Slashdot Mirror

← Back to Stories (view on slashdot.org)

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

Posted by timothy on from the smart-people-vary-in-their-conclusions dept.

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."

63 of 379 comments (clear)

  1. No winners here. by Anonymous Coward · · Score: 4, Insightful

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

    1. Re:No winners here. by Austerity+Empowers · · Score: 4, Interesting

      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.

    2. Re:No winners here. by nycsubway · · Score: 3, Insightful

      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.

      --
      http://github.com/gbook/nidb
    3. Re:No winners here. by Anonymous Coward · · Score: 2, Insightful

      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.)

    4. Re:No winners here. by Grishnakh · · Score: 2, Insightful

      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.

    5. Re:No winners here. by mysidia · · Score: 2

      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.

    6. Re:No winners here. by EndlessNameless · · Score: 3, Informative

      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.

      --

      ---
      According to the latest ruleset, this post should be modded as Vorpal Flamebait +5.
    7. Re:No winners here. by Znork · · Score: 3, Insightful

      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...

    8. Re:No winners here. by epine · · Score: 4, Insightful

      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.

    9. Re:No winners here. by dbIII · · Score: 2

      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.

    10. Re:No winners here. by shutdown+-p+now · · Score: 2

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

    11. Re:No winners here. by Tough+Love · · Score: 2

      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.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    12. Re:No winners here. by Grishnakh · · Score: 3, Interesting

      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.

    13. Re:No winners here. by TheRaven64 · · Score: 2

      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.

      --
      I am TheRaven on Soylent News
  2. bunch of hot air - use FreeBSD then? by ecloud · · Score: 2

    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.

    1. Re: bunch of hot air - use FreeBSD then? by Billly+Gates · · Score: 2

      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

      --
      http://saveie6.com/
    2. Re: bunch of hot air - use FreeBSD then? by brambus · · Score: 2, Interesting

      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).

  3. SFS? More like FFS... by Bearhouse · · Score: 5, Insightful

    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...

    1. Re:SFS? More like FFS... by edtice1559 · · Score: 4, Informative

      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.

    2. Re:SFS? More like FFS... by edtice1559 · · Score: 2

      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.

    3. Re:SFS? More like FFS... by brambus · · Score: 2

      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.

    4. Re:SFS? More like FFS... by lkcl · · Score: 2

      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.

  4. I don't see the advantage of objecting by davecb · · Score: 2

    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.

    --
    davecb@spamcop.net
    1. Re:I don't see the advantage of objecting by brambus · · Score: 2, Interesting

      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.

    2. Re:I don't see the advantage of objecting by oh_my_080980980 · · Score: 2

      Which makes this totally fucked.

  5. It isn't features, it is stability + features by poet · · Score: 4, Insightful

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

    --
    Get your PostgreSQL here: http://www.commandprompt.com/
    1. Re: It isn't features, it is stability + features by BellyJelly · · Score: 2

      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.

    2. Re:It isn't features, it is stability + features by bheading · · Score: 3, Insightful

      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.

    3. Re:It isn't features, it is stability + features by rl117 · · Score: 2

      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.

  6. SFS/FSF does not get to rule on GPL by Anonymous Coward · · Score: 4, Insightful

    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.

    1. Re:SFS/FSF does not get to rule on GPL by jedidiah · · Score: 4, Insightful

      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...

      --
      A Pirate and a Puritan look the same on a balance sheet.
    2. Re:SFS/FSF does not get to rule on GPL by slashping · · Score: 4, Insightful

      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.

    3. Re:SFS/FSF does not get to rule on GPL by edtice1559 · · Score: 4, Insightful

      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.

    4. Re:SFS/FSF does not get to rule on GPL by Olipro · · Score: 4, Informative

      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.

    5. Re:SFS/FSF does not get to rule on GPL by MachineShedFred · · Score: 2

      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.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    6. Re:SFS/FSF does not get to rule on GPL by Lord+Crc · · Score: 5, Informative

      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...

    7. Re:SFS/FSF does not get to rule on GPL by pr0fessor · · Score: 3, Informative

      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

    8. Re:SFS/FSF does not get to rule on GPL by pem · · Score: 2

      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.

  7. Re:Software Freedom? by slashping · · Score: 5, Insightful

    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.

  8. Linus doesn't object to binary kernel modules by edtice1559 · · Score: 4, Insightful

    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.

  9. Hilarious by Anonymous Coward · · Score: 4, Insightful

    >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.

  10. What's the Conservancy's role in this? by bradley13 · · Score: 4, Informative

    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?

    --
    Enjoy life! This is not a dress rehearsal.
  11. A grey area Oracle might exploit by evolutionary · · Score: 3, Interesting

    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.

    --
    "Imagination is more important than knowledge" - Einstein
  12. Re:Software Freedom? by Marginal+Coward · · Score: 2

    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...

  13. Re:Software Freedom? by drew_kime · · Score: 4, Insightful

    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.

    --
    Nope, no sig
  14. Software Freedom Conspiracy by wjcofkc · · Score: 4, Insightful

    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.

    --
    Brought to you by Carl's Junior.
  15. "Illegal" is misleading by coldsalmon · · Score: 2

    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.

  16. is it a derived work? by bradleykuszmaul · · Score: 2

    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.

  17. btrfs by spectrum- · · Score: 2

    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.

    1. Re:btrfs by rl117 · · Score: 2

      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.

    2. Re:btrfs by allquixotic · · Score: 2

      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.

  18. Enforcement of GPL is a gain by jbn-o · · Score: 2

    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.

    --
    Digital Citizen
    1. Re:Enforcement of GPL is a gain by bug1 · · Score: 2

      The difference is using internal API vs external API.

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

  19. Re:Software Freedom? by TsuruchiBrian · · Score: 4, Interesting

    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.

  20. Re:Software Freedom? by squiggleslash · · Score: 3, Insightful

    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.

    --
    You are not alone. This is not normal. None of this is normal.
  21. It is no different than Minix AFS by mlwmohawk · · Score: 2

    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.

  22. Lawyer: Linux is not *quite* GPL by hawk · · Score: 5, Informative

    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.

  23. The CDDL cannot apply to the Linux kernel by tetraverse · · Score: 2

    "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

  24. Re:Lawyer: Linux is not *quite* GPL by hawk · · Score: 3, Interesting

    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

  25. Re:Lawyer: Linux is not *quite* GPL by hawk · · Score: 3, Interesting

    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

  26. Re:Lawyer: Linux is not *quite* GPL by shutdown+-p+now · · Score: 3, Insightful

    "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.)

  27. Re:Lawyer: Linux is not *quite* GPL by Tough+Love · · Score: 3, Informative

    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.

    --
    When all you have is a hammer, every problem starts to look like a thumb.
  28. Linking by kbahey · · Score: 2

    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.

    --
    2bits.com, Inc: Drupal, WordPress, and LAMP performance tuning.