Submitted by Sun Microsystems is the latest company to become a patron of the Free Software Foundation. The FSF’s corporate patron program allows companies to provide financial sponsorship for the FSF in return for free license consulting services. High-profile FSF patron affiliates include prominent technology companies like Google, Nokia, IBM, Cisco, and Intel. FSF involvement represents Sun’s latest attempt to take a more active role in the open-source software community.
This is good news for the the Free Software community but it could be bad news for the Linux kernel. If there is a wholesale switch to GPLv3 the Linux kernel will be left out in the cold while the Solaris kernel will look a lot more enticing to Free Software distributors.
Interesting. I wonder how hard it would be for a distribution to switch kernels, and who might be the first?
Oh. A bit of googling says this is hardly a new idea. There’s already a Debian based distro that uses Solaris
http://www.gnusolaris.org/gswiki
I bet Debian will become more involved when GPLv3 drops.
This is sure to add some excitement to the Free OS scene (And complaints about fractualization). Bring it on!
It will make no difference to Linux development.
The vast majority of the people who actually develop the Linux kernel do so because they care about the Linux kernel or are paid to work on the Linux kernel. The number who switch because of gplv3 versus gplv2 will be trivial.
It will make no difference to Linux development.
The vast majority of the people who actually develop the Linux kernel do so because they care about the Linux kernel or are paid to work on the Linux kernel. The number who switch because of gplv3 versus gplv2 will be trivial.
I said nothing about development. I am talking about distribution. If there is a whole sale switch to GPLv3 (which I am NOT counting on, I am only making a point), distributors would be more likely to make a switch to the Solaris kernel to alleviate any possible license incompatibilities.
I said nothing about development.
Nobody said you did.
I am talking about distribution. If there is a whole sale switch to GPLv3 (which I am NOT counting on, I am only making a point), distributors would be more likely to make a switch to the Solaris kernel to alleviate any possible license incompatibilities.
Distributors of what, Solaris?
None of the Linux distros are going to switch kernels just because of GPL versions. It’s just more work than it’s worth.
Besides, how could the GPLv3 create a license incompatibility with the GPLv2 when the FSF says it’s only trying to solve the same problems?
> how could the GPLv3 create a license incompatibility with the GPLv2
Section 6 of GPLv2 says:
> You may not impose any further restrictions on the recipients’ exercise of the rights granted herein.
GPLv3 is expected to impose more restrictions than GPLv2. Therefore, GPLv2 and GPLv3 are expected to be incompatible.
>when the FSF says it’s only trying to solve the same problems?
The spirit of (any version of) the GPL is to ensure that all users have the right to share and change the licensed program; GPLv2 is doing a great job at ensuring this. However, there are people that have managed to nullify the user’s right to share or change a GPLv2 licensed program without breaking the letter of GPLv2.
One of the intentions of GPLv3 is so that the loopholes found in GPLv2 that nullify the users’ rights won’t happen in GPLv3 licensed programs.
That’s all old news. How does this make GPLv2 software unusable with GPLv3 software?
If there is a wholesale switch to GPLv3 the Linux kernel will be left out in the cold while the Solaris kernel will look a lot more enticing to Free Software distributors.
How is linux less enticing for free software distributors? And more to the point, who among the free software distributions, or even the freedom-or-nothing side of the community, are actually committing code to the kernel?
Are IBM and HP going to switch to supporting openSolaris with code contributions? Is Red Hat going to drop the linux kernel for openSolaris?
And considering the dual-licensing, will the free software community be contributing to the openSolaris kernel knowing that their contributions must be dual-licensed under CDDL/GPL, along with copyright assignment, in order to be accepted?
Or will the free software community, which makes up but a portion of the OSS community, manage to sustain a fork of openSolaris that’s pure GPL, breaking away from the advantage of upstream support from Sun? Is that an incentive for Sun to adopt the GPL? So that the free software community can immediately strip it of CDDL licenses and prevent any community contributions back to Sun? Or from the community, will the Debian maintainers, who are having a hard enough time getting linux releases out the door, adopt a forked openSolaris kernel? Will the FSF be able to manage a forked openSolaris kernel as a GNU project?
I’m certainly not saying Sun’s adoption of GPL v3 is a bad thing, but I am having a hard time envisioning why this is such a monumental win for the free software community when the fact of the matter is it will remain dual-licensed with Sun reserving the right to close future versions? Everybody keeps saying “Woo hoo! GPL!” without explaining what the really means in the overall scheme of things. If free software supporters are going to be comfortable contributing to a dual licensed CDDL/GPL project, what was the opposition to CDDL in the first place? Seems to me that openSolaris is fairly viable right now with CDDL licensing, and it’s starting to build up a community of it’s own, so I’m not sure where the change is going to come into play. There’s a lot of sound and fury here, without much clarity, and I suspect that’s sort of the intent from a corporate marketing point of view.
Empty arguments about Tivoisation aside (and I say empty because the people arguing against Tivoisation weren’t actually producing the code that was Tivoized), linux is still the free software community’s most viable kernel. Don’t underestimate the advantage of it’s de-centralized ownership. Heck, even the Hurd kernel will take a step back with v3 since they won’t be able to relicense the linux driver code they have incorporated over time.
Again, not to take away from Sun, but let’s wait and see how things shake out before speculating on the future of free software. v3 is not even out yet, and will take some time before it’s impact can realistically be judged.
Very well said. In addition to this and my argument below (Congrats to Sun for missing the point), it’s worth clarifying that the GPLv3 is almost completely ineffective when combined with the CDDL. The latter is nearly always more permissive than the GPLv3, so the more restrictive parts of it can be legally side-stepped on the grounds that the CDDL grants a license to do what the GPLv3 prohibits. Essentially:
CDDL + GPLv3 == CDDL + under-informed fanboys
Very well said. In addition to this and my argument below (Congrats to Sun for missing the point), it’s worth clarifying that the GPLv3 is almost completely ineffective when combined with the CDDL. The latter is nearly always more permissive than the GPLv3, so the more restrictive parts of it can be legally side-stepped on the grounds that the CDDL grants a license to do what the GPLv3 prohibits. Essentially:
CDDL + GPLv3 == CDDL + under-informed fanboys
Dual licensing isn’t a combination of licensing. It is the process of distributing something under two different licenses. If you only want to abide by the GPL you are free to. SUN will not accept the changes into their tree but they don’t have to. It no longer belongs to Sun. It belongs to everyone.
How is linux less enticing for free software distributors? And more to the point, who among the free software distributions, or even the freedom-or-nothing side of the community, are actually committing code to the kernel?
That’s besides the point. If there is a wholesale switch to GPLv3 then most likely Debian would switch to the Solaris kernel as they are closely tied with Free Software, which in turn could impact distributions based on Debian, of which there are many including the popular Ubuntu. Gentoo could also switch easily or at least provide the Solaris kernel as the default eventually, as they are already attempting to support multiple kernels.
Are IBM and HP going to switch to supporting openSolaris with code contributions? Is Red Hat going to drop the linux kernel for openSolaris?
It’s a definite possibility. You don’t understand that if Solaris goes GPLv3 it no longer belongs to SUN except in name only. It is then Free Software. It doesn’t matter where it came from.
And considering the dual-licensing, will the free software community be contributing to the openSolaris kernel knowing that their contributions must be dual-licensed under CDDL/GPL, along with copyright assignment, in order to be accepted?
This is not the problem you make it out to be. We don’t have to let Sun keep it if we don’t want to. We can fork it and call it something else. Once it is GPLv3 no one can stop that from happening.
Or will the free software community, which makes up but a portion of the OSS community, manage to sustain a fork of openSolaris that’s pure GPL, breaking away from the advantage of upstream support from Sun? Is that an incentive for Sun to adopt the GPL? So that the free software community can immediately strip it of CDDL licenses and prevent any community contributions back to Sun? Or from the community, will the Debian maintainers, who are having a hard enough time getting linux releases out the door, adopt a forked openSolaris kernel? Will the FSF be able to manage a forked openSolaris kernel as a GNU project?
The FSF doesn’t have to maintain a fork. Someone else can, and it wouldn’t be difficult considering the fact that they can fold all official Solaris kernel changes into their own kernel if they so wish. It would be rather trivial in fact because you have all the power of the official Solaris kernel release PLUS any additions that you want to include.
I’m certainly not saying Sun’s adoption of GPL v3 is a bad thing, but I am having a hard time envisioning why this is such a monumental win for the free software community when the fact of the matter is it will remain dual-licensed with Sun reserving the right to close future versions?
Once GPL always GPL. It’s completely out of Sun’s control after that, just like Linus and Linux. If Sun is a good shepard of Solaris then it will stay “Solaris”. If they are a bad shepard then it will be “Phoebus” or something else.
linux is still the free software community’s most viable kernel
This is true…at the moment. All I am saying that if Solaris goes GPLv3 and so does everyone else, this may change.
Again, not to take away from Sun, but let’s wait and see how things shake out before speculating on the future of free software. v3 is not even out yet, and will take some time before it’s impact can realistically be judged.
Hmm. I’m not judging GPLv3. I’m merely speculating about the effect of a possible scenario, one that includes both the wholesale switch to GPLv3 and relicensing Solaris under the GPLv3.
>>That’s besides the point. If there is a wholesale switch to GPLv3 then most likely Debian would switch to the Solaris kernel as they are closely tied with Free Software, which in turn could impact distributions based on Debian, of which there are many including the popular Ubuntu. Gentoo could also switch easily or at least provide the Solaris kernel as the default eventually, as they are already attempting to support multiple kernels.<<
Your argument is based on the flawed assumption that Opensolaris will be dual licensed under the GPLv3. From everything i’ve seen from Opensolaris community there is alot of opposition to any such dual licensing. One of the main reasons been the fear of GPL only forks.
>>It’s a definite possibility. You don’t understand that if Solaris goes GPLv3 it no longer belongs to SUN except in name only. It is then Free Software. It doesn’t matter where it came from.
…snip….
Once GPL always GPL. It’s completely out of Sun’s control after that, just like Linus and Linux. If Sun is a good shepard of Solaris then it will stay “Solaris”. If they are a bad shepard then it will be “Phoebus” or something else.
<<
You logic is flawed, the main difference between Linux and Solaris in this regard is Sun owns the copyright on all the code they’ve released as “OpenSolaris” as result they can do whatever they want license wise with it. They could cease dual licensing it for example. In which case any GPL only fork would need a dev team to maintain, given the FSF progress with HURD I don’t see this been a huge success do you?
The point been that after such a “fork” everything new and interesting going into Solaris would not be available to your theoretical GPLv3 fork.
Your argument is based on the flawed assumption that Opensolaris will be dual licensed under the GPLv3. From everything i’ve seen from Opensolaris community there is alot of opposition to any such dual licensing. One of the main reasons been the fear of GPL only forks.
I never said it was definite. I was responding to the article which mentions the possibility.
You logic is flawed, the main difference between Linux and Solaris in this regard is Sun owns the copyright on all the code they’ve released as “OpenSolaris” as result they can do whatever they want license wise with it. They could cease dual licensing it for example. In which case any GPL only fork would need a dev team to maintain, given the FSF progress with HURD I don’t see this been a huge success do you?
There is a huge difference between HURD and Solaris. The Solaris kernel isn’t alpha software. It is release quality already with enterprise features. If SUN stopped dual licensing Solaris there is a chance that it could be killed but if a community has sprung up before that happens there is nothing Sun can do to stop it.
The point been that after such a “fork” everything new and interesting going into Solaris would not be available to your theoretical GPLv3 fork.
Only if dual licensing is stopped. With a full featured kernel available already and the possibility of a sustained community springing up around Solaris, there is nothing anyone could do to stop the GPL version from becoming a standard kernel amongst open source operating system distributors.
Wrong thread.
Edited 2007-02-28 22:47
Great news.
I look forward to a usable alternative to Linux on desktops and laptops within 2 or 3 years. Right now the installers for Linux and the driver availability are more user-friendly and reliable than anything available for Solaris.
But having a mature operating system joining the Free Software family is valuable and appreciated. Obviously, Sun wants more exposure and legitimacy through this and they are bound to get it.
They have gone from the stuff I would not touch with a 10-foot pole to stuff I am interesting in and may soon test on some test boxes.
It would be paradoxical if Linus and company let their current success in the market blind them as to the long-term benefits of GPLv.3, thereby creating a huge hole in the market for someone like SUN to fill.
I look forward to a usable alternative to Linux on desktops and laptops within 2 or 3 years. Right now the installers for Linux and the driver availability are more user-friendly and reliable than anything available for Solaris.
Pardon, but I’ve been using Solaris on my desktop and Laptop for the past two years. It was a usable alternative for developers like me then, and still is now. Just as Linux deskstops are not useable for some, Solaris desktops are not usable for others. It is all a matter of perpsective and need.
But having a mature operating system joining the Free Software family is valuable and appreciated. Obviously, Sun wants more exposure and legitimacy through this and they are bound to get it.
Solaris joined the Free Operating System family over a year ago. This SUN officially associating themselves with a corporate sponsorship and has nothing to do with the OS.
Everyone assumes that just because SUN is doing this means that OpenSolaris will be licensed or dual-licensed under the GPLv3. That is not true. That remains undecided.
“Should OpenSolaris be dual licensed via CDDL and GPLv3”
http://www.opensolaris.org/jive/thread.jspa?threadID=23699&tstart=0
Sun is increasingly a good citizen of the free software community, OpenSolaris is a great free software OS, and the CDDL is a damn good free software license for the pragmatic OSS developer. Sun is beginning to see the light when it comes to free software, and their new relationship with the FSF is representative of this newfound appreciation for (and confidence in) free software licensing.
But Sun is still years–even decades–away from coming around on the other key aspect of free software development–the community. Merely opening the code and bringing the issue tracker public is not enough. This is just a play for the hearts and minds of the free software community. What good is this if Sun isn’t responsive to externally contributed bug reports or patches? Free software is about community and its participation. That’s what makes the model so successful, and that is what’s missing from OpenSolaris (and the rest of Sun’s free software plays).
Sun has this habit of creating free software projects that don’t foster any community. First came OpenOffice. By all accounts, a breakthrough in the competitiveness of the free software desktop. But after almost 7 years as a free software project, upwards of 90% of the commits still come from inside Sun. The sprawling nature of a productivity suite isn’t to blame, it’s mostly the stifling complexity of the codebase. This isn’t a problem for KOffice, since they embrace high-level abstractions and modular programming techniques.
The clumsiest aspect of free software is the process of taking a proprietary product and evolving it into a successful free software project. If you consider the big projects donated to the community by the likes of Sun and Netscape, it becomes clear that they still haven’t fostered strong community development efforts after the better part of a decade. They’re incredibly popular and successful products that would fall into major disrepair if the core developers moved on to other things.
Sun seems to think this is easy. They seem to think that they can throw the doors open and make OpenSolaris compete for developers head-to-head with Linux. Linux has gotten to this point by evolving over the past 15+ years as a codebase that must be accessible to new developers and suitable for distributed development. Many technical decisions over the years were decided on the basis of these central requirements. Solaris has never evolved with these considerations in mind, and it has never seen the great and the ugly of distributed development over mailing lists. It hasn’t been a community effort, and if history is any indication, it won’t be a community effort 10 years from now either.
Congratulations, Sun, for diving head-first into a swimming pool you neglected to fill beforehand. You’ve completely missed the point of free software, and so you will never reap the rewards of a strong development community. You’re VPs will stay up nights wondering how they’re going to continue to match pace with Linux kernel development. Hardware vendors will wonder if it makes sense to develop and maintain a Solaris driver when the Linux kernel community maintains their Linux driver in-tree.
Congratulations on successfully releasing your OS product as free software. But try not to look surprised when you realize that this free software thing is give-and-give for you and a take-and-take for everyone else. The community provides the give-and-take, and you simply haven’t figured that out yet.
Sun seems to be quite aware that community building is the #1 issue with opensolaris and java – and that’s what they do.. slowly, but you described already, how complicated such a process is.
as for “Hardware vendors will wonder if it makes sense to develop and maintain a Solaris driver when the Linux kernel community maintains their Linux driver in-tree.”, this is just an issue with linux (that having drivers in-tree is preferable), because they don’t have any stable interfaces.
Hardware vendors have few issues with writing a driver, if it still works with the system 3-5 years down the road.
And I regularily use solaris 8 drivers (~2000 era) in solaris express (2 weeks old).. and some people go back even more.. so 7 years and counting – that makes any argument of “we would even maintain it” pretty pointless.
They didn’t miss the point. When they Open Sourced Java, they had Richard and Eben giving glowing endorsements. Yes the projects aren’t really that open but they have still managed to make RMS into their Free Software spokesperson… okay, maybe they did miss the point.
Well, you may have noticed the typo in the subject (the spurious “the”). I had originally titled it “Congrats to the FSF for missing the point,” but my ideas evolved while writing this comment. I changed FSF to Sun but forgot to remove the extra “the.”
Edited 2007-02-28 19:49
You’ve tried to disect the problems nice and academically by claiming its because of licencing or lack of community development – a simpler, more direct reason is this; terrible marketing and in the case of Solaris, bad PR from past decisions made.
It is the same reason why Darwin is nothing more than a show off of Darwin code rather than being a community where people can disect the code and improvement – it is company policy of ‘only running on Mac’ which stops people.
Those who want to contribute, might not necessarily want to run a Mac, but unless they purchase one, they’re either left with a rather stunted, in terms of functionality, operating system – which lets remember, a programmer who might contribute isn’t going just programme all day, he or she might want to surf the web, play games, listen to music and the like.
Transpose that to the situation of OpenSolaris – look through the Navada change log relating to the development of what will eventually get known as “Solaris 11” – http://www.opensolaris.org/os/community/on/flag-days/all/
Go through the list from top to bottom – look at who has contributed most of the code, and what is the code relating – most of the code is submitted by Sun and is only for Sun related hardware – if I am a developer, like the Darwin experience, I am not going to go out of my way, trash my current development box to purchase a new Sun one just so I can get some resemblence of a decent workstation/desktop experience, and if it requires me to do that, when I can achieve the same result using Linux, as well as having access to the source code as an enticement, which am I going to go for? lets remember folks, there aren’t too many people attracted to train wrecks and try to see the beauty behind the carnage.
Linux has developers because the desktop/workstation experience is at that stage where you can dump it on a generic machine, and away you go – if you look at the number of developers, it has grown as the desktop/workstation experience improved – as part time coders for opensource, they could use the same operating system for their work whilst at the same time when they’re in their own time, work on their hobby.
If Sun wish to improve their image and the number of contributors, they first need to step out and make some much needed improvements to Solaris in regards to running it in a desktop/workstation setup by improving hardware support – nick drivers of FreeBSD if required, how you do it is immaterial, just get it done; improve the GUI experience – SMC is a usability nightmare which appears to be written when everyone had a Java fetish – GTK is your friend, use it.
Fix up your API’s; the sound API is a complete joke, it is stuck 20 years ago when basic stereo sound was considered ‘acceptable’ – if it means buying out OpenSound and merging the OpenSound API into the mainline Solaris source code, then so be it.
Fix up your hardware support – if Ati/AMD won’t support the hardware then dam well approach the company and PAY for it to be ported and PAY for it to be supported – same goes for Nvidia.
Get involved with software OUTSIDE Solaris – yes, you port GNOME but I see NO development on NEW applications or actually fixing the existing applications that exist – it seems like “lets just port it, but leave it up to the GNOME developers to correct any problems we might encounter” – sorry, thats unacceptable.
If Sun want people to jump onboard the OpenSolaris gravy train, how about Sun stepping up to the crease and actually doing something about it – simply sitting back and only fixing problems that relate purely to your own hardware smacks of a complete disconnect between Sun and what is means to be as a community player – as a community player, it might actually require you, at times, develop technologies that don’t directly benefit you, such as third party hardware support, but will benefit you indirectly, in the form of a new customer who will come forward, knowing their hardware is supported, for a Solaris maintainance and support contract.
Go through the list from top to bottom – look at who has contributed most of the code, and what is the code relating – most of the code is submitted by Sun and is only for Sun related hardware – if I am a developer, like the Darwin experience, I am not going to go out of my way, trash my current development box to purchase a new Sun one just so I can get some resemblence of a decent workstation/desktop experience, and if it requires me to do that, when I can achieve the same result using Linux, as well as having access to the source code as an enticement, which am I going to go for? lets remember folks, there aren’t too many people attracted to train wrecks and try to see the beauty behind the carnage.
That reasoning is flawed for several reasons.
You are aware that SUN sells x86 hardware, right?
You are also aware that most of the changes are *NOT* for SUN hardware specifically, right?
You are also aware that SUN has publicly stated that they want Solaris to be available and supported on Dell, HP, and many other systems, right? (doesn’t sound like SUN hardware to me…)
You are also aware that many people can run Solaris on everything from laptops to desktops to workstations all which do not come from SUN, right?
For example, I’ve been running Solaris on my Dell Inspiron 5150 and three different custom built systems that I had long before I had Solaris and originally ran Linux and Windows. Now almost all of them run Windows and Solaris.
You are also aware that SUN recently announced a partnership with Intel for further expanded support of Intel hardware?
You are also aware that a driver with 3D acceleration will be integrated soon for Intel’s new 3D chipset, right?
You are also aware that nVidia drivers for Solaris are available from http://www.nvidia.com directly, or out-of-the-box with recent Solaris Express builds, right?
If Sun wish to improve their image and the number of contributors, they first need to step out and make some much needed improvements to Solaris in regards to running it in a desktop/workstation setup by improving hardware support – nick drivers of FreeBSD if required, how you do it is immaterial, just get it done; improve the GUI experience – SMC is a usability nightmare which appears to be written when everyone had a Java fetish – GTK is your friend, use it.
If you were following the community, you would know the following:
1) There are projects going on right now to significantly increase the number of network drivers available
2) A significant amount of GUI experience improvement is going on, in fact GNOME 2.16 is available right now on Solaris Express and an OpenSolaris KDE Project was just started a few weeks ago
3) SMC will be replaced, this was stated almost two years ago
4) nVidia drivers are available out of the box on Solaris Express or from http://www.nvidia.com for others
Fix up your API’s; the sound API is a complete joke, it is stuck 20 years ago when basic stereo sound was considered ‘acceptable’ – if it means buying out OpenSound and merging the OpenSound API into the mainline Solaris source code, then so be it.
One could argue that Linux’s sound API is a complete joke too. Rather, it’s a joke that obviously no one can make a good one so Linux has ended up with several.
The current level of audio support is fine for SUN’s target market. Their target market is not desktop users. It is workstations, and so on that do not need 3D audio. Basic stereo audio is all that is required. Users that want or require more can download *FREE* drivers from 4Front Tech for personal use or purchase them otherwise.
Fix up your hardware support – if Ati/AMD won’t support the hardware then dam well approach the company and PAY for it to be ported and PAY for it to be supported – same goes for Nvidia.
People make the mistake of assuming that money will always get them what they want. This is not true. ATi in the past has refused any offers even *for money* to obtain specifications needed to write drivers or to provide drivers. You want proof of this? Just go talk to the guys at http://www.xig.com.
Get involved with software OUTSIDE Solaris – yes, you port GNOME but I see NO development on NEW applications or actually fixing the existing applications that exist – it seems like “lets just port it, but leave it up to the GNOME developers to correct any problems we might encounter” – sorry, thats unacceptable.
Then you haven’t been following the community. In fact, you haven’t even been following the GNOME community either. If you had, you would know SUN is one of the biggest contributors to the GNOME project.
If Sun want people to jump onboard the OpenSolaris gravy train, how about Sun stepping up to the crease and actually doing something about it – simply sitting back and only fixing problems that relate purely to your own hardware smacks of a complete disconnect between Sun and what is means to be as a community player – as a community player, it might actually require you, at times, develop technologies that don’t directly benefit you, such as third party hardware support, but will benefit you indirectly, in the form of a new customer who will come forward, knowing their hardware is supported, for a Solaris maintainance and support contract.
Your statements are outright false. Please research the situation accurately before making these claims.
You are aware that SUN sells x86 hardware, right?
Which has nothing to do with the conversation – I didn’t come down with the last shower, and of course i know about their Opteron/AMD64 based machines, and some time this year, Intel based ones as well.
You are also aware that most of the changes are *NOT* for SUN hardware specifically, right?
I’ve just had a look through the changes, most of them relate to problems between Solaris and SPARC based hardware – when are we going to see improvements on the x86 side?
You are also aware that SUN has publicly stated that they want Solaris to be available and supported on Dell, HP, and many other systems, right? (doesn’t sound like SUN hardware to me…)
And they fail to provide an adequate level of hardware support for those third parties – tell me when Solaris x86 can run on my Toshiba A100/PSAA9A with the hardware supported out of the box.
You are also aware that many people can run Solaris on everything from laptops to desktops to workstations all which do not come from SUN, right?
Which has nothing to do with the conversation – generalisations about ‘many people’ constute nothing more than a giant size hyperbole when it comes to contributions from outsides – those outside Sun’s own pool of employees.
For example, I’ve been running Solaris on my Dell Inspiron 5150 and three different custom built systems that I had long before I had Solaris and originally ran Linux and Windows. Now almost all of them run Windows and Solaris.
Good for you *solitairy clap* now get my problems fixed – chop! chop!
You are also aware that SUN recently announced a partnership with Intel for further expanded support of Intel hardware?
They had 2 years to fix the problems, they chose to do nothing about it – I don’t care how many ‘alliances’ the have, as they have produced jack squat when it came to hardware support – what about the AMD link up? when are we going to see some decent bloody drivers for Ati graphics cards? what about the Nvidia link up, when are we going to see them support consumer grade graphics cards that *do* appear in laptops and low end workstations?
You are also aware that a driver with 3D acceleration will be integrated soon for Intel’s new 3D chipset, right?
Which is completely irrelevant to the conversation.
You are also aware that nVidia drivers for Solaris are available from http://www.nvidia.com directly, or out-of-the-box with recent Solaris Express builds, right?
Which only supports workstation class graphics cards and worse still, lacks DRI/DRM support thanks to the non-existant effort on Sun’s part to fix up the problems with Solaris.
One could argue that Linux’s sound API is a complete joke too. Rather, it’s a joke that obviously no one can make a good one so Linux has ended up with several.
Bullshit – Linux had used OpenSound because it was DONATED by OpenSound; yes, it wasn’t the best thing on the planet, but since no one came up with an improvement over it, it was kept as the status quo.
Then ALSA was developed with the sponsorship of SuSE which addressed many of the issues that plagued OpenSound – in 2.4 it was mature, and many vendors started merging it into the mainline kernel – 2.6 saw the removal of the OpenSound modules in favour of ALSA – again, where are these ‘several’? there is only one SoundAPI – its called ALSA.
The current level of audio support is fine for SUN’s target market. Their target market is not desktop users. It is workstations, and so on that do not need 3D audio. Basic stereo audio is all that is required. Users that want or require more can download *FREE* drivers from 4Front Tech for personal use or purchase them otherwise.
Oh please, what a pathetic excuse, I won’t even entertain coming up with a reply given the short sighted narrow mindedness that came through in that post, and the failure to realise that there is more to ‘office work’ than just pumping out spreadsheets and compiling code.
People make the mistake of assuming that money will always get them what they want. This is not true. ATi in the past has refused any offers even *for money* to obtain specifications needed to write drivers or to provide drivers. You want proof of this? Just go talk to the guys at http://www.xig.com.
What happened to Sun’s relationship with AMD? non-existant? went down the toilet? hot air being spoken by Sun executives again?
Then you haven’t been following the community. In fact, you haven’t even been following the GNOME community either. If you had, you would know SUN is one of the biggest contributors to the GNOME project.
Then how come I can’t download and compile GNOME out of the box?
Your statements are outright false. Please research the situation accurately before making these claims.
Where is Intel 3945abg support? Where is RealTek HD support? where is go 7300 support? where is DRI/DRM? where is bluetooth support? where is infered support?
Edited 2007-03-01 03:42
You are also aware that nVidia drivers for Solaris are available from http://www.nvidia.com directly, or out-of-the-box with recent Solaris Express builds, right?
Which only supports workstation class graphics cards and worse still, lacks DRI/DRM support thanks to the non-existant effort on Sun’s part to fix up the problems with Solaris.
The Solaris drivers use the same code as the Linux & FreeBSD ones – they work on all the cards, consumer and workstation, Sun just only sells support contracts for the cards that it sells in it’s workstations.
No one has DRI for nVidia cards – it’s never been written before since nVidia’s driver uses it’s own kernel module instead. (The nouveau project is working to fix that, but isn’t usable yet.) Solaris does have DRI for Intel graphics because Sun has been working to solve this problem.
(BTW, the ChangeLog you linked to covers only the kernel and core OS utilities – it doesn’t include X, GNOME, Firefox, or any of the higher level parts of the OS.)
Maybe, since you’re a Sun employee, you can inform me when this so-called ‘driver’ or atleast NDIS support for 3945abg is going to be finished – 6 months worth of ‘its being developed my our chinese team’ is enough to make me go batty.
I’d like to know some answers now – is Sun actually working on it? I mean, if Sun just simply want to tell customers to go f*ck themselves, why don’t they just simply say it, rather than beating around the bush. All the actions so far by Sun have basically said to customers, “screw you and the horse you rode in on”.
I’ve just had a look through the changes, most of them relate to problems between Solaris and SPARC based hardware – when are we going to see improvements on the x86 side?
The changelog you linked to is not the changelog for the entire OS. Secondly, I suspect much of the development you see is for SUN’s new SPARC based systems. I don’t know why you expect a huge flurry of x86 specific development given that Solaris already supports a lot more x86 hardware than you seem to give credit for.
And they fail to provide an adequate level of hardware support for those third parties – tell me when Solaris x86 can run on my Toshiba A100/PSAA9A with the hardware supported out of the box.
Sorry, but that’s a “generalisation” to use your term. I have systems that many Linux distributions won’t run on. Does that mean that Linux doesn’t provide an adequate level of hardware support? Just because Solaris doesn’t work on *your* configuration doesn’t mean it doesn’t have an “adequate level of hardware support.” There will *always* be hardware that doesn’t work with Solaris and Linux. Quite frankly, I would *never* expect all hardware to work with Solaris or Linux. It is an unreasonable expectation, given that even newer versions of Windows don’t support a lot of “relatively recent” hardware.
Bullshit – Linux had used OpenSound because it was DONATED by OpenSound; yes, it wasn’t the best thing on the planet, but since no one came up with an improvement over it, it was kept as the status quo.
I never mentioned OpenSound. I am well aware of ALSA. So this is irrelevant to the discussion. As far as several? I am talking about the fact that many applications today don’t use OSS, ALSA, or some other API directly. Instead, they use SDL, JACK, NAS, eSD, aRTS and others. That means a different audio API.
Then ALSA was developed with the sponsorship of SuSE which addressed many of the issues that plagued OpenSound – in 2.4 it was mature, and many vendors started merging it into the mainline kernel – 2.6 saw the removal of the OpenSound modules in favour of ALSA – again, where are these ‘several’? there is only one SoundAPI – its called ALSA.
There may be only *one* KERNEL sound API. But there are *many* sound APIs. Any free software developer that chooses to use ALSA is limiting his application to Linux. ALSA is not a portable API and should never be used in my view. A portable library over the top of it should be used instead.
I would say the same regardless of platform as well. OpenAL is a better choice…
Quite frankly, I don’t like ALSA, and I’m not the only developer. Before you accuse me of not being a Linux developer, I am. In fact, I’ve helped develop and port games to Linux for years now and I still actively maintain one of those ports even now.
Which only supports workstation class graphics cards and worse still, lacks DRI/DRM support thanks to the non-existant effort on Sun’s part to fix up the problems with Solaris.
WRONG. SUN only offers *support contracts* for the workstation class cards. Consumer level cards work just fine with the driver, however. Such as my GeForce GO 5200, or my GeForce 7800 PCI-E GT KO and so on. DRI/DRM support has never been part of nVidia’s drivers because they use a completely different architecture and that is true even on Linux. I don’t think you actually know what is there and what is not.
As far as *general* DRI/DRM support, that is coming soon, and is starting with the new Intel 3D chipset.
Oh please, what a pathetic excuse, I won’t even entertain coming up with a reply given the short sighted narrow mindedness that came through in that post, and the failure to realise that there is more to ‘office work’ than just pumping out spreadsheets and compiling code.
It is not shortsighted at all. It’s called being realistic. As I said before, if you want *more*, there’s a driver that’s *free* for you to use out there. What is so hard to understand about that?
What happened to Sun’s relationship with AMD? non-existant? went down the toilet? hot air being spoken by Sun executives again?
I fail to see how SUN’s relationship with AMD *a processor company at the time* has any bearing on *ATI a graphics division of AMD only recently*.
Considering that ATi won’t even fully support the Linux platform, I fail to see how you can lay *any blame at all* on SUN for ATi’s failings.
You know what ATi said the last time ATi owners complained about the level of support for other operating systems and even on Linux? That the amount of their support was proportional to the market. Or in other words, customers aren’t worth their time. Go direct your venom at ATi, not SUN.
Then how come I can’t download and compile GNOME out of the box?
Because you don’t have a proper development environment setup for building GNOME?
I can show you many Linux distributions and BSD distributions and others where I can’t compile things “out of the box.” Does that mean that they suck, don’t do enough, etc.?
You do realise that almost every Linux distribution has a special build environment for GNOME, KDE and so forth, right?
There is one for GNOME as well:
http://opensolaris.org/os/project/jds/contributing/building/
Just because Solaris doesn’t come with the tools out of the box needed to build some random open source application you picked out is not a reliable indicator in any fashion of the viability of a platform.
Where is Intel 3945abg support? Where is RealTek HD support? where is go 7300 support? where is DRI/DRM? where is bluetooth support? where is infered support?
Wireless support is being worked on. Just because every device you want to be supported today isn’t doesn’t mean anything about platform viability. For example, Atheros based wifi devices are very well supported right now. Also, other devices work using the ndiswrapper.
I could ask the same thing about my Broadcom 4309 wireless adapter and Ubuntu. It doesn’t work out of the box on Ubuntu either even though the laptop was first made four years ago…hrm….
HD audio support begin integration into Solaris in June 2006 of last year, behold the code:
http://cvs.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/c…
The GeForce 7300 GO works just fine. Go download the driver from http://www.nvidia.com.
A prototype of bluetooth support has been posted on the OpenSolaris website, go here for more information:
http://opensolaris.org/os/project/bluetooth/
IrDA support is being discussed right now and it is possible that work will begin soon:
http://www.opensolaris.org/jive/thread.jspa?messageID=83227𔔛
The changelog you linked to is not the changelog for the entire OS. Secondly, I suspect much of the development you see is for SUN’s new SPARC based systems. I don’t know why you expect a huge flurry of x86 specific development given that Solaris already supports a lot more x86 hardware than you seem to give credit for.
I expect a HUGE flurry of activity fixing up the deficiencies in their x86 port! I expect them to fix up their base library support so that applications can compile out of the box without needing to dick around with settings – 2 years and still not addressed, quite frankly, that is unacceptable.
Sorry, but that’s a “generalisation” to use your term. I have systems that many Linux distributions won’t run on. Does that mean that Linux doesn’t provide an adequate level of hardware support? Just because Solaris doesn’t work on *your* configuration doesn’t mean it doesn’t have an “adequate level of hardware support.” There will *always* be hardware that doesn’t work with Solaris and Linux. Quite frankly, I would *never* expect all hardware to work with Solaris or Linux. It is an unreasonable expectation, given that even newer versions of Windows don’t support a lot of “relatively recent” hardware.
Windows Vista runs wonderfully, Linux, coded by a bunch of piffy part time coders support my hardware better than Solaris.
Explain to me why a multi-billion dollar organisation like Sun can’t support the most basic of hardware on my laptop, and yet, a little cadre of programmers scattered around the world, developing an opertating system in their part time, can do a better job?
I never mentioned OpenSound. I am well aware of ALSA. So this is irrelevant to the discussion. As far as several? I am talking about the fact that many applications today don’t use OSS, ALSA, or some other API directly. Instead, they use SDL, JACK, NAS, eSD, aRTS and others. That means a different audio API.
And all those API’s are available on Solaris – infact, Solaris DOESN’T have a default one at all! Has it also ever occured to you that each have a different purpose? has it occured to you that Sun hasn’t even decided which one it is going to support?
Wireless support is being worked on. Just because every device you want to be supported today isn’t doesn’t mean anything about platform viability. For example, Atheros based wifi devices are very well supported right now. Also, other devices work using the ndiswrapper.
Over sixth months Sun have been promising support for 3945abg, and no delivery – sorry, they lost me as a customer, and in what I do, they lost my willingness to pitch a Sun product over Apple or Microsoft.
If Sun want to be an arrogant prick, then so be it; but don’t try to make them out to be the second coming or gods gift to computing by virtue of the rantings and ravings from clueless managers.
I expect a HUGE flurry of activity fixing up the deficiencies in their x86 port! I expect them to fix up their base library support so that applications can compile out of the box without needing to dick around with settings – 2 years and still not addressed, quite frankly, that is unacceptable.
Many applications *can* compile out of the box. Most of the ones that can’t are because they are *NOT* portable. I don’t see why you think NON-portable applications are SUN’s problem. Solaris has a very capable set of development tools that comply with widely available sets of standards which some platforms have chosen not to follow.
Windows Vista runs wonderfully, Linux, coded by a bunch of piffy part time coders support my hardware better than Solaris.
Explain to me why a multi-billion dollar organisation like Sun can’t support the most basic of hardware on my laptop, and yet, a little cadre of programmers scattered around the world, developing an opertating system in their part time, can do a better job?
As I said before, just because your hardware doesn’t work doesn’t mean anything about the Solaris platform as a whole, no more than some of my hardware not working on Linux means anything about the Linux “platform” as a whole.
And all those API’s are available on Solaris – infact, Solaris DOESN’T have a default one at all!
Actually, Solaris DOES have a default audio API. See “man audio.”
Has it also ever occured to you that each have a different purpose?
Yes, has it occurred to yet that I do know about all of these things as a developer involved in game ports for the Linux and Solaris platforms as I already explained I was?
has it occured to you that Sun hasn’t even decided which one it is going to support?
Yes, though I fail to see how that relates to this discussion.
Over sixth months Sun have been promising support for 3945abg, and no delivery – sorry, they lost me as a customer, and in what I do, they lost my willingness to pitch a Sun product over Apple or Microsoft.
They have? Where? I have seen no such promises, I especially have never seen a specific delivery date promised. I’ve seen Linux distributions promising better wireless support for years, yet my four year-old broadcom 4309 wireless adapter still doesn’t work “out-of-the-box” almost every major Linux distribution. Should I hold them just as accountable as you hold SUN?
If Sun want to be an arrogant prick, then so be it; but don’t try to make them out to be the second coming or gods gift to computing by virtue of the rantings and ravings from clueless managers.
Your venom seems so personal and unfounded. Solaris is an operating system, and SUN is a business. I think that there are far worse things in the world that are far more deserving of such venom.
Get involved with software OUTSIDE Solaris – yes, you port GNOME but I see NO development on NEW applications or actually fixing the existing applications that exist – it seems like “lets just port it, but leave it up to the GNOME developers to correct any problems we might encounter” – sorry, thats unacceptable.
If you think this is true, you obviously haven’t been following the community at all. For example, SUN has ported HAL to Solaris so removable desktop devices work on GNOME. That’s proof enough that they don’t leave “any problems we might encounter” to GNOME developers. What you say is also completely untrue because SUN has people that *are* core GNOME contributors. If you had taken the time to read the changelogs for JDS, you would know that SUN has fixed *many* things along the way in maintaining and porting GNOME to Solaris.
If Sun want people to jump onboard the OpenSolaris gravy train, how about Sun stepping up to the crease and actually doing something about it – simply sitting back and only fixing problems that relate purely to your own hardware smacks of a complete disconnect between Sun and what is means to be as a community player – as a community player, it might actually require you, at times, develop technologies that don’t directly benefit you, such as third party hardware support, but will benefit you indirectly, in the form of a new customer who will come forward, knowing their hardware is supported, for a Solaris maintainance and support contract.
I suppose that supporting Intel systems before they ever shipped them doesn’t count as 3rd party hardware. I suppose that assigning resources to the development of bluetooth, wireless, and *consumer-level 3d nVidia cards* that SUN doesn’t even ship doesn’t count either then?
Your accusations that SUN does nothing for third party hardware support, especially when it does not directly benefit them, are patently false.
Edited 2007-03-01 10:28
What good is this if Sun isn’t responsive to externally contributed bug reports or patches? Free software is about community and its participation. That’s what makes the model so successful, and that is what’s missing from OpenSolaris (and the rest of Sun’s free software plays).
Pardon, but they are quite responsive to contributed bug reports and patches. I invite you to become involved with the community to discover this for yourself.
OpenOffice. By all accounts, a breakthrough in the competitiveness of the free software desktop. But after almost 7 years as a free software project, upwards of 90% of the commits still come from inside Sun.
Perhaps because SUN has thousands of software engineers which makes it easy to dwarf anyone else’s contributions?
Really, I don’t understand this claim. Other companies open source something and people accuse them of merely throwing code over the wall and expecting others to be free labour to write their product. SUN not only opens the code, but spends millions of dollars in labour, legal fees, documentation, etc. for that process to happen and then still remains the largest contributor and they get accused of not doing enough?
This isn’t a problem for KOffice, since they embrace high-level abstractions and modular programming techniques.
Meanwhile, KOffice remains in the barely measurable amount of impact category. I fail to see how they can be considered successful given how little uptake their project has among users.
If you consider the big projects donated to the community by the likes of Sun and Netscape, it becomes clear that they still haven’t fostered strong community development efforts after the better part of a decade.
With SUN I can understand how you might claim them as not having fostered a strong community…but Netscape? You do realise the Mozilla project is considered one of the largest and most successful open source projects, right?
Sun seems to think this is easy.
Pardon, you obviously haven’t been following *any* of the discussions among SUN and the community. They have been saying since *day one* on all of these projects that it would be a gradual process that was painful and was causing a significant number of changes internally and externally.
It hasn’t been a community effort, and if history is any indication, it won’t be a community effort 10 years from now either.
I flatly deny your claims and invite you to become involved in our community. There are quite a few community efforts surrounding OpenSolaris right now that are quite active.
Congratulations, Sun, for diving head-first into a swimming pool you neglected to fill beforehand. You’ve completely missed the point of free software, and so you will never reap the rewards of a strong development community.
Before accusing others of diving in without doing the proper research, perhaps you should take some time to get involved with our community yourself.
You’re VPs will stay up nights wondering how they’re going to continue to match pace with Linux kernel development.
You are aware that SUN has over a 1,000 engineers working on Solaris right? I don’t think they’ll be up at night worrying
Hardware vendors will wonder if it makes sense to develop and maintain a Solaris driver when the Linux kernel community maintains their Linux driver in-tree.
Hundreds of vendors already choose to maintain drivers for Solaris, and one of the reasons they do so is because it makes sense. Are you aware that many drivers that were written years ago for Solaris still work without modification today? Unlike Linux, but like Windows and other operating systems, Solaris actually has a well-documented, reliable, and stable driver API.
Hardware support for Solaris is now greater than it ever has been before. Your statements are of little import in that context.
Congratulations on successfully releasing your OS product as free software. But try not to look surprised when you realize that this free software thing is give-and-give for you and a take-and-take for everyone else. The community provides the give-and-take, and you simply haven’t figured that out yet.
This doesn’t even make sense to me. What do you mean by give-and-take? SUN has given more than any other company, individual, or person to the open source community. This is true in terms of lines of code, value of software, “man years” of engineering and many other ways.
The community is also giving back to that effort.
What are you trying to say?
Your point is that Sun (SUN?) is contributing a lot to free software and therefore that they will be successful in building a community around OpenSolaris. My point is that Sun is contributing a lot to free software and therefore that they haven’t figured out how to build communities around free software.
I’m not accusing Sun of not contributing enough to free software. I’m accusing Sun of not being effective at building free software communities in the past, which makes me believe that their chances of success with an entire free software operating system are slim.
Sun has 1,000 engineers working on Solaris, while there’s around 1,500 contributors to the Linux 2.6.x series. And that’s just the kernel. Combine that with all of the commercial Linux vendors, Debian, and other community efforts, and the disparity becomes quite large.
Hardware support in Solaris might be the best it’s ever been, but that’s not good enough. Hardware support in Linux is the best it’s ever been as well, better than Solaris, yet it still doesn’t support everything. The in-kernel APIs in Linux have been settling down and will continue to settle down as the kernel maintainers are increasingly happy with the current state of things.
I won’t be joining your community. First and foremost, I’m happy with Linux, and I see no reason to switch. Also, I don’t believe OpenSolaris is really going anywhere in the long-run, whereas it’s hard to argue the same thing for Linux. Finally, I work for IBM, and they don’t especially like Sun. On these forums I speak my own mind, though.
Sun has 1,000 engineers working on Solaris, while there’s around 1,500 contributors to the Linux 2.6.x series. And that’s just the kernel. Combine that with all of the commercial Linux vendors, Debian, and other community efforts, and the disparity becomes quite large.
I don’t know where you get this numbers (hell, sun has 38,600 employee). Even so, Sun has engineers working FULL TIME on Solaris, linux has contributor (working on their free time)…
Anyway, if you think that having a lot of developpers means better results, Microsoft is making the best os out there.
Your point is that Sun (SUN?) is contributing a lot to free software and therefore that they will be successful in building a community around OpenSolaris. My point is that Sun is contributing a lot to free software and therefore that they haven’t figured out how to build communities around free software.
I fail to see how contributing a lot of free software correlates with not being able to figure out how to build a community. I also fail to see how projects that they only recently started can be fairly compared to other open source projects which have existed for a lot more years.
Sun has 1,000 engineers working on Solaris, while there’s around 1,500 contributors to the Linux 2.6.x series. And that’s just the kernel. Combine that with all of the commercial Linux vendors, Debian, and other community efforts, and the disparity becomes quite large.
Consider how many of those contributors actually work on Linux full time, then compare to the fact that SUN can have all of theirs working full time. If Linux has such a huge advantage, why didn’t they have anything like zones until recently? Why don’t they still have an answer to ZFS? Why isn’t SystemTap done yet while SUN has DTrace already? I could go on, but suffice to say that the number of contributors is not a valid point of comparison.
Hardware support in Solaris might be the best it’s ever been, but that’s not good enough. Hardware support in Linux is the best it’s ever been as well, better than Solaris, yet it still doesn’t support everything.
Better is always relative. For example, I was able to run Solaris on my Intel Core 2 DUO system before I was able to run it on Ubuntu. So for that period of time, for me, it was better.
Not only that, Linux will likely always support more hardware, and quite frankly, I don’t care. I always find the claim of Linux supporting so many devices to be fascinating at best. Having spent many years building my own Linux kernels, having helped write a Linux sound driver, and other trials and travails, I’m quite aware of what most people consider “support” for hwardware. I’m also aware that most of the hardware that is claimed as part of that “superior” hardware support is fairly old.
Would I like to see more hardware supported for Solaris? Sure, and the community is working towards that, as well as SUN. But for now, all my hardware is supported (older and newer systems) and I’m quite happy…
The in-kernel APIs in Linux have been settling down and will continue to settle down as the kernel maintainers are increasingly happy with the current state of things.
On Solaris, a user can install drivers that were written *years* ago and they still work today perfectly. When the Linux driver API has “settled down” in a few years, come back and let me know.
I won’t be joining your community. First and foremost, I’m happy with Linux, and I see no reason to switch.
That’s a shame, since obviously you’re a highly motivated and technical person that could bring good things to a community. I also fail to see how becoming part of the OpenSolaris community would somehow imply that you would have to switch operating systems. For example, I still boot to Ubuntu occasionally because I maintain and support an adventure game runtime engine Linux port that was originally written for Windows. I also run Windows XP Pro SP2 as well.
Also, I don’t believe OpenSolaris is really going anywhere in the long-run, whereas it’s hard to argue the same thing for Linux. Finally, I work for IBM, and they don’t especially like Sun. On these forums I speak my own mind, though.
Well, I believe you’re wrong about that. Our community is vibrant and growing, and has continued to grow since the first day the project was launched. If you speak your own mind, I don’t see how IBM’s opinion about SUN matters. Finally, I wouldn’t expect them to like their direct competitor, SUN, very much at all
“But Sun is still years–even decades–away from coming around on the other key aspect of free software development–the community.” [snip] “What good is this if Sun isn’t responsive to externally contributed bug reports or patches?” Free software is about community and its participation. That’s what makes the model so successful, and that is what’s missing from OpenSolaris (and the rest of Sun’s free software plays).”
You haven’t been paying attention for awhile now it seems. Since this information you gave is just plain wrong. Sun has been very responsive to external bug reports, and outside contributors have been submitting code to Java for a couple of years now–even before it became GPL.
Has anyone seen the Sun Laptops?
http://www.sun.com/desktop/workstation/ultra3/
They’re no longer available.
The Sun Laptops had a sparc processor I guess there was not enough demand or they were to pricey?
They looked really nice and I bet the I/O was awesome!
I still haven’t gotten my free Solaris 10 DVD.
I still haven’t gotten my free Solaris 10 DVD.
I fail to see how that relates to this topic. But to reassure you, SUN simply had more demand than they anticipated and so had to produce more media, etc. to send out. They have started to send out the backlog of orders this week.
Not so long, linux users was saying that Sun politic toward open source was bullshit and not ‘real’ open source.
What do you have to say now ?
What’s the next troll linux users will throw against Sun this time ?
Will this include the installation of Java in Linux as in linked to the browser sometimes getting this setup can be very troublesome to say the least. Also linking it to apps like LimeWire and such to.
it should, before too long – the new JRE is GPL.
> How does this make GPLv2 software unusable with GPLv3 software?
Using a GNU GPLv3 licensed software with GNU GPLv2 licensed software won’t make the resulting software unusable. The problem is that it is unacceptable if the licensee links GPLv3 software together with GPLv2 software and share copies of the resulting software. The licensee is allowed to link GPLv3 software with GPLv2 software if the licensee does not share the resulting software.
Why isn’t the licensee allowed to do this? The answer is in Section 6 of GPLv2. GPLv2 has a certain set of restrictions. The GPLv3 is expected to have more restrictions than GPLv2. Therefore, the licensee unable to comply with Section 6 of GPLv2 if he mixes GPLv3 and GPLv2. Because the licensee cannot fulfill Section 6 of GPLv2, Section 7 of GPLv2 says the licensee loses the right to distribute the GPLv2 licensed software.