that by saying that X1 is not Y. Which therefore contradicts OP's original statement.

Re:Software Freedom?

[Anonymous Coward]

You're all faggots. Fuck off.

Re:Software Freedom?

[edtice1559]

This is what I get for feeding trolls. The OP to which I'm responding seemed to think that the GPL somehow limited what you can do with software. I was pointing out that it only limits *distribution*. Combining two pieces of software to make a derivative work *is* changing the software albeit not in the way we normally think about it. There's no doubt that some licenses cannot be combined with GPL to make a work that is derivative of the two. SFC argues that Linux kernel + ZFS = derivative work that can't possibly be distributed in compliance with both licenses. Smarter heads argue that this isn't a derivative work. But neither is relevant in terms of the OP since neither license limits what you can do with the software only how you can distribute it.

Re:Software Freedom?

[Anonymous Coward]

Yes, just like personal freedom doesn't let me find out where you live and murder you in your sleep.

Re:Software Freedom?

[edtice1559]

Well if we want people to use software, you're unlikely to achieve it by setting a record for the most number of f bombs in a single post. TFA (yes I read it) asserts that what Canonical is doing qualifies as creating a derivative work. Other posters (in less colorful language) have managed to make a distinction between a distribution that combines multiple works and a derivative work. Hopefully moderation will pull those posts up and yours down.

Re:Software Freedom?

[bluefoxlucid]

I've had arguments with this guy. He has definitely said things to that effect. A less-direct statement to that effect is written into the preamble of the GPLv2; I've only ever gotten the full potato out of RMS by arguing with him directly about the MIT and BSD licenses. The man could feed the Irish.

Re:Software Freedom?

[macs4all]

Of course the flip-side to that is that I could be telling the truth and you could still sue me for slander

In the U.S. (unlike some other nation/states (I'm looking at you, U.K.), the Truth is a Complete Defense to an allegation of Slander.

IOW, in the U.S., at least, if you, as the Defendant in a Defamation of Character Suit, can provide evidence that what you said/wrote WAS in fact, true, the Plaintiff's lawsuit will be Dismisssed.

Re:Software Freedom?

[Anonymous Coward]

Freedom of speech is meaningless if it's merely the freedom to say what's uncontroversial. Everyone, everywhere has had freedom of uncontroversial speech.

Re:Software Freedom?

[pem]

So which part of the Linux kernel did the FSF or SFLC guys write?

That's right. Those bitches don't get a say.

Re:Software Freedom?

[fnj]

I take it then that the use of words you find naughty offends you more than the douchebaggery of the Software Freedom Conservancy?

Re:Software Freedom?

[Anonymous Coward]

Both of you are wrong. You don't have to agree to the license to use the software. You only have to comply with the license if you distribute the software.

If you support copyright you should support the restriction. I don't necessarily support copyright generally speaking- but to the extant that I do its only of works licensed under copyleft free software licenses. Copyleft licenses enforce in law the intent of copyright (to promote the arts and sciences). Proprietary licenses don't do that in my opinion. I believe that to the limited extant that copyright is allowed these copyleft licenses should only be enforceable against commercial activity. That is to say you can do what you like (ie piracy of GPL software is not a crime, unless it's in combination with commercial activity of any sort, so adding advertising to a GPL program and giving it away would be commercial activity if your profiting off the sale of advertising) so long as you do not gain from it financially. I don't believe we should be utilizing violence (which the law is) against people. We should only utilize it against commercial activities (ie to seize the assets or goods derived from such activities).

Re:Software Freedom?

[Your.Master]

Freedom of Speech is a concept that is thousands of years old, and is not specific to governments.

You're probably thinking of the US First Amendment, which is a limitation on government that is broadly aligned with the concept of Freedom of Speech, and mentions the concept by name, but does not invent it or control all uses of the term.

Re:Software Freedom?

[Anonymous Coward]

You'd actually be wrong. The FSF and SFLC have paid for and represent several developers who have made contributions to the Linux kernel. The SFLC is actually representing a developer in Germany whose got kernel contributions and whom copyright was violated by Oracle.

Re:Software Freedom?

[TsuruchiBrian]

If I may adapt the quote above from Mr. Orwell, I think you just said "Freedom is Licensing Restrictions."

I definitely didn't say that. I said that it's a semantic debate (i.e. rather than a substantive debate).

I presented both sides in order to show that, and I didn't even take a side.

Re:Software Freedom?

[TsuruchiBrian]

I am just trying to better explain slashpings claim, which you did not seem to understand. Slander laws are a good example of the complimentary nature of freedoms. Slander laws grant a limited freedom to silence people (when they are being slanderous), thereby reducing "the freedom of speech" to "the freedom of non-slanderous speech".

Re:Software Freedom?

[TsuruchiBrian]

You can also show that your claim was satire, and that no reasonable person could mistake your claim to be serious. For example, you can say that Donald Trump's father was an orangutan. That said, I think slander laws are pretty stupid, and I don't think the benefit we gain from them is worth the diminished state of our freedom of speech.

Re:Software Freedom?

[Anonymous Coward]

Of course the flip-side to that is that I could be telling the truth and you could still sue me for slander

In the U.S. (unlike some other nation/states (I'm looking at you, U.K.), the Truth is a Complete Defense to an allegation of Slander.

IOW, in the U.S., at least, if you, as the Defendant in a Defamation of Character Suit, can provide evidence that what you said/wrote WAS in fact, true, the Plaintiff's lawsuit will be Dismisssed.

They can still sue, but they likely won't win.

Re:Software Freedom?

[Marginal Coward]

Sorry, my bad. I should have read your response more carefully. Then again, as Bill Clinton said, "It all depends on what the meaning of the word 'is' is." ;-)

Re:Software Freedom?

[allquixotic]

It's worth mentioning that a subset of the Linux kernel copyright holders, including some of the most influential and prolific contributors, believe that the situation around the binary modules from Nvidia, AMD, etc. is similar to this ZFS module.

Just because it's gone unchallenged for a number of years doesn't mean that the issue is settled or buried. It's never been challenged or ruled-on in court AFAIK, and until it has, it's anyone's guess what the law will actually decide when interpreting the GPLv2 and copyright law.

It really doesn't matter if you want to label individuals as "extremist" for thinking a particular way about the law, though I'll admit that's a better term than Barbara's use of the term "freetard" -- again, namecalling just weakens one's argument. The only thing that matters is the way that a judge will ultimately interpret the relevant texts to determine if there is any violation of law, and what the penalties are if so.

Re:Software Freedom?

[Anonymous Coward]

No - what he just said was "Some licensing restrictions can prevent you from giving up other freedoms that are more important"

Freedom is not an absolute concrete concept - neither is "Software Freedom". The "Freedom" that Stallman wants, is the freedom to modify and fix any software he uses. In order to achieve that _particular_ freedom, you may need to add restrictions to yourself - like not using any software that _doesn't_ allow you to modify and fix it. That's not doublethink. That's just understanding the actual context of the word, instead of deciding it means something different to what it was supposed to in this case.

However, Stallman is still a dick, most of the time - and I don't buy his argument that "Someone might take this and make it so I can't modify and fix it" as a reason to not use it right now. I won't use that hypothetical closed future version, but I will use the actually existing open version right now.

Re:Software Freedom?

[Anonymous Coward]

You just can't *distribute* the changes.

You can't help your neighbor by compiling this for them and giving it to them you *must* make them do it themselves and if they are not in a position to do that then it's bad luck for them. That is a poor choice because what you would produce for them exactly is the same thing you expect them to produce for themself but they must do the work even though you have already done it. It is the same "you wouldnt steal a car" idiocy propagated by the RIAA/MPAA except in this case there is not even any percieved "loss" (by definition of "lost sales" due to piracy), it is just other people imposing their will on you and your neighbor despite their not even being the flimsiest claim of loss (like the RIAA/MPAA have with their claims of lost profits).

Re:Software Freedom?

[Anonymous Coward]

The SFC holds lot of copyrights given freely by kernel developers. Is it really that hard to read the internet, or are you just a trolling drolling idiot all the time.

Re:Software Freedom?

[exomondo]

The comparison to slavery is completely ridiculous. You are totally free to not use proprietary software and/or to discontinue its use if and when you see fit and to even use it only for specific cases and in specific envrionments for specific timeframes. Slaves were not free to just not be slaves when they didn't want to be anymore.

Sounds more like you're arguing that bondage should be illegal because it allows people to enter and exit "slavery" as they see fit.

Re: Software Freedom?

[ceoyoyo]

Mmm, you seem correct up until that last bit. Stallman's problem with non-GPL open licenses is that they lack or don't have as strong protections against redistribution. He does have an essay where he says that proprietary software is morally wrong. IIRC he also makes the link to licenses that don't prevent proprietary software. If not in that particular essay, then in other writings.

Re:Software Freedom?

[macs4all]

You can also show that your claim was satire, and that no reasonable person could mistake your claim to be serious. For example, you can say that Donald Trump's father was an orangutan. That said, I think slander laws are pretty stupid, and I don't think the benefit we gain from them is worth the diminished state of our freedom of speech.

Yeah, you're right, I wasn't thinking. And also, it is also a rebuttable defense that the target is a public/famous figure, and that they lose a measure of defamation-protection.

Re:Software Freedom?

[macs4all]

Of course the flip-side to that is that I could be telling the truth and you could still sue me for slander

In the U.S. (unlike some other nation/states (I'm looking at you, U.K.), the Truth is a Complete Defense to an allegation of Slander.

IOW, in the U.S., at least, if you, as the Defendant in a Defamation of Character Suit, can provide evidence that what you said/wrote WAS in fact, true, the Plaintiff's lawsuit will be Dismisssed.

They can still sue, but they likely won't win.

Exactly. I meant to make that point, but forgot.

Re:Software Freedom?

[Zontar The Mindless]

No, but insistence on unnecessarily colourful invective might lead others to suspect that your issues are emotional rather than rational in nature.

Re:Software Freedom?

[Grishnakh]

What's important is if anyone is going to sue over it. The only people who can sure are those who have standing. Arguably, Linus himself has the most standing, and his position is the pragmatic one: he doesn't mind modules linking to the kernel, though they can't declare themselves GPL and use the GPL-only interfaces if they're not. He's even given Nvidia the middle finger in one interview because he wasn't a big fan of their proprietary route, but he's never talked about suing them or causing any legal problems, and since then from what I've read they've actually helped out with the Nouveau drivers some.

The issue isn't going to be ruled on in court until someone with standing sues over it. And if they haven't done so by now with Nvidia and AMD, I don't see why they'd start now. They'd only be hurting the cause for Linux proliferation. Having more software working in Linux is a good thing for the users. The only people who are against this are the extremists, and none of them seem to do any kernel work. So they can whine all they want, but no one cares, and they'll be laughed out of court if they try to bring a lawsuit.

Re:Software Freedom?

[TsuruchiBrian]

I think it's pretty clear that "freedom" means different things to different people. I think "is" means the same thing to everyone except Bill.

Re:Software Freedom?

[TsuruchiBrian]

In this example slavery is not analogous to the use of proprietary software, it's analogous to agreeing to the license of proprietary software, which you can't unagree to.

Sounds more like you're arguing that bondage should be illegal because it allows people to enter and exit "slavery" as they see fit.

I am arguing is that it's a semantic debate, and neither side is right or wrong.

Re:Software Freedom?

[exomondo]

In this example slavery is not analogous to the use of proprietary software, it's analogous to agreeing to the license of proprietary software, which you can't unagree to.

Of course you can. You can't use the software without agreeing to its license stipulations, if you no longer agree you no longer use it. Just like your license to use the road, you must agree to the license conditions and if you don't then you do not use the road and if you "unagree" you cease to use the road. Slaves can't just "no longer agree" and then not be bound by slavery anymore. Using proprietary software - or the road for that matter - under the specified licensing conditions is a choice, slavery is not.

Re:Software Freedom?

[TsuruchiBrian]

Of course you can. You can't use the software without agreeing to its license stipulations, if you no longer agree you no longer use it.

That's may be true for some conditions. Other conditions might promises not to reverse engineer the software. I don't think this means that you are allowed by the license to reverse engineer the software once you stop using it.

Using proprietary software - or the road for that matter - under the specified licensing conditions is a choice, slavery is not.

Historically, slavery, for the vast majority of people, has not been a choice (neither entering nor exiting it). That doesn't mean it can't be a choice. There are no doubt people who makes decisions for various reasons that they know for whatever reason will lead to their enslavement. There are also probably people who forego the opportunity to become freed from slavery for various reasons.

This historical aspect of slavery rarely involving choice, is not the focus of my analogy.

If you really can't let it go, then try to imagine some other kind of slavery like thing that involves the freedom to restrict your own freedoms.

Re:Software Freedom?

[exomondo]

That's may be true for some conditions. Other conditions might promises not to reverse engineer the software. I don't think this means that you are allowed by the license to reverse engineer the software once you stop using it.

So you think you can reverse engineer the software so long as you don't agree to the license?

Historically, slavery, for the vast majority of people, has not been a choice (neither entering nor exiting it). That doesn't mean it can't be a choice.

But it isn't, that's why it's not analagous to slavery because in order for it to be you need to redefine slavery to mean you have a choice about whether you want to be a slave or not.

If you really can't let it go, then try to imagine some other kind of slavery like thing that involves the freedom to restrict your own freedoms.

I have control of my freedoms, I have all the choice, that is the complete opposite of slavery. You think I feel enslaved just because I can't come into your house and just pee all over your walls? I - like most people - are quite happy to restrict my freedom to do that, to suggest this is anything even close to slavery is just complete idiocy.

Re:Software Freedom?

[Anonymous Coward]

Basically, Richard Stallman's line is that the MIT license restricts your freedom as a software developer and a software user, and the GPL protects it. It's backwards double-think.

He says no such thing. He says that liberal free software licenses (MIT, BSD, XLib) are not immoral of themselves. He says that anybody who takes advantage of forking such software into a proprietary software distribution are immoral: these distributors are immoral, not the users of the software. Stallman says that user freedoms are not guaranteed by liberal free software licenses; users must judge their freedom for every instance of the liberal-licensed software on a case-by-case basis.

With the GPL this is not the case. In all cases of GPL3+ software, users can expect to have the four specific users freedom with the software. If there is ever a case of a GPL program that doesn't respect the users' freedom, the distributor of that specific program is liable to copyright infringement of the copyright holder.

Re:Software Freedom?

[TsuruchiBrian]

So you think you can reverse engineer the software so long as you don't agree to the license?

According to who? The license or the law?

But it isn't, that's why it's not analagous to slavery because in order for it to be you need to redefine slavery to mean you have a choice about whether you want to be a slave or not.

That's definitely not true. Nearly all cars have steering wheels. That doesn't mean a car is defined by a having a steering steering wheel. It doesn't mean that a car without a steering wheel is redefining what a car is. It shouldn't be that hard to imagine a scenario where a person makes a choice to be enslaved.

I have control of my freedoms, I have all the choice, that is the complete opposite of slavery. You think I feel enslaved just because I can't come into your house and just pee all over your walls? I - like most people - are quite happy to restrict my freedom to do that, to suggest this is anything even close to slavery is just complete idiocy.

I really don't see how this statement has anything to do with anything I've said.

In some places you don't have control over *all* your freedoms. In some places you are prevented from entering into a contract where you enslave yourself (e.g. like in the US). This is maybe a good freedom to restrict, as sacrificing this freedom will ensure the preservation of your other freedoms, but it is nonetheless a freedom that is denied to you.

Denying a person the freedom to shoot themselves in the foot may help to preserve other freedoms that person may have (e.g. freedom to walk), but it is nonetheless a restriction on that persons freedom.

Re:Software Freedom?

[Anonymous Coward]

It is very different, in that Nvidia owns the copyrights to their driver, and are willing to jump through the hoops necessary to avoid the pitfalls, including keeping the driver and the kernel separate and have a shim with a license that allows linking to both.

Canonical, on the other hand, does not own ZFS, and Oracle (who does) are not willing to jump through any such hoops. Quite the opposite, they (or rather Sun, which is now owned by Oracle) specifically wrote a license to make sure such a shim could not be distributed (it would have to be GPL and CDDL at the same time, and the licenses are incompatible).

Re:Software Freedom?

[Marginal Coward]

You folks never fail to disappoint me. Every time I express this general point of view, my post gets marked as a "Troll." Did anybody ever consider that I might just be expressing my honest opinion? It amazes me that folks who are all enthused about the abstract concept of "software freedom" don't really much care about a more concrete thing "freedom of speech."

(Note to "Moderators": Go ahead, make my day.)

Re:Software Freedom?

[Marginal Coward]

Fair enough. I just wish Mr. Stallman would coin a new word for whatever it is he means by "freedom." Like so much of the terminology he co-opts, I have a hard time wrapping my head around his concepts when he uses words for them that mean something different than the meanings we were all taught in grade school. For example, in grade school, "free" meant "free, as in the lunch poor kids got" (we didn't drink beer back then - well, maybe once...) and "freedom" mean something like "the ability to do what you want to the maximum extent practical."

Re:Software Freedom?

[david_thornley]

Let's look at it this way.

We have two similar pieces of software, B, which is BSD-licensed, and G, which is GPLed. Neither clearly leads to more freedom than the other.

B can be used in more situations, and is freer in that way. However, a Very Large Corporation can take the code to B, change it so the original B is incompatible, and widely publish it in proprietary form. The original isn't really useful anymore, because of the ubiquity of the incompatible version, and so people with the original B can't change and use it effectively any more. This couldn't happen with G, since VLC would have to publish its version of G with source code and permission to hack, so the GPL is freer there.

Even the GPL is subject to "freedom" arguments. Linus Torvalds has no problem with what Tivo did with the Linux kernel, since he can look through their source code and take all the good bits. Richard Stallman does, since he can't reprogram his Tivo (assuming he has one).

There are cases where something is more free than another, but BSD-style vs. copyleft licenses aren't among them. You need to pick which freedoms you most care about, and choose a license accordingly.

Re:Software Freedom?

[Anonymous Coward]

Stallman believes that freedom is a matter of self-control. The owners of proprietary software do not practice self-control. The owners of proprietary software ask users to take control of their computers and their lives. Users actually choose to grant them this control. Users accept this choice cannot have freedom.

Re:Software Freedom?

[Anonymous Coward]

It's not hard to understand the intent - it's to preserve the "Four Freedoms":

* The freedom to run the program as you wish, for any purpose (freedom 0).
* The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is a precondition for this.
* The freedom to redistribute copies so you can help your neighbor (freedom 2).
* The freedom to distribute copies of your modified versions to others (freedom 3). By doing this you can give the whole community a chance to benefit from your changes. Access to the source code is a precondition for this.

The GPL is necessary to ensure that when you get the code, you have these freedoms and can't keep them from anyone you distribute code to.

The part I really disagree with is that dynamic linking creates a derivative work. Part of the freedom to "run the program for any purpose" in my book includes using the program with other programs that may not be GPL licensed, including plug-ins and extensions.

Re:Software Freedom?

[Anonymous Coward]

The comparison of proprietary software to slavery is analogous. People choose to be bound to a software master. People who accept that choice choose to subject their computing lives to the goodwill of the software owner - if the user wants the software modified, the only thing that the user can do is beg and hope. The fact that you can choose to end the bondage at will does not change the fact that you are bound while you agree to it. Having the choice of one or many masters to be subject to doesn't change the fact that you cannot be free while you have accepted the choice.

Having the choice to stop being a slave doesn't imply you have freedom when you choose to be one.

Re:Software Freedom?

[Marginal Coward]

BTW, I just noticed that my "Troll" at top generated quite a lot of interesting discussion, including one comment (by someone else) that got a "5."

You're welcome.

Re:Software Freedom?

[exomondo]

According to who? The license or the law?

The law, DMCA.

Nearly all cars have steering wheels. That doesn't mean a car is defined by a having a steering steering wheel. It doesn't mean that a car without a steering wheel is redefining what a car is.

Yes but you don't just point to a submarine and say it is a car. The idea of "choice" is directly contradictory to slavery, though it is becoming evident you don't understand the meaning of the word which is why you're so confused.

It shouldn't be that hard to imagine a scenario where a person makes a choice to be enslaved.

Except for anybody who knows that the definition of "enslaved" is precisely about choice and if you remove freedom of choice then you are no longer enslaved. The "choice" to be enslaved is directly contradictory to the concept of slavery.

In some places you don't have control over *all* your freedoms.

Right, and I have the freedom to choose whether I want to go there or not, that is exactly the scenario you laid out and it is absolutely not slavery because I have choice.

In some places you are prevented from entering into a contract where you enslave yoursel

Except that isn't what we are talking about because in terms of use of proprietary software you always have the freedom of choice to stop being under those conditions at any point you like, because you are not a "slave".

I like that I have the freedom to visit a museum despite not having the freedom to smash the place up when I'm there and I also like that I have the freedom to leave and go somewhere else should I feel the need to smash something up. You have to be extremely intellectually dishonest or extremely stupid to think that is in any way anything like "slavery".

Re:Software Freedom?

[drew_kime]

Both of you are wrong. You don't have to agree to the license to use the software. You only have to comply with the license if you distribute the software.

Yup, fair point. You don't have to comply with any terms just to use it. It would have been more accurate to say that the only reason you have it at all is that the person who distributed it to you was complying with the terms.

Re:Software Freedom?

[TsuruchiBrian]

The law, DMCA.

The DMCA does not prohibit reverse engineering. It prohibits circumventing a copy protection system. And yes I think you can reverse engineer software if you did not specifically agree not to.

Except for anybody who knows that the definition of "enslaved" is precisely about choice and if you remove freedom of choice then you are no longer enslaved [oxforddictionaries.com]. The "choice" to be enslaved is directly contradictory to the concept of slavery.

That's only true if you can't wrap your head around the idea of choosing to limit your own choices.

Right, and I have the freedom to choose whether I want to go there or not, that is exactly the scenario you laid out and it is absolutely not slavery because I have choice.

A person could choose to go to a place that has debtors prisons and choose to rack up a huge debt, and be forced into a slave labor camp. Despite the fact that this person decided to become a slave, they are a slave who no longer has freedoms nonetheless.

Except that isn't what we are talking about because in terms of use of proprietary software you always have the freedom of choice to stop being under those conditions at any point you like, because you are not a "slave".

Yeah you already said that, and I said, there are some conditions in software licenses that you can not unagree to.

I like that I have the freedom to visit a museum despite not having the freedom to smash the place up when I'm there and I also like that I have the freedom to leave and go somewhere else should I feel the need to smash something up.

I don't see what this has to do with anything I've said. The fact that you bring up this seemingly irrelevant example, makes me feel liek you aen;t understanding what I am saying.

You have to be extremely intellectually dishonest or extremely stupid to think that is in any way anything like "slavery".

Or maybe you just lack the comprehension necessary to understand what I am saying.

Re:Software Freedom?

[exomondo]

The comparison of proprietary software to slavery is analogous. People choose to be bound to a software master.

But you aren't "bound by a software master" any more than you are bound by the rules of a museum when you visit. Sure I can't go in there and smash the place up or change things, but calling that "slavery" is just idiotic.

Re:Software Freedom?

[exomondo]

The ultimate freedom lies with the user, you can debate the merits of the licenses all you want but ultimately the user has the choice and that is what is important. It doesn't matter if you are aligned with Linus' view of software freedom or RMS' view of software freedom or the BSD view of software freedom or you don't care about software freedom so long as you have choice. If you care about a software ideology then you should support and invest in that ideology, but most people don't care and those who do rarely care enough to limit themselves to that ideology alone.

Re:Software Freedom?

[exomondo]

And yes I think you can reverse engineer software if you did not specifically agree not to.

And that is wrong, thankfully the EFF has put together a guide here about it.

That's only true if you can't wrap your head around the idea of choosing to limit your own choices.

Wrong! I perfectly understand that I am free to visit a museum while not being free to smash the place up, why can you not understand this?

A person could choose to go to a place that has debtors prisons and choose to rack up a huge debt, and be forced into a slave labor camp.

Yes and a person could choose to murder somebody and be forced into a prison, we are all bound by the consequences of our choices. But of course in terms of software you can just delete it and its provisions no longer affect you.

I don't see what this has to do with anything I've said. The fact that you bring up this seemingly irrelevant example, makes me feel liek you aen;t understanding what I am saying.

Then perhaps you need ot review what you wrote. A person has the right to choose to limit their own freedoms in certain circumstances, you can't take that away.

Denying a person the freedom to shoot themselves in the foot may help to preserve other freedoms that person may have (e.g. freedom to walk), but it is nonetheless a restriction on that persons freedom.

Right, that's my point. It sounds like we agree then!

Re:Software Freedom?

[Anonymous Coward]

You forgot the best part: "I know that I can't have babies! But I must have the right to have a baby!"

In French, there is the famous"Permis de conduire un orchestre"...

Re:Software Freedom?

[GLMDesigns]

The issue comes in defining speech. It's not contradictory to put limits when defining a term. Language is broad and the same word or phrase can mean different things to different people. Freedom of Speech refers to expressing ideas - political, religious, philosophical and not being punished by government for saying it. Government cannot prevent you, nor punish you for expressing your ideas even if these ideas are

hateful: Examples include overt racist speech (blacks | white| asians) are X

or wrong: Example: the earth sits on a tortoise; the earth has 4 corners; the earth was created 6000 years ago.

or inciteful and twisting facts: Example: Muhammad was a pedophile

Freedom of Speech does not, and never has, applied to saying untruths about a specific individual (slander). That is outside the scope of the definition of Freedom of Speech.

Re:Software Freedom?

[david_thornley]

The ultimate freedom lies with the user

Ah, but which user? The GPL was inspired by a printer driver Stallman couldn't rewrite, so he was the user. Microsoft adopted the BSD TCP/IP stack, which I think was probably a Very Good Thing, so Microsoft was the user there. It probably would not have mattered if Stallman's driver had been BSDed, he wanted the source code. Microsoft could not have used a GPLed stack.

Re:Software Freedom?

[TsuruchiBrian]

Everything you've just said is irrelevant to the singular point I am reiterating, which is that for every right that is granted a complimentary right is removed. This is not a claim about any specific laws in any specific countries.

It is certainly common for countries to place their own combination of restrictions on free speech as they see fit. But nearly everyone views these as restrictions on free speech (albeit necessary restrictions). It would be silly to treat the United States version of freedom of speech as the benchmark, and treat all deviations from this benchmark as added restrictions or extraneous rights from the gold standard definition for freedom of speech. Laws can change. In fact the UN has urged members of the International Covenant on Civil and Political Rights to decriminalize libel.

Freedom of Speech does not, and never has, applied to saying untruths about a specific individual (slander). That is outside the scope of the definition of Freedom of Speech.

Freedom of speech wasn't invented by the United States and it is a concept that transcends US law.

There was probably a time when people said that the constitution did not, and never has applied to slaves, and it was true, until it wasn't. Eventually people recognized that the institution of slavery was incompatible with the morals they wanted to embody. I believe that eventually the same thing will happen (and is starting to happen) with defamation laws.

But all this is besides my point that rights are complimentary. The granting of every right is the rejection of it's counterpart. Sometimes it is clear which rights we want (e.g. the right to life vs. the right to murder), but sometimes it isn't so clear (e.g. the right to free speech vs. right to not be slandered). This is a perfectly legitimate and widely accepted rights model.

Re:Software Freedom?

[GLMDesigns]

The concept right is something that you have qua living. It's not anything given to you - but it is something that can be infringed on or taken away. The concept "Right" is an enlightenment term; a term that had little meaning before and has been under attack from the 19th C German Romantic period on. It's a term that can only exist within the context of individual freedom and living in a state where one is a citizen (as opposed to a subject); where government is limited by a matter of law as well as culture.

I would argue that the Freedom of Speech did NOT exist in 15th C Europe; or China; or Japan or in the Muslim world. In fact the very concept of it where by a human being could speak their mind and the government COULD NOT (by law) interfere would have been considered foolishness beyond the realm of conversation. You might as well have been talking about flying and breathing under water.

While Freedom of Speech was not invented by Americans it is a relatively new phenomenom. And we see very clearly the anti-enlightment philosophies that unite in opposition to it; we also see these same philosophies appropriating enlightment terms (rights, liberal) and attaching new meanings to them. Examples include the concept of positive rights and civil rights.

What rights may I ask do you have that comes from being part of a group that you don't have as an individual? Racists may have denied individual rights to "others" but they did not create rights for themselves. If you are outraged by this - think about what you're outraged about. Reflect that what happened was the negation of rights for some; not the creation of rights for the privileged.

A key test to what is and what is not a right is by looking at where "it comes from." You have the right to speech (nobody gives you the thought or the impulse to promote that view) as opposed to the right to healthcare (in which others must provide it for you - even against their will). It you are promoting the later than you are part of the anti-enlightenment; anti-freedom; anti-liberal crowd who has appropriated sanctified terms in order to bolster the value of their positions.

The right to life is not in opposition to the right to murder. There has never been a right to murder another. Think about the definition of rights - the enlightment definition of rights to get this. (There is a right to self-defense though; and for good reason.)

Re:Software Freedom?

[TsuruchiBrian]

I think you have a very narrow personal definition of what a right is, and that is limiting your ability to have a conversation that doesn't fall within your strict definition.

The right to life is not in opposition to the right to murder. There has never been a right to murder another.

The right to murder has certainly existed for various individuals throughout history. They were allowed to kill people as they saw fit according to the laws of their lands. You might argue that while their laws allowed them to murder, they still did not have the "right" to murder.

I am quite familiar with the concepts of positive and negative rights, so I really don't need you to keep trying to explain them to me. In fact the concepts of positive and negative rights fits quite well into my model of complimentary rights (i.e. where one is positive and one is negative).

You seem to be someone who is of the opinion that only positive rights are legitimate. Fine. You are free to have that opinion, and you have a lot of company. And in general I agree that positive rights tend to be the better choice when deciding which side of the dichotomy we want. But what you seem to be denying is that the dichotomy even exists.

Where does the right to life come from? I suspect I know what your answer will be, but I'll let you answer for yourself. My answer is that it comes from the government's laws against murder and it's prosecution of crimes that violate these laws. This then becomes a right that requires government action to manifest, blurring the line between positive and negative rights a little.

I don't believe in the abstract concept of self evident rights bestowed by our creator. I find this to be a pretty lazy explanation that creates a convenient yet flawed oversimplification.

Re:Software Freedom?

[exomondo]

Ah, but which user?

Any user. If RMS doesn't want to use a non-free printer then he doesn't have to, if Microsoft doesn't want to use a GPL'd TCP/IP stack because they don't want to release their code then they don't have to. But situations like this where the only difference is the user has to put the bits together because the GPL says somebody else can't do it for them then distribute it to them is just silliness, it doesn't grant any additional freedoms or have any different result except for making it more difficult for the user.

Re:Software Freedom?

[GLMDesigns]

Re: "narrow personal definition of a right" I agree that is narrow, but not that it is personal. :-) Meaning that I hold a concept of rights that is more Enlightment / Individualist rather than Collectivist.

Re right to murder - now we're getting into a matter of semantics that is very difficult to get into in a setting such as this (quick posts as opposed to lengthy conversations or extensive, elaborate explanations. Let it be said that in my "narrow personal definition" of a right I make a distinction between their power to act without consequences to right.

Now source of rights. In the 18th C they used the concept God and Natural Law. I'm an athe!st (a spaghetti-monster sort of athei!t) so the God part is not convincing. What is convincing is the concept that *I* have a right over my own body (not an 18thC idea) as well as my ideas. The government's role in our life is many things - but what it is not is the violator of my rights. Government is there to enforce the social contract - "I promise not to kill you and take your stuff if you promise not to kill me and take mine."

Government did not create my ideas, did not "breathe life into me" (an 18th C phrasing - not mine). It did not give me my rights but it certainly can prevent me from exercising my right: freedom of conscience - called religion back when; freedom of speech, etc...

Re:Software Freedom?

[TsuruchiBrian]

Re: "narrow personal definition of a right" I agree that is narrow, but not that it is personal. :-) Meaning that I hold a concept of rights that is more Enlightment / Individualist rather than Collectivist.

By personal, I mean that there is no universally agreed upon model of what rights are and where they are derived and how they should be presented (positive vs. negative or, of complimentary rights, etc).

Re right to murder - now we're getting into a matter of semantics that is very difficult to get into in a setting such as this (quick posts as opposed to lengthy conversations or extensive, elaborate explanations. Let it be said that in my "narrow personal definition" of a right I make a distinction between their power to act without consequences to right.

Exactly. So the right to murder would be the ability to legally murder people (i.e. without legal consequence). This would not stop another private citizen from exercising their right to murder you. Like freedom of speech it only protects you from state consequences of your speech.

What is convincing is the concept that *I* have a right over my own body (not an 18thC idea) as well as my ideas.

To me this doesn't explain anything about where rights come from, anymore than explaining where consciousness comes from by saying "I have a soul because I find the concept that I have a soul convincing".

Government is there to enforce the social contract - "I promise not to kill you and take your stuff if you promise not to kill me and take mine."

I agree. But I also think that it is the enforcement of this contract that causes these rights to manifest. I think we could have chosen a different social contract and different rights would have manifested. I don't think there are any rights that transcend a government's willingness to uphold them. Is there the right to life under the sea? I would say no, because there is no King Triton prosecuting undersea murders.

Government did not create my ideas, did not "breathe life into me" (an 18th C phrasing - not mine). It did not give me my rights but it certainly can prevent me from exercising my right: freedom of conscience - called religion back when; freedom of speech, etc...M

No government did not create the concept of rights. But it is nonetheless where rights come from. If you ask where does a Boeing 737 come from. One answer might be that it came from Boeing engineers and visionaries. Another answer would be that it comes from the Boeing factory, and the raw materials came from wherever they came from. Obviously human beings created the concept of rights. I am saying that governments are rights factories (where rights come from in a literal sense).

We wouldn't have commercial airliners without engineers, we also wouldn't have them without the factories (or facilities where they are actually produced). By the same token, rights would just be an abstract concept without governments to make them a reality.

Re:Software Freedom?

[TsuruchiBrian]

BTW I accidentally switched positive and negative rights. Everything else I still stand by.

SFS? More like FFS...

[Bearhouse]

Why oh why do we have to keep shooting ourselves in the foot?
OK I'm a BSD user so, well, stones and glass houses, but even so the open source community's continuing ability to why things should not be allowed is depressing...most people in our crowd want our stuff to be USED by as many people as possible...

Re:SFS? More like FFS...

[edtice1559]

The goal of the GPL is not to get the most widely used software (although that's a noble goal). The goal is to ensure that the users of the software are able to maintain and secure it so they don't get left hanging out to dry if something happens to their upstream vendor. When you receive software under the GPL you expect to have these freedoms. When software with multiple licenses is combined, users have to become lawyers to know what rights they actually have and whether these freedoms are protected. I'm not sure it's a concern in this particular case since both GPL and CDDLv1 protect the user's rights. (It's only distributors who are affected by this). But knowing that you can secure and maintain software is valuable. It would be undesirable to get into a situation where you think you have that right, but it turns out that you don't.

Re:SFS? More like FFS...

[Anonymous Coward]

I'm not sure it's a concern in this particular case since both GPL and CDDLv1 protect the user's rights. (It's only distributors who are affected by this).

But there are a lot of "distributers" out there.

If I'm running a small computer repair shop, and I decide to take an old PC and sell it, then I'm a "distributer" of whatever OS is on that machine. I'm not necessarily an expert on software licensing. All I want to know is that I'm legally safe if I put Linux on the PC.

Anyone who sells any kind of computing device, even if it's just to a neighbor, or to someone on e-bay, is potentially a "distributer".

Re:SFS? More like FFS...

[oh_my_080980980]

So that's why this instance is bull shit. Software Freedom Conservancy can go fuck itself. CDDLv1 isn't not some license that's going to prevent someone from using Linux. The fact that any user can download and use ZFS proves the point. There's no reason not to include it.

Re:SFS? More like FFS...

[edtice1559]

IANAL so I can't really answer the question about the small PC shop. In terms of selling to your neighbor, you would be considered a casual not commercial distributor of the software. If you received the software in compliance with the GPL, your sale to your neighbor will remain in compliance. Commercial distributors have a higher burden because they can't rely on a third-party offer for compliance. Again, IANAL, but the the GPL FAQ does talk about this. As a small PC shop, you should consider installing the software as an agent of the purchaser rather than distributing it. I'm not entirely sure how the mechanics of that would work. I do know from experience that some defense customers are still buying Windows/XP machines. The customer procures the license and then the seller installs on their behalf. Almost no difference on the surface except the customer has to do more paperwork. The reality is that, as a small seller, who isn't creating any derivative works, you're not a target for anybody and it probably doesn't matter either way.

Re:SFS? More like FFS...

[brambus]

From a user perspective, the only practical distinction between GPL and CDDL is that CDDL code doesn't infect code linked with it. That's all. You can use the code, modify it, link it in your product and release it, but you are only obliged to release code for the original files (/w your modifications). The reason for the incompatibility is that the GPL would really like to infectiously spread itself over the CDDL'd code and take that freedom away, but luckily the CDDL has a clause not allowing the license to be overridden.

Re:SFS? More like FFS...

[dbIII]

Linux is not an FSF or gnu project.
They are shooting at somebody else's feet.

Re:SFS? More like FFS...

[Beezlebub33]

The goal of the GPL is not to get the most widely used software (although that's a noble goal). The goal is to ensure that the users of the software are able to maintain and secure it so they don't get left hanging out to dry if something happens to their upstream vendor.

Which would be the case if the source code was available. And it is. (Crap like this is why I avoid GPL'ed code like the plague. Because it is the plague: it infects everything and brings no benefit. Having developed open source software for a long time, I'm a huge fan of permissive licenses, which is why we always use Apache or MIT. )

The GPL license is the problem here. I know that SFC wants Oracle to change, but maybe changing the GPL would be better. Add a clause that says that you can distribute the binaries if you also distribute the source.

Re:SFS? More like FFS...

[edtice1559]

Sun licensed ZFS in a way intended to make it incompatible with the GPL. They achieved their goal. How can you fault the FSF for this? If the GPL terms had been different, CDDL would have been adjusted to maintain the incompatibility. An organization set out to create this situation, they succeeded in creating it, somebody points out their success and somehow it's a third-parties fault. I'm having trouble following that logic. This seems more like an emotional reaction to the GPL. When I first heard of it I certainly had one. I made my living writing software and these "free software" guys were driving the value of software to zero dollars. It's only later, now that I understand a lot of about software security, that I realize source code distribution is the only real way to make software secure.

Re:SFS? More like FFS...

[Bengie]

When you receive software under the GPL you expect to have these freedoms. When software with multiple licenses is combined, users have to become lawyers to know what rights they actually have and whether these freedoms are protected.

You already have to be a lawyer to understand the GPL. If it's more than a small paragraph long, it's too long.

Re:SFS? More like FFS...

[Bengie]

Let me know how your Linux ISO with no ZFS handles booting from ZFS without modifying the ISO. As soon as something doesn't work out of the box, it's broken. Decide to upgrade and ZFS somehow breaks, now your system is fubar.

Re:SFS? More like FFS...

[david_thornley]

If you can't understand the GPL, I don't want you working on software I work on. It's not that difficult. It requires a little concentration and a little care, but less than I exert daily with the code base I work on.

Re:SFS? More like FFS...

[lkcl]

Why oh why do we have to keep shooting ourselves in the foot?
OK I'm a BSD user so, well, stones and glass houses, but even so the open source community's continuing ability to why things should not be allowed is depressing...most people in our crowd want our stuff to be USED by as many people as possible...

copyright law and the GPL are there for very good reasons. i've posted about this before, but a simple scenario which can easily be demonstrated as a real-world "nightmare scenario" that should tell you why you are mistaken is: security vulnerabilities in smartphones.

let's take a simple GPL-violating Mediatek smartphone (for example the Fairphone 1). Fairphone created their first phone to reduce e-waste and also to tackle conflict materials problems and to provide factory workers with fair working conditions and pensions. except... two years on, the GPL-violating linux kernel that was used turned out to have a severe security flaw.

so, end-users contacted Fairphone and said, "any chance of some security updates, please? our product's less than 18 months old". so Fairphone contacted the factory and said, "can we have the source code please?" and the factory owner went, "what source code?" [actually he probably said "what's source code"].

it turns out that they were supplied with a binary-only "solution". there IS no way to fix the security vulnerabilities, leaving end-users with the stark choice of throwing away a perfectly good phone that's SUPPOSED TO BE REDUCING E-WASTE.

now, can you *honestly* tell me that in this scenario - which is extremely common - that you "Just Want Your Stuff To Be USED By As Many People As Possible"?

if you do, then you are, if i may be absolutely bluntly honest in my assessments, a self-serving ego-maniac who wants fame more than he wants freedom. spreading ILLEGAL SOFTWARE to "as many people as possible" cannot POSSIBLY be the ultimate end-goal that you wish to see happen, so that possibility is ruled out. spreading software with CRITICAL SECURITY FLAWS cannot possibly be the end-goal either. the only thing left is that you wish to be famous, by virtue of having "reached more people".

if the above assessment is wrong, please do correct me, i am happy to be told i am wrong.

Re:SFS? More like FFS...

[edtice1559]

This is reminiscent of Microsoft FUD from the 90s. If you find GPL code valuable and want to use it, don't begrudge giving your changes back. That covers about 90% of the cases. If you're out combining multiple pieces of different software, you have to be aware of the license terms. There are lawyers who specialize in this and even entire software products dedicated to it. But the problem isn't a function of any particular license it's a function of having disparate licenses. The FSF has actually gone to great lengths to make this easy. The way I handle this is I look at the list of GPL-compatible free software licenses at http://www.gnu.org/licenses/li... and only accept code with licenses on the list.

SFS?

[Nighttime]

Shirley, SFC for Software Freedom Conservancy?

Re:SFS?

[Anonymous Coward]

No, my name is not Shirley, but I do agree that S.F.S. makes no sense for an abbreviation of Software Freedom Conservancy.

I don't see the advantage of objecting

[davecb]

One wonders if SFC sees a risk to the community from Oracle acting against Canonical. If not, they may inadvertently be playing dog-in-the-manger.

Re:I don't see the advantage of objecting

[brambus]

But the joke here is that it's not Oracle who would sue. It would have to be the Linux Foundation (or some copyright holder of the Linux kernel), because it's not the CDDL that's imposing itself on the Linux kernel. It's the other way around.

Re:I don't see the advantage of objecting

[oh_my_080980980]

Which makes this totally fucked.

Re:I don't see the advantage of objecting

[brambus]

I think it'd be absolutely hilarious if the Linux Foundation sued Canonical for including another piece of open source with the kernel. Frankly, I think it'd be so much bad PR for the Linux Foundation that they won't touch it with a 10' pole.

Re:I don't see the advantage of objecting

[pem]

It won't be the foundation. It'll be some lame-ass dev who has 3 bugfix lines somewhere.

Watch how quickly those lines get replaced, and how that guy can't get a job afterward.

Re:I don't see the advantage of objecting

[Anonymous Coward]

Oracle can totally sue end users using ZFS if they violate the CDDL, or if they determine that you are violating their patents and that the CDDL patent grant does not apply.
Not that Oracle has been known to ever do anything like this of course.

Re:I don't see the advantage of objecting

[Anonymous Coward]

Or perhaps a competing distribution like Redhat that has enough code lines committed to not make it easily replaceable and that would actually benefit from making things hard for Canonical.

Re:I don't see the advantage of objecting

[Anonymous Coward]

Noooooo...each is 'imposing' itself on the other. E.g. you can't distribute ZFS with Linux without putting all the Linux source under the CDDL license so how is that the GPL imposing itself on ZFS? Equally there are terms of the GPL incompatible with the CDDL...

Re:I don't see the advantage of objecting

[Anonymous Coward]

Noooooo...each is 'imposing' itself on the other. E.g. you can't distribute ZFS with Linux without putting all the Linux source under the CDDL license so how is that the GPL imposing itself on ZFS? Equally there are terms of the GPL incompatible with the CDDL...

You're either painfully ignorant or outright lying.
The CDDL says no such thing.

All Linux kernel code would be able to remain under the GPL. All changes to ZFS would, however, have to be committed under the CDDL.

It isn't features, it is stability + features

[poet]

Linux does not have a stable file-system and volume manager that can come anywhere near the performance and stability of ZFS.

Re: It isn't features, it is stability + features

[BellyJelly]

So don't use it. Use a BSD or opensolaris, or pay for your software. IANAL, but if this breaches the GPL and you don't like it then don't use GPL software. Or add the module to the kernel yourself and don't re-distribute. I don't understand why some people bitch about linux or the GPL. No one is forcing anyone to use it. In the meantime, millions of others will carry on using it successfully while not giving a flying fuck about re-distribution with incompatible software licenses.

Re: It isn't features, it is stability + features

[poet]

Sure and I don't have a problem with BSD nor am I complaining about Linux or the GPL. The problem here is we have different attorneys both are biased saying different things about including ZFS and really the only people that get an opinion is Canonical and Oracle.

Re: It isn't features, it is stability + features

[oh_my_080980980]

And I can add it after the fact Potsy. So what then. All they are doing is providing people with the option of using it.

Re:It isn't features, it is stability + features

[danbob999]

All filesytems have the same stability. It's their implementation that can be buggy.
Is the linux ZFS implemention more stable than its BTRFS implementation? And if so, does it matter for the 99% of us who are fine with EXT4 (and even EXT3)?

Re:It isn't features, it is stability + features

[poet]

Yes, Linux ZFS is more stable than BTRFS.

And if ext4 works for you great but comparing ext4 to zfs is like comparing a pinto to a corvette.

Re:It isn't features, it is stability + features

[danbob999]

Also, is there anything blocking a clean-room GPLv2 implementation of ZFS for Linux?

Re:It isn't features, it is stability + features

[Anonymous Coward]

Time. Money. Testing. Stability. Naming. Maybe patents.

Re:It isn't features, it is stability + features

[danbob999]

Good example. A corvette is a niche, expensive car suiting very few people.
If a corvette is not compatible with my roads (or fuel, or whatever) I'll change the car not the road.

Re:It isn't features, it is stability + features

[danbob999]

But they had time to create BTRFS? I understand it was easier than to clone ZFS?

Re:It isn't features, it is stability + features

[Anonymous Coward]

It's for totally different use-cases. ZFS only works fine if any upgrade is done by a complete set of disks. If you think of needing more storage, and answering that by pushing another rack into the shelf. btrfs is much more flexible, making it a lot more useful for smaller installations. Where "smaller" still means much bigger than a NAS from the supermarket next door.

Re:It isn't features, it is stability + features

[bheading]

Sure. BTRFS is, arguably, a disappointment. It has a lot of the features of ZFS, but it doesn't seem to be anything like as user friendly or logical. When I played with it for the first time about a year ago, I found no out-of-the-box automated snapshot features, and had to install a package to handle this, which didn't seem to work reliably. It's also an enormous missed opportunity that BTRFS doesn't have the ability to natively do RAID-6, or the ability to use an SSD as a fast block cache in front of standard HDDs. ZFS has done all of this for many years now.

I'm sure someone will reply with details of how to do these things. But with ZFS it was obvious and I didn't need to spend a lot of time googling.

That said, there's nothing that says that ZFS on Linux is automatically stable either. It's been necessary to extensively modify it in order to make it work. Running either of these two filesystems in production on Linux would be a risky proposition. It's no wonder the major enterprise vendors haven't switched to use them yet.

Unfortunately at the moment it looks like the short term future of filesystems on Linux is based around XFS. Longer term, bcachefs looks interesting.

Re:It isn't features, it is stability + features

[bheading]

That's not a fair characterisation. I tested an upgrade of ZFS; it could do live, in place upgrades just fine. So you could swap out a RAID array of disks just by switching each disk and waiting for the rebuild. Of course it is an enterprise FS, and intended for people who'd rather add a new shelf of disks than run the risk of downtime.

ZFS never made any secret of the fact that it was designed to use lots of (cheap) RAM and an SSD backend to get consistent performance.

Re:It isn't features, it is stability + features

[mysidia]

I would liken Ext4 more to a 1932 Ford V8; XFS is a Reliant Robin
BTRFS is a 1981 DeLorean DMC-12.

Finally, ZFS is a 1990 Lexus LS 400

Re:It isn't features, it is stability + features

[Anonymous Coward]

Bullshit. It's more like comparing a station wagon to a main battle tank. ZFS is quite possibly the most massively over-hyped, over-designed and resource wasting piece of software ever. Anyone who says different is just a deluded cool-aid drinker. Totally over the top in almost every conceivable situation, but it gives nerds hardons so slashdot loves it.

Re:It isn't features, it is stability + features

[Tough Love]

comparing ext4 to zfs is like comparing a pinto to a corvette

Great, a car analogy, just what the world needed. How do you incorporate the fact that Ext4 is in general, faster than Zfs?

Re:It isn't features, it is stability + features

[guruevi]

Please explain further. The ZFS codebase is very simple and lean compared to some other bases with similar features (look at BTRFS or EXT4).

BtrFS already has double the amount of code than ZFS does and it's not even feature-complete or competitive. XFS has even more lines.

If anything is over-designed, I would say BtrFS is.

Re:It isn't features, it is stability + features

[dbIII]

Is the linux ZFS implemention more stable than its BTRFS implementation?

Currently yes, it only gets criticism because it's a bit behind the other ZFS ports.

And if so, does it matter for the 99% of us who are fine with EXT4

Since you can have ext4 on top of ZFS and just use ZFS as RAID it may, but it's probably easier just to have one or the other.

Re:It isn't features, it is stability + features

[Anonymous Coward]

Not the original poster, but... ZFS: To Dedupe or not to Dedupe.... The short and long of it? If you want to keep your dedup table in RAM for a 2TB drive? You might need ~12GB of RAM just for the table. Sure, it doesn't have to remain entirely in RAM but you can imagine how slow that can be potentially...

This is, btw, no means of dismissing ZFS or claiming BTRFS is better. BTRFS currently sounds like a mess as far as, you know, that all important "not losing data". But clearly there's some parts of ZFS that could use some work upon being actually usable. :/

Re:It isn't features, it is stability + features

[Anonymous Coward]

If getting from point A to point B is the task, then a used but well maintained Ford Pinto can be just as good or better then a Chevy Corvette as for exapmple, you have to consider gas mileage as a factor, in that case the Pinto may well out perform the Corvette (hey I've seen the way many Corvette drivers, drive!). What is it about ZFS that you find superior to EXT4?

Re:It isn't features, it is stability + features

[rl117]

Well, the reason for that is that the filesystems make different tradeoffs. ext4 has journalling, but no means to detect corruption of data blocks. It's faster because it's simple and less safe. ZFS is slower because it provides additional features, such as block-level checksumming or hashing, including transparent repair when the volume type allows it, and can also compress blocks. It does a whole lot more as well, and while those features are really nice, they don't come for free--there's a performance cost attached.

I use both filesystems. I use ZFS, for example at home with a RAID10 setup, for storing important data that I care about not losing. Here, I have the integrity guarantees, plus on-line snapshotting, separate datasets for different things, and the ability to zfs send my backups elsewhere. Where I care about performance but with reduced data integrity guarantees, for example scratch space for builds or large amounts of temporary data, I'll go with ext4 or ufs.

ZFS can however scale to be vastly faster than ext4 as you add more disks. Since it can stripe writes over multiple volumes (RAID sets), its performance scales up as you add more disks. While you can run LVM on RAID to approximate this, ZFS does a much more intelligent job of managing this, since unlike conventional RAID, it knows exactly which parts of the disk are used or unused, and can resilver (rebuild) an array much faster that you could with RAID.

Re:It isn't features, it is stability + features

[Tough Love]

Well, the reason for that is that the filesystems make different tradeoffs. ext4 has journalling, but no means to detect corruption of data blocks. It's faster because it's simple and less safe.

I don't know about that. Ext4 has in fact saved me countless times and never failed me, whereas the net is full of anecdotes like this where the complexity of Zfs apparently becomes its undoing.

ZFS is slower because it provides additional features, such as block-level checksumming or hashing... ZFS can however scale to be vastly faster than ext4 as you add more disks.

Where's your evidence of that? I don't know of any. Ext4 does just fine with big md arrays. Anyway, remember, this thread is about a car analogy. So... you scaled up the car by adding more wheels? Seems like we need to move on to a train analogy.

Re:It isn't features, it is stability + features

[rl117]

Ext4 hasn't let me down either. It's a solid filesystem. But it's also true that ZFS does a lot more checking and can pick up single bit errors which ext4 wouldn't (and can't) notice.

And yes, you can run ext4 on big md arrays, but at that scale you then have the issue of sanely managing the storage, and that ext4 has size and scalability limitations which are going to make it perform badly and put your data at risk. While I could (and have) layered ext4 on LVM on md, ZFS is much better from all sorts of angles. Having a unified set of tools (zfs, zpool, zdb) to manage everything, having the one system manage the disks, volumes, pools, datasets, and being able to snapshot the lot at will, with all those extra data integrity guarantees. I've used both extensively, and after having done that I'm vastly happier with ZFS than anything that I used before. As for the complexity of ZFS being its undoing, and that it did have bugs in its distant past, I've yet to encounter a single issue in ~2.5 years of using it heavily on Linux and FreeBSD.

Re:It isn't features, it is stability + features

[Tough Love]

What about my train analogy? Derailed.

SFS/FSF does not get to rule on GPL

[Anonymous Coward]

The Linux Foundation, as the copyright holder of the Linux kernel, gets to determine what is and is not allowed.

This is matter does not involve Software Freedom Conservancy (SFS) or Free Software Foundation (FSF), they are are not even parties to the copyright license used for the Linux kernel.

In my professional opinion, Canonical can feel free to tell SFS to go pound sand.

Re:SFS/FSF does not get to rule on GPL

[jedidiah]

Agreed. The SFC is not the owner of either of the works in question and is not the author of the license in question. They have no stake or standing in the matter.

As authors of the GPL itself, I would say that Stallman or the FSF do have some say in the matter.

The SFC can go pound sand...

Re:SFS/FSF does not get to rule on GPL

[slashping]

SFC only advises and warns, just like the FSF could do. Ultimately it's the copyright holders that can sue in court, and then it will be a judge or jury that decide the matter.

Re:SFS/FSF does not get to rule on GPL

[edtice1559]

IANAL but it's not required to assign the copyright to the Linux foundation. Many people contribute to the Kernel. It's GPL so you could fork it. And the Linux Foundation would definitely *not* automatically get copyright to your version. It's ultimately the Linux Foundation who could prove copyright to large swaths and take action bu the foundation could not take the kernel as is and relicense it under GPLv3 for example. They'd have to rewrite all the parts for which they don't hold the copyright.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

The intent of the authors of the GPL is not relevant in a court. Stallman is simply the author of a license, and has chosen to distribute that license for others to use. That Torvalds chose to use that license doesn't mean he adopts Stallman's intentions. It boils down to what Torvalds intended by that license.

The intention of the copyright holder does matter, and the polices and practices they have provided verbally and informally also matter. Of course in a court the default interpretation tends to be the most legal literal interpretation. But if both parties (copyright holder and licensee) have a different interpretation than what the contract literally states, it is up to a court to determine the proper action if any.

This could all be cleared up if one digs up a few emails on how to handle binary modules in Linux, then gets the Linux Foundation to agree to sign off on that as policy. If they won't prosecute licensees for using binary modules according to their arbitrary definition of what is allowed, then it is effectively legal for Canonical to do so.

Re:SFS/FSF does not get to rule on GPL

[aristotle-dude]

Agreed. The SFC is not the owner of either of the works in question and is not the author of the license in question. They have no stake or standing in the matter.

As authors of the GPL itself, I would say that Stallman or the FSF do have some say in the matter.

The SFC can go pound sand...

The authors of the GPL can have their say all they want but ultimately copyright supersedes any interpretation of a license by a third party even if that third party authored the license. They are not lawyers or judges. If the copyright holders have no problem with it then tough.

Re:SFS/FSF does not get to rule on GPL

[Olipro]

Copyright of the Linux Kernel is actually spread across all its contributors, The Linux Foundation doesn't have any overriding ownership. Though of course any particular author(s) code could be excised from it in order to make their copyright irrelevant.

In any case though, the copyright holders get to determine what licenses or permissions to grant over the work, they don't get to then decide how the license should be interpreted - that's up to an arbitrator or a court. If Canonical's lawyers are incorrect in their interpretation then someone will need to bring a court case to have it resolved.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

But it's like the SFC never read the posts that Linus has made on this subject. It's been made pretty clear that they won't sue under these circumstances.

Re:SFS/FSF does not get to rule on GPL

[guruevi]

Since when does anyone hold the copyright to the Linux kernel? Anyone that has contributed to the kernel is copyright holder as well as anyone who has 'purchased' it (even for free).

Re:SFS/FSF does not get to rule on GPL

[MachineShedFred]

They do have a stake in creating press releases, for the sake of creating press releases though.

And this article and subsequent conversation is the direct result of that.

Re:SFS/FSF does not get to rule on GPL

[Lord Crc]

The SFC is not the owner of either of the works in question and is not the author of the license in question. They have no stake or standing in the matter.

It seems they do have a stake:

The GPL Compliance Project for Linux Developers is comprised of copyright holders in the kernel, Linux, who have contributed to Linux under its license, the GPLv2. These copyright holders have formally asked Conservancy to engage in compliance efforts for their copyrights in the Linux kernel. In addition, some developers have directly assigned their copyrights on Linux to Conservancy, so Conservancy also enforces the GPL on Linux via its own copyrights in Linux.

From https://sfconservancy.org/copy...

Re:SFS/FSF does not get to rule on GPL

[slashping]

Keep in mind that there is another party.

Re:SFS/FSF does not get to rule on GPL

[fbobraga]

Man, I'm in love with you sig!

Re:SFS/FSF does not get to rule on GPL

[pr0fessor]

Canonical only announced their intention and encouraged SFC to speak out because if law suites because now is the time to find out if any interested parties are going to sue... prior to actually starting and getting sued

Re:SFS/FSF does not get to rule on GPL

[armanox]

Oracle's terms aren't violated, so there is no reason for them to sue.

Re:SFS/FSF does not get to rule on GPL

[BarbaraHudson]

No, neither the FSF nor Stallman have standing to sue. They do not hold ownership of either the kernal or zfs. They would be laughed out of court if they tried.

Re:SFS/FSF does not get to rule on GPL

[BarbaraHudson]

They would then have to prove that the license is violated of code shipped in the same ISO is a "combined work", which it obviously isn't, since they are separate files. They are no more a combined work than having a can of motor oil and an oil filter both sitting in the same box are a combined work.

(automotive analogy) :-)

Re: SFS/FSF does not get to rule on GPL

[Anonymous Coward]

Likewise, the GPL is not violated. If canonical places a licence over ubuntu as a whole distribution then perhaps that could be violated - except that by distributing it themselves canonical have given it an ok.

This whole argument is a bit of a waste IMHO.

Re:SFS/FSF does not get to rule on GPL

[pem]

As authors of the GPL itself, I would say that Stallman or the FSF do have some say in the matter.

They can say what the fuck they want, and it has whatever weight the Linux kernel devs want to assign to it.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

Oracle could find a way... even if it involved somehow bringing in the SCO Unix butt nuggets and reopening that can of fail.

Re:SFS/FSF does not get to rule on GPL

[lkcl]

Agreed. The SFC is not the owner of either of the works in question and is not the author of the license in question. They have no stake or standing in the matter.

As authors of the GPL itself, I would say that Stallman or the FSF do have some say in the matter.

The SFC can go pound sand...

from the article:

"However, our conclusion is simple: Conservancy and the Linux copyright holders in the GPL Compliance Project for Linux Developers" https://sfconservancy.org/copy...

you will note that the SFC *advises* its members who *ARE* Linux copyright holders. so your conclusion that "the SFC can go pound sand" is misleading.. the SFC has advised its members. the members - who are Linux copyright holders - willl consider that advice, and will consider whether to take legal action against Canonical.

Re:SFS/FSF does not get to rule on GPL

[lkcl]

also from the copyleft link above:

"In addition, some developers have directly assigned their copyrights on Linux to Conservancy, so Conservancy also enforces the GPL on Linux via its own copyrights in Linux."

so unfortunately you are factually wrong. the SFC does in fact now directly own the copyright on Linux kernel source code, having been assigned that Copyright by the original authors.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

A fork would be a derivative work and would have to comply with the terms of the existing license, or else it would be a copyright infringement.
Moderation: -5 Dead Wrong.
Try again.

Re:SFS/FSF does not get to rule on GPL

[vux984]

That is entirely correct.

The Linux foundation holds copyright to large parts of the kernel, but not all of it. Individuals can hold copyright to their contributions. And SFC by the sounds of it also holds copyright to some parts.

Any copyright holder could theoretically have standing in a lawsuit for GPL violation; and bring a lawsuit. The linux foundation is just one party of many who could have standing in a lawsuit.

And as you said, NONE of them could unilaterally alter the license; they'd either need permission from all other copyright holders, or they'd need to excise those holders contributions and rewrite that functionality.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

The Linux Foundation does not have copyright over the entire Linux kernel, there are a lot of individual copyrights. There is no system where you hand over your copyright to the Linux Foundation (which is a trade organization of companies, do you really want that?)

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

Yes they are in so much as you can't put the kernel under the term of the CDDL but it requires it...in distributing the combined software Oracle has standing to sue if they like.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

I don't think the Linux Foundation holds copyright to any parts of the kernel at all.

Their employees retain their own copyrights (i.e. Linus and Greg).

I don't think anyone can or has donated their copyrights to the LF. The SFC yes, but not the LF.

Re:SFS/FSF does not get to rule on GPL

[Bruce Perens]

Linux foundation is not the copyright holder. Individual developers are. A significant amount of the developers are represented by Software Freedom Conservancy.

Linux Foundation actually represents copyright infringers of the Linux kernel such as VMWare. They recently voted to keep the Linux developers off of their board because of this. They have effectively made themselves into a piracy organization.

Re:SFS/FSF does not get to rule on GPL

[lkcl]

The Linux Foundation, as the copyright holder of the Linux kernel, gets to determine what is and is not allowed.

whoa, whoa, whoa - that is flat out FALSE. individuals who write the CODE are the Copyright holders. ONLY if every SINGLE individual and corporation who has written code has EXPLICITLY ASSIGNED that copyright to the Linux Foundation (which they most definitely have not - i am a minor copyright holder and i have NOT assigned my rights to the Linux Foundation) would your statement be true.

i am kinda staggered that your anonymous comment received a +5 "insightful" when there is not a single factual statement that's been made which is correct. i've posted other replies on the other two sentences which are also wrong and factually incorrect.

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

Yes they are in so much as you can't put the kernel under the term of the CDDL but it requires it...in distributing the combined software Oracle has standing to sue if they like.

Not what the CDDL says. Stop lying already!

Re:SFS/FSF does not get to rule on GPL

[Anonymous Coward]

It is already generally accepted that LGPL and CDDL are compatible.

While the kernel is not LGPL, there are statements made by the copyright holder of the GPL'd Linux Kernel (Linux Foundation) that they allow exceptions to certain kinds of linking that don't use APIs they consider to be private. There is a whole set of macros in the Linux kernel to help document the allowable interfaces and track down license abuse.

Using ZFS would likely set the tainted flag on the kernel, and it is unlikely the community would react to your bug reports. This would make the distro vendor (canonical) responsible for the bug reports. But they have already been the front line in accepting kernel crash reports from users of their distro so this is not likely a big deal to them.

Go die alone.

[Anonymous Coward]

How could someone make a movie just look like shit all of the sudden?? And why????

Linus doesn't object to binary kernel modules

[edtice1559]

Regardless of the legal situation, the holder of the copyright to the vast majority of the kernel doesn't object to binary modules. There are those who would argue that even certain user-space binaries are a GPL violation. Fortunately those voices haven't prevailed. I guess, at some point, to squelch any doubts, ZFS will have to get turned into something akin to a hybrid driver where code to load interchangeable binary blobs with a certain interface is contributed to the kernel under GPL and the blobs come with their own license. In this case it will be CDDLv1. This is somewhat of a PITA but it's actually good in the medium-term because it means that users face less ambiguity in terms of knowing under which license they are receiving code.

Re:Linus doesn't object to binary kernel modules

[Creepy]

Yeah, I'm not sure if they really have any grounds to stand on, but it really depends on how you look at it.

If you had this in application space, it would be something like this
GPL_shared_library + Non_GPL_shared_library = GPL application
The GPL application in this case would be the kernel, but just because the kernel is GPL, the Non_GPL_shared_library doesn't have to be. In application-space that is a completely legal setup and doesn't violate the GPL (copyleft only applies to the application). What they are arguing is that the kernel objects are not like shared libraries, they are part of the kernel and therefore they have to be GPL. You could make an argument from either position, but I personally agree with the shared library view.

Re:Linus doesn't object to binary kernel modules

[Anonymous Coward]

Ouch. If by "Non_GPL_shared_library" you mean "library using license terms incompatible with the GPL" then you are flat out wrong, it would be GPL violation.
Why else would the GPL have the system library exception part?

Hilarious

[Anonymous Coward]

>The FSF, stewards of the GPL, have stated many times over the past decades that they believe there is no legal distinction between dynamic and static linking of a C program and we agree.

The FSF is almost certainly wrong in this. Linking isn't going to even be considered in a court room. What they *will* consider is whether or not zfs.ko contains derivative work from Linux. Headers are now subject to copyright (as is my understanding) so it's very possible that it does. However, if they use "api-only" headers, such as posix stuff and it does not rely on being compiled against the GPL-ed version of the headers, then they are most likely going to be free and clear of any issues.

Personally, the SFC can go eat a bag of dicks and die in a fire.

Re:Hilarious

[edtice1559]

I wish you had posted this in a way that somebody could moderate it up. The headers are subject to copyright but there is a way to make hybrid kernel modules. Doing so is tedious and the result, from a technical standpoint, is something almost identical. But it creates a clear line where one license ends and the next begins and that's valuable.

Re:Hilarious

[BarbaraHudson]

Don't forget, they also consider php scripts that call each other as "linking", even though there is absolutely no linking stage - just an interpreter loading data. They truly are masters of the over-reach.

Remember what Stallman said about Jobs - ""I'm not glad he's dead, but I'm glad he's gone." Let that be Stallman's epitaph.

Re:Hilarious

[Creepy]

With the GCC runtime exception I seriously doubt anything in zfs.ko has to be GPL as compiled. The only way the argument works is if a kernel object (ko) is considered part of the kernel vs being a shared library. In the "kernel" case, SFC is right. In the shared library case, Canonical is right. I personally agree with Canonical - if it looks like a duck and it flies like a duck, it's a duck.Kernel objects look and behave exactly like shared libraries in that the kernel doesn't need them to function, only to extend functionality, which is exactly what a shared library does.

Re:Hilarious

[Anonymous Coward]

They could call 'linking' whatever the hell they want, even ioctl calls from userspace.

Re:Hilarious

[Anonymous Coward]

Uh, you are arguing the wrong way round.
If the kernel modules absolutely and without doubt need the Linux kernel to work, function or otherwise be useful that likely makes it a derivative work, and thus it probably must be distributed under the GPL. One way out is that of course if you don't distribute the Linux kernel, at least not together with it, then the license of the kernel has no impact (for development/debugging the GPL allows you to do whatever you want).
I can't quite figure out what if anything would change if the kernel actually needed those modules. But my gut feeling is that it really doesn't make a difference at all.

Re:Hilarious

[Anonymous Coward]

Where does the article suggest zfs.ko is a derivative work of Linux?

The article describes distributing the combination of Linux and zfs.ko as a combined work is a derivative of Linux. It's the Ubuntu Linux kernel distribution with zfs.ko that violates the GPLv2 (and CDDL as stated) that disallows distributing both together.

Re:Hilarious

[Anonymous Coward]

I keep hoping there will be some court decision that decides once and for all that dynamic linking does NOT make a module a derivative work - at least if it only adds capability to the program, and the program does not rely on it.

Static linking is different; it's obviously building a "single program" for the purposes of the GPL. Dynamic linking could be interpreted broadly enough that any program that talks to the kernel at all would be a "derivative work" even if it has nothing to do with kernel functionality.

What's the Conservancy's role in this?

[bradley13]

Reading their piece, they sound like they are lawyers looking for a client, like they want to take Canonical to court over this. One likes to think that anyone who defends OSS must be on the side of the light, but I note that the Conservancy sued companies using Busybox over the objections of at least some of the copyright holders. This quote is kind of interesting:

"But that didn't stop them from creating a self-funding legal machine where they never found any actual useful code that should have gone upstream, but they still demanded $15k or so in legal fees each time so they could go sue the next company."

For those who didn't RTFA, the core of the question is this: Is a Linux distro that contains a ZFS module a combined product, derived from Linux and/or ZFS? Or are they two independent items that happen to be delivered in the same package. The Conservancy states "we have yet to encounter a Linux module that — when distributed in binary form — did not, in our view, yield combined work with Linux", and then they go on to say that the intend to "exhaust every diplomatic option...before seeking resolution from the courts".

Only: nobody is asking them to take anyone to court. If Oracle doesn't like this (they are the primary holder of ZFS copyrights), I'm sure they have their own lawyers. So WTF is going on here? Instead of a patent troll, maybe we have a GPL troll?

Re:What's the Conservancy's role in this?

[danbob999]

they are asking Oracle to release ZFS under GPLv2 (or anything compatible)

Re:What's the Conservancy's role in this?

[BarbaraHudson]

Of course they're trolling - it's good for donations.

Re:What's the Conservancy's role in this?

[brambus]

Given that Oracle hasn't released ZFS even under the CDDL, SFS knows full well this has a snowball's chance in hell of happening.

Re:What's the Conservancy's role in this?

[dbIII]

I think their aim is to try to get software under rival licences under theirs instead.
That was ultimately behind the gtk VS qt (gnome VS KDE) storm in a teacup and what was telling is the fight didn't stop when the complaints were addressed and the qt licence changed, the fight continued until the rival licence was abandoned and the GPL was adopted.
So I think it's more of the same of that tired old fight.

A grey area Oracle might exploit

[evolutionary]

Oracle has been looking to profit form the free labor from open source software advocates ever since it was losing clients to MySQL which they later purchased. The original developers were so concerned they created MariaDB. And from this LibreOffice sprung to protect "OpenOffice" and after realizing Oracle was never going to make money on OO They give it to the Apache Foundation. Oracle has not had a history of playing nice with the open source community and their development on MySQL and related products has not exactly been dedicated. Suspect they are only doing it because it was a condition for the EU to allow the European portion of the take over of Sun Microsystems. Why hasn't Oracle just put put ZFS as a fully open GPL2 compliant code for all to use and improve? Probably, like OpenOffice, to see if they can exploit it to their benefit in some way, which could even sabotage Linux (or at least Ubuntu). Debian has the right idea in distributing the source only of ZFS. We'll see where this goes but I think The Software Freedom Conservancy is in the right, and in cases like this far better to err on the side of caution. If everyone lobbied Oracle, we could get a proper and compatible ZFS license that oracle couldn't do a poison pill or "about face" on later. Why not give that a try.

Re:A grey area Oracle might exploit

[DamonHD]

I'm afraid that I don't see GPL(2) as "fully open" because it deliberately nobbles some (to my mind entirely legitimate) uses of the software, as in this whole issue in fact. Which is why I prefer to license under Apache 2 (or BSD, etc) to give users of code I create (or pay to be created) as wide a range of choices as to what they do with it as I can.

Cue GPL vs permissive licence wars... 3... 2... 1...

The point here is the GPL2 is not some global universal agreed maximum of "freedom". It is a local maximum for some users and use cases at best.

Rgds

Damon

Re:A grey area Oracle might exploit

[armanox]

Except in this case the GPL is the restrictive license, not the CDDL. Oracle has no grounds to file suit since their terms aren't being violated. It's that pesky GPL getting in the way, since CDDL isn't their brand of 'free' software it must be evil.

Re:A grey area Oracle might exploit

[Anonymous Coward]

Oracle still own the rights to Virtualbox and so far, they haven't done anything to it but let the project continue as it was before Oracle.

Re:A grey area Oracle might exploit

[Anonymous Coward]

GPL has no bearing on how you use the software as this isn't a software usage license. The GPL is a software conveyance license that governs the terms at which you convey the licensed software to another party.

Using Linux?!

[Anonymous Coward]

Using Linux?! even after systemd? why?
oh... nevermind... FreeBSD is going to switch to using NOSH (systemd for BSD)... using Linux makes sense then.
I answered my own question :)

Don't forget...

[Anonymous Coward]

Don't forget to pay your $699 licensing fee.

GODDAMMIT TIMOTHY!

[Anonymous Coward]

License violations are not illegal. The word illegal does not appear anywhere in the story. What the fuck is it that you do here, other than fuck up stories?

Re:GODDAMMIT TIMOTHY!

[Anonymous Coward]

License violations are not illegal.

In and of themselves, no. However, the license dictates the terms under which you can distribute something. If you're not abiding by the license, it's copyright infringement if you distribute it, which most definitely *is* illegal. You're splitting hairs.

*tableflip*

[Anonymous Coward]

This is the kind of bullshit that has kept me away from any kind of involvement in open source/free software etc.
This kind if pointless shithead quibbling.

So can I use CD drive?

[Anonymous Coward]

Sermon on the mount

Re:So can I use CD drive?

[BarbaraHudson]

mkdir /home/sermon
mount /dev/sdb1 /home/sermon

Re:So can I use CD drive?

[omnichad]

That's a mount on the sermon.

Re:So can I use CD drive?

[Anonymous Coward]

# mkdir /home/sermon
# mount /dev/sdb1 /home/sermon
mount: /usr/home/sermon: No such file or directory
# mount /dev/ada1p3 /home/sermon

It's gray area

[Anonymous Coward]

If user compiles the shim layer, they're violating the copyright. This is no different from Nvidia and ATI/AMD binary blobs.

Re:It's gray area

[Anonymous Coward]

Umm, no. The GPL covers distribution of binaries, not use of source code.

Software Freedom Conspiracy

[wjcofkc]

I've read both arguments, and since I know everyone posting here has also (right?), I won't bother with too much of a critique.

I have to side with Canonical on this one. Their short and sweet post pretty much puts to bed the legal ramblings of the Software Freedom Conservancy. The SFS article reads like they started out trying to make a point, then has to fallback on an academic lesson on GPL Incompatibility and Combined/Derivative works. I would go so far as to say their verbosity defeats their own argument, while managing to inset the letters "ZFS" wherever they can find room.

Past that, as someone who uses ZFS on FreeBSD, ZFS is pretty fucking useful. Canonical's legal inclusion of ZFS as self-contained file system module is a big step forward for Linux.

Re:Software Freedom Conspiracy

[Anonymous Coward]

Funny that you side with Canonical based on reading both posts. I also read both posts, and I have to side with the SFC; their post backed up their arguments with background information while Canonical's post just disagreed with the SFC and fell back on sloganeering.

Re:Software Freedom Conspiracy

[Anonymous Coward]

Too sad you totally forget about the netapp patents that ZFS violates, or that Oracle could easily find a way to say that their patent grant to you does not apply either, and sue you.
Not that Oracle has sued anyone before of course.

Re:Software Freedom Conspiracy

[Anonymous Coward]

Arguments that are incredibly boring, and basically boil down to: "You can't do this because we're trying to emotionally blackmail Sun's successor into relicensing zfs."

Meanwhile, everyone else is expected to give up some of their software freedom to make a bunch of nerdy lawyers happy.

Please STFU, Copyright is not "FREEDOM"!

[Anonymous Coward]

Maybe they should remove the word "Freedom" from their organization. Look up what freedom means in the dictionary, please. Open Source developers want unrestricted use of software for everyone, "FREEDOM", well put it in the public domain and be done with it. BSD, GPL, MIT, Apache, etc... are all copyright and place restrictions on what you can and cannot do so it's not "FREEDOM". So stop associating "FREEDOM" with restrictions.

Re:Please STFU, Copyright is not "FREEDOM"!

[Anonymous Coward]

Maybe they should remove the word "Freedom" from their organization. Look up what freedom means in the dictionary, please. Open Source developers want unrestricted use of software for everyone, "FREEDOM", well put it in the public domain and be done with it. BSD, GPL, MIT, Apache, etc... are all copyright and place restrictions on what you can and cannot do so it's not "FREEDOM". So stop associating "FREEDOM" with restrictions.

Shut the fuck up. Freedom doesn't not mean anarchy.

"Illegal" is misleading

[coldsalmon]

More accurate would be: "could give rise to a civil cause of action." We usually say that something is "illegal" when it violates criminal law, or some other statue passed by the legislature for a public purpose. Violating contract or license terms is not illegal in this sense. Any legal risk comes from the willingness of the aggrieved party to pursue a remedy. Crucially, there is no public stake in enforcing these rights: if the rightsholder does not want to pursue a remedy, nobody else will care. This is in contrast to activities which are prohibited by statue: the public at large has an interest in prohibiting these activities because they are bad for everyone for one reason or another. It's true that this claim could be based in copyright, which is a "creature of statue" so to speak, but copyright itself is designed to be enforced only by private parties. And more importantly, in this case the text of the license would be the determinative factor.

Anyhow, it's just a semantic niggle, but it really annoys me when people write deliberately misleading headlines like "flashing your firmware is now illegal," when they are really just talking about private causes of action based on licenses or private contracts. In fact, the word "illegal" does not appear at all in TFA.

Re:"Illegal" is misleading

[Anonymous Coward]

There are criminal parts of copyright law to do with commercial copyright infringement. If Ubuntu are selling it commercially (and they are) then it could apply.

https://en.wikipedia.org/wiki/Criminal_Copyright_Law_in_the_United_States

Re:"Illegal" is misleading

[Anonymous Coward]

It is now and always has been a criminal offense to willfully breach a contract.

Insane!

[fbobraga]

Insanity like this one makes the free x opensource software fight...

is it a derived work?

[bradleykuszmaul]

TFI says: "The FSF, stewards of the GPL, have stated many times over the past decades that they believe there is no legal distinction between dynamic and static linking of a C program and we agree. Accordingly, the analysis is quite obvious to us: if ZFS were statically linked with Linux and shipped as a single work, few would argue it was not a “work based on the Program” under GPLv2." And that's where it all falls apart. How do we know that it's not a "mere aggregation". See http://www.gnu.org/licenses/gp... which argues that determining whether a separate module is a covered work is actually a tricky question requiring some thought and analysis.

Re:is it a derived work?

[BarbaraHudson]

There's a hell of a difference between static linking (the code is included in the final binary) and what we refer to dynamic linking, which is code we can optionally load into memory, executed, and then dumped, mix, lather, repeat. That same code can also be loaded by other programs, so it certainly is not part of the first one to load it.

But then again, what do you expect from a couple of lawyers?

Re:is it a derived work?

[Anonymous Coward]

Actually there's even more difference here. A kernel module, is dynamically linked, but that is more like running a program with another program, than the dynamic linking of say zsh against libpcre. The kernel binary can run just fine without zfs.ko, just as it can run just fine without /usr/bin/zsh. We don't say that zsh is a derivative work of the kernel because it runs on the kernel ABI, so why would zfs.ko be a derivative work for the same reason.

The whole argument the FSF has saying dynamic linking is as much a derived work as static linking is as thin as "they both have linking in the name", because by any other measure it's possible to create and distribute a program that dynamically links with a library that is no way derived from the library, and can be distributed (though is unusable) without that same library.

I think Canonical are in the right here, but I think they could make it even less ambiguous by creating a "free as in not telling other people what they can and can't do" licensed API and ABI and loader, hack zfs to be compatible with that loader, and then add that loader to Linux. I think this is unnecessary and people should just stop being so penisy about software licenses, and focus on improving people's software freedom in other ways: like making the software easier to hack on, making it easier to user and get things done.

The thing that grinds my gears, is that there are all these OSS Linux distributions, but they all make it hard as nails to get the source, debug symbols, debug it, modify it and install the modified binaries. The latest thing is Canonical and Redhat pursuing Secureboot which will make it impossible for the user to replace their own kernel: because freedom (yea right).

Compare that to the BSDs, which aren't "free as in force other people to be free through state sponsored violence", where to change the software it's as easy as opening emacs, browsing to /sys/src (or /usr/ports/whatever),making a change and typing m-x, compile.

Aside from a name and a ring, how is a ko any more a derived work than /sbin/init? They're both *separate* executable code that is copied to memory by the kernel and then execute.

Re:is it a derived work?

[Anonymous Coward]

You should learn to read before spouting inane bullshit.

zsh runs in userspace, there is an exception in the kernel license for userspace applications. There is no exception for modules.

But hey it's easier to be a retard than read the fucking licenses.

Re:is it a derived work?

[Anonymous Coward]

Why would I waste my time with inane bullshit like licenses.

I prefer to exercise my software freedom and actually get shit done, rather than censuring my own software freedom due to the wording of some dumb licenses. This is the fucking disease in the free software community right now. People who are more concerned with autistic interpretations of licenses and would rather be lawyers and pedants than actually move things forward with technical improvements, and solving some of the problems that are actually relevant to users.

there is an exception in the kernel license for userspace applications.

Also I call bullshit on that. What exact license file has that exemption text, what is the exemption text so I can grep for it.

Re:is it a derived work?

[Anonymous Coward]

Yes, there is a big difference between dynamic and static linking, and this is acknowledged by having two different licenses, the LGPL which allows dynamic linking, and the GPL which does not. Guess what... The Linux kernel is under the GPL, not the LGPL.

The intent of putting the Linux kernel under the one that does not allow dynamic linking should be clear even to a lawyer or judge.

Re:is it a derived work?

[BarbaraHudson]

I'm tired of replying to bullshit arguments from anonymous cowards. Life is too short to waste on ignorami. Want a real response - go to the effort of creating an account.

Glad...

[Anonymous Coward]

I run Windows... I just pay the mindless overlord in Redmond to keep my life happy.

New Moderators wanted anyone?

[williamyf]

The summary dedicated 3x more space to a previous Canonical blog, than to the actual article being discussed in the Headline. Way to go!

btrfs

[spectrum-]

ZFS can be replaced by btrfs in many cases. Yes ZFS is more mature and has more of a track record. But only by people using btrfs will it gain that level of testing in production environments. It is quite stable now and OK some may have data that they just can't bring themselves to trust away from XFS or ext4 or ZFS but I think it's time to look at alternatives to ZFS.

Re:btrfs

[Anonymous Coward]

Yeah you could replace zfs with btrfs in some situations. However, once they have experience with ZFS, you'll see that most people become reluctant to make such a move.

Re:btrfs

[Anonymous Coward]

btrfs doesn't duplicate all the features of ZFS.

This is because the Solaris kernel is different from the Linux kernel in that it has different abstractions, and ZFS duplicates a bunch of features that in Linux are expected to happen across an abstraction boundary from the filesystems.

Re:btrfs

[Anonymous Coward]

Yeah! Let's use an unstable, unproven, experimental filesystem that's trying to be a copy of another open source filesystem instead of the one they try to copy! That'll fit right into a production environment! I can't wait to lose all my data!

You're a fucking moron. btrfs has no reason to exist nor is there a reason to use it over other, more open filesystems like ZFS.

Re:btrfs

[rl117]

I've tried them all many times, and thrashed them all. Btrfs trashed its filesystems on more than one occasion. How many times do I get seriously burned by it before I say "enough is enough", and turn my back on it. Well, about four times to date. Then I discovered ZFS.

ZFS rules. I've used it on Linux and now for the last two and a bit years on FreeBSD as well. It's everything Btrfs should have been, and much more. It works. It's robust. It has actual documentation and tools that work as documented. On the type of systems I use, it's slower than ext4 (the robustness doesn't come for free and I don't have huge striped arrays to make it super fast), but still faster than Btrfs by a good margin. It hasn't eaten any of my data, unlike Btrfs. In my book, it's doing everything I want from a filesystem. I can trust my data to it, and I can be confident in my ability to administer it safely.

I won't be trusting Btrfs again anytime soon. I'll leave that "experience" to the masochists who want to get their fingers burned.

Re:btrfs

[allquixotic]

Anyone who trusts their data to ext4 is grossly misinformed and playing with fire. Almost any other filesystem with a journal is less likely to eat your data than ext4. If you're using ext4 it should be for data that you really don't care if you lose / if the system running it crashes until you have a chance to reboot. In other words, it's fine for laptops where people check mail and browse facebook, but it has absolutely no place in the datacenter.

Re:btrfs

[dbIII]

but it has absolutely no place in the datacenter.

You can boot off it, use it for scratch storage, use it for an OS disk, keep software on it - a lot of stuff in datacentres does not need to last forever or be spread over a raidz2 array or whatever. You comment is about as over the top as saying that there is no place for a cheap keyboard in a workplace and we should all have an IBM Model M.

I use ZFS a lot but I'm not so blinkered as to think it has to be used everywhere even in situations where it really does not matter.

Oh so Sandler again?

[Anonymous Coward]

Yeah I'm not paying that fucking idiot any attention and neither should anyone else.

Enforcement of GPL is a gain

[jbn-o]

Copyright holders who license under a free software license gain compliance and social respect for choosing a license that lets us run, share, and modify covered software. There is nothing to gain in saying popular GNU/Linux distributions don't have to comply with licenses but other distributors do. There is also a lot to gain by showing that licensing under a popular free software license (such as the GNU GPL) is enforcible. One such benefit: Every enforcement action taken showing how enforcible the GPL is helps silence critics who claim otherwise.

Re:Enforcement of GPL is a gain

[AvitarX]

The general (not unanimous, and untested in courts) opinion is a binary linked to GPL code is required to adhere to the license.

This isn't bundling a stand alone binary, it is a binary linked to the Kernel.

Re:Enforcement of GPL is a gain

[Anonymous Coward]

It's not linked to the kernel.
It's loaded by the kernel at run-time.
Userspace applications are also loaded by the kernel at run-time

Ok, it can't run on its own and needs the kernel to load it for it to run - but userspace applications also can't run on their own, they're written to make use of a kernel API, and need the kernel to load them in order to run.

The only difference between a kernel module and a userspace program is really an esoteric techincal one. Not one that should matter in a legal sense.

Re:Enforcement of GPL is a gain

[bug1]

The difference is using internal API vs external API.

If its using an internal API its part of the kernel.

Re:Enforcement of GPL is a gain

[AvitarX]

NOTE! This copyright does *not* cover user programs that use kernel
  services by normal system calls - this is merely considered normal use
  of the kernel, and does *not* fall under the heading of "derived work".
  Also note that the GPL below is copyrighted by the Free Software
  Foundation, but the instance of code that it refers to (the linux
  kernel) is copyrighted by me and others who actually wrote it.

Maybe the court sees it that way, maybe not. Only one way to find out, and we shall soon I suspect.

Re:Enforcement of GPL is a gain

[Anonymous Coward]

Why? That's just an arbitrary distinction. There's no _technical_ difference - certainly nothing that matters from a copyright point of view.

Re:Enforcement of GPL is a gain

[bug1]

Exactly...

And there is a file called syscall.h that i believe defines "normal system calls". If thats the only link between the kernel and another program then its all good.

Re:Enforcement of GPL is a gain

[Anonymous Coward]

First, not only technical differences matter.
Second, there is a huge technical differences.
The APIs the Linux kernel provides to user space programs not only can but in fact are implemented by others, for example the Linux compatibility layer of FreeBSD.
That is a core point that could be used (though I don't think there is a legal precedent) to distinguish between "coded to follow an API" (we'll see if Oracle ruins that for us though) and "coded against and only usable in conjunction with another specific program and thus a derivative work".
I am sure there is huge grey area in-between those though.

mod uP

[Anonymous Coward]

from within. study. [rice.edu] this is consistent Writing is on the

It is no different than Minix AFS

[mlwmohawk]

Historically, there's been things like the original Andrew filesystem
module: a standard filesystem that really wasn't written for Linux in the
first place, and just implements a UNIX filesystem. Is that derived just
because it got ported to Linux that had a reasonably similar VFS interface
to what other UNIXes did? Personally, I didn't feel that I could make that
judgment call. Maybe it was, maybe it wasn't, but it clearly is a gray
area.

Personally, I think that case wasn't a derived work, and I was willing to
tell the AFS guys so.

http://yarchive.net/comp/linux...

ZFS was clearly developed for a different operating system, and I don't think Linus would care. If he does, I'd like to see something he has written on the subject.

Unless there is a copyright holder with reason and "standing" to sue, there is no violation.

Poor Oracle!

[pem]

From the SFLC post:

We are also concerned that it may infringe Oracle's copyrights in ZFS.

IOW, Oracle is just like a woman who is weak and lacks agency and needs an SJW to take up her cause and be offended on her behalf.

So what?

[Anonymous Coward]

Almost all of the code isn't his.

Lawyer: Linux is not *quite* GPL

[hawk]

I am a lawyer, but this is not legal advice. If you want legal advice, pay my retainer.

There is a *really* big hole in the analysis.

Linux is *not* quite GPL; it, like many others, is better understood as "quasi-GPL", or QGPL.

Since pretty much the beginning, the Linux developers approved, condoned, and encouraged binary, non-GPL modules. Standard legal analysis means that this trumps the boilerplate of the license/contract.

The second serious error is arguing about the FSF position on linking. Under the rules of legal analysis, the author of a document's opinion is weighted at pretty much nothing: the author had his chance, and later comments are irrelevant. That is, there are about 7 billion people whose opinions on interpreting it come first.

Now whether distributing Linux with that module violates Sun's CDL could be an entirely different issue; I've never looked at it.

hawk, esq.

Re: Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

Go fuck yourself, lawyer scum. Your whole profession is an invented need by people who don't want to produce anything, but be parasites in the rest of us.

Live contamination free or die

[Anonymous Coward]

GPL/Copyleft advocates can say this is illegal; but must also admit that fears of contamination are well-founded. Alternatively, they can permit proprietary modules and continue to argue (as they usually do) that fears of contamination are not well founded. Any other combination of arguments is hypocritical, inconsistent, whatever you want to call it.

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

"I am a lawyer, but this is not legal advice. If you want legal advice, pay my retainer."

The software world could learn a lot from that statement. Of course, if we acted like lawyers we'd have no Open Source or GPL to argue about in the first place.

"I am a software developer, but this is not software I'm writing for you. If you want software, pay my salary/consulting fee."

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

Yes and no.

"the Linux developers approved, condoned, and encouraged binary, non-GPL modules". You could argue that Linus has known all along that Linux distributions included binary, non-GPL modules (say NVIDIA video drivers?) They've known products were being distributed that aren't under the GPL license.

But no one is arguing that these products (NVIDIA) SHOULD be under the GPL license. So...no harm, no foul?

Flip side of the coin. Other products (NVIDIA) isn't arguing that the Linux Kernel should be invalidated against the GPL. (In theory, they could...but that wouldn't grant them a license to use the Linus Kernel. They could argue that their product (NVIDIA) can't be distributed with the Linux Kernel however. They could try to argue damages - but that would be pretty difficult given that they give product away for free as well.

So...now - let's look ZFS. Licensed under CDDL. Given away fro free.

SFC is arguing that ZFS owners COULD argue that giving away ZFS with Linux would violate the GPL. Yeah....so what? At that point Canonical apologies and pulls ZFS.

Damages? Maybe. I'm not sure I see how...but if Canonical is up to take the risk...the OWNER of ZFS could only attack Canonical for their actions. They CAN'T get any additional rights to Linux...Canonical doesn't own 'em.

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

The thing is "Linux" isn't that of a single party and the copyrights are held by many. I'm not sure I'd jump to the conclusion that there isn't a legal argument here. Though I'm not a lawyer- but lawyers often have different legal opinions too so....

I also think you jumped to the conclusion that they actually thought that the author of the license opinion matters. It may not, but that doesn't make the author legally wrong either.

Obvious Solution

[laing]

All Canonical needs to do is to is to distribute the ZFS module as source-only, and have their install scripts build and install the binary module during the OS installation process. Doing this should get around the Oracle licensing restrictions and allow distribution with the GPLv2 kernel (and GPLv3 GNU utilities).

Re:Obvious Solution

[Anonymous Coward]

Why are you pretending that the restrictions comes from the CDDL? It's not the case. It's the GPL., or rather a controversial interpretation of it that makes this situation polemical.

The CDDL cannot apply to the Linux kernel

[tetraverse]

"The CDDL cannot apply to the Linux kernel because zfs.ko is a self-contained file system module — the kernel itself is quite obviously not a derivative work of this new file system." ref

Re:Lawyer: Linux is not *quite* GPL

[bug1]

Since pretty much the beginning, the Linux developers approved, condoned, and encouraged binary, non-GPL modules. Standard legal analysis means that this trumps the boilerplate of the license/contract.

I am NOT a lawyer, and i know better than to make very stupid statements like that;

- Non GPL modules where not tolerated in the beginning, the issue didnt come up until 10 years after the project started. (Linux wasnt created the moment corporations notices)
- Can you cite an example where there has been any sort of consensus of developers have encouraged non-GPL or binary modules ?
- Tolerating copyright violators does not invalidate copyright law in any way, it might justify lower penalties.
- In what jurisdiction do you claims apply ?

Sigh

[Anonymous Coward]

I really wish the GPL zealots would stop trying to limit the freedoms of those of us who don't subscribe to their particular brand of fundamentalism.

Re:Sigh

[Anonymous Coward]

Sounds like you want the 'freedom' to force others to use a license which suits you so you can do what you like with *their* code.

You're free to fuck off and write *your own software* with any license you choose.

The SFS is now liable (TPP)

[Anonymous Coward]

Under the provisions of the TPP, the SFS is now guilty of enabling violations of copyright, as they espouse modifications of copyright held by parties != themselves.
Nicely done SFS! Happy prison time!

Re:The SFS is now liable (TPP)

[nomentanus]

Oy veh. The owner can modify the terms - see the specific case law cited by the article.

Re:The SFS is now liable (TPP)

[Anonymous Coward]

Specific US case law is now irrelevant because of the TPP. The relevant law is now International Treaty, UN Regulations, and other international restrictions.

The GP is correct. SFS has now encumbered themselves with a violation of international law.

Re:Lawyer: Linux is not *quite* GPL

[amorsen]

Alan Cox has made clear from pretty much the moment that this issue came up that he considered binary modules to be derivative of the Linux kernel. While he has AFAIK not taken this view to court, he has certainly neither approved, condoned, nor encouraged binary, non-GPL modules.

Ubuntu kFreeBSD then?

[Kryptonut]

If it's more important to try and push an open source but restrictive license down everyone's throat (hey, the GPL is great for some things, but horrible for others), then perhaps Ubuntu could come up with a variant that uses Debian's kFreeBSD distribution as a base instead?

It's a shame though, I thought the general spirit was to make great things available to the masses and to drive innovation.

Re: Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

You're a dipshit.

The profession was "invented" because a few hundred years ago most people were completely uneducated, frequently didn't know their rights, and were unable to read them IF they managed to get access to them. A lawyer's job was to put those people on equal footing with the remainder of the judicial system.

Default Install

[EEPROMS]

From what I can see as long as the default distribution install is GPL compliant anything else is a user selectable customisation. For example cars that meet strict design regulations when sold by a dealer will often if prompted by the buyer modify the car that often leads to the car not comply with the regulations the car was certified under (just changing the exhaust can do this).

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

I mostly agree with your assessment. I believe NVIDA has a pass because they use a totally standard kernel interface to do everything (I thought in the posts where Linus discusses it, it was read/write). So they developed an entirely independently and then pass data back and forth through a stable API. Otherwise, it includes now Linux source code.

This is very different than other kernel drivers which are tightly hooked into the kernel and use lots of kernel functionality and APIs (like the days when the Hypervisor wasn't distributed with mainline kernel source). I believe this standard, and Linus's assertion that he doesn't consider it a derived work are what makes NVIDA special. The question is does ZFS meet the same standard. In this case, I know the code was developed entirely independently, but it isn't clear to me how cleanly separated it is from the rest of the Linux internals.

Re:Lawyer: Linux is not *quite* GPL

[Tough Love]

Since pretty much the beginning, the Linux developers approved, condoned, and encouraged binary, non-GPL modules.

Being a lawyer doesn't make you out of touch. "Linux developers" have never approved of or encouraged binary, non-GPL modules, quite the contrary. Tolerated, or looked the other way is the most you could say. Linus has at times expressed a permissive attitude, but he has also said this. What you need to remember is that it's not Linus's kernel, it's only partly his. Anyone who holds copyright in it (thousands) is entitled to make an issue of illegal distribution if they wish.

Re:Lawyer: Linux is not *quite* GPL

[hawk]

It's definitely a mess, but so is just about anything where the GPL has been around.

As for authority and jurisdiction: the Common Law of England goes back to the twelfth century, and has been passed on to substantially all English speaking countries (I forget the exceptions). The principles of construction predate this country, and are pretty much the same through the english speaking world.

Frankly, if someone wanted to litigate this, it would be an utter mess. The unwritten changes that *did* become part of the license would be binding upon all later contributions, and attach to them. It is quite possible that different parts have different licenses--and that the whole body of the kernel couldn't be distributed together. *noone* wants to open *that* can of worms . . .(except maybe redmond :)

hawk

Want ZFS without whiny GNU Freetards?

[Anonymous Coward]

Just run FreeBSD or PC-BSD. Then you won't have whiny prepubescent Freetards whining that everything included isn't free as in communism adhering to "the one true license". A separate kernel module is not the same as compiling it inline in the kernel. This type of crap is one reason I avoid the Linux community. The other reason is that most Linux distros are a sloppy patchwork fragmented mess in perpetual beta and BSD is just so much cleaner.

Re:Lawyer: Linux is not *quite* GPL

[hawk]

Also, while I'm at it, to the extent that some invited, while others tolerated, aside from introducing different licenses with the problem that that creates, leads to the issue of "estoppel"--a situation in which one cannot assert a position, even if legally entitled to do so, do to his prior actions and/or the reliance of another upon those actions. (and for those who care, estoppel is an equitable principle, not a legal principle, having come from the Chancery Courts of England).

All in all, anyone who thinks that they would like the results of the litigation is deluding himself . . .

hawk

Derivative work matter

[Anonymous Coward]

It will be a length battle if this case go to court. I think Canonical have more to gain than SFS
The matter of non-GPL binary loadable kernel modules is of big importance to any Linux distributor.

There are multiples examples of this usage on Linux, especially graphic binary blobs (NVIDIA, AMD,etc).
I don’t think, arbitrary, any binary module can be argued as a “derivative work” of Linux kernel, it will be very difficult to hold this argument on court.
The ZFS module has been ported from Solaris not developed from scratch for Linux, arguing it is derivative is not obvious (the same apply to NVIDIA and AMD graphic drivers, ported from Windows).

The interpretation of FSF or SFS is mere one of many. Only in a court this interpretations will be sedimented.

What the what?!

[Anonymous Coward]

ZFS is the killer app for the BSD folks... If it gained widespread distribution you could kiss BSD goodbye... Ok that's an exaggeration, these SFC loonies need to go check in at the front desk. Who let them out of their cage??

Re:Lawyer: Linux is not *quite* GPL

[shutdown -p now]

"Legal advice" is a formal term, defined by law, and it's actually illegal to offer it if you're not a lawyer in many places. Furthermore, for a lawyer, giving such advice can be constituted as establishing an attorney-client relationship, which results in legal obligations for the lawyer. So, even ignoring the time and effort, it's never free for them. I would assume there is some kind of insurance that exist specifically to indemnify lawyers in cases where they become subject to such obligations; and, naturally, that insurance is not free.

Software developers, on the other hand, don't establish this kind of relationship with the user merely by virtue of writing software for them. At least, not any more so than anyone else providing any other service does.

(I am not a lawyer, and this is not legal advice.)

Effing GPL

[Anonymous Coward]

I can read a binary as well as a CPU can, so why _must_ I have the source? In any case, I don't want the GPL dictating what I can and can't run on my computer, and disrupting distribution is the surest way to do that.

Re:Lawyer: Linux is not *quite* GPL

[Tough Love]

You seem to be unaware that in every case where the GPL has been litigated, it has been upheld. You talk about common law, surely you understand how this rich body of legal precedent has buttressed the strength of the GPL. As for estoppel, it's not enough for some of the copyright holders to make a habit of looking the other way. All of them would need to, and that simply hasn't happened.

Re:Lawyer: Linux is not *quite* GPL

[Tough Love]

...you could try relying on the statute of limitations and hope that nobody notices some quiet little misdeed for a sufficient number of years...

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

I am a lawyer, but this is not legal advice. If you want legal advice, pay my retainer.

Can I just shoot you int he fucking face instead?

Re: Lawyer: Linux is not *quite* GPL

[Entrope]

So explain EXPORT_SYMBOL vs EXPORT_SYMBOL_GPL.

Same kernel developers recognized that there are standard parts of how loadable modules interface to the kernel. Ardent wishes by others (like Alan Cox) do not set aside that reality.

If you think that's bad regarding ZFS then...

[Anonymous Coward]

Don't look at Nutanix. They are flat out embedding ZFS with KVM in their next 4.6 product release.

Re:Lawyer: Linux is not *quite* GPL

[Kjella]

Also, while I'm at it, to the extent that some invited, while others tolerated, aside from introducing different licenses with the problem that that creates, leads to the issue of "estoppel"--a situation in which one cannot assert a position, even if legally entitled to do so, do to his prior actions and/or the reliance of another upon those actions.

I find it highly unlikely that estoppel would stick to anyone but those that have done it to themselves. If I write some GPLv2 code without contributing it to the kernel, somebody else says hey GPLv2 that's compatible and add it to the kernel they got no authority to alter the license to "quasi-GPL" and whoever is relying on the "quasi-GPL" can't use their actions as estoppel to halt my lawsuit since it's not my actions. And since it's open source and free for distribution to anyone, there's no technical action or inaction to indicate permission. The only question is if legal inaction could lead to some form of estoppel, but even then it'd probably be a limitation on damages. If it's in violation of copyright, they still have to cease violating it. Unless you have some kind of promissory estoppel, but I don't think anyone has made that kind of promises on behalf of the whole project.

Fork yeah!

[epine]

Totally by chance, I'm watching an old Cantrill presentation (2011) and he'd got a slide at 24m28 which reads:

"Contrary to public claims by some ex-Sun employees, this was not done to be deliberately GPL incompatible."

Fork Yeah! The Rise and Development of illumos

He calls CDDL "file-based copyleft" and says it was explicitly designed to allow combined works with proprietary elements.

PHP programs are linked

[Anonymous Coward]

It's true that PHP include statements do not involve the merging of executable binary objects into a single native executable object. This does not change its fundamental nature that executable programs of PHP are loaded together (linked together) to form a single program. The fact that the defacto implementation of the PHP language is an interpreter rather than being a native machine language compiler does not change this. In fact language interpreters/translators/compilers are not an inherent property of any language, these are implementation specific decisions; one can write a C language interpreter and one may also write a Perl language -> machine executable code compiler if they wanted. Well the "PHP include system" loads in PHP files in the same fashion as a binary linker: the result is a full program that is loaded into memory as a single program as opposed to being dumped to disk file and a native executable object.

Re:PHP programs are linked

[BarbaraHudson]

You know something - I'm tired of replying to bullshit arguments from anonymous cowards. Life is too short to waste on ignorami. Want a real response - go to the effort of creating an account.

Re:PHP programs are linked

[Anonymous Coward]

You know something - I'm tired of replying to bullshit arguments from vanity accounts. Life is too short to waste on ignorami. Want a real response - go to the effort of replying anonymously.

Linking

[kbahey]

Reminds me of a related issue: the FSF's position on linking (which will not impact the issue at hand: ZFS in Ubuntu, but has been raised in different contexts).

Basically, WordPress allows non-GPL modules even though WordPress itself is GPL. The FSF does not like that, and they hold that to extend a GPL application, every extension must be GPL, and they invoke the linking interpretation. Drupal on the other hand, takes the position that all modules must be GPL.

The linking interpretation makes sense when you have A depend on B, and B is proprietary and you can't run A without B, or you can't inspect B at all since you don't have the source code for it. But an extension is the other way. It is not essential for the main application to run, it is optional. Also, the linking interpretation was done in the days before dynamic linking (.so) was possible, and everything was static (.a). And now, we have things like WordPress and Drupal which are written in interpreted languages such as PHP, and you have the source code already.

That linking interpretation is archaic and needs to be expanded or reevaluated.

Re:SFS/FSF/LF does not get to rule on GPL

[Anonymous Coward]

The Linux Foundation is not "THE" copyright holder of the Linux kernel. Linus has never required copyright assignments from contributors and therefore every single programmer/organization who contributed code retains copyright in their contribution. They as individuals get to decide when (or if) they want to bring legal action against potential infringers of their copyright. Your lack of knowledge of this basic fact makes me wonder what your profession is. Some of those contributors undoubtedly support SFC's interpretation.
Whether they have the time/resources to actually sue is a different matter.

Re: Lawyer: Linux is not *quite* GPL

[Tough Love]

So explain EXPORT_SYMBOL vs EXPORT_SYMBOL_GPL

Walling off non-GPL-compliant modules from a good part of the kernel is a rather direct expression of disapproval, don't you think? This in no way blesses the practices of violators, it is simply a pragmatic way to decrease their comfort. Tainting is in the same category.

So YAAL.

[Anonymous Coward]

So You Are A Lawyer. That's nice for you. But...

> It's definitely a mess, but so is just about anything where the GPL has been around. ...you clearly seem to be a biased lawyer. Now guess which grade of trust I put on biased lawyers.

Static Link = Dynamic Link? Says Who?

[softcoder]

The Conservancy seem to have glossed over this point. It is not obvious to me. If Canonical were to distribute two, CDs, tarballs, whatever, with Linux on one and ZFS.ko on the other, with a script to load ZFS.ko at install time would _that_ be different?

How about if they distribute Ubuntu on one CD, the ZFS source on the other and a script that builds and loads ZFS.ko at install time? Would _that_ be legal?

If you can distribute linux with binary blob graphics drivers I don't see a big difference with a filesystem module.

softcoder

Re:Static Link = Dynamic Link? Says Who?

[david_thornley]

The number of CDs or whatever is irrelevant. It's always been legal to supply GPLed code alongside proprietary code, as long as they don't interact too closely. Exactly what "too closely" means is one of the big issues with versions of the GPL for a long time.

Local SEO Module

[Anonymous Coward]

Really great post. This answered the majority of my questions. When I read this I actually opened up a word document and started taking notes haha. also go through this link http://localclienttakeover.com...

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

* your

Re:Lawyer: Linux is not *quite* GPL

[TheRaven64]

The problem with your analysis is that this:

Since pretty much the beginning, the Linux developers approved, condoned, and encouraged binary, non-GPL modules

Is not even remotely true. Greg KH and Alan Cox have both stated publicly that this is not the case, but more formally the kernel has an explicit set of symbol annotations for things that they consider to be public interfaces (which do not therefore invoke the GPL, because - possibly modulo the recent Oracle ruling - interfaces can not be copyrighted and so being a derived work of something that is not copyrighted can not spread the GPL, and even without this the Linux kernel license contains an explicit exemption for these interfaces). Anything that interacts with code that is not part of these interfaces is a derived work of the kernel and covered by the requirements of the GPL. Given how deeply ZFS must hook into the VM and VFS subsystems to work, it's pretty likely that it requires touching GPL'd parts.

Companies like nVidia sidestep this, because the GPL is a distribution license. They provide a binary blob, most of which is developed as part of a cross-platform effort and so is not a derived work of the kernel. They also distribute a shim. The shim is a derived work of the kernel, and the parts of the blob that talk to the shim could possibly be argued to be derived works of the shim, but the only license that would be violated if they were would be the GPL on the shim, and so the only person with standing to sue nVidia would be nVidia. The fact that they've had to jump through these hoops to do so is evidence that not doing so would not be condoned by the various copyright holders of the kernel. I challenge you to find one example of a binary-only kernel module that is distributed with the awareness and consent of the Linux copyright holders and uses anything other than the interfaces declared public.

The CDDL is a per-file license, so may be linked with code under any license without issues unless those issues come from the other license.

I am not a lawyer, but I've spent enough time with IP lawyers to have had most of these discussions before.

Re: Lawyer: Linux is not *quite* GPL

[Entrope]

No, I think the distinction expressly codifies what parts of the kernel are generic "kernel" interfaces and what parts are Linux-specific (and thus imply derivative work status). Otherwise there would be no reason to make that distinction, or to allow modules to declare non-GPL licence status.

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

Not only that, but the NVidia binary driver is based on their Windows driver. It doesn't even work with Linux.

On top of both the kernel and the binary driver they build a shim, which is licensed as GPL + exception for linking with the binary driver.

Oracle could do the same for ZFS (or they could change the license, since - unlike NVidia - the source is available), but they won't. ZFS is under the CDDL license which does not allow linking with GPL code, so a shim like NVidias cannot be made without Oracles blessing. To comply with both licenses, such a shim would have to be (GPL and not CDDL), and (CDDL and not GPL) at the same time.

TL;DR: NVidia can do it because they can set the license for their half of the code to work around the problem. Oracle could do the same, but Ubuntu can't because they don't own either part of the code.

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

the parts of the blob that talk to the shim could possibly be argued to be derived works of the shim

Don't be in doubt that the NVidia lawyers made sure that only the shim is a derived work of the driver, never the other way around.

Likely the people who develop the driver are not even allowed to LOOK at the shim, to ensure that nobody could argue that the driver is in any way derived from the shim.

Adhering to the respective licenses

[ghodder]

I don't see where they're not adhering to the respective licenses. "Distribution" of a piece of software is reasonable to be restricted to the distribution of that specific piece of software, not it's inclusion in a larger distribution through a combination of software. If modifications are made to a specific piece of a "distribution" of a paricular project then that is what the license on that specific piece of software covers. When combined within a collection of pieces of software published under the GPL or any other license that does not apply to the collection just because one of them might have been released under a GPL or CDDL license, just the specific elements. The incompatibility does not come from the projects. GPL zealots beware!

Weak argument

[Sloppy]

Whether or not they're correct, their argument comes down to "we think we're right." The issue isn't whether the licenses are incompatible (I doubt anyone disputes that) but whether or not a kernel module is a necessarily derivative work of the kernel (because that must be true for the license incompatibility to matter). And here is what they say about that:

our lawyers have analyzed these situations with the assistance of our license compliance and software forensics staff for many years, and we have yet to encounter a Linux module that — when distributed in binary form — did not, in our view, yield combined work with Linux. The FSF, stewards of the GPL, have stated many times over the past decades that they believe there is no legal distinction between dynamic and static linking of a C program and we agree. Accordingly, the analysis is quite obvious to us. .. Canonical has found some lawyers who disagree — a minority position, from our understanding of community norms.

And there you have it. If you agree with them (and why wouldn't you? It's "quite obvious" so therefore that settles it) then they're preaching to the choir and you can applaud now. If you disagree with them (and side with Canonical) then .. oh my, you're in a "minority position."

Re:Lawyer: Linux is not *quite* GPL

[david_thornley]

I'm not a lawyer, but what the heck. The following isn't even illegal advice.

What are the limits to estoppel? I write some software and notice that it's being distributed illegally. I decide I don't care, and people distribute it freely. Then, I decide that I want to stop the illegal distribution, and start firing off cease-and-desist letters. Can I stop the illegal distribution? I'm not in a good shape to sue for damages, but do I have to let the distribution continue?

Also, suppose I write some software and GPLv2 it. Over the years, some people link proprietary blobs to it and I do nothing. Then someone comes along with a new proprietary blob, and I decide "enough is enough". Does estoppel apply when a party is counting on what the copyright holder has allowed in similar situations in the past?

Re:Lawyer: Linux is not *quite* GPL

[j-beda]

Yes and no.

"the Linux developers approved, condoned, and encouraged binary, non-GPL modules". You could argue that Linus has known all along that Linux distributions included binary, non-GPL modules (say NVIDIA video drivers?) They've known products were being distributed that aren't under the GPL license.

But no one is arguing that these products (NVIDIA) SHOULD be under the GPL license. So...no harm, no foul?

Flip side of the coin. Other products (NVIDIA) isn't arguing that the Linux Kernel should be invalidated against the GPL. (In theory, they could...but that wouldn't grant them a license to use the Linus Kernel. They could argue that their product (NVIDIA) can't be distributed with the Linux Kernel however. They could try to argue damages - but that would be pretty difficult given that they give product away for free as well.

So...now - let's look ZFS. Licensed under CDDL. Given away fro free.

SFC is arguing that ZFS owners COULD argue that giving away ZFS with Linux would violate the GPL. Yeah....so what? At that point Canonical apologies and pulls ZFS.

Damages? Maybe. I'm not sure I see how...but if Canonical is up to take the risk...the OWNER of ZFS could only attack Canonical for their actions. They CAN'T get any additional rights to Linux...Canonical doesn't own 'em.

I was thinking similar thoughts, but not so completely.

Who has standing to enforce the potential breach of license? If nobody, then it doesn't really matter. If the people with standing are the ones doing the breaching, again there is not much result.

I suppose if I ever submitted a patch to the project, I could potentially have standing, but am I likley to try to get the districution halted?

Why should Oracle change it's license?

[rhyous]

Why should Oracle change it's license?

If Linux is incompatible and it is Linux that wants this, then Linux needs to change to a more permissive license. The copyleft things is kind of considered a virus by a lot people.

Enforced freedom is not freedom.

Re: Lawyer: Linux is not *quite* GPL

[Tough Love]

I think the distinction expressly codifies what parts of the kernel are generic "kernel" interfaces and what parts are Linux-specific (and thus imply derivative work status).

Incorrect argument. Many, in fact, the vast majority of the GPL symbols are clearly Linux-specific. So the distinction between GPL and non-GPL symbols tells you nothing about whether a work is derivative.

Otherwise there would be no reason to make that distinction, or to allow modules to declare non-GPL licence status.

Creating discomfort for GPL violators is a perfectly valid reason to make that distinction. It's very cost-effective.

Re: Lawyer: Linux is not *quite* GPL

[Entrope]

What part of my argument is incorrect? EXPORT_SYMBOL_GPL(foo) means that, technically, the kernel will only expose "foo" to modules that declare a GPL-compatible license, and that, legally, the kernel developers think using that symbol means a kernel module is a derivative work of the kernel. EXPORT_SYMBOL(bar) means that, technically, the kernel will expose "foo" to any loadable kernel module, and that, legally, the kernel developers think it is a fairly generic interface for a Unix-like kernel. That they make the distinction, and allow modules to declare a license, implies that they expect GPL-incompatible kernel modules to exist. The distinction provides clarity and a measure of safety for distributors of such kernel modules, not discomfort.

Re: Lawyer: Linux is not *quite* GPL

[Tough Love]

What part of my argument is incorrect? EXPORT_SYMBOL_GPL(foo) means that, technically, the kernel will only expose "foo" to modules that declare a GPL-compatible license, and that, legally, the kernel developers think using that symbol means a kernel module is a derivative work of the kernel

There's your incorrect argument. You don't know what kernel developers actually think, and further, you don't know how it would affect legal standing even if you did. See "thought crime".

Re: Lawyer: Linux is not *quite* GPL

[Entrope]

Wow, you really know how to double down on the stupid. See "Donald Trump".

Or, if you prefer, see what Linus said about it.

Re: Lawyer: Linux is not *quite* GPL

[Tough Love]

Funny, I was thinking the same thing about you. Linus isn't "programmers", Linus is just one programmer of thousands who hold copyright in the kernel. In a legal sense, what Linus posted is "hearsay", and arguably a nice demonstration that he might be right when he claims programmers making lousy lawyers. As far as I know, intent ("willful infringement") may matter in computing copyright damages but not in deciding whether infringement actually occurred. Keep pursuing your theories about intent by all means, I doubt you will find anything of value down that path but somebody needs to do it.

Re: Lawyer: Linux is not *quite* GPL

[Entrope]

What Linus said may be hearsay in that one could not submit it as evidence in court, but I am not trying to do that. It merely illustrates that he, the most prominent copyright holder for Linux, agrees with me that usage of EXPORT_SYMBOL() versus EXPORT_SYMBOL_GPL() symbols is significant when it comes to determining whether a loadable kernel module is a "derivative work" of Linux in the GPLv2's sense. That kernel developers regularly debate which of those to use for which interfaces (e.g. dma-buf) indicates that his view is very widely shared.

An alleged infringer's intent has nothing to do with those annotations. They express the copyright holders' intent about the interface, and that is the intent that is relevant to figuring out whether a work is a "derived work" under the kernel's license. Keep throwing up irrelevant claims if you want, but realize that most of us can see right through them.

Re: Lawyer: Linux is not *quite* GPL

[Tough Love]

I have never heard any kernel developer state that the intent of non-GPL symbol exports is to encourage GPL infringement, or distribution of binary-only modules.

Re: Lawyer: Linux is not *quite* GPL

[Entrope]

Boy, you sure beat the hell out of that strawman!

Re: Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

Go fuck yourself, lawyer scum. Your whole profession is an invented need by people who don't want to produce anything, but be parasites in the rest of us.

Interestingly, that seems to be the whole reason the GPL is around in the first place.
To offer something to play with for an anti-social neckbeard and his lawyer friends.

I'm not normally fond of lawyers, but if a lawyer and RMS were both about to drown, I'd take a very long time deciding on which sandwich to eat...

Re:Lawyer: Linux is not *quite* GPL

[Anonymous Coward]

The problem with your analysis is that this:

Since pretty much the beginning, the Linux developers approved, condoned, and encouraged binary, non-GPL modules

Is not even remotely true. Greg KH and Alan Cox have both stated publicly that this is not the case, but more formally the kernel has an explicit set of symbol annotations for things that they consider to be public interfaces (which do not therefore invoke the GPL, because - possibly modulo the recent Oracle ruling - interfaces can not be copyrighted and so being a derived work of something that is not copyrighted can not spread the GPL, and even without this the Linux kernel license contains an explicit exemption for these interfaces). Anything that interacts with code that is not part of these interfaces is a derived work of the kernel and covered by the requirements of the GPL. Given how deeply ZFS must hook into the VM and VFS subsystems to work, it's pretty likely that it requires touching GPL'd parts.

Companies like nVidia sidestep this, because the GPL is a distribution license. They provide a binary blob, most of which is developed as part of a cross-platform effort and so is not a derived work of the kernel. They also distribute a shim. The shim is a derived work of the kernel, and the parts of the blob that talk to the shim could possibly be argued to be derived works of the shim, but the only license that would be violated if they were would be the GPL on the shim, and so the only person with standing to sue nVidia would be nVidia. The fact that they've had to jump through these hoops to do so is evidence that not doing so would not be condoned by the various copyright holders of the kernel. I challenge you to find one example of a binary-only kernel module that is distributed with the awareness and consent of the Linux copyright holders and uses anything other than the interfaces declared public.

The CDDL is a per-file license, so may be linked with code under any license without issues unless those issues come from the other license.

I am not a lawyer, but I've spent enough time with IP lawyers to have had most of these discussions before.

1. Neither Cox nor Hartmann are majority owners of the kernel code.
2. Public statements on community platforms aren't a valid replacement for common legal instruments.
3. Go ahead and check out the ZoL source yourself. You won't find a single "EXPORT_SYMBOL_GPL" in there.
4. The GPL doesn't explicitly state anything about "cross-platform efforts". Judging by the FSF's/GPL's faulty adaption of common law as well as computer science, NVIDIA would be in violation of the GPL as well, since the blob is a part of the kernel in their flawed mindset.