The Nexenta OS project has released its first publically available release. NexentaOS pre-alpha 1 has been released (also trying to beat this, I guess). You can download it from this page. This release contains OpenSolaris kernel build 26, GNOME 2.12.1, GNOME Office, Xorg 6.8.2, and much more. Update by AS: Screenshots!
Look’s realy nice..
Can’t wait to try this out…
Nice project! It’s nice OpenSolaris growing and some more nice projects also using free DEs.
I’m not trying it right now because I’m out of discs (unless someone packs a WMware player disk)… But I’ll try this one soon as I get some!
I hope they embrace Tango project too for a nice standard by the way! =]
Nice to see a release from this project it looks really promising.
The only thing I don’t get is why Thom Holwerda thinks why they are trying to beat SkyOS – those projects are miles apart.
I think he means that Nexenta OS is trying to beat them by being first with there release.
Although your right. It is unrelated in an OS nature kind of way.
I believe he was referring to the developer preview of the alpha release candidate. Either that or a comparable vowel to consonant ratio. Take your pick.
He’s talking about their, uh, naming convention.
The rendering looks a bit washed out, say in comparison to Fedora. Why is that?
I think it’s the jpeg compression.
Gamma.
I’ll try it, good job, keep it up.
perfect for workstation
I see so many announcements about this, but every time I try to download the ISOs and get a “file not found” error. Am I looking at a wrong place or the project has yet to release anything other than the release notes?
are there any KDE packages? when i tried the package search on the site it seemed broken.
Also, it seems like ethe distros is based around Unbuntu rather than Debian… and while I have heard of Unbuntu multiple times I am unaware of it’s reputation (unlike Debian whose faults and virtues I know).
Are the packages up-to-date? are they as stable as Debian? do they tweak and mangle each package with their own patches (a la Mandrake^H^H^Hiva)? Is Solaris x86 JVM in the package collection? Is the standard Solaris userland and manpages available (did they even open-source those?)?
I’d ask on there forums but trying to submit anonymous post it vomitted on me (mysql and php).
PS- I’m excited… but judging from the brokeness of the webpage I’m not hopeful (several mysql vomits).
PPS- I like giraffe… but just like beastie and konqi they make good logos do not.
First of all, the trademark issue: In choosing to name their product “GNU/Solaris” they’re using “Solaris”, a Sun trademark, in a way that is in contravention of Sun’s trademark policy (http://www.sun.com/policies/trademarks/). Given that trademark law *requires* trademark holders to aggressively police use of their mark or risk losing it altogether, I think we can expect a response from Sun’s lawyers pretty soon.
Additionally, there is a large body of thought which holds that by compiling a piece of code against a given C library, the result is a derivative work of both the code *and* the C library. Sun’s CDDL-ed libc is the one being used in this project; much of the code being compiled against (and distributed with) said C library by Nexenta is GPL-ed, and the GPL and CDDL are incompatible with one another. Hence, it is not possible to legally distribute something that is a derivative work of both CDDL and GPL codebases in the way that they are doing.
“But wait!” the Nexenta devs claim – The GPL has an explicit exception to make this okay:
“However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.”
But of course, in this case the CDDLed components (most notably the C library) *are* themselves accompanying the GPLed applications, so the above exception does not actually apply.
The most damning evidence of all is from the Nexenta devs own posts to the debian-devel mailing list. They first tried to persuade Debian to relicense DPKG under CDDL-compatible terms, with no success, then threatened to reimplement DPKG’s functionality under CDDL-compatible terms (see http://lists.debian.org/debian-devel/2005/11/msg00258.html). But obviously that turned out to be too much effort, so they thought they’d just nick it and use it illegally instead. If they really thought that there were no issues with incorporating the GPL-ed DPKG in the way they have, why would they have even bothered with all the preceding discussion about relicensing or reimplementing the code?
From the gnusolaris.org front page: “At the moment, Nexenta is not part of the Debian Project. Our packages are not present in the Debian database. We are hoping that in the future this will change and our packages will get their “upstream acceptance”.” – Yeah, good luck with that, guys! I’m sure the Debian Project will embrace you with open arms after stealing DPKG, and even more so when you illegally incorporate the debian-installer as well further on down the line. 😛
hmm, a Gentoo/OpenSolaris (I assume SolairsTM will be reserved for Sun’s releases to distinguish between supported and unsupported distros) would then be legal like Linux binary kernel-modules. That said I hope the GPL focus don’t start a big fuss over this since CDDL is just another OSI approved license. As for asking for dkpg… well they’d probably sleep better having it but it’s not obvious they need it.
As for compiling against == derived, well thats a like a bit of a stretch and it doesn’t seem like something that could be decisively decided without a length and expensive legal battle which I can’t see how that would be productive considering they are both OSS projects, it’s not Sun doing it, and there are more important fights.
steal the installer? if it’s oss why would they need to steal it?
“a Gentoo/OpenSolaris (I assume SolairsTM will be reserved for Sun’s releases to distinguish between supported and unsupported distros) would then be legal like Linux binary kernel-modules.”
No, the exact same murky legal issues exist with regards to binary kernel modules, as well. See for example LWN’s quote of the week (http://lwn.net/Articles/157225/), the words of kernel developer Greg Kroah-Hartman:
—
Oh, and at least one major distro has been served with legal papers due to them shipping closed source kernel drivers, and more are on the way. That’s the direction some developers are taking.
—
“As for asking for dkpg… well they’d probably sleep better having it but it’s not obvious they need it.”
They want to base their project on Debian and Ubuntu, and hence they need DPKG. Obviously they feel that Debian’s packaging technology is superior to that of Gentoo.
“As for compiling against == derived, well thats a like a bit of a stretch and it doesn’t seem like something that could be decisively decided without a length and expensive legal battle which I can’t see how that would be productive considering they are both OSS projects, it’s not Sun doing it, and there are more important fights.”
No link since it’s too recent to have been publicly archived, but one of the most recent posts to debian-devel consisted of the Nexenta developer Alex Ross providing evidence of Sun doing the exact same thing wrt to mixing GPL-ed and CDDL-ed code, and asking “are you going to pursue a legal action against Sun Microsystems?” To which the Debian developer Matthew Garrett responded “”yes”. I’m not sure how that’s supposed to excuse you in any way.”
“steal the installer? if it’s oss why would they need to steal it?”
Because Open Source != Public Domain. You still have to abide by the terms of the license if you want to make use of the work. I already elaborated on how Nexenta are in violation of the license terms in my last post. And, as I pointed out, Nexenta must *know* they are as well or they wouldn’t have bothered to ask Debian to relicense DPKG in the first place. Their only means ofs justifying their current actions seems to be “Sun’s getting away with it, so we can too” which doesn’t seem all that watertight (or moral, given that the Debian developers made their opinions fairly clear) to me.
The CDDL is not GPL compatible. Like the OpenSSL library, CDDL licensed code cannot be linked together with GPL licensed code.
Therefore distributing any GPL software linked together with OpenSolaris’ libc is a big no-no.
See
http://www.gnu.org/licenses/license-list.html#CDDL
Ok, one thing is putting gpl applications to run on a non-gpl compatible kernel and another diferent thing is to link gpl libraries with non gpl applications. If you couldn’t use gpl apps on a non-gpl kernel then we wouldn’t se any gnu apps on Darwin/MacOSX or even Windows
And IIRC these have bee ported to the two OSes mentioned above.
“Ok, one thing is putting gpl applications to run on a non-gpl compatible kernel and another diferent thing is to link gpl libraries with non gpl applications. If you couldn’t use gpl apps on a non-gpl kernel then we wouldn’t se any gnu apps on Darwin/MacOSX or even Windows”
You’ve missed the point. Once again, the crucial clause in the GPL reads as follows:
“However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.”
The problem is that Nexenta are distributing the GPL-ed applications, and the CDDL-ed libc those applications are linked against, *together*. Thus, the above exception simply does not apply to them, however much they might claim that it does.
If the GPL-ed components were distributed separately from the CDDL-ed base system, there would be no problem. But of course, were they to do that it wouldn’t be possible to integrate DPKG and debian-installer into the base system… Thus, they chose to deliberately disregard the license terms and go ahead and incorporate DPKG et al illegally.
“However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.”
——————————————————-
The problem is that Nexenta are distributing the GPL-ed applications, and the CDDL-ed libc those applications are linked against, *together*. Thus, the above exception simply does not apply to them, however much they might claim that it does.
——————————————————-
Do you mean that they’re linked against each other?
Or do you mean that they’re linked together with the CDDL LibC?
And IMO, if the CDDL is also another free-software license there should be no problem. And besides, the problem with these licences (if i read them correctly) is that they’re to similar(in terms of relicensing)
“Do you mean that they’re linked against each other?
Or do you mean that they’re linked together with the CDDL LibC?”
The GPL-ed applications are linked against the CDDL-ed libc, and the libc and applications are then being distributed together.
“And IMO, if the CDDL is also another free-software license there should be no problem.”
That’s not the way it works. Not all licenses are compatible with each other. For instance, the four-clause BSD license (the one with the “advertising clause”) is incompatible with the GPL because the GPL stipulates that no additional restrictions may be placed upon the work on top of the requirements specified within the terms of the GPL. Mandatory advertising within documentation would be just such an additional restriction and hence the two licenses are incompatible.
“And besides, the problem with these licences (if i read them correctly) is that they’re to similar(in terms of relicensing)”
No, that isn’t the problem. The problem is similar to that which I outlined above with regards to the 4-clause BSD license. The GPL and CDDL have different requirements, and it isn’t possible to simultaneously satisfy the terms of both licenses at the same time for any given codebase. Which is what you’d have to be able to do in order to legally distribute anything that’s a derived work of both GPL-ed and CDDL-ed code. *Unless* you can invoke the
“However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.”
portion of the GPL, which Nexenta cannot take advantage of because in their case the CDDL-ed components (the libc etc) *are* accompanying the GPL-ed executables (such as DPKG).
Hopefully that makes things somewhat clearer. 🙂
Syntaxis,
Since you know a lot about the License, can you explain to me why “OpenSolaris” can use GCC instead of Sun Studio to build “OpenSolaris” from Solaris Express?
GCC is GPL so what does that mean?
While your at it, please explain
BeleniX & SchilliX, I believe both contain CDDL and GNU/GPL’d software.
THanks,
And OSX, Don’t forget OSX, it’s also built with gpl tools
“can you explain to me why “OpenSolaris” can use GCC instead of Sun Studio to build “OpenSolaris” from Solaris Express?
GCC is GPL so what does that mean?”
This quote from the GPL itself seems to adequately explain this:
“Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does.”
The act of simply *using* the compiler (as in, to compile code) is not restricted in any way by the GPL, and the compiler’s output is unlikely ever to be considered a derivative work of the compiler itself.
“While your at it, please explain
BeleniX & SchilliX, I believe both contain CDDL and GNU/GPL’d software.”
I haven’t used either of them. But if they’re doing the same thing as Nexenta (you tell me) then yes, they, too, are violating the license terms of the GPL-ed software. As, apparently, is Sun itself (as I mentioned in one of my earlier posts).
Welp, Bryan Cantrill is using it, so thats good enough for me…….
http://www.opensolaris.org/jive/thread.jspa?messageID=14159㝏
zip Syntaxis.zip Syntaxis-fud.conf
jlc-
I assure you that Sun — and Apple and HP and IBM and every other OS vendor — spent plenty of time and money on in-house counsel before shipping GPL’d software. You can be sure that those companies do not believe that they are violating the GPL. The claim that Nexenta is violating the GPL because they are also distributing libc is absolutely absurd — libc is in no way a derived work of GPL’d software, and in no way has dependencies on GPL’d software. You need to understand the basics of copyright law; if I write a book, I have no right over the other books in your bookshelf. Similarly, if I write software, I have no right over software that isn’t a derived work. Seriously, talk to a lawyer — ANY lawyer — about this; this is an open-and-shut case, and it’s NOT a violation of the GPL.
Note that the GPL makes no claim to the libc being a derived work of the GPL’d work, so your argument does not work. The GPL solely prohibits redistribution of the GPLd work in this specific case, which is entirely valid within the context of rights granted to the author within the copyright law.
In essence: if you can satisfy the obligations under the GPL, you can redistribute. If you can’t satisfy them, you may *not* redistribute.
While pointing to other people’s behaviour and lawyers is nice and well, it doesn’t help Nexenta in court. The copyright holders, in this case Debian developers, can chose against whom they’ll play hardball.
In any case, it does not excuse Nexenta’s developers from the obligations towards the authors of the very source code they intend to build their distribution upon. If Febian developers tell them they are not satisfied with their license compliance, and should pull the binaries until they have the licensing issues sorted out, then they should have enough respect for the people who wrote their distribution to consider doing that for now, and working with Debian to fix the legal issues.
You can bet Sun Microsystems will be pretty pleased if a company (with no contact information to be found anywhere) manages to drag Solaris into the gutter with them.
cheers,
dalibor topic
Note that the GPL makes no claim to the libc being a derived work of the GPL’d work, so your argument does not work. The GPL solely prohibits redistribution of the GPLd work in this specific case, which is entirely valid within the context of rights granted to the author within the copyright law.
No, it is not within the rights reserved for the author. Or rather, what’s the difference between libc and, say, the kernel? That one happens to be in your address space? Sorry — libc is not a part of the Program as specified by the GPL. That is, the “special exception” doesn’t actually matter in libc’s case, because it isn’t part of the Program in the first place. (And this is why the fact that it is not a derived work matters — not in terms of the GPL per se, but in terms of the rights afforded (and not afforded) to any copyright holder to begin with.)
I believe that the rights to reproduce, copy and distribute one’s works are rights granted exclusively to authors via copyright laws. The author may agree to grant a license to others to do the same, and he may attach some conditions to it. The CDDL, for example, attaches certain conditions to redistribution rights granted to CDDL licensed code. I believe the authors of CDDL-licensed code are well within their rights when they attach certain conditions on redistribution of their works, and I don’t see why the same wouldn’t hold for authors GPLd works. YMMV.
But then, I think the whole ‘licensing controversy’ could have been avoided easily, had the Nexenta devs not decided to push it despite being warned that some developers of the source code they distribute and base their distribution upon disagreed with their line of legal reasoning.
Heck, the disagreeing Debian developers may even be wrong, too, but then it’s more polite and productive to try to convince them, rather than to work around or against them.
cheers,
dalibor topic
It’s a shame that disagreements always end up with a few histrionic people freaking out and name calling. I would like to know who is really behind Nexenta. Is this an actual company? Who is putting up the money?
“It’s a shame that disagreements always end up with a few histrionic people freaking out and name calling.”
Assuming you are referring to me:
Liars:
They promised to create a CDDL-compatible reimplementation of DPKG to resolve the licensing concerns raised by the Debian developers, and failed to carry out on that promise. They lied.
Thieves:
They then incorporated the GPL-ed DPKG in contravention of the license terms, despite having been explicitly told that it would be illegal for them to do so. That is theft, plain and simple.
That isn’t mudslinging. I believe the labels are accurate and entirely justified in light of their actions. If you believe otherwise, please feel free to state your case, together with the evidence to support it.
Unless you know the situation first hand and know these people personally, you’re jumping to conclusions and judging by your abbrasive tone, seems to be based motly on stray emotions…. I really can’t stand people like you.
Assuming you are referring to me: (I’m sure he/she was)
Liars:
*** They promised to create a CDDL-compatible reimplementation of DPKG to resolve the licensing concerns raised by the Debian developers, and failed to carry out on that promise. They lied.
A lie must also stem with an *intent* to decive, or distort truth. It is entirely plausible that these people simply have not completed the work yet. Do you know first hand as a FACT that these people simply will not deliver on that promise, or are you simply applying your time frame to this and jumping to conclusions and falsely accusing these people of being liars?
Thieves:
*** They then incorporated the GPL-ed DPKG in contravention of the license terms, despite having been explicitly told that it would be illegal for them to do so. That is theft, plain and simple.
Are you personally aware of any conversations that took place between them and the Debian folks, or are you speculating this? If you are speculating, then you *thieve* statement of fact, is really an assumption. Which of course would make you quite an ass if your opinion would turn out to be false.
*** That isn’t mudslinging. I believe the labels are accurate and entirely justified in light of their actions. If you believe otherwise, please feel free to state your case, together with the evidence to support it.
Seems to me it is in fact mudsligning and self agrandizment based on pure speculation.
“Are you personally aware of any conversations that took place between them and the Debian folks, or are you speculating this?”
As I already mentioned, the conversations in question took place in public on the debian-devel mailing list. I’ve already provided a link to at least one post in the thread, the start of which can be found at http://lists.debian.org/debian-devel/2005/11/msg00051.html – feel free to read the thread in its entirety for yourself.
“Seems to me it is in fact mudsligning and self agrandizment based on pure speculation.”
Only to one who hasn’t bothered to read the thread…
It’s such a joy to watch the perpetual internal strife of the people frequenting these forums. I wonder when some of the “fiery” individuals around will step up and propose possible solutions to the problems “discussed”, like some of the more constructive minds tend to do; solutions to license incompatibility in this case.
It seems to me that bickering is a more acceptable passtime, or perhaps it is just the tone one can expect from people entrenched in the IT-business, except, of course, if one is a paying customer with an agenda outside of the IT-business, in which case one does not wish to spend ones time on returning the mud slung.
Is this really the case? Or is it because you care so much about your operating systems and their licenses that you would rather be a minority of enemies than a constructive communitive? I, for one, do not recognize the community spirit you’re supposed to represent. That is far more prevalent amongst the many people of those “other communities” I consult, loving their Macs and Wintels.
I wonder whether you would use the same language if you were looking each other in the eyes. Maybe you would. That would make you about equal to the rabble I meet on their way to the local soccer game, drunk and loud-mouthed, full of confidence with little else on their minds but brawls and constant bickering.
Thieves? Liars? Whoever you are, those are harsh words you use and obviously haven’t bothered to look up the definition of:
“Lie”:
– tell an untruth; pretend with intent to deceive
(wordreference.com)
1) A false statement deliberately presented as being true; a falsehood.
2) Something meant to deceive or give a wrong impression
(dictionary.com)
“Thief”:
– a criminal who takes property belonging to someone else with the intention of keeping or stealing it
(wordreference.com)
One who steals, especially by stealth
(dictionary.com)
Neither of which can be said of people who take something (the property of whom is disputable, as this discussion proves) but do not intend to keep it and do this in the open, nor can be proven to have told an untruth or intend to deceive (again shown by the very discussion you are engaging in – it is not closed matter nor a logically decidable fact).
This is not to say that I personally think what they did is necessarily right, but I fail to see how bickering back and forth about it, not to mention calling people names helps either party, of which none of you are apparantly a member.
So easy to sling the mud at people you never have to meet in person, isn’t it? Fortunately they are busy coding while you are busy bickering.
“Neither of which can be said of people who take something (the property of whom is disputable, as this discussion proves)”
DPKG is the property of its copyright holders. That is indisputable. I already linked to a public statement of one of the copyright holders, Adam Heath (http://lists.debian.org/debian-devel/2005/11/msg00258.html) in one of my earlier posts. From that post, its context and his other contributions to the thread, I think it’s plain to see that he considers Nexenta’s use of his copyrighted material to be illegal. Such illegal misappropriation of his intellectual property can, IMO, legitimately be referred to as theft.
“but do not intend to keep it”
Really? Nexenta intend to stop using DPKG? Marvellous!
“and do this in the open, nor can be proven to have told an untruth or intend to deceive”
Let’s quote from a post by Nexenta developer Erast Benson (http://lists.debian.org/debian-devel/2005/11/msg00202.html):
“Nexenta community willing to make appropriate changes to the system and make it absolutely Debian legal OS. And more I’m looking into it, i’m sure it is quite easy possible by making main Nexenta OS CD to be GPL-free. All GPL software will be distributed on Nexenta Companion CD, if user wants to.”
There you have it. Categorical proof direct from the horse’s mouth that Nexenta promised that the main ISO would be GPL-free. But, since they included the GPL-ed DPKG, they obviously went ahead and broke that promise. Hence, I once again reiterate that it is completely justified to refer to them as liars.
And, as I’ve already pointed out, they also claimed that they would reimplement DPKG in a CDDL-compatible license since incorporating the original GPL-ed version would be in violation of the GPL. But obviously that was too much work, so they just nicked the original instead.
“I fail to see how bickering back and forth about it, not to mention calling people names helps either party”
Again, it isn’t name-calling when the use of the terms are justified by the actions of the accused party. You also completely miss the point of my argumentation. I simply believe that the community at large have the right to know what dastardly characters the Nexenta developers are. I want as many people to be aware of their immoral actions as possible.
“of which none of you are apparantly a member.”
Check your facts. At least Dalibor Topic is an active member of the Debian community, and I’ve also made (admittedly rather minor, compared to him) contributions.
I’d avoid getting into meta-bickering, myself.
On a side note, I am not a debian developer, although I do some work together with Debian developers and other volunteers in the debian-java part of the project.
cheers,
dalibor topic
DPKG is the property of its copyright holders. That is indisputable.
Since we are talking about the GPL here I think it would be appropriate to quote the FSF [1] here:
One effect of the term is a bias that is not hard to see: it suggests thinking about copyright, patents and trademarks by analogy with property rights for physical objects. (This analogy is at odds with the legal philosophies of copyright law, of patent law, and of trademark law, but only specialists know that.) These laws are in fact not much like physical property law, but use of this term leads legislators to change them to be more so. Since that is the change desired by the companies that exercise copyright, patent and trademark powers, these companies have worked to make the term fashionable.
and might I remind you that the whole rationale behind the GPL is to circumvent and overthrow copyright. [2]
My work on free software is motivated by an idealistic goal: spreading freedom and cooperation. I want to encourage free software to spread, replacing proprietary software that forbids cooperation, and thus make our society better.
So not only it is wrong in a legal sense to claim copyright as “proerty” but especially in a case where we are talking about the GPL it is down right dangerous.
[1] http://www.fsf.org/licensing/essays/not-ipr.xhtml
[2] http://www.fsf.org/licensing/essays/pragmatic.html
“So not only it is wrong in a legal sense to claim copyright as “proerty””
http://www.fsf.org/licensing/essays/not-ipr.xhtml most certainly does not claim that. You have completely misinterpreted the essay from which you quote. It merely argues against the use of “intellectual property” as a catch-all phrase to refer to patent, trademark and copyright law all at once. Hence its conclusion:
—
If you want to think clearly about the issues raised by patents, or copyrights, or trademarks, or even learn what these laws say, the first step is to forget the idea of lumping them together, and treat them as separate topics.
—
Whilst that’s good advice, it’s completely irrelevant, since I am not guilty of using the term “intellectual property” in so broad and generic a fashion as the one the essay describes.
It’s the same lumping and it’s the same word. The passage I quoted clearly refers to the “property” part of the word. Wether you call it “intellectual property” or just simply “property” is irrelevant. If anything it is a sign that the ideal of free software (as in free culture) has been completley lost on you.
If you are going to argu against me you could at least have done it properly. The second link i refer to is all about how you don’t have to agree with the FSF to use the GPL. Whish destroys in effect destorys my premise that usage of GPL implies agreeing with the free software ideal…
Edited 2005-11-09 06:49
“It’s the same lumping and it’s the same word”
No, it really isn’t. Once again: http://www.fsf.org/licensing/essays/not-ipr.xhtml is protesting only that the phrase “intellectual property” has become an overly broad, confusingly generic umbrella term which is being used to refer to copyright, trademark and patent laws as though they are all one and the same. It is arguing only against the use of that specific phrase (“intellectual property”) and even then, only when it is being used as a catch-all term that would lead to over-generalization in the way that is described in the essay.
The only way that your own blinkered interpretation holds is if one chooses to read only the tiny fragment you quoted whilst steadfastly ignoring the whole of the rest of the essay. I think that doing so is intellectually dishonest in the extreme, but perhaps that’s just me.
Your quote actually states that making a separate CD with the GPL-ed code was their INTENTION, not that they promised to do so.
Immoral actions? Another strong statement. Is it immoral to use code that someone else wrote and published under a public ownership license like the GPL? Isn’t the very reason that Linus chose the GPL for Linux that he wished to relinquish control of the OS? (Although I do think he should have used the BSD-license – see below). Isn’t the whole point of the GPL that people should have the freedom to use the code for their own benefit, as long as they don’t keep others from doing the same – with the added “benefit” of forcing others to publish their work under the same license?
It seems to me that, in this discussion, that freedom is now not as important as the “owners'” rights to revoke that basic freedom. Protecting the right of the authors of the code – their right to dictate who uses it and who doesn’t – is what is the most iterated topic here.
Have Nexenta kept code from you or other people? Is the CDDL viral? Will what Nexenta did change the licenses of the DPKG-code into CDDL? Please enlighten me, and once you have – as an aside, try looking at the BSD license, which complies with what I understand as true freedom: The freedom to do whatever you like with whatever somebody ACTUALLY gives away. Anything else is less, and – to my mind – a result of people pointing moral fingers the way kids do at kindergarden.
Again, what other people do with something once I ACTUALLY relinquish control (as do the BSD-licensing coders) is non of my business, which is exactly what freedom is about: Other people’s lives are theirs to live, not mine.
But maybe, by stating these views, I just fuled another endless discussion on the same tiring subjects.
The indesputable fact remains, some people of the open source community in the OSNews forums spend more time discussing the details of IP than I hear within corporate environments, but maybe that’s because I live outside of the States (I’m a Dane). Or maybe it’s because by introducing the many complicated and certainly restrictive licenses of “freedom”, the open source community – as well as the Free Softare Movement – have created their own license hell to deal with.
I, for one, wish to license this comment under the LPBL – the Less Puclic Bickering License, to, hopefully, guard it against too much flaming.
-Mike
“Is it immoral to use code that someone else wrote and published under a public ownership license like the GPL?”
If you don’t respect the terms of the license under which the code was released (as is happening in this case) then yes, it is.
“It seems to me that, in this discussion, that freedom is now not as important as the “owners'” rights to revoke that basic freedom.”
The freedom to distribute GPL-ed and CDDL-ed code in the way that Nexenta are doing was never granted to them in the first place. As has been mentioned countless times already, the
—
However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.
—
clause in the GPL was written precisely to *deny* them the ability to do so.
As has been mentioned countless times already, (OS exception clause) in the GPL was written precisely to *deny* them the ability to do so.
Well, to be fair, it was written to prohibit distributing GPL code together with *non-free* libc, not to prohibit distributing GPL code together with *free but GPL-incompatible* libc. I believe that current situation was unforeseen by GPL authors.
I wonder, how many OpenSource projects are ported to windows with MS propietary libs?
I wonder, how many OpenSource projects are ported to windows with MS propietary libs?
Lots and lots of them. That’s fine: they aren’t distributing MSVCRT together.
MSVCRT is normally distributed with the major components of the operating system on which these GPL codes run, which is the case explicitly allowed by GPL.
I tend to agree, though it’s hard to judge what the FSF may or may not have considered during the drafting time. It will be much more fun when the drafting for GPL3 begins, and I guess this issue will pop up there to be resolved either way in GPL3.
In any case, the traffic on debian-devel suggests that the Nexenta devs are trying to work together with Debian developers now to fix potential GPL-compliance problems. Storm is over, so I guess it’ll only take a few days now until it hits /. and all hell breaks lose again.
best of luck to everyone,
dalibor topic
Afaict from reading the threads on the debian-devel list, a few possible solutions have been proposed:
a) Nexenta could relicense Sun’s code, if they obtain the copyrights form Sun, or a license to do so
b) Nexenta could ask Sun to change the CDDL
c) Nexenta could ask Sun to relicense the libc alone
d) Nexenta could ask Debian developers to relicense their tools
e) Nexenta could port glibc to OpenSolaris
f) Nexenta could rewrite dpkg and other debian tools under a license of their choice
g) Nexenta could stop redistributing works from Debian they do not have the consent of copyright holders for
cheers,
dalibor topic
They then incorporated the GPL-ed DPKG in contravention of the license terms, despite having been explicitly told that it would be illegal for them to do so. That is theft, plain and simple.
Copyright infringement is not theft. It’s, uhm, copyright infringement. Whats next, are you going to call them pirates and bloody murderers?
Microsoft?
cheers,
dalibor topic
Post removed (I should learn to read, lol)
Edited 2005-11-08 02:46
“First of all, the trademark issue: In choosing to name their product “GNU/Solaris” they’re using “Solaris”, a Sun trademark, in a way that is in contravention of Sun’s trademark policy (http://www.sun.com/policies/trademarks/). Given that trademark law *requires* trademark holders to aggressively police use of their mark or risk losing it altogether, I think we can expect a response from Sun’s lawyers pretty soon.”
Actually I do not beleive their is anything wrong with this name other than they might need to add a trademark legend after Solaris? In this case Solaris is being used as an adjective to describe what the kernel is. Using the phrase “GNU/Solaris” implies saying “the GNU system running on top of the Solaris operating system kernel”. If anything, they should change their name because the name of the kernel technically is SunOS and not Solaris. Not to mention, not only is my post technically in violation of their trademark, there are many other posts that are as well. I guess they will be coming after me too…
Nexenta is a company making an operating system. Sun is a company making an operating system. Both are marketed as Solaris.
$insert obligatory highlander music
cheers,
dalibor topic
Nextenta is marketing Nexenta OS. They are using the word Solaris to describe the kernel.
All this thread about some developers being thieves and liars made me a bit depressed (like to many others, I guess). The project looks interesting, and those developers have probably put a lot of work to give this OS for free, as Debian and Ubuntu do. Shouldn’t they be brothers beyond the words of licenses they use?
Can’t a license be overridden by the owner’s permission? In that case, couldn’t Nexenta developers could just ask Debian developers “Hey, we’d like to use dpkg, but there seems to be a conceptual problem with the license. Can we use it anyway?” And Debian developers should have answered: “Sure, no problem”.
Is it so difficult?
Can’t a license be overridden by the owner’s permission?
Of course it can.
In that case, couldn’t Nexenta developers could just ask Debian developers “Hey, we’d like to use dpkg, but there seems to be a conceptual problem with the license. Can we use it anyway?”
That’s exactly what happened. They asked to debian-devel mailing list.
And Debian developers should have answered: “Sure, no problem”.
No. Debian developers, specifically some of dpkg authors, firmly said, “No”. Syntaxis already mentioned it:
http://osnews.com/permalink.php?news_id=12569&comment_id=57597
http://lists.debian.org/debian-devel/2005/11/msg00258.html
It doesn’t seem to be a consensus though. Nexenta developers are more than free to ask 24 authors listed in dpkg copyright, to get relicensing when possible, and to rewrite parts that authors didn’t relicense.
“No. Debian developers, specifically some of dpkg authors, firmly said, “No”. Syntaxis already mentioned it”
But in that thread they’re asking to relicense dpkg, which I understand the authours don’t want to do. But did they just ask permission to USE it (without changing the license)? I don’t see why Debian devs would have say no to that.
Anyway, it’s a sad thing. I hope it’s solved in the right way.
This thread is fun 😉
But in that thread they’re asking to relicense dpkg, which I understand the authours don’t want to do. But did they just ask permission to USE it (without changing the license)? I don’t see why Debian devs would have say no to that.
There is no law that requires a work to be licensed by one and only one license. Dpkg is availible to everyone under the terms of the GPL. To grant permission to USE it, as you say, is by definition to license it. So if the nexenta developers would be granted permission to use the software in non GPL-compatible ways, they would by definition be granted a license to do so.
If that be the case, then do explain GAIM, OpenOffice, Mozilla/Firefox/Thunderbird, MySQL Community Edition and other OSS for Windows that is at least dual-licensed under the GPL. I guess for some reason in your mind it’s ok for a GPL’d program to be released for Windows, but not Solaris. Do you even **have** the source for Windows, let alone a right to look at it?
I guess the big difference is that in the cases you mention it is the copyright holders that distribute the software.
does this contain dtrace/zones/smf?
if not all we need is a standard solaris 10/express with a good package manager.
Not sure yet, installing it now, but I would say yes to dtrace and smf, just not zones.
OpenSolaris can’t use zones at the moment
http://opensolaris.org/os/downloads/releasenotes-20051103.txt
9. Zones are not usable on OpenSolaris, due to missing package
support and tools.
GNU’s Not Unix, but Solaris is. So, is this system Unix or not?
Yes the Solaris Operating System is UNIX, this as far as I can tell just uses the Solaris Kernel and GNU and knowing that GNU is not unix, that means that this is not UNIX…
The same rules applies to GNU/Linux, the Linux Kernel is POSIX Compliant, so it could be made into a UNIX if the rest of the operating system could be classed as UNIX.
started the livecd on my lap to have a looksie , but this thing won’t start up . i can log in and it started the xserver and after 30 min all i got is the debian splash and a forever rotating cdrom .
hmmm …..
well that was commented down, wonder who voted……
Search through the opensolaris forums and you will see a lot of SUN dev’s looking at it.
The theme in that screenshot lookd very Ubuntu-ish, and yet better. I may have to give this a try. I love Debian, but the whole CDDL vs. GPL thing makes me wonder how long this project will be around.
My thoughts at this point would be: stay away from Nexenta.
Nexenta’s post to Debian regarding changing apt’s license could actually be read as a “do this or else” type of message which seemed kind of strange.
I would not be surprised if those funding Nexenta have an alteria motive, which is to spark FUD regarding the GPL and attempting to get GPLed projects relicensed to less free licenses.
http://www.gnusolaris.org/gswiki/Download#head-63aab7b98f7efd7d413f…
Our “Elatte” Alpha 1 release is distributed via ISO images only. We provide two versions of our software: a LiveCD and an InstallCD.
We are in need of ISO mirrors; if you can help us by mirroring our software, please contact us. We currently have a limited bandwidth contract with our provider; we appreciate your patience while downloading the ISOs directly from our main site.
Heh, sorry: that’s what BitTorrent is for. Combined with the quality of the site (the aforementioned MySql vomits), this leaves me worried that this is run by amateurs.
<stdlib.h>
<iostream.h>
<etall.h>
Oh sure and they all derive from ANSI/ISO
troll
GPL is infectus code and there isnt going to be 10,000 copies of open solaris floating around.
As I said dpgk porting of apps etc why not build one driver for solaris under dual licence release. Rome was not built in a day and seems like lots of solaris based roge products are showing up. GPL is easy to understand and is much more effective than patents. Sun microsystems may actualy stand up and kick some buts for abuse of licencing.
… about license stuff. Is Open Source, nobody is *stealing*. They are *NOT* distributing XXXX package and saying they did it or that it belongs to them. All these FOSS advocators drive me nuts. One day Apple is Evil, the next day is Microsoft, now FOSS is evil… because it *steals* ?
C’mon… get a grip. (and a shot of thorazine)
Yes, I’ve read your posts, I know what happened, I’ve seen the Site. Yet who cares? We’ve got more important things to do.
There is no such thing as GNOME office. It simply don’t exist. So please stop making false articles.
The GNOME Office people appear to think otherwise:
http://www.gnome.org/gnome-office/
And some people wonder why the masses aren’t embracing FOSS.
With regards to linking, I believe the following is really more accurate:
a) ILLEGAL = CDDL application linked with GPL lib
b) LEGAL = GPL application linked with CDDL lib
.. think about. If (b) was not ok, then that means anyone can FORCE everyone else to convert to GPL simply by making a one line program that links with the other persons’ non-GPL lib. Can you imagine that? An unstoppable viral clause!
This is why BSD folks CANNOT link against GPL libs, because their BSD work will *effectively* turn into GPL code but GPL folks can link and use source from BSDL’d code. .. and please no excuses about seperate copyrights or licenses since the entire work would now be considered derived from the GPL therefore anyone who uses such a combined work of BSD + GPL from then on must apply the GPL to their own
Remember that linking or more specifically interface usage is UNIDIRECTIONAL! A library cannot control nor know who will link with it. Derived work means that final product (executable or binary) *using* (i.e. calling or branching to) GPL code must be also be GPL — that does not imply the inverse!
Another example: the linux kernel which links with kernel modules. There are already of LEGAL proprietary non-OSS binary kernel modules.
It does not work that way. The GPL is not the magic wand of relicensing. Only the author of a chunk of code can legally relicense.
cheers,
dalibor topic
I wish everyone would shut up about the GPL the GNU and all that junk and talk about whether the OS WORKS or not. Geez!!!!
the downloads have been removed from the website.
The downloads are temporarily removed. Our intent is to provide complete source code that exactly corresponds the binaries. To achieve that, we need to expedite some of the unfinished testing (to be able to check in the remaining local source), and clean up parts of the APT and SVN repositories. Sorry for the inconvenience!
yup.
It does not work that way. The GPL is not the magic wand of relicensing. Only the author of a chunk of code can legally relicense.
I do realize that. I’m more or less explaining in terms of what would happen if it was and wasn’t like that. That is, if said author decides to write code that combines or links with GPL code into his own, then the author is obligated to release his finished product under the GPL. But if the author has existing GPL code that he wrote, and decides to link it with another developer’s non-GPL code, GPL does not apply to the non-GPL code because the usage direction is such that: GPLfunc() calls special_func() in libnongpl.so
Such is the case with GNU tools linking with Solaris’ CDDL libc. Otherwise, if it were as suggested earlier about Nextena “stealing”, then such a link in that direction would cause unresolvable issues I mentioned before about “forced” relicensing.
I do realize that. I’m more or less explaining in terms of what would happen if it was and wasn’t like that. That is, if said author decides to write code that combines or links with GPL code into his own, then the author is obligated to release his finished product under the GPL. But if the author has existing GPL code that he wrote, and decides to link it with another developer’s non-GPL code, GPL does not apply to the non-GPL code because the usage direction is such that: GPLfunc() calls special_func() in libnongpl.so
No you don’t. If said author decides to write code that combines or links with GPL code into his own, then the author can legaly distribute the result if he complies with the GPL meaning, in part, that the parts of the code that is his must be availble under the GPL to those he distribute to. But if the author has existing GPL code that he wrote, and decides to link it with another developer’s non-GPL code, he would have to have a license from the other author to be able to do so legally. He would also not be able to license the whole work as GPL without permisson from
the other author.
http://shots.osdir.com/slideshows/slideshow.php?release=495&slide=3
Great project.
In what way is this in competition with SkyOS? this is the second time you’ve used this subtle link in a totally irrelevant place what’s going on?
Am I missing something?
He’s making fun of the crazy naming scheme of pre release versions. SkyOS has Beta 8.6 Alpha 2 and Nexenta OS has Pre-Alpha 1. This is my interpretation and I could be wrong though. Now if you’ll excuse me, i’m off to work On Uber1337 OS Beta 5 Alpha 4 Beta 1 Alpha 5…
No you don’t. If said author decides to write code that combines or links with GPL code into his own, then the author can legaly distribute the result if he complies with the GPL meaning, in part, that the parts of the code that is his must be availble under the GPL to those he distribute to.
I stated the author is obligated to release his code which *uses* GPL code (as in calling from source or calling from linking to a GPL lib) as GPL. Where’s the disagreement? Secondly, that entire derived work i.e. source of all binaries which have such links, of said author’s must also be GPL’d. He won’t be able to keep non-GPL source in such binaries.
But if the author has existing GPL code that he wrote, and decides to link it with another developer’s non-GPL code, he would have to have a license from the other author to be able to do so legally. He would also not be able to license the whole work as GPL without permisson from the other author.
Well you certainly don’t need to get permission if the license that the other author uses already allows you to do so!
Anyways, my whole point is this:
THERE IS NOTHING WRONG WITH GPL APPS LINKING TO CDDL LIBS because the CDDL allows it! The CDDL is less restrictive in that regards; the other way around is not true meaning that a it is wrong for CDDL apps to link with GPL libs because of the viral of the GPL license.
(and before some nitpicks, yes I’m assuming this is for work that will be publicly distributed.. and this does not apply to *LGPL* libs)
Where’s the disagreement?
That he is not obligated to GPL his code. If he does not comply with the terms of the GPL license the license is simply null and void, which means that the case reduces to simple copyright infringement. He can thus choose to a)cease distributing or b) keep distributing illegally.
Anyways, my whole point is this:
THERE IS NOTHING WRONG WITH GPL APPS LINKING TO CDDL LIBS because the CDDL allows it! The CDDL is less restrictive in that regards; the other way around is not true meaning that a it is wrong for CDDL apps to link with GPL libs because of the viral of the GPL license.
The confusion here is about who is copyright holder and who is licenser or licensee. In this case the copyright holder of some code whishes to create a “Larger Work” (as defined in the CDDL) with his code and some other code under the terms of the CDDL. Whether he makes his own code availible under the GPL or not has absolutley nothing to do with anything in this case. So yes the CDDL allows “Larger Works”, quote: You may create a Larger Work by combining Covered Software with other code not governed by the terms of this License and distribute the Larger Work as a single product. In such a case, You must make sure the requirements of this License are fulfilled for the Covered Software.
If he did not have copyright for the code and could only license it under the terms of the GPL (as is the case with dpkg and Nexenta, and which you imply when you say GPL APPS LINKING TO) he would have no legal way to distribute a “Larger Work” as the GPL requires that the whole work also be licensed under the GPL while the CDDL requires that Covered Software that You distribute or otherwise make available in Executable form must also be made available in Source Code form and that Source Code form must be distributed only under the terms of this License.