Slashdot Mirror
CDDL Project Leader on the CDDL
SunFan writes "Claire Giordano, Sun's lead on CDDL development, gives the inside story on how the new license was developed. She discusses how people within Sun debated the various licenses, including the GPL, and shows why the GPL, BSD, and MPL licenses were found to come up short in meeting the needs of Sun's broad customer and developer base."
I find that there are just not enough licenses out there, so I'm thrilled to hear there's a new one. Eventually there may even be more licenses than things to be licensed, and what then?
I'm just sayin' is all...
Starsucks
' is your friend.
We announced a great $2-billion we-both-respect-IP deal with Microsoft, and the F/OSS guys won't be giving us $2 billion soon.
How do we write a license that keeps Microsoft happy and trys as well as we can to divide and conquor the Linux developer base."
sun, and their servers, are not going anywhere, unless the military goes bankrupt, and the banking industry, and ...
What a lame excuse. Why not do the same thing as the Linux community, and release it without those drivers until someone writes an open-source replacement.
Hmmm ... you might want to learn how to punctuate "we're" before you start running smack...
In case you didn't notice, by the end of the comment I did start using a "'"....
According to the article, the project follows the mantra of "It's not about the license, it's about the community."
To paraphrase Hillary Clinton, it takes a village to raise a software project.
I am scientifically inaccurate.
What the author of "Failed as in succeeded wildly." meant was LGPL-style licenses. LGPL has been around a lot longer than MPL and actually fits what he describes.
"Wow were SUN... we've lost all the other markets... so lets act like we're open source!"
Proof of "everyone loves a winner", and it's correllary "everyone boos a perceived loser".
Q - Why do we need so many licenses, aren't the ones we have enough already?
A - Should at 16 year old be allowed to drive a bus or semi after showing they can parallel park, of course not. Same goes here, different licenses suit different needs, now quit yer bitching.
It is open source. What more can you ask for?
It's not GPL-compatible. I appreciate you want to be able to mix in proprietary code. If you need that, you can do one of two things. You can use LGPL, or you can include a clause "May also be sublicensed under the GPL, version 2.0 or newer." The only thing you lose is the patent litigation protection. You can also grant yourselves rights to change the main CDCL without alienating the mainstream FLOSS community.
Either way, a large number of people, for whatever reasons, distrust Sun. Releasing under a license that is widely percieved to be incompatible with the greater free software community is fanning those flames. Most people percieve this as a simply ploy to make it impossible to move code back and forth between the Linux kernel and Solaris. That's not really big on freedom.
One of the great parts of free software is rapid development. About half of my apps are built by glueing together code from other people's programs. Making Solaris not compatible with the GPL breaks that, and it's not good for building a developer community around Solaris.
I wish you guys the best of luck, but I think that given a GPL-incompatible license, you'll have a really hard time building a large developer community.
With a GPL-compatible license, if Solaris got to be better than Linux, virtually no one would think twice about switching from Debian GNU/Linux to Debian GNU/Solaris.
Also, notice the trend. Virtually everyone (Netscape, TrollTech, etc.) starts with their own crappy license, and eventually, switches to a dual-licensing model that includes GPL. Follow the lesson, and start off on the GPL side from the start.
...so lets act like we're open source!
Gee, I didn't hear people complaining when they opened the source to Open Office, their grid control, or any number of other things.
For [ insert deity of choice here ]'s sake, the CDDL IS AN OPEN SOURCE LICENSE. Certified by the OSI no less! "act like". Ppffttt....
Just because it isn't GPL, or BSD doesn't mean it isn't open source or that they're "acting".
People like you are going to get a few million lines of code stuffed in their mouth as revenge when OpenSolaris comes sometime before the end of this quarter.
....those who can't work in marketing and legal and come up with licenses.
These posts express my own personal views, not those of my employer
Anyone think they are putting their version of an open source license out there to get developers to work on hardware compatability for the x86 market ? Sounds like a call for free labor.
Does it depend on how you look at it?
Per the article: I realized how important it was for us to allow OpenSolaris kernel source files to be compiled and linked with other open source files and even with proprietary source files in the kernel. GPL would not allow this. On Day 1 of OpenSolaris, because some OpenSolaris IP is encumbered by other companies (example - 3rd party drivers), we're going to have some source files in the kernel that will remain proprietary. Hence GPL was out of the running.
Does this mean that the reason OpenSolaris isn't GPL'd is that it needed to include some non-open stuff?
If that's the case, well fine, I support the right of anyone to license their stuff any way they want ... but calling it "open" seems to re-define "open".
Correct me if I'm wrong, please!
--- Attorneys Assisting Citizen-Soldiers & Families -
From TFA: On Day 1 of OpenSolaris, because some OpenSolaris IP is encumbered by other companies (example - 3rd party drivers), we're going to have some source files in the kernel that will remain proprietary. Hence GPL was out of the running.
The whole point of opening the source and creating a community is so that people can develop the things they need, free of the problems proprietary code brings. I find it hard to believe that much Solaris kernel code belongs to anyone other than Sun.
Linux started out with very few drivers, but now supports most common hardware. Sometimes this support takes the form of binary or wrapped drivers, but that hasn't prevented Linux from remaining under the GPL.
This is just more of the usual Sun guff.
Vino, gyno, and techno -Bruce Sterling
The license may be opensource, but Sun sure as hell isn't, they sold java to M$ for chump change for fucks sake, that small amount of cash isn't going to make a difference, even to a company falling as fast as Sun, its clearly indicative of a greater plan to retake the Unix world. Sun is very clearly anti-open-source, after all they are Sco's secret licensee.
They are afraid they will lose control, not that it matters much.
Get a grip, will ya? Sun hates Microsoft--enough to extract $2 billion and then some out of them, which is the biggest MS payout yet. Sun has no love for SCO's business model, either (Solaris 10 shows they can and do compete on merit).
I just wish it was GPL compatible.
I know, I know, I shouldn't complain....it just seems like such a waste.
if the GPL is more restrictive it would have been nice to have a conversion clause or something. Though even if that was legally possible, it might still be a strategically bad move for Sun, since code could not go back once it was converted to GPL.
Sigh...such a waste.
Tharkban (It is a signature after all)
Kinda disappointing write up...
She says that "early in the planning for OpenSolaris that we needed to use a "true" open source license - a license that complies with the terms of the Open Source Definition and that encourages royalty-free use, modification and distribution."
As opposed to all those other false ones out there that completely fail in the above goals. They considered all options, she claims, except the glaringly obvious LGPL, it would seem.
We at least know that Sun recognises "it's all about the community" which is no doubt why they have left themselves such great scope for slagging off the huge base of *successful* commmunities which just happen to use those awful GPL licences.
-- Free software on every PC on every desk
If only Sun fans weren't so embarresed they need to be anonymous...
The main purpose of the license is to enable Sun to claim tens of millions dollars as a charitable tax credit for making the 'donation' to their own charity organisation, while still allowing them to retain some modicum of control. I think this license achieves that purpose admirably. When in doubt, follow the money trail.
It smells like someone is trying to get free labor on their OS.
Either that or research labs and universities will have a field day tweaking the fastest OS to be even faster for their workloads.
Gee, I didn't hear people complaining when they opened the source to Open Office, their grid control, or any number of other things.
Open Office was a great contribution, but sadly this is not more of the same. See PJ's comments on the CDDL for details.
Vino, gyno, and techno -Bruce Sterling
I just read the definition of the definition of the CDDL. Outside of the fact that it's not compatible with the GPL (which can always be solved by dual-licensing), I find it reasonable. Of course, I may have missed some details, so I'd be happy if someone could point out what the problem is (outside of "it's from Sun" or "it's another license"). Also, I see that they distinguish between "Initial Developer" and "Contributor", but I don't see any difference between the rights given to one and the other. Can anyone enlighten me on this too?
Opus: the Swiss army knife of audio codec
Well, you're certainly welcome to write your own drivers for something that had to be held back in OpenSolaris due to a 3rd-party's constraints which are external to Sun. Come on, isn't that one of the beloved Linux mantras?
/. posters seem to be missing here is that Sun /did/ look at the GPL... they didn't outright dismiss it. Sun found that they were encumbered by 3rd-party constraints which the GPL was incompatible with and so they had to short-circuit to the next best thing - the CDDL. There's no single magic wand waving that can make that catch go away.
The thing cynical
Solaris is an awesome technology. You would be - no pun intended - CLOSING yourself off if you think being able to look at and use the Solaris source is irrelevant. But lots of people here seem to be taking after Henry Ford, in that your open source is painted "any color that he wants, so long as it is black" (read: Linux/GPL)
The term "myopic" comes to mind.
It's OSI approved i.e. OPEN SOURCE. Some slashdotters are only getting their knickers in a knot because Solaris is not Linux. Kind of sad that many people have this view of open source.
Sun could open source their entire software line (and are well on their way to doing so (with the exception of Java)) and these folks would still not be satisfied.
Some of us actually have/had a job so it's a sensible idea for us to post AC under these sort of articles. You on the other hand can quite happily come out with these laughable statements from your parents basement.
Raise a software project? I'm intrigued by your ideas and would like to subscribe to your newsletter. Where is this village? Is this like the Amish version of software developers? Can they finish a bug-free n-tier software development project in a single iteration?
Just some facts:
Sun's revenue Q3: $2.625 billion
Net loss for the third quarter of fiscal 2005 on a GAAP basis was $9 million
The cash and marketable debt securities balance at the end of the quarter was $7.357 billion.
http://www.sun.com/aboutsun/investor/earnings_rele ases/pr/2005-q3.html
So, your remark should be modded Troll....
http://www.openbsd.org/lyrics.html#33
There are a lot of threads about the issue of GPL incompatability. But not including that the author of the article mentions that he wants the CDDL to become a license in line with the GPL or the BSD license. The problem is:
1) Its not as simple as the BSD license thus manages to duck the hard issues
2) Its not as well thought out as the GPL which addresses the hard issues dead on.
1) CDDL definition of source code: "Source Code" means (a) the common form of computer software code in which modifications are made and
(b) associated documentation included in or with such code.
GPL definition of source code: The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. 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.
What does documentation included with such code mean? Say for examples most changes are made in response to bug tickets do you need to include them? Very vague
2) Obligation to distribute source
CDDL - Any 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.... You must inform recipients of any such Covered Software in Executable form as to how they
can obtain such Covered Software in Source Code form in a reasonable manner on or through a medium customarily used
for software exchange.
GPL -- Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
You can see the differences.
-- How long do you have to provide source code
-- What about casual copying
-- What if the information in the executable is no longer true (like you can download this code from www.xyz.com/abc/bcd)?
etc...
3) CDDL on "modifications" -- The Modifications that You create or to which You contribute are governed by the terms of this License. You represent that You believe Your Modifications are Your original creation(s) and/or You have sufficient rights to
grant the rights conveyed by this License.
GPL on derived works -- Each time you redistribute the Program (or any work based on the Program),
The GPL uses the notion of derived work which has tons of case law behind it. The CDDL uses "modifications" which opens the door to which sorts of derivations are modifications. I don't see any point in this change.
Anyway I'm not sure what the CDDL offers over the GPL other than lots of chat about patent claims.
This is what happens when you violate the GPL.
What about OpenSolaris implies lock-in, and who modded this insightful? In fact, there was a mention at Blastwave that some people are interested in a PowerPC port of OpenSolaris. That's on top of all the x86 hardware it runs on, along with all the SPARC hardware.
-- Microsoft is the most expensive commodity operating system and office suite vendor in the marketplace.
The CDDL was mentioned on an earlier slashdot article: Sun's Schwartz Attacks GPL.
The gist of what he had to say: "GPL sucks! CDDL rulez!"
Well, you're certainly welcome to write your own drivers for something that had to be held back in OpenSolaris due to a 3rd-party's constraints which are external to Sun. Come on, isn't that one of the beloved Linux mantras?
/. posters seem to be missing here is that Sun /did/ look at the GPL... they didn't outright dismiss it. Sun found that they were encumbered by 3rd-party constraints
which the GPL was incompatible with and so they had to short-circuit to the next best thing - the CDDL.
Under the CDDL they would have to be rewritten anyway if we wanted to make our own Solaris distros, because the 3rd party patent deals / licence grants apply only between that party and Sun. About which, more below...
The thing cynical
Is the next best thing the CDDL, or is it the MPL upon which the CDDL was based? Under the MPL (section 3.4), Sun would have had to disclose the extent to which the code was encumbered by third parties. One of the few differences between the CDDL and the MPL is the removal of this clause. The implication of this is that Sun (who already has IP deals with the third party vendors) can use CDDL code in commercial products without any legal worries. But open source developers distributing CDDL cannot, and Sun's modification to the licence means that the encumbered parts of the code cannot be distinguished from the unecumbered parts.
The point of the CDDL seems to be to supply Sun with free development and debugging for code which only it can safely distribute. Lest you think this is me donning my tinfoil hat, please read the relevant Groklaw commentary.
I am not asking Sun to release other people's encumbered code. I just want them to identify (and, for bonus points, encapsulate it) so that it can be replaced with unencumbered code.
Solaris is an awesome technology. You would be - no pun intended - CLOSING yourself off if you think being able to look at and use the Solaris source is irrelevant.
Solaris does have a lot of cool stuff in it. Sadly, learning from some of that stuff could get us into legal trouble. It might be better for people who also do GPL development not to even look at the code.
Vino, gyno, and techno -Bruce Sterling
The same way that Nvidia drivers work with a hostile GPL.
Sun needed a way to present themselves as going open source without allowing the open source community to get any benefit from this. The CDDL was their tool to do this.
Everything else, any rationalizations for why they engineered their license the way they did, is just PR.
Gee, I didn't hear people complaining when they opened the source to Open Office, their grid control, or any number of other things.
Yes. In fact, one would not have heard complaining about those things. In fact one would have generally noticed they were welcomed quite enthusiastically.
Yet people seem to be complaining about the CDDL releases quite a lot.
Do you think this maybe might indicate there is some sort of difference between the two situations?
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
I have been wondering lately, when Sun do open up Java a little more (which process has begun by making some subtle changes to the license of Java-1.5 to allow re-implementation), presumably they'll use the CDDL or something. Hopefully they will simultaneously submit Java to the EMCA or other standards body.
Anyway, my question: Why doesn't IBM just take their own Java implementation, put it under the (L)GPL, modify the language slightly in a backwards-compatible way (like C# only backwards-compatible with Java), and then release *that* to a standards body under a new name (with no reference to Java, as was the case when C# came out, in order to eliminate the need to claim compatibility to use the name), and release all their ex-Java tools under the GPL?
IBM is much more likely to do that than Sun, and we wouldn't have to worry about the CDDL ramifications.
We need Open Java for the GNU/Linux desktop.
OpenSolaris is completely open! You can create your own OpenSolaris Distro! Casper H. S. Dik, one of the OpenSolaris community advisory board members talks more about the CDDL. He basically states the CDDL is the MPL without some of the restrictions..
o la ris/search?group=comp.unix.solaris&q=Alan+Coopersm ith+cheapbytes+Opensolaris&qt_g=1&searchnow=Search +this+group
An example he gives is the right for the Mozilla foundation to revoke the MPL and the requirement to have lawsuits settled in California (which would be bad for international users). That is why SUN didn't use the MPL.
SUN enginner Alan Coopersmith points out in comp.unix.solaris that anyone can create an OpenSolaris distro.
http://groups-beta.google.com/group/comp.unix.s
If the GPL wasn't so anal it wouldn't be a problem The problem is with the GPL not everything else.
Many readers who complain about the CDDL don't even bother to read it or think about the reasoning behind it. They just read "Sun" and "GPL incompatible" and those sign stimuluses are the only base of their negative opinion.
Tell them that the Mozilla Public License used by Firefox is GPL incompatible and they have no problem with it. Or what about Apache's or the license of LaTeX. No problem. In my opinion those people don't reflect very much upon the consistency of their reasoning.
I for myself welcome the CDDL. It certainly represents a small improvement over the MPL and is in pratice not very different from the LGLP except for the added patent protection which is not bad either. Note that I am not saying the CDDL == LGPL + Patent protection. What I mean is that in practise it doesn't hurt to think of it in those terms. And if you consider all this--how can you blame Sun for creating such a powerful license?
One more thing. Joerg Schilling, famous for creating Linux's cdrecord program is creating his own OpenSolaris Distro called SchilliX. http://schillix.berlios.de/ Blastwave.org, a website that contains free software for Solaris users is also creating an OpenSolaris distro according to the regulars of the usenet group comp.unix.solaris. A benefit of OpenSolaris is that it will have a proper cdrecord functionality. Again SUN Engineer and OpenSolaris community advisory board member Casper H. S. Dik says this about OpenSolaris compared to linux. In message ID Casper Dik writes: "I think [Joerg Schilling (author of cdrecord)] prefers neither but rather has a proper USCSI interface; note that this isn't really a "IDE SCSI" emulation layer; ATAPI is SCSI-over-ATA. " The problem with the linux kernel is that a scan by cdrecord may not find all of your cdrecord devices because of the changes in the 2.6 kernel. Even Alan Cox mentioned that it was not optimal in one of the long linux.kernel threads. :-)
Losing 9 mil out of 7.3 billion is called "breaking even". Sun is doing pretty well considering their post-bust blues. The fact that they've turned around so much in only a couple years is awesome.
It may be an "Open Source" licence, certified by the OSI - but it's not "Debian-free" according to the Debian Free Software Guidelines (on which the OSI is based) by my understanding. The OSI is currently trying to limit the number of Open Source licence variants. In accepting this one, they've accepted something that most of the community either can't or won't use. I would suggest that dual licensing of code will only increase or that some people will actively exclude OpenSolaris from using their code [in the same way that OpenMotif (R) is licensed only for use on free software - all else you pay for].
[oops, fixed the formatting!]
:-)
One more thing. Joerg Schilling, famous for creating Linux's cdrecord program is creating his own OpenSolaris Distro called SchilliX.
http://schillix.berlios.de/
Blastwave.org, a website that contains free software for Solaris users is also creating an OpenSolaris distro according to the regulars of the usenet group comp.unix.solaris.
A benefit of OpenSolaris is that it will have a proper cdrecord functionality. Again SUN Engineer and OpenSolaris community advisory board member Casper H. S. Dik says this about OpenSolaris compared to linux.
In message ID 42611331$0$148$e4fe514c@news.xs4all.nl Casper Dik writes:
"I think [Joerg Schilling (author of cdrecord)] prefers neither but rather has a proper USCSI interface; note that this isn't really a "IDE SCSI" emulation layer; ATAPI is SCSI-over-ATA. "
The problem with the linux kernel is that a scan by cdrecord may not find all of your cdrecord devices because of the changes in the 2.6 kernel.
Even Alan Cox mentioned that it was not optimal in one of the long linux.kernel threads.
If you follow the debate over the CDDL in the OSI mailing list, you'll notice that the OSI seemed to approve the license pretty quickly. There were many licenses waiting for approval that were submitted earlier, yet were never approved. Doesn't that seem odd?
1 35
1 37
Could it be that Sun paid off members of the OSI community to push their license through?
http://www.crynwr.com/cgi-bin/ezmlm-cgi?3:mss:9
http://www.crynwr.com/cgi-bin/ezmlm-cgi?3:mss:9
Hmmmmm..... interesting.
No sense defining that in the submission. That would be too much like actual reporting.
Because IBM is not going to open anything they are making money with. They won't open web sphere, db2 or AIX.
n .html
They are not pro open source, they just have better PR department.
This is enough to see how IBM is "pro open source":
http://swpat.ffii.org/gasnu/ibm/index.e
Robert
Oh, because your job was as a PR spin doctor for Sun? I see.
Linux potentially violates almost 200 patents in total.
Nothing has been mentioned about Microsoft, and no-one has been told whether or not there are any actual violations (potential and actual are two different things).
Then why not BSD it? IIRC the code used to be BSD, both Sun and Microsoft love the BSD licence rather than that viral and communistic GPL, so release as BSD already.
Or is that just a convenient excuse for not GPLing it?
It is up to Sun to do what they want with their code, however, if they say "it is to be of benefit of you our customers", then they should listen to the customerss if they give suggestions.
especially for commercial companies:
If you add or change the code for your use, Sun owns thosechanges as well as you.
So you are sharing your competitive edge with Sun for no return. With the GPL, you share your cmopetitive edge with your competitors (all of them), but under the less palatable (from a commercial POV) GPL. You also get THEIR changes back free to yourself.
IBM Will NEVER contribute to CDDL.
..Expect that LGPL allows re-licensing under GPL, which may cause that someone's modifications CANNOT be merged back to original version, if the one who makes modifications decides to release modified version under GPL.
Please, people. Do you invenstigations before you bash.
"Without a GPL compatible license, I think Solaris will have significantly less hardware support in the long term, especially for the random hardware a developer might have sitting around at home. Resulting in a smaller userbase, smaller developer base, and less developer mindshare than Solaris might otherwise have enjoyed."
You're forgetting that Nvidia is much more likely to sign a NDA and release specs to Sun, as opposed to Joe Random GPL Hacker. The same applies to a lot of other companies. So while the overall pool may not include a piece of hardware, pulled from the dustbins of history? It has a greater chance of having access to more modern hardware, and better feature support overall than what's generated basically by guessing.
For classic corporate licenses, the priorities of license convenience are like this:
1. The corporation. It's all about the profit to the corporation, isn't it?
2. The customer. Sure we want to attract the customer to buy the product, and the license is partially responsible...
3. The developer. Actually, pretty much the license only takes all the rights away from the developer... They are compensated for that - in form of salary.
For Open Source/Free Software, the licenses must be written in exactly opposite order.
1. The developer. If there's nobody to write the software, nobody will be able to use it.
2. The user. It's for users, and the idea is user is free to do what they want with the software, so let's give it.
3. The corporations. Actually, protecting the program against abuse by corporations.
Of course, SUN as a corporation, wanted the first kind of license. GPL, as the second kind, obviously didn't meet their expectations. What SUN forgot though, is that the goals their license meets are straight opposite to goals -expected- by Open Source developers. A corporate license in service of a corporation, taking some rights away from the customer and many rights away from the developer, while the developer isn't compensated for that in any way?
The difference is that a GNU developers still "own" code they wrote. They retain all rights to the code. SUN's license says, "You will be assimilated. All your code are belong to us."
Anagram("United States of America") == "Dine out, taste a Mac, fries"
Is this the reason you could only burn CDs as root in the 2.6.9 kernel? I know when scanning the bus with cdrecord, not all of my drives are found.
"Why isn't it free? What does it fail to meet in the FSF's Free Software definition?"
It makes RMS cry. Seriously read this guys other posts. Nice to know that fundamentalism has come to Open Source.
Top 10 Reasons It's Better to be a Bartender than an Engineer:
10. You get to throw out annoying customers
9. The tips are bigger than the sunshare bonus.
8. You don't have to go to a steering committee to mix someone a drink.
7. Drunks are not nearly as long winded as Tony.
6. It doesn't take 6 signatures to hire a new bartender.
5. It's in a growth industry.
4. It's a lot more fun to work from home.
3. You don't need ARC approval for a new Cosmo recipe.
2. You never look at one of your fellow bartenders and think "what the heck does that guy do all day?"
1. Being a bartender helps you get action.
hilarious
Assuming you are referring to my blog entry discussing the way that the success, not failure, of MPL-style licenses is at the root of the license proliferation problem, I'm afraid I don't agree with you. LGPL does not include the explicit patent grant that the MPL includes, nor does it establish ground rules for maintaining a patent peace, and thus does not serve as the archetype for MPL licenses and their (many) derivatives. Instead it includes an exception to the scope of the GPL which depends on the language and architecture of the software in use and makes assumptions on how dynamic linking will take place. In all other respects it is the same as my third license category, GPL licenses.
I wonder why they didn't just just say, "Hi, this is Sun and we're developing a new liscense..."
Maybe they don't understand OSS.
abuot the "determine damages of GPL code". When sharinng music, it is expected that you get full commercial recompense, independent of whether it would realise that profit. Also, in the US, expecting reciprocal sharing for free is considered moentary equivalent, so the GPL requesting sharing other code IS commercial (case law would be the RIAA suits).
GPL would not have zero damages and so far as current case law shows you can make up any figure you like for damages.
The company itself seems to have its heart in the right place, but Schwartz is stuck in the old pre-Cluetrain ways, making out that THE COMPANY is all that is good and everyone else is just a minor nuisance and have it all wrong anyway.
He's just delivering the wrong public message, and it does Sun a lot of harm. If it wasn't for Schwartz's antics, Sun could be as well-regarded within FOSS ranks as IBM is now.
CDDL may fit OSI's definition of "Open Sorce" but it has severe problems.
0 91 41718941
http://www.groklaw.net/article.php?story=200504
"The proprietary mind can't swallow the FOSS concept all the way down, I guess, so Sun's response to the GPL is to offer a license under which programmers can write Brand X Open Source software, software that ends up not open at all on a whim, which the engineers get to write for them in the open and then the company gets to take closed and proprietary, and not only that, you don't get any code back from them in return for the code you donated, unless they feel like it. No money either. The company makes all the money.
What CEO wouldn't love that? The only thing better would be slavery. No. Slavery is worse, because you have to pay to feed slaves. "
You got first post!
And the problem with this is what exactly? If a developer decides they're going to release their mods under the GPL, you just ignore the work that developer did if you're still interested in making your OS proprietary.
...but have a look at the http://opensolaris.org/ site:
sure. This stands on that page since months or so. Where is the code?
"First they ignore you, then they laugh at you, then they attack you, then you win." -- Mahatma Gandhi
"... contributions must be licensed under this license ..." and I'm not quoting the GPL.
The GPL at least says any GPL compatible license will do.
This attitude is pervasive in the CDDL. It's like Sun deliberately distilled out the worst of the MPL and put it in the CDDL.
Sun is gonna let me see the source. If I have a Solaris customer with a problem and need the source to fix his problem, I'll have access. All I have to do is make the fix available to the planet. I don't have to worry about getting sued either.
Is this not a reasonable trade, in the context of taking care of ones customer?
The liense promotes what has been going on all along anyway. It just makes it easier to take social credit for ones work without fear of getting sued. It also makes it easier for a little guy to make a fix because he can see the source without months of negotiation.
1)It's good for your customers. They have the fix.
2)It's good for Sun's customers. If the problem is pervasive, Sun has the fix too.
The only downside is that Sun has never seemed particularly honorable when it comes to respecting your relationship with your customer. Not a problem until you get into the $100k plus deals and a real problem once you pass the $500k mark.
With the CDDL license, you won't be able to sue Sun over glomming your customer based on the work you did, and released.
CDDL has potential if Sun changes it's stripes when it comes to respecting third parth sales of Sun stuff. It's a cultural change. Much more difficult than a license.
Well dude, if they expect to get MY hard work for free, why should i get THEIRS in return?
I'm curious in what environment Solaris is faster than Linux. I'm sure there must be one, but around here, with ordinary sorts of loads like compiles and filesharing, Linux is about twice as fast as Solaris on the same Sun hardware.
Socialism: a lie told by totalitarians and believed by fools.
As far as I can make out, OSI was formed when Eric Raymond went on an ego trip, and decided to form OSI. He took himself and a bunch of nitwits, and they declared themselves the spokespeople of the FLOSS movement. Later, Eric got kicked out, leaving just the nitwits. The second to last nitwit to head it was a libertarian wacko who got kicked out for posting to Slashdot about how people don't care about freedom, since they allow themselves to be taxed.
Why anyone lends OSI any credibility, I do not understand. It's not the FSF, EFF, SPI, or any organization like that. It's a bunch of self-proclaimed spokespeople, who outside of having given themselves the name OSI, have no credibility whatsoever anywhere.
Someone care to prove me wrong? I know I'll be marked troll or something, but as far as I can tell, it's the truth. If you've got actual evidence to the contrary (read: facts), I'd love to be proven wrong.
Except now it has to do with whose license is "bigger and badder." Jesus, is crap like this going to drag down the community AGAIN!
-- uh...
Must give Flamebait a positive modifier some day. Anything with that moderation is usually brutally true.
Handeling multiple threads across multiple processors is Solaris's claim to fame. Running Solaris on a single processor system does degrade the performance.
"I use a Mac because I'm just better than you are."
https://mustang.dev.java.net/
Then don't use the GPLed code. Simple.
But if you are sticking GPLed code in your proprietary product (contrary to the terms of the GPL), and getting caught in that situation leaves you only a few choices, and you don't like any of them... well, my heart fails to bleed for you.
> Sometimes this support takes the form of binary or wrapped
> drivers, but that hasn't prevented Linux from remaining under the GPL.
Binary drivers are technically a violation of the GPL because they link into a GPL executable. Such linking normally requires all involved components to be GPL. Fortunately, Linus is not objecting much to this, so Linux is doing just fine. If Linux were being developed by Richard Stallman, however, it wouldn't have lasted a day.
Just because they release their code in a manner specifically designed to be useless to the open source community
Useless? How so? Just because you can't stick the code in the Linux kernel doesn't make it usesless.
You also imply that the BSDs don't count by your uninformed statement. Since the BSD license has no prohibitions in place that would prevent it linking with CDDL code at all.
So if anything, a very large portion of the Open Source community can benefit from it.
Linux doesn't define Open Source.
Do you think this maybe might indicate there is some sort of difference between the two situations?
Yes, it indicates to me that *some* people when it comes down to it are just GPL zealots. The fact is, it's open source. Just because it doesn't work with their favourite license doesn't make it any less a valuable contribution.
Two words: Solaris 10.
Assuming you are referring to my blog entry
That's the one.
I'm afraid I don't agree with you. LGPL does not include the explicit patent grant that the MPL includes, nor does it establish ground rules for maintaining a patent peace, and thus does not serve as the archetype for MPL licenses and their (many) derivatives.
Well, I consider the patent clauses to be more or less refinements to deal with the deteriorating patent system in the US rather than core features, but I digress.
Instead it includes an exception to the scope of the GPL which depends on the language and architecture of the software in use and makes assumptions on how dynamic linking will take place
That's true, it isn't quite "file by file" it's more "module by module". (Which I think is a better model, but may be too difficult to nail down properly. I do agree that the LGPL certainly falls apart when stretched too far beyond its homeground of C/C++ on UNIX-like OS's.)
In all other respects it is the same as my third license category, GPL licenses.
These differences are enough that when I see a GPL'ed program I know I can't use it as any part of a production release at work, but when I see LGPL'ed code, I know I can make use of it if I'm sure to follow the guidelines.
I can certainly agree that LGPL may fall somewhere between MPL and GPL in actual license grant, but I definitely felt like a mention of the LGPL was suspiciously absent from your article. (And I would still guess that the MPL was very much inspired by the LGPL in many regards... Well, that and even earlier "vanity" licenses.)
In any case. Here's hoping the CDDL (or some other derivitive) becomes a common standard useful for both commercial entities and the community that helps bring an end to this "license du jour" nonsense.
I read your post and blog entry and I understand where you're coming from. Sun does not appear to be in a position to use GPL for OpenSolaris, even if it wanted to. You're right that MPL-class licenses are ideal in this situation. However, I think some of the controversy could be alleviated by using a selective multiple license policy. For example, the Mozilla project now requires that all new source files added to their CVS tree be tri-licensed as MPL/GPL/LGPL to maximize cross-pollenation with other related projects. They are in the process of obtaining permission from authors to re-license existing code under this tri-license, but this does not hinder Mozilla from being distributed today under MPL. I see this as a best of both worlds approach.
y .html
So here's my question: Have you folks at Sun considered using a similar approach with OpenSolaris? License everything under CDDL, but gradually dual/tri license certain components under a GPL compatible license as well. (preferably as much as you legally can) The Open Source community has historically welcomed this approach and I think it would boost contribution dramatically.
Regardless of license choice, there is one big fly in the ointment: software patents. Lets just put it on the table.. the primary reason many members of the Open Source community have a distrust of Sun is because of this issue. From Java to the secret settlement terms with MS, there is perhaps legitimate concern that Sun would someday turn on the OSS community -- particularly its competitors (Linux, GNU Classpath, Apache/Jakarta, Mono, etc.) Maybe the current leadership would never even think of such a thing, but leadership can change. Or what if Sun was bought out by a company hostile towards Open Source? These are unknowns, and unknowns breed fear. When Sun says, "here are 500 patents that we're turning loose.. oh.. but.. you can only use them with our CDDL licensed software," that doesn't jive with community ethos. It implies an attempt to maintain a style of proprietary control long deemed unacceptable not just by GPL zealots but by anyone aware of the potential legal risks to developers.
I think it's admirable that Sun has been increasingly moving in the direction of Open Source, but I caution that there are still some barriers. Be careful not to label these as just silly ideological nuances. Some of them are, but others, like software patents, are not. Until the menace of software patenting in the US is destroyed, Sun needs to assure the OSS community that it will have no part whatsoever in any offensive use of its patent portfolio.
See this page for details on the Mozilla policy:
http://www.mozilla.org/MPL/license-polic
I couldn't agree more. For what it's worth, I am not too worried which MPL-derivative license becomes the one 'template' license that everyone uses. I would be very pleased to work on MPL 2, if that would be more 'politically acceptable'. As I have made clear on my blog, I believe that the MPL license family brings the greatest freedom while preserving the value of the commons, so I think it would be a mistake to allow zealots to misguidely extinguish it.
At Sun, we couldn't wait for a template MPL to be created independently for OpenSolaris because we were very keen to get the existing global Solaris community in a position to work with the code, but there's frankly no obstacle in the future to working on a 'blessed' reusable license in the MPL family. What I hope the CDDL makes impossible is the creation of any more 'vanity' clones of the MPL. If anyone tries, they deserve the full wrath of the meta-community.
It's something of a halfway contribution. It's open in the sense that anyone can get the source, and modify it freely. But if the code can't be taken out and used along other common open licenses, the value of that contribution is lessened.
Most projects with licenses incompatible with the GPL offer some form of dual licensing to allow cooperation between projects - Sun aren't doing that.
So let me be perfectly clear. Sun has never been a patent terrorist and does not intend to start now. Sun accumulates patents in parallel with software development in just the same way that all US corporations do - it would be irresponsible to do otherwise. However, our policy is to grant these patents to the communities around the software they relate to.
If you read the license that you are granted by the Java specifications, you'll find that they give you an unrestricted blanket grant of all the patents Sun has that might be infringed by Java implementations. The same is true of the CDDL and OpenSolaris - if you work in the OpenSolaris commons, you have the full use and protection of Sun's patents. As more companies work in that commons, the CDDL will force them to donate all their patents too. Those grants survive anything that may happen to the companies involved, so even a hostile acquisition would not result in the loss of patent protection. Blanket grant provisions incrementally build a meshed patent commons that promises safety for developers and as it grows offers protection against patent terrorism.
Until we're able to reform the patent system (and Sun is a firm believer in patent reform and is working behind the scenes in Europe too to promote sanity) the smart thing to do is not to neglect patents, any more than it would be smart for a policeman who opposed gun ownership to protest by not wearing body armour. However, I think the F/OSS meta-community should show zero tolerance for patent holders who don't give the communities in which they work blanket protection from their patent holdings.
This does not mean I want to see lots of individual patents in some way made public. Software patents are deeply flawed and breach the social contract implied by patenting because they do not usually provide the know-how in a way that the public commons is enriched. Software patents are not sources of either sample code or of inspiration. Instead, I want to see corporations forced to donate all patents that are used by the various code commons they care about, with a blanket, unenumerated, blind grant and an enforceable patent peace. That's why I am a fan of the MPL/CDDL!
That's why I'm also not a proponent of multiple licensing - yes, I have considered it. If people could use the GPL as their license, they would not have to contribute to the patent commons surrounding OpenSolaris or be subject to its patent peace provisions. Multiple licenses are great all the time they share values, but I am so committed to creating a patent-safe developer commons around OpenSolaris that I feel it would be a bad idea to multi-license CDDL-licensed code until the GPL has suitably strong patent provisions. Once the GPL becomes combinable and includes patent protection, I would be very happy to revisit this view.
What version of Solaris are you running? Most of the x86 performance issues I'm aware of have been addressed in Solaris 10 and the people I hear from tell me the speed differential you're describing has gone away.
Exactly.
However, maybe after things cool down, they'll realize that they're just trying to do the same thing Apple was originally, and they'll step back and realize they could at least yield as much as Apple did.
That's a very interesting perspective. It seems to me that there are two "camps" regarding how the problem of software patents should be handled.
:) But it seems possible to create an "anti-patent commons" if you will, negating the need to build up patent arsenals while still providing extensive protections for both users and developers. If everyone is really that serious about getting rid of software patents, a license with these provisions would seem to be the most potent way to help this happen.
The GPL/BSD camp holds that software patents should be ignored unless a threat emerges. Their reasoning follows: All significant software unknowingly infringes on many patents, regardless of how trivial. Nevertheless, most patents wouldn't hold up in court or could be invalidated by prior art. It isn't worth developers' time and it is in fact dangerous to go looking for possibly applicable patents when writing software. Companies who support software patents realize that a direct challenge to a major open source project could end up going all the way to the Supreme Court and result in software patents being invalided altogether. Holding our own patents in defense would be hypocritical, costly, and may even undermine our arguments against them. Even if we did hold patents for defense, there are so many out there that any patent-grant-covered code/project would still be vulnerable to patents we don't hold. For the last 15-20 years or so, this approach has appearently worked flawlessly, weathering many waves of legal FUD.
The MPL/CDDL/Apache(?) camp holds that software patents are a much bigger threat than commonly perceived and that we must assume they will be around for awhile. Attacks are inevitable as proprietary vendors resist the continued success of OSS. Corporate users are presumed to be leary about patent issues with OSS and thus we need to create protective measures for them. The best defense is a strong offense. We should thus join into the software patent "cold war" by building up our own "arsenals" to deter a real war from actually starting. If projects do not produce their own defensive patents, they should at least use licenses that allow/require contributors to do so, effectively creating a 'patent commons' safety net. Once the stakes are high enough, everyone involved should realize how ridiculous the situation is and collaborate to mutually disarm by lobbying legislators to invalidate software patents altogether. Until that time, we can also use our patent arsenal to enforce compliance with our open specifications and protocols so that there are no proprietary, incompatible forks.
Maybe that's a bit of a simplification, but I think this closely sums up both camps from what I've heard/read over the last few years. I just think it's a shame that it has come to this. Neither camp is a significant majority, so there's little hope of either ideology winning out through dominance. Somehow, both camps need to meet somewhere in the middle. Unfortunately, nobody has figured out how to make this happen yet. This is all the more reason to push hard for patent reform!
But I do have one idea. Suppose that the GPL was modified to include one or both of the following patent provisions:
- "your rights to use software licensed under GPL v3 shall terminate immediately should you engage in a patent lawsuit against any software project covered by an OSI or FSF approved open source license"
- "by distributing, in original or modified form, software covered by this license, you implicitly grant a worldwide, non-exclusive, royalty-free right to use any patents you may hold which are necessary to implement features in this software, or similar features in any software covered by an OSI or FSF approved open source license. **"
** This could also be narrowed to only those licenses which require modifications to be given back to the community upon re-distribution.
Yes, obviously that would all need to be translated into proper legalese.
It isn't the hardware but the platform lock-in that I meant. The codebase is of, for, and ever-shall-be Solaris.
As I mentioned earlier, it is the platform lock-in that I refered to. You can "create your own OpenSolaris Distro", but it will be OpenSolaris. I was moderated down as flamebait for expressing my total lack of interest, but I sure didn't mean to flame anyone. I just don't see how this will advance the state of the art of Computational Science (which is something I do care about.) I guess at heart I'm an academic. So drive a stake through my heart for my distain at commercial barriers to the free exchange of ideas. It is better to control, and profit, than evolve and grow. Ok, fine.
So try the LGPL? "The problem is with the GPL not everything else." So you expect to see code sharing with Apple or the BSDs? I think not.
Yes, those are the sort of provisions I'd want to see. Combinability also needs to be addressed. The other strength of MPL-style licenses is that they allow a binary to be built from a mix of licenses.
Putting licensing peeing contests aside, Solaris is UNIX complaint, implements POSIX, and lots of other standards (listed in the standards(5) manual page). Even better, OpenSolaris will give people access to the implementations of those standards.
Both hardware _and_ software, there really is no lock-in, especially in the Microsoft sense.
-- Microsoft is the most expensive commodity operating system and office suite vendor in the marketplace.
"At this point the entire product is GPL. Care to point out where I am mistaken?"
No, at that point the entire product would have to be released under the GPL if it is to be legally distributed under the terms of the GPL.
If the "entire product" is only used internally, no licensing issues arise, because the GPL covers distribution, not use (the position of the FSF is that there is no legal requirement to simply use software, any more than a publisher can readers to have a license in order to read a book).
If the "entire product" has been distributed in violation of the terms of the GPL, then the company has broken the law, but the proprietary code still isn't GPL, unless the company decides to make it so. They would have to stop distributing the derivative product and they would likely have to pay a fine. There is no reason to assume that a court would force the company to release their code under the GPL as the remedy, although I guess it is imaginable.
In any event, linking does not "automatically" make proprietary code into GPL code, it just makes the derived product illegal to distribute.
It's something of a halfway contribution. It's open in the sense that anyone can get the source, and modify it freely. But if the code can't be taken out and used along other common open licenses, the value of that contribution is lessened.
Except the GPL is about the only major open source license that I know of that it's incompatbile with.
There's nothing stopping those using BSD, MIT, and many other licensed projects from using CDDL stuff.
Most projects with licenses incompatible with the GPL offer some form of dual licensing to allow cooperation between projects - Sun aren't doing that.
But, no one said they have to. As far as cooperation, one could say that GPL projects should cooperate with others and offer dual licensing as well. The point is that there are many projects out there that can use CDDL code because they're NOT GPL. So it's silly to say that it's useless.
Agreed that it isn't MS level lock-in. I'd even suggest that from a historical perspective, it is significant that Sun is going Open, even if it isn't Free compatible. The fact is that this is going to look good for Open Source. Since Open is a subset of Free, this is not bad PR in the long run. You've got to start somewhere, I suppose. I'd just hate to see one of the survivors of the Unix Splinter Wars start splintering the landscape again.