Slashdot Mirror

← Back to Stories (view on slashdot.org)

VMware May Violate Linux Copyrights

Posted by Zonk on from the what-about-its-copylefts dept.

Nailer writes "Bloomberg believe VMware's IPO today may the largest technology offering since Google. But doubts have been cast over the company's supposedly proprietary ESX product, as top 10 Linux contributor Christopher Hellwig claims the software may violate Linux kernel copyrights. 'Is Hellwig right, and is VMware a derived product of Linux? Unless vmkernel can be loaded without the Linux kernel, it would appear so. VMware was developed from another, long ago OS created as a research project, but it's unclear whether vmkernel was ported from that OS or rewritten as the Linux-requiring binary blob. What's more of an issue is that VMware had these serious questions posed directly to them a year ago, repeated in a public forum many times since, but have yet to respond at all.'"

22 of 443 comments (clear)

  1. They made a movie about this with Charlie Sheen by captnitro · · Score: 3, Interesting
    • Wait for big, innovative company to IPO.
    • Watch as share price goes up 90% on a day when the Dow is losing 100 points. Feel bad I don't work for that company. Boo.
    • Blog about possible copyright violations that would surely bring down EMC or VMW. Make investors nervous. Buy low.
    • Profit.
  2. Incorrect wording in title by JosefAssad · · Score: 5, Interesting

    Copyright gets infringed, licenses get violated.

    --
    The Banjo Players Must Die!
  3. Re:Adds to Perception of GPL as Viral by sH4RD · · Score: 4, Interesting

    So? If they want to make it closed source, they certainly shouldn't use Linux or GPL software in their products.

    Before you all massacre me: I see your real point, that they will fear using Linux as a base operating system for their products, even when that usage wouldn't cause their code to fall under GPL. But should that stop people from protecting their IP? Contributors to the Linux kernel and other GPL products have issued an exclusive license under which their copyrighted material should be released. Allowing corporations to desecrate this for the lofty goal of popularizing Linux doesn't make sense. GPL is what it is, and if it doesn't become any more popular because of it's "viral" nature or even perception of such, so be it. Otherwise you will just be destroying the authors goals - to keep the software free and open at all costs.

    Bottom line is, if it adds to the negative perception of GPL, it's worth advertising the positive, but certainly *not* worth dismissing the issue. Stand by the GPL principals, or don't use them in the first place.

    --
    WASTE - The Secure P2P
  4. Not necessarily a violation. by sl3xd · · Score: 4, Interesting

    There is, of course, a way that they aren't violating linux copyrights: They may simply be using the Linux Kernel to get the hardware into a known state prior to loading the VMkernel. Similar projects include LinuxBIOS, and Linux's own kexec (kexec lets you boot a new linux kernel without actually 'booting').

    Of course, it is a violation if ESX is actually running a modified Linux Kernel, instead of using the Linux Kernel as a bootloader. Using the Linux Kernel as a bootloader is a done deal; just look up 'kexec' for proof of it. (Though I'm fairly certain kexec isn't what VMware uses).

    But even then, remember that ESX is their "enterprise" product, which acts more like a hypervisor, and is not to be confused with VMware Workstation, VMware Player, or VMware Server.

    --
    -- Sometimes you have to turn the lights off in order to see.
    1. Re:Not necessarily a violation. by Waffle+Iron · · Score: 2, Interesting
      This whole debate gets tiresome. To put this to rest, somebody ought to write an NdisWrapper-style shim for BSD that will run any Linux kernel module, then release it under a BSD-style license. Then any binary Linux-compatible kernel module will by definition no longer be "derived" from Linux, since it could run without Linux.

      Of course, a better solution would be for the courts to rule that this silly legal theory that "won't run without X" === "derived from X" is bunk. "Derived" works should really only mean works where you take some source code and edit it. I find it ironic that in this gray area, the FSF is on the vanguard of expanding the scope and onerousness of copyright law as much as they possibly can.

  5. Re:Old news? by morgan_greywolf · · Score: 2, Interesting

    It could be viewed that Linus' statements constitute a verbal license. Licenses don't always need to be granted in writing. A license is a permission. I lend you my car, I don't necessarily need to write you a note saying that it's okay to borrow my car for such-and-such time period for the purposes of so-and-so...I just say "Here, borrow my car. Here's the keys." That's a permission, and that's exactly what a copyright license is - a permission to do something normally reserved for the copyright holder. So the GPL is a license, but so is Linus saying "You can write a binary driver so long as it can stand on its own without the kernel".

    Note that IANAL.

    --
    My blog
  6. Re:If it cannot be loaded without the linux kernel by TheRealFixer · · Score: 2, Interesting

    From what I understand, the Linux 2.4 kernel (Service Console) does the initial boot, but only addresses a limited amount of RAM (usually between 256M and 384M, depending on how many VMs you plan on running). It then loads the vmkernel, which takes the rest of the RAM and takes over scheduling functions from the Service Console, and directly addresses the hardware. With the exception of the initial RAM, everything else the Service Console does is through the vmkernel. The way it was explained to me was that the SC essentially becomes a pseudo-VM itself, relying on the vmkernel for hardware access.

    In turn, the Service Console has some hooks into the vmkernel to perform management functions, monitoring, etc. The SC functions as a bridge between the vmkernel and the outside world. That's the way I understand it, anyway. Could be totally wrong. I have heard rumors in the past, though, that VMware is planning to ditch the Linux SC in future versions in favor of their own service console OS. Which would render the whole argument moot.

  7. Help me understand... by jdray · · Score: 1, Interesting

    Maybe someone can help me understand something I'm unclear on: How does one go about creating proprietary software that runs on Linux (the kernel) without "violating the GPL?" Based on the blurb, and many similar stories I've read here over the years, it almost seems like the FOSS community expects every piece of software that is written to run on a Linux platform to be Open Source. If that's the case, why? I understand not liking proprietary device drivers for the want of controlling the hardware you own, and I understand liking a project/product like MySQL for the desire to be able to tune the software that controls your data. But there should be room for proprietary software that leverages the significant platform that is Linux.

    I harken back to the days of Netware NLMs. Netware didn't seem to want some sort of ownership for people loading things onto their kernel, nor did Microsoft demand rights for people distributing TSRs. So why is this different?

    --
    The Spoon
    Updated 6/28/2011
    1. Re:Help me understand... by Anonymous Coward · · Score: 2, Interesting

      Under that logic almost all FSF software would be the property of Sun, AT&T, HP, IBM, etc. since for a long time it required SunOS, Solaris, HP/UX, AIX to run on. I'm pretty sure that none of those companies made public "linking exception" statements. Not to mention Linux used the minix filesystem. Oh the irony of A. Tennenbaum owning Linux due to the GPL advocates glommy perception of linking. Anyone who looks at linking as creating a derivative work is seeing the world through a very narrow slit in the blinds.

      The reality is is that a lot of kernel developers and linux advocates don't actually understand the law involved and spout what they would like the terms of the code to be and have some mental convolution that fits those terms in the the GPL.

    2. Re:Help me understand... by the_B0fh · · Score: 2, Interesting

      There's a couple of things you have to understand first. You *can* pay Microsoft for the privilege of writing code to their kernel.

      However, there's this bunch of freaks who release their code to the world, for others to use. Their payment is that if you *derive* stuff off their work, you have to pass your source code on as well.

      You pick your payment plan, and off you go.

      Not too many people have to write things that hook into the kernel itself. MYSQL certainly doesn't. For the most part, people who do that either need it internally (you did not re-distribute, so do not have to pass your source code along), or do it for another reason (to support whatever). If you are doing it to support whatever, then just obey the license. Again, your choice of licenses - pay Microsoft, or pay by passing along your source code.

      Why should anyone develop their code, and let you use it for free? *YOU* certainly are not willing to do so.

    3. Re:Help me understand... by whmac33 · · Score: 2, Interesting

      Does anyone know of this actually being tested in court. The GPLv2 only speaks of derivative work. But I don't think that linking follows the traditional concept of derivative work. Just because it's in their FAQ doesn't make it law.

    4. Re:Help me understand... by jonatha · · Score: 2, Interesting
      it's very clear that they'll take you to court if you violate what they deem a derived work. My hunch is that their definition of infringement is vastly looser then any court in the U.S. would ever interpret it to be.

      I agree.

      As their definition is vastly looser then any standard definition used by a court, it's not like the court will use a stronger interpretation. It is also very difficult for someone to argue that they felt in compliance of the license, while using a different definition of "derived work".

      I disagree.

      If I create a non-GPL'd program that dynamically links to a GPL'd (not LGPL'd, GPL'd) library and I distribute that non-GPL'd program without the GPL'd library, I am not distributing the GPL'd library and so the legal proscriptions against distributing someone else's copyrighted work do not come into play.

      I also do not need to accept any terms of the GPL in order to distribute my non-GPL'd program because I am not doing anything with the GPL'd library (the work to which the GPL applies) that I have to accept the GPL in order to do - I am not distributing the GPL'd library, or a modified version of it.

      The only question is whether I have to accept the GPL in order to distribute my non-GPL'd program because my non-GPL'd program constitutes a derived work according to the legal definition of "derived work". I don't have to use the FSF's broader definition because unless the courts agree that my non-GPL'd program is in fact a derived work I am not doing anything that is prohibited by copyright law, and all the GPL does is give me additional privileges that would otherwise be disallowed by copyright law.

      I think if the law and case history were anywhere near clearly supportive of FSF's position on this, they would plaster citations all over their website. AFAIK they don't, and they don't seem particularly eager to defend their legal theory concerning derived works.

      --
      The SCO lawsuit makes me wish my company were in Utah. We need a new building.
  8. Cannot use linux as a bootloader? by Timothy+Brownawell · · Score: 1, Interesting

    So if I'm reading this right, anything that uses Linux as a bootloader is "derived from" Linux (because it depends on Linux, because nobody bothered to implement another bootloader) and must be GPLed? That seems very, very bogus, and I will be very annoyed if such nonsense is upheld. Just because something looked at your code funny once, does not automatically make that thing derived from your code.

  9. Derived Works by Sloppy · · Score: 2, Interesting

    But some proprietary modules do exist, and they do that on one premise: Linus Torvalds (the copyright holder for the Linux kernel) has repeatedly stated that he doesn't consider drivers ported from other operating systems to be derived works of Linux. After all, if something can load without Linux, it can't really be considered a derived work.

    Sometimes I have to ask: where the fuck are the lawyers? Did we finally kill them all? ;-)

    Folks, software creators like Linus or the FSF people, can put whatever terms into licenses that they want, but one thing they can't do is define derived works. Congress does that (very poorly, so the courts end up mostly stuck with the job). And unless you make something that is a defined work, do never need to get bound to the license in the first place, so.. words in the GPL do not matter, and Linus' opinion does not matter. Well, it matters in the since that we're talking about smart people who have obviously given the issue some thought. But that's all.

    What I'm getting at, is that Linus is making an argument. He is not giving an authoritative declaration as a copyright holder or licensor. He can't.

    Linus has determined? (inferred? decided?) that if something works w/out Linux, it's not a derived work, and if it doesn't work w/out Linux, it is a derived work. I think that's very arbitrary, and brings up so many (apparent?) counter-examples that it would terrify everyone in the software industry except for maybe the BIOS guys.

    --
    As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
  10. Hardware Support by JumboMessiah · · Score: 3, Interesting

    I've never used ESX, but I would imagin that no matter what, ESX needs driver support.

    Being a hypervisor, it has to access all kinds of devices like VGA consoles, serial ports, Fibre Channel HBAs, SCSI HBAs, IDE controllers (for CDROMs), Ethernet adapters, etc., etc. So my question is, where does the ESX hypervisor (vmkernel) get these? Does it pull them from Linux or did they write their own? This hardware _HAS_ to be setup, initialized, and arbitrated. Does vmkernel have it's own stack of device drivers, or does it conveniently run the ones in the Linux "bootloader"?

    I'm thinking there's more to this than just the binary blob issue...

  11. Re:Philosophy by Anonymous Coward · · Score: 1, Interesting

    Considering ESX runs on a number of platforms my *guess* is they probably favor the later.
    Incorrect.
    Per VMWare, ESX IS the platform - it runs on it's own custom OS. The issue raised here is not whether or not an application only runs on Linux - lots of those exist, and they are not derivative works by anyone's reckoning. The issue is not whether binary blobs are allowable by a strict interpretation of the law and license - they interface to the kernel through a non-public interface, they run in kernel-space, and use kernel code directly, even if they do not incorporate kernel source. The issues are these.
    Linus (who, like it or not, effectively sets a lot of policy regarding the Kernel, even bits of it he does not own the copyrights to) has stated he does not feel he has the moral right to keep people who wrote drivers for their hardware on another OS from porting them to Linux under whatever license they choose. Since he, personally, does not consider them derivative he refrains from action - regardless of whether or not he would have legal standing to do so.
    Using Linux as a "boot-loader" is nonsensical. It's an analogy for what is going on, and a poor one.
    ESX uses a real boot-loader to load the Linux Kernel. The system loads, using insmod, the vmkmod kernel module. A kernel module is part of the kernel - simply modularized for convenience. Once it is loaded in kernel-space the difference is moot. There is a fairly strong case to be made that vmkmod is therefore a derivative work of the linux kernel. Also, the question is relevant whether vmkmod and the rest of the ESX proprietary binaries are separate works or parts of the same work. IANAL, so I can't say. I don't necessarily think that this is trolling, and I don't think it is FUD to raise the issue in a public forum, even if it would take a team of kernel engineers and IP lawyers to actually sort out whether or not there is something to worry about. I DO think that the timing of this article is very suspect.
  12. Re:Adds to Perception of GPL as Viral by AchiIIe · · Score: 2, Interesting

    > So, as you see, the GPL is clearly not viral.

    Are you that asinine? Is a "viral video", or "viral marketing" not viral ?
    Anyhow, what the grandparent was right, it's viral.

    (at this point I took some time to research it myself to prove you were stupid about this)

    Oh wait, the gpl is not viral. (no kidding I just realized this myself)

    > http://yro.slashdot.org/article.pl?sid=03/12/15/20 41236
    > Pamela Jones of Groklaw has put together a short FUD-killer on the General Public License that explains why you can't lose your proprietary code if you inadvertently incorporate GPL code. This is not the only text of its kind, but it is so well explained that you might want to bookmark the page for future reference."

    The article:
    > http://www.groklaw.net/article.php?story=200312142 10634851

    Suppose a company really did mingle GPL code into a program with their own proprietary code and then distributed the merged product under a proprietary license or without living up to the terms of the GPL? Now what happens? What will the judge do now? Order the code released under the GPL over the wishes of the owner?

    Stop and think. What happens if you violate the terms of a fishing license? For example, the license may restrict how much fish you can catch on a particular day or what kinds of fish you can keep, what sizes, etc. Suppose you violate the terms of the license. What happens? You lose your license to fish. There may be a fine to pay, right? That's essentially the same thing that happens under the GPL, except it's nicer, because the company gets to choose what it wishes to do under the terms of the GPL. If it still isn't resolved, and it goes to a judge, however, it's enforced as a violation of copyright law, not contract law. Here is Professor Moglen's explanation of what happens:

    Sorry about the insults, cheers.

    --
    Nature journal lied in Britannica vs Wikipedia Ask to retrac
  13. I don't understand this by Master+of+Transhuman · · Score: 2, Interesting

    I read the article and as far as I can see, while there may be a technical legal issue with the VMWare kernel being loaded by a Linux kernel (or some part thereof - insmod), thereby in some extremely technical sense making the VMWare kernel a "derived work" - this is really reaching in my opinion.

    Particularly since it would seem obvious that that they could easily rewrite the thing to do its bootloading in some other way. The Linux kernel appears to be have been used only as a convenience to make the system more portable than their original development OS. And this was probably done "back in the day" since they're using a 2.4 kernel.

    And if said Linux kernel being used is described as a "badly hacked 2.4 kernel", then who the hell cares? Hellwig seems to be pissed that VMWare asked the kernel maintenance list for some support or something, but basically seems to be on a "crusade" like the FSF fanatics. He's all pissed off about something that nobody else in their right mind couldn't care less about.

    Perhaps VMWare should rewrite their boot loader (they certainly have enough money and smarts to do so), but basically I agree with the first poster - this appears to be either FSF fanaticism or an attempt to influence the VMWare IPO or both.

    It's really beginning to seem like a religious crusade for some "fundamentalists" to root out "heretics" in the OSS world. The same socialists who deride proprietary companies for preventing "freedom" are more than willing to use a state-enforced license to drag people into line with their ideology. This is not "freedom". It is coercion.

    --
    Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
  14. Re:VMware, ESX and a bad smell by Anonymous Coward · · Score: 1, Interesting

    JUst as a response to this comment. I work at VMware. This is not an official represntation of the companies stance. It is my personal opinion and mine alone. Firstly to the poster I am replying to, If you interview here and are asked about your linux skills you are required to have them (I too did the Robert Half thing and they do ask for that when you talk to the rep). And you will be quizzed on it. Bottom line is for desktops and and servers we basically have users who want to use linux so we set it up for them. They have that option here. So therefore everyone from the helpdesk up must know how to use and troubleshoot linux. As far as BO issues you would not be the first. You can bash us not getting in but that is the head hunters fault for wasting your time and our time in interviewing someone who did not have qualifications WE specified to them. Now back to the show. Honestly I am pretty sure VMware has got this covered legally (again not any kind of official response). This whole article smells of FUD from someone who neglected aspects of the situation at hand. If anyone in this article had bothered to go to the source, mainly our vmware.com website you would see this http://www.vmware.com/download/vi/open_source.html The source code for the products in question are posted right on our website so I do not see how VMware is violating the GPL. There are certainly proprietary portions of the product. Internally I will say there is a deep respect for the FOSS community and we are working to make efforts to work better with the community. It is by no means perfect, but at least we try. The OP never did any real looking. This would probably explain why the OP never got an answer to his posts. If there is something further beyond that is being violated then I am not exactly clear on what it is. The code is being posted in what seems to me in accordance to the GPL. As for the other products they are built very differently and as far as I am aware do not contain any GPL code in them. If they do they get posted on the website in the same manner as the link I posted. Just in case that link was missed

  15. Umm they do distrube the code!! by opieum · · Score: 2, Interesting

    http://www.vmware.com/download/vi/open_source.html So what is the problem? I bet the reason the forums have not gotten an answer is because the user never bothered to look for the answer. He wanted it answered for him instead of actually you know doing some of the legwork himself. That or just posting FUD. The code is there for all to see.

  16. Re:nothing mysterious here by marcushnk · · Score: 2, Interesting

    Freemware is now called Plex86, and is alive and well :-)
    http://www.plex86.org/

    --
    "Consider how lucky you are that life has been good to you so far. Alternatively, if life hasn't been good to you so far
  17. Re:Vmware the driver writers? by Paul+Jakma · · Score: 2, Interesting

    Apparently the source to the Linux bits /is/ available. It's the "depend on Linux for hardware IO" that Hellwig (and others?) object to as a licence violation.

    --
    I use Friend/Foe + mod-point modifiers as a karma/reputation system.