Submitted by Dissatisfaction has surfaced in the OpenSolaris community. The dispute centers around how derivatives can use the OpenSolaris name and branding. Sun says that permitting broad downstream use of the OpenSolaris name would risk diluting the Solaris trademark, and it has stated that a policy needs to be established on how the name can and cannot be used. A member of the governance board has resigned. This looks similar to what happened with the Mozilla trademark.
Summary?
Sun is shunning the current niche community in favor of developing something that attracts a bigger NEW community. Thus the “friction”.
In reality the niche community doesn’t matter much to the software itself because Sun is entirely responsible for its development. Though some people have tried and hoped that Sun would allow that to change, development has really remained behind closed doors.
To me it comes down to Sun Microsystems being a fair weather friend of the open source movement. They do things the “open source way” only when it suits them. Hypocritical for a company that touts itself as, what was it again, “the largest commercial contributer to open source software in the world”?
This “friction” came about because some people thought otherwise until the truth bonked them on their heads.
Personally, I’m surprised that it’s trademark what it’s causing “friction”, and not the requeriment to grant Sun full copyrights and “joint ownership” of all the code you want to push to solaris, java, or openoffice. Yeah, the FSF does it aswell for all the GNU projects, but the FSF is just a foundation, not a company that wants/needs to earn money. I don’t mind (even if i don’t like the idea) giving my copyright to a foundation, but why should I give my copyright to a corporation that earns so many millions every quarter?
To me, it seems unfair: in a “true” community project every part benefits from the process, be it a individual or a big company: both collaborate, the company is who makes money but that’s fine because you benefit from it. The opensolaris model feels to me like i’m giving too much – joint ownership of my code!! Sun says that it wants it to “protect the OpenSolaris code base, enable alternative licensing models, and protect the flexibility to adapt the project to the changing demands of the community”. Coincidentally, it also allows Sun to release closed-source versions of Solaris with portions of propietary code not released in opensolaris.
Edited 2008-02-22 21:31 UTC
Sun, amd any company in a similar position, would be idiots not to insist on a single point of control for the products their business is built around. Nothing’s stopping you from forking the code if you take issue with the requirements to get your additions into Sun’s main tree.
Yes, I say that too. The source is out there and anyone can use it. This article isn’t about that.
This news article is about the open source community that Sun created turning out to be shallow and powerless, governing body and all. People were mislead to believe that it is something that it is not.
Sure. But that doesn’t makes the “opensolaris community” more friendly.
That’s probably the only workable model for Sun to release their code under OSS license. I thought we established that long time ago. What’s the alternative?
So why is it that community-developed software and open source have to go hand in hand? Just because they frequently do, and in certain cases it might be in a company’s best interest to establish an ecosystem of 3rd-party developers around their software, Sun did spend hojillions of dollars developing the software, and then released the code so all could benefit. Sure it might not be exactly what you want, but if you have the resources, you could take the software in the direction you’d like. If you don’t, I guess you can’t expect something for nothing.
So call it what it is: shared source. Sun releases source code and claims “open source” superiority, yet they require copyright assignment on contributions that reserve the right to change the license. They may still technically meet the definition of OSS, in that the code is released and it is forkable, but it’s certainly not an embrace of OSS. They’re patronizing the community at best.
Don’t get me wrong, the community should still be grateful when a company like Sun releases that significant a volume of code to the community, but the community needs to realize it’s up to them to do with it what they will. Sun isn’t embracing two-way partnership in any sort of a joint-relationship sort of way. They’re using OSS in much the same way Apple does.
Sun is entitled to do what they do, all the more power to them, and the community should be happy to have what they have. But Sun’s OSS commitment is one-way, they’re not looking for community collaboration, they’re looking for community adoption.
> They’re using OSS in much the same way Apple does.
Sun’s doing much much more than Apple in terms of open development.
I fully disagree. Suppose that Sun would have licensed Java under the GPL + license exception and did not require copyright attribution. The community could have redistributed Java under the GPL, it would not be easy to change the license because the copyright is spread. The current situation is not different, the community can distribute it under the GPL, but can not change the license. So, for the free software community dual-licensing or no dual-licensing does not really matter.
But now they can also let customers pay who want some customized closed-source version of Java. And to be honest, you need some “bargain” like that. Be it Sun, ex-Trolltech, ex-MySQL, or ex-Sleepycat, it’s hard to justify investing millions of dollars in a body of code, and putting the competition on exactly the same feet. It is not easy to make money through support only (especially for libraries like Qt or Berkeley DB). Or, for existing companies to switch fully to a service model.
So, dual-licensing deals seem fair to me, and fully in the spirit of free software: the community gets code as they would have without dual licensing. The company that contributes the code still has a tool to reasonably leverage profits.
I don’t disagree with anything you’re saying, but then I wasn’t really arguing against dual-licensing. I have nothing against it.
I guess I should have clarified by simply saying that Sun does not utilize a community approach to open source development. Basically, they own the project, they don’t really want or solicit community involvement in directing the project, and if you want your code considered for inclusion, then you must be willing to sign a copyright assignment so that they can reserve the right to relicense at will.
And certainly I’m not singling Sun out. But there’s a difference between throwing your code out to the public and saying “here you go”, versus throwing your code out and saying “let’s work together”. Hence my implication that Sun’s approach is more akin to shared source than open source.
In other words, Sun is not really looking to build a community to leverage shared development. They’re looking to disperse their product through the OSS community to help strengthen the Solaris brand. And frankly try to dilute the momentum of linux. There’s nothing wrong with that, and the community can still benefit, but let’s call it for what it is.
The linux kernel is very likely the most popular OSS project, and it has a decentralized development structure (at least compared to corporate-sponsored projects) with no attribution required. I don’t think those facts are coincidental. If you’re going to deny developer participation and require free software developers to provision code for proprietary use, you’re going to limit your pool of developers.
Opening Solaris, and Java for that matter, was a marketing effort to generate developer buzz. There’s no leveraged developmental gain for Sun, which doesn’t matter since they’re not looking for one.
Certainly the community still gains, but it’s a one-way street. They can’t help shape the project(s) unless they fork, or are willing to commit their code for proprietary use.
I think the reason that people are so mad is because SUN gave the impression that they wanted to build a community project around opensolaris. I don’t think anyone would be upset if SUN would not have done that. There was a lot of effort put out and a lot of hupla made about build a “community” with it’s own charter and powers to manage the opensolaris project. It is frustrating to the people that committed time and resources to the project under this agreement.
Everything would have been fine if SUN would have just managed their opensource community like mySQL does right out of the gate. Moves like this would have been expected, and probably even welcomed in such a case.
On the outcome is Roy Fielding’s resignation from the OGB. And that outcome will probably go they way [friction] initiator wants.
Many people resign from various posts/positions when friction exceeds reasonable limits and they do it on a principle.
I always applaud people who stick to their guns and principles. However, there’s a huge downside to these departures. When those who initiated the friction see their main resistance gone thy many times get their way as the opposition is no longer an obstacle.
It’s a shame that this happened.
There are a lot of Linux distros. They dilute the stream of talent and lose focus on many items. Many times when an issue crops up it’s “fork! fork! fork!” vs. “Hmm, let’s sit down and discuss this like rational adults”.
Sun is trying to avoid this and keeping the Solaris and OpenSolaris name something corporates will latch on to and be able to ask for support on and Sun being able to answer with a modicum of reliability.
Personally, it’s both a boon and a bane to Linux as a whole that it’s able to be so easily forked and modified. It’s such a moving target. Sure, the Novel’s and Red Hat’s of the world try and create stable and non-moving targets, but even they often have issues with this.
I do think Sun is a team player in the OSS community, but perhaps they like to keep their playbook close to the chest.
No, it’s not about that. Copyright don’t forbid people from forking opensolaris, they just forbid them from naming it “opensolaris”. This happened in the linux world too – red hat wanted to protect their brand, so they stopped releasing “red hat linux” as a freely distributable distro, and they asked people to stop distributing “red hat” isos -they were allowed to do it because the isos contained red hat’s “propietary” copyright- and then created a distro called “fedora”
Edited 2008-02-22 21:30 UTC
The term you’re looking for is trademark, not copyright. Whereas the GPL covers the latter, the former is still up to the holder.
(Also, unlike with copyrights, you have to enforce your trademark to retain it.)
I’m sorry, but I disagree with the logic here. How many forks of Linux exist? Last time I checked, a patchset does not a fork make. Or do you consider Apache + mod_python a fork? Does mod_python dilute Apache’s credibility and brand recognition?
Your repetition of the same tired characterization of the FLOSS community as a bunch of “fork early, fork often” knee-jerkers does nothing to lend credence to the view. I’d love to see some proof of your alleged dilution of talent and lost focus. All those upstream patches by such “forkers” as Red Hat and Novell serve as a strong rebuttal to your assertion of fractured effort. And the only group for whom Linux is a “moving target” are device driver writers, and even they are only occasionally affected (typically for the better).
The brand power of Linux is hardly diluted by the many distros that bear the name. They’re mere wrappers around the same kernel. Different strokes, as they say, for different folks.
rhavenn is very very clearly talking about Linux distributions, not the kernel. He even says so in the first sentence:
This is correct: there are hundreds. Many of them are simple re-packing of larger distributions with perhaps one or two changes which could perhaps be done just as easily by providing a set of packages for an existing distribution instead.
The ability to make these sorts of changes is great if you’re a geek but it’s a major headache if you’re an ISV.
There are a lot of Linux distros. They dilute the stream of talent and lose focus on many items. Many times when an issue crops up it’s “fork! fork! fork!” vs. “Hmm, let’s sit down and discuss this like rational adults”.
I don’t see cries of “fork!” as a bad thing. Look at some of the major forks in OSS history. Xfree vs Xorg, gcc vs egcs, gimp vs cinepaint. In the case of X, Xorg ended up developing at a much faster rate than Xfree and it pushed development further along. Egcs merged back into gcc along with all of its development. Cinepaint forked form GIMP and created a new tool focused on film post processing while GIMP remained a more generic tool for general photo editing.
There are many “forks†of Linux, this much is true (although your wording is distinctly ambiguous).
There are are a select few popular distributions, but most die off within a few months. Sometimes, though, one of the forks becomes popular itself and lives on: take Mandrake, for example, which is now on its second (I think?) name and thirty-eleventh year.
Forks drive development. Developers get complacent if there’s no competition—not a criticism, incidentally.
Holy crap does OSNews need to learn UTF-8.
And maybe stop changing normal doublequotes into ampersand-quot-semicolon, or doing strange stuff to apostrophes, that would help. 🙂
I feel that the submitted summary of this story doesn’t catch the essential of the current debate, and is misleading.
“policy needs to be established on how the name can and cannot be used.” – I haven’t seen ANYONE in the debate disagrees with Sun on that.
The _real problem_ is, before work with the community (both inside & outside of Sun) to establish such policy, Sun all the sudden unilaterally named their own distribution (created under the project name ‘Indiana’) THE OpenSolaris Distribution, by calling it “OpenSolaris Developer Preview”, and put it on OpenSolaris.org homepage.
Today, it’s still listed under such name on the download page:
http://www.opensolaris.org/os/downloads/
That, of course, pissed off many community members.
Many weeks (months?) later, Sun responded with following:
http://mail.opensolaris.org/pipermail/ogb-discuss/2008-February/004…
Community reaction on that response can be found here:
http://mail.opensolaris.org/pipermail/ogb-discuss/2008-February/thr…
Roy T. Fielding’s resignation:
http://mail.opensolaris.org/pipermail/ogb-discuss/2008-February/004…
While Sun certainly has legal authority to do so, John Sonnenschein explained why some members are furious:
http://mail.opensolaris.org/pipermail/ogb-discuss/2008-February/004…
John Plocher put it this way:
“I think it is safe to say that most (if not all) of us are distressed
by the *way* that the Indiana project was initiated and developed
outside the view of the community and by the *way* it was positioned
as the sole recipient of the OpenSolaris name. “Legalisticly”,
according to the charter and constitution, what Sun did was perfectly
OK; from a social values/”equal footing community participant”
perspective, however, many feel betrayed because the above principles
were violated, and they feel that Sun no longer is playing by the same
rules it set down 3 years ago.”
http://mail.opensolaris.org/pipermail/ogb-discuss/2008-February/004…
Hope everyone can do some research before comment.
Now I’ve done it for you, all you need to do is read those links.
I hope it is resolved sensibly, and we dont end up with the mess of Ubuntu/Foobuntu/Barbuntu
Its truly a shame that Ubuntu let its name get diluted like that
“What’s in a name? That which we call a rose \ By any other name would smell as sweet”
If it’s good, it will get the credit it’s worth, no matter the name, that’s what’s [also] good in the open source world. Unlike with closed source companies where we don’t always know where a code is coming from, in FOSS we always know the origins, and there’s only a fairly limited possibility to undeservedly grab credit.
Sun does not now, nor ever has, and unlikely will ever understand community. I have a great respect for Sun and the many technologies they have created and share with the world. Yet it is abundantly obvious that the developer culture inside Sun is incapable of tearing down the walls that it builds around itself to protect its privileged position-absolute control . Only when these walls come down, replete with the loss of total control, in exchange for two-way sharing, learning and improvement, will Sun be something other than hostile towards community development. Sun does not want to be a part of anything else-that other things are a part of Sun is no problem.
Most of this is an “attitude” problem. Sun has cultivated a feverish developer culture where Sun developers are indoctrinated in all the glory of Suns technology prowess-they are the best developers, working on the best technologies, creating the best software, etc. This of course is a very good way of motivating their employees. But this comes at the expense of rather large degree of alienation from other, non-sun, developers and stands at odds to the kind of community solidarity found in FLOSS projects.
Code is not important for sun-Process is. Everything is process, and these processes are totally incompatible with volunteer/non-Sun communities. The small steps towards community that Sun has made also appeals to many Sun employees who become jaded by “everything is Process”, with the allure of actually coding for coding sake-but this kind of meritocracy, where code and code speaks, is itself at loggerheads with the Process.
Changing the culture inside of Sun-and this is what is required of Sun to be part of any real community-is going to be very difficult and take a long time. A lot of this will only be changed by a generation shift in Sun employees.
#
Apropos:
Substitute the word Policy* for Process and this describes much of Microsofts hostility towards FLOSS. Policy is even more abstract than Process-developers(coders) are almost completely exchangeable at Microsoft. Great coders do not stand out at Microsoft with very few notable exceptions, Policy nullifies such distinctions and makes all coders “equal” which breaks the pride and identification with ones own work, alienating many along the way. Microsoft corporate projects like .NET offer a temporary outlet for pent-up frustrations and allows a degree of code identification (identify oneself with ones contribution) but only so long as what is being created is “new”, “innovative” etc.
*Process, as in Suns case, also contains many policies, but the policies shift according to changes in Process. Where Policy rules, the processes change according to Policy. Policy rules, means that everything is dictated and individual coders are essentially voiceless. Process from this vantage point is more “communitarian”. Communitarian does not equate to openness and participation in greater communities.
I say that it takes time for Sun to change everything. It is hard, but they are trying to do that. Look at Microsoft, that tries to opens protocols. MS is just in the beginning of this opening process, Sun has come farther. With some more time, everything will settle out.
But I agree, I dont like Ubuntu/Zubuntu/Flossbuntu etc thingy. The Solaris is code is available. It is not to much to ask for, to find a new name? RedHat did this and there are no RedHat/BlackHat/GreenHat/etc clones out there. That is good.
Incorrect. Roy Fielding was a member of the original CAB (community advisory board) over a year ago. He was not a current member of the governance board and did not resign any such position.
He resigned his status as a community member; not as a board member.
The source article is wrong.
Well, if they are to chose a new name, they better go egotripping. Staying with the star theme (“Solaris”), they might rename it to Cephei, which is the second largest known star (the first one’s called Canis Majoris, sounds dumb).
Supernova would be nice new name for OpenSolaris
Don’t know what the big deal is.Mozilla took the same stance not so long ago.
If one wanted to play the devils advocate one could argue that the mozilla foundation asks for donations,doesn’t pay taxes and on top of that gets a load of money from different deals while sun is a company that pays taxes and officially has to make money to survive.
Now which of the two would depend on keeping their brandname ‘clean’ so there is no confusion with other brands?
Sun that has to post a profit or mozilla that doesn’t?