Submitted by Yesterday, the OpenSolaris team released OpenSolaris 2008.05, the fruit of Project Indiana. The first review we found was published over at Blogbeebe, which is overall fairly positive. At the same time, Practical Technology believes that “OpenSolaris has finally been released just in time to die”.I won’t bore you with the details of the review itself (you can read for yourself), but the conclusion sums it up pretty well:
OpenSolaris shows a lot of intriguing features with this release. It isn’t perfect, and it’s lack of support of many audio subsystems required for full multimedia (such as the M685’s Intel 82801G (ICH7) high definition audio controller) makes this release more suitable for OS geeks and developers than the average Joe Sixpack. But it’s definitely a release worth investigating further, and certainly one worth keeping an eye on for future development.
This words my feelings about this release perfectly. For a first attempt at a proper desktop-oriented distribution for ‘normal’ users, they seem to have done a pretty good job. If I compare my experiences with Solaris 9 on my Sun Ultra 5 with the first bits of information about OpenSolaris 2008.05, it’s like comparing Fiona Apple to chewed and dried up gum; Solars 9 was incredibly difficult to administer for someone without extensive knowledge about the OS. It seems as if the 2008.05 release is light years beyond that – and it’s only the beginning.
The comments at Practical Tech feel a bit like an attempt at ‘fear, uncertainty, and doubt, to me.
What the article fails to mention is that Novell already has a lot of problems with its image in the open-source community. Its deal with Microsoft irked a lot of people in the Free software community, and I’m fairly certain Novell will not want to annoy them even further by blocking the open source release of OpenSolaris. Sure, OpenSolaris is a competitor to their Linux business, but doing something as drastic as the article suggests seems like a very bad idea for Novell.
But we’ve seen companies do weirder things. Like suing Linux users because Linux is an exact copy of UNIX.
From the blogbee review:
“One feature that finally (finally!) worked for me is the gears inside the cube effect. I’ve never gotten it to work in any of the Linux distributions, but it worked like a champ on OpenSolaris. I know it’s silly, but hey, if you’re going to have eye candy, then it should all work.”
The Practical Tech article was just lame FUD and didn’t even talk about the 2008.05 release. That article could easily have been written when OpenSolaris was first announced years ago. Hardly relevant now.
Agreed. It is almost 100% FUD only (= Fear, Uncertainty and Doubt. That is what that article is all about, isn’t it?), not much relevant content that could be taken very seriously.
Also I don’t like the choice of using that article’s title for both reviews here on Osnews (especially as I’ve been marked as the article submitter although I submitted only the Blogbeebe review link and not the FUD article..
). Why not use a neutral title like simply OpenSolaris 2008.5 reviews or something like that? Or maybe only publish the Blogbeebe review that has some real content worth reading?
Edited 2008-05-07 07:36 UTC
I don’t think would be in the best Novell interest to stop OpenSolaris.
They already pissed off lots of people from the community, when they made the deal with Microsoft.
Now, suppose they move on to stop OpenSolaris. It would put them in the very strange situation where they use, or have free access to, code from lots of projects (FOSS code) and at same time make impossible to the same community to have access to code NOT MADE by Novell. That would be really awful.
Only a lunatic would act like that and pretend to be part of the community. And if they do act, this would really trigger I way bigger response from both, developers and users.
There’s zero evidence that Novell has any intention of trying to take legal action to impede publication of Solaris source code. After all, source availability for Solaris is in no way new.
In addition, since I’m not a kernel hacker, I’m not affected in any way if Solaris beats Linux or vice versa, since either way I get a full free Posix-compatible system that runs on cheap commodity hardware and has Gnome, KDE, or whatever else you want on top for desktops, and runs whatever server apps you want as well. Even Linus will tell you that all of the exciting work is happening in userland these days.
It might even be optimal to run both in your home or small office network: maybe you prefer the Linux hardware support for your desktops and laptops, but want to try out a ZFS file server.
The point of that piece is speculation that Novell might sue Sun such that Sun will stop developing OpenSolaris.
I think that the speculation in question is stupid
“OpenSolaris has failed to develop a strong developer community.”
It has a lot of developers. They are Sun employees working in Sun offices!
OpenSolaris development happens totally differently than it does for linux. Does anyone think this is by accident?
The fact that the author of this article assumes that everyone is striving to be like him (and therefore are failing when they are not) really shows off his sickly biases.
OpenSolaris is obviously trying to copy the success of linux and ubuntu and red hat. But they are not obviously trying to copy the development model.
I think you’ve just touched on exactly what the article is talking about.
No. The only problem for some people who have looked to get involved with OpenSolaris is the fact that if Sun wanted to do a MySQL type open source model, then they should just come out and admit it.
If Sun wants to do the MySQL development model?
Sun owns MySQL.
cheers
I like some features of Solaris .. but the whole package leaves a lot to be desired. I think trying to position Solaris as a linux wannabe with a few cool features ( Dtrace is nice, ZFS is blingy although it uses Vistaesques amounts of RAM ( like 0,5 GiB ) ) and like 1% of the drivers is not wise.
People are well aware of the possiblity of vendor lock-ins and OpenSolaris just seems a bit like the first free fix. Once you are hooked it will be hard to switch .. this kind of fragmentation does not help anybody.
What vendor lock in are you refrering to?
I have not seen a modern piece of software that runs on OpenSolaris that can’t be compiled for Linux also.
Mate, obviously you seem to have issues finding code or reading sentences. He has said that by OpenSolaris existing it stops vendor lock in. That vendor lock in being software written in the opensource world that only compiles on Linux. Want proof? go grab a generic wine tarball and try to compile it on OpenSolaris out of the box – it won’t compile.
Lame, up until decently, same situation – it didn’t compile unless you used patches (3.98beta8 works without patching). Code is still being locked into specific operating systems – even when that code is opensource. Programmers using Linuxisms, gcc’isms and any other possible ism I might have forgotten.
As for the value of OpenSolaris, it is nice to do something CPU intensive and not find that it is impossible to surf the internet because the network connection dies. Yes, I’ve had happen with Ubuntu, OpenSuSE, Fedora – its pathetic; are we supposed to believe that in 2008 one shouldn’t multitask?
I have admined about 80 Fedora, CentOS, and Ubuntu boxes of widely varying harware configs and OS versions for some years now, and I call bullshit on that one. I have never observed such a problem.
I did not bother with the ^A¨DOA^A¨ article, and I^A'll take your word for it that it was bad. But that is no reason to start gratuitously talking baseless trash about Linux.
This is usually a sign for beta quality software, some people with nice experiences and some people with fallouts. This extreme you’ll get especially with Fedora _and_ Ubuntu and this is a know fact if you have a look into bugzilla and co.
Oliver, I^A'm posting this from OpenSolaris 2008.05, and believe me, you don^A't want to get me talking about alpha and beta quality software. I have too much ammunition handy just now.
Edit: There is absolutely no reason that these threads have to turn into Linux<->Solaris bashing sessions. It is detrimental to both our communities. Evil spreads as good men do nothing but bicker among one another. In the words of Flash Gordon, ^A¨Let^A's team up!^A¨.
Edited 2008-05-07 12:00 UTC
Excuse me, how is it bullshit when I have a bloody wireless internet connection that drops faster than Britney spears on a crotch?
Excuse me, maybe the next time you’re in New Zealand I’ll invite you over for tea, coffee and ‘why Linux sucks’ classes. Point out the joys of ripping a cd and finding that the wireless connection slows to a crawl or simply stops responding.
Errrr, it’s wireless, and you should probably be looking elsewhere?
Hmmmmm. Whatever.
Kawai,
The tone of my post was probably a bit sharp, and for that I apologize. You did not mention that you were specifically referring to a wireless connection. That I can believe, although I find it unlikely^Wodd that processor load would be related. The quality of wireless drivers under Linux is currently rather variable. As the owner of a laptop with a bcm4318 chipset, and another with an Intel chipset, I know this well. (Actually, they both work quite well now, but this has not always been the case for the broadcom.) Due to primarily external factors, like vendor support[1] wireless driver quality in Linux today is somewhat like the girl with the curl in the middle of her forehead. I also have deployed a number of linux routers using a broadcom chipset and broadcom^A's binary driver, and wireless reliability does not get any more demanding than basing the wireless infrastructure on it. These applications have worked well.
Of course, I^A've yet to see truly rock solid wireless support in any OS, simply due to the nature of the beast. 2.4Ghz signals are fickle things. Line of site range is amazing… until a cow walks past. However, I must admit to a bit of sketpticism regarding any claim that the wireless driver support situation is better in OpenSolaris. I^A'm just getting my feet wet with it, but so far my perception has been that driver support, in general, is not as broad as is currently the case with Linux. (These things take time.) If you have a wireless chipset that works well with OpenSolaris, but not well with Linux, I would be curious as to what it is. And it is nice to hear that wireless can work well in OpenSolaris. This is not something with which I yet have experience. Come to think of it, I do have a linksys wireless pci card I can put into this machine for testing.
Let us hope that the wireless situation continues to improve for all of us in the Linux/*BSD/*Solaris world.
[1] I would be remiss not to mention that Linux^A's internal infrastructure for wireless has been suboptimal until recently. So not all the reasons are external.
Edited 2008-05-07 13:35 UTC
Umm…. I love blanket statements with no real numbers to back them up.
Currently, the destop that’s being used to post this message is running boinc (180% CPU), two vmware virtual machines (one is checking for updates), email client, firefox, large number of VI’s and being used as a file and printer server for my wife’s desktop.
…. And this is a mid-end Athlon64 5000X2 machine; The dual DC Opteron that sits next to is far more busy.
– Gilboa
Edited 2008-05-07 04:06 UTC
“As for the value of OpenSolaris, it is nice to do something CPU intensive and not find that it is impossible to surf the internet because the network connection dies.”
Maybe you shouldn’t run your IE in Wine as root, pal
Who is using IE on wine? I’m using the Firefox included with the distribution – Why the heck would I want to use Internet Explorer anyway?
Did I mention I’m also selling a clue-bat?
That vendor lock in being software written in the opensource world that only compiles on Linux. Want proof? go grab a generic wine tarball and try to compile it on OpenSolaris out of the box – it won’t compile.
Lame, up until decently, same situation – it didn’t compile unless you used patches (3.98beta8 works without patching). Code is still being locked into specific operating systems – even when that code is opensource. Programmers using Linuxisms, gcc’isms and any other possible ism I might have forgotten.
Wow, is this a straw man argument or what :O You do realize that it’s open code and it just might not compile on a new platform out-of-the-box unless someone has taken the time to test and make it work? And well, talking about wine..it works on SkyOS, Windows, Linux, BSDs…so, what were you saying about vendor lock-in?
As for the value of OpenSolaris, it is nice to do something CPU intensive and not find that it is impossible to surf the internet because the network connection dies. Yes, I’ve had happen with Ubuntu, OpenSuSE, Fedora – its pathetic; are we supposed to believe that in 2008 one shouldn’t multitask?
As others have already said, it seems no one believes what you have written here. And neither do I. I’m not saying it’s not possible…but heck, I have never come across such issues. And believe me, compiling things like FireFox on a Athlon 1ghz IS CPU intensive yet network connection never died and it was still very much possible to multitask. I don’t own a single multicore machine yet multitasking has always worked just fine, no matter if I’ve had Blender rendering in the background, compiling stuff, backing up the system or anything such.
And you realise perky buns, unless you’re using incredibly low level parts of the operating system – for the sake of argument, a cd player that needs to reach deep into the system – then sure, parts will need to be ported. But one shouldn’t need to ‘port’ lame, one shouldn’t need to port numerous user space applications that don’t read even the slightest in the operating system.
What it tells me that the person spent very little time using the documented open interfaces such as POSIX and UNIX 2003 and instead did the old, “if it compiles, ship it”.
Quite frankly you little piece of gutter snipe, no one cares what you or your entourage of Linux fanboys think, or more correctly, ‘feel’ about the situation. The fact is, I experienced a shit time with Linux. You can either accept that it was my experience or you can live in denial like a religious fundamentalist who is too high on their own delusion to see the reality standing before them.
Edited 2008-05-07 12:56 UTC
You might want to tell that to Sun’s reps the next time they start telling people to recompile everything in Forte versus gcc as a matter of course.
The vast majority of open source software out there is built for a GNU toolchain. Solaris has traditionally had major problems with that. Speak to Sun about it.
Hmmmm, good for you. I haven’t experienced anything you’ve described, and what you’re describing is not something you can throw into the ‘Linux’ basket.
There’s a reason that they recommend the Sun Studio toolset over the gcc one.
The gcc one is horrid for performance. The gcc compiler is so bad, that you can take as much as a 50% performance hit on sparc systems. It’s still fairly substantial on Intel systems as well. Sun has the better compiler.
That’s the reason why stuff doesn’t work, things aren’t ported and stuff isn’t packaged for Solaris. No one uses Sun’s compiler studio, and if you come on an open source list with a problem with Solaris, you’ll generally be given short shrift. That’s why Solaris has comparatively little software written and supported for it in the open source community.
No one in the open source community cares about SPARC systems. That’s why people running Python and lots of other open source software on Solaris simply moved to Linux and x86. It’s just plain easier.
Actually, quite a few people use Sun Studio, and it’s also available for GNU/Linux.
As far as very little software is ported and packaged, that is also not true.
You can find *thousands* of packages between sunfreeware.com and blastwave.org.
No one in the open source community cares about SPARC systems. That’s why people running Python and lots of other open source software on Solaris simply moved to Linux and x86. It’s just plain easier. [/q]
It’s also poor on some Intel platforms as well. It’s not just sparc. Therefore, that argument is irrelevant.
What it tells me that the person spent very little time using the documented open interfaces such as POSIX and UNIX 2003 and instead did the old, “if it compiles, ship it”.
And this is wrong in exactly how? It allows people to use the software anyways, and it didn’t seem like it required much work to make Lame compile under Solaris. You’re just making a bull out of a fly here. Would be very interesting to see you write something even half as functional as Lame and have it compile and work on *BSD, Solaris, Linux and any other POSIX compliant system without ever trying it on those systems yourself.
Quite frankly you little piece of gutter snipe, no one cares what you or your entourage of Linux fanboys think, or more correctly, ‘feel’ about the situation. The fact is, I experienced a shit time with Linux. You can either accept that it was my experience or you can live in denial like a religious fundamentalist who is too high on their own delusion to see the reality standing before them.
Wow how bitter you are, instantly resorting to personal attacks. No, sorry to have to break it to you but I am not a Linux fangirl. I do like using it but I also use Windows at home, and I do defend whatever OS in question if I see people trying to smear their image by spreading misinformation. I have had “shit time” in every OS, including Linux, but I still don’t resort to lurking on OSAlert and posting bull..
Lurking on osnews.com? sugar buns I’ve been on this website for well over 4 years; infact IIRC I used to post on this forum when I was in Australia under I think this one (or some other pseudonym). So whilst you’ve been here for 5 minutes. I’ve been on this site alot longer, and thus, cannot be considered a ‘lurker’.
So whilst you’ve been here for 5 minutes
Been here since 2006.
And yes, I see now, you again omitted all the things I said and try to change the subject. I have never seen you actually put your money to where your mouth is yet you preach to people that Linux users act like they are holier than thou.
Oh, and what was it..vendor-lockin through sources? Hmmmmmm…Lemme think. Oh yes, Wine. And _open_ sources, available for anyone to port to the OS they wish to use.
May I suggest you speak to Solaris and FreeBSD programmers who have tried to get their patches accepted into the main tree. Things aren’t all ‘free love and coding’ – there are those who do their upmost to ensure that non-Linux distributions do not succeed by making life as difficult as possible for maintainers of other operating systems.
May I suggest you speak to Solaris and FreeBSD programmers who have tried to get their patches accepted into the main tree. Things aren’t all ‘free love and coding’ – there are those who do their upmost to ensure that non-Linux distributions do not succeed by making life as difficult as possible for maintainers of other operating systems.
Give concrete examples as to such software instead of giving all-round-and-unspecified claims. As far as I can see, GNOME and KDE folks are working very hard to make sure their code is highly portable, Wine just gains advantage if their code is also useable on as many platforms as possible and so forth.
I never said they weren’t. I just said that there are examples of those who deliberately make their code unworkable.
KDE infact have been very hospitable when it comes to supporting alternative platforms, same can be said for GNOME. But with that being said, just because there are a few cases where there is co-operation doesn’t mean there aren’t projects who stifle other projects because of personal operating system biases.
I never said they weren’t. I just said that there are examples of those who deliberately make their code unworkable.
Then GIVE those examples instead of just keeping on saying they exist. If you can’t give such examples then it just clearly proves you’ve been blowing smoke here all along.
KDE infact have been very hospitable when it comes to supporting alternative platforms, same can be said for GNOME.
Both KDE and GNOME has been useable on BSDs for who knows how long, and the devs are actively seeking to enlarge the platforms on which those environments are useable. Have you f.ex. heard that they are aiming to bring KDE to Windows? Nah, I can’t think of any single app that is actually worth mentioning that would be Linux-specific.
Excuse me, but do you even know what hospitable means?
I have asked for specific examples three times now and since you still fail to provide those and instead try to change the subject I declare case closed. Thank you, see you later.
I did. You asked for example, and I said wine. You want to bury your head in the sand – good, go ahead and suffocate yourself on a big pile of sand. Its not going to change the fact that I gave an example and you chose to ignore it.
Edited 2008-05-07 14:31 UTC
I am not specifically familiar with the wine situation. But I believe that at a certain point, a developer or developers with an interest in a certain platform need to stand up and address the issue. Concerned users of a platform should make their presence known via polite postings to the relevant mailing lists. They squeaking wheel gets the grease, and all that. But I^A'm not certain that OSAlert is the most effective forum in which to ^A¨squeak^A¨.
I did. You asked for example, and I said wine.
After googling for a while, I still can’t find anyone complaining that wine devs would be even somewhat against BSDs or any other OSes. And as said, they are actively trying to improve how well Wine works under Windows, too. So that just proves your point utterly false. Go ahead, google around. Wine has been available and working under BSD for years and years, you can easily find messages about the devs trying to make it work under Windows and so forth…
In related news, I hear that Crossover Office is now, or shortly will be, available for FreeBSD. And since I am already posting about this… I believe that they are also spinning off a more fast moving gaming version in addition to the more conservative Crossover Office product.
Edited 2008-05-07 15:02 UTC
In related news, I hear that Crossover Office is, or shortly will be, available for FreeBSD. And since I am already posting about this… I believe that they are also spinning off a more fast moving gaming version in addition to the more conservative Crossover Office product.
I do not know if any BSD will actually be officially supported but there is an unsupported version available for download on their website. It sounds a good thing, seen lots of positive feedback. Haven’t tried it out myself, though.
And Crossover Gaming has been available for a while now
That’s a new contender for Cedega, and I personally like Crossover better. They contribute to Wine much much more.
Yes, “Crossover Games” already has a “Buy Now” link on the codeweavers front page. (With OpenArena, why would anyone need anything else? ) After having thought a little more, I believe that the BSD version is to be officially supported on PC-BSD, with assurances that it will unofficially work on FreeBSD.
Time for the OpenSolarians to start lobbying, I guess.
Edited 2008-05-07 15:14 UTC
I don’t think it would be too difficult to compile it on OpenSolaris either. Atleast Wine compiles just fine as long as you have GNU toolchain installed. (for those interested, look here http://blogs.sun.com/psum/entry/let_s_check_out_wine )
But unless they have some Solaris advocates in-house willing to do the compiling et al I doubt it’s going to happen.
We often ignore the fact, but writing portable code, and doing it well, does entail significant extra effort and care. If one platform offers a feature, e.g. inotify, and another supported platform does not, functionality must be duplicated. And that means extra work for the programmer. And that is aside from just getting things to compile properly on all platforms. All things considered, I think it reflects well on the OSS developer community in general that even the imperfect level of cross-platform compatibility which currently exists has been achieved. None of it came for free. It is to be expected that, depending upon the developer resources available, biases as to target platform are going to come into play.
Edited 2008-05-07 14:35 UTC
“You can either accept that it was my experience or…”
I completely believe about that because it’s obvious you don’t know much about computers. But that’s no problem as long as we’re all having fun
Linux took how long to get the mass of developers it has now? At the beginning, Linux didn’t have many developers, either. To compare the two is nonsense.
OpenSolaris doesn’t have the mindshare that Linux has, nor the license. It is certainly not dead on arrival.
I can’t foresee Novell starting litigation against Sun regarding Solaris. I don’t anticipate that Solaris is in a doom position either.
But OpenSolaris is attempting to become a Linux clone with a few intersting features and that doesn’t score many points in my book.
I think businesses that deploy Linux will opt for Ext4 or FUSE ZFS (devel versions are somewhat usable) and SystemTap, which is now implementing most of the kernel and user-space application debugging/tracing features of DTrace.
http://sourceware.org/systemtap/wiki/SystemtapDtraceComparison
Embracing similar technologies of the current platform is more cost-effective and efficient than migrating to another platform and re-training the staff.
Edited 2008-05-07 00:37 UTC
I’m all for accepting the short comings of OpenSolaris – but at the same time Practical Technology established nothing in their article. The article jumped all over the place, it claims there is no opensource development community around it and yet provides no evidence to substantiate such claims.
Infact, if you read the article, most of it was dedicating to digging up FUD over the legality of OpenSolaris and putting the contract between USL/Novell and Sun (the original one, the one made when Novell used to own USL/Novell) where by the purchased a broad reaching agreement which entitled them to rights which no other UNIX vendor has.
Is this really the quality of articles these days, when web rags like Practical Technology can’t even mount a logical argument, so they skip between different topics in the illusion that they’ve actually addressed some of the issues?
Sure, OpenSolaris has challenges, but if Practical Technology got off their fat chuff and had a look at the discussion lists on OpenSolaris they would see that development is happening rapidly.
Practical Technology also mounts a pathetic case of participating being the benchmark for the success or failure of OpenSolaris – open sourcing Solaris was never a matter of simply dumping code out there, firing programmers and using free labour to reduce costs; to make such claims (opensource dumping) without evidence seems to me that a Linux advocate has their nose bent out of shape because its no longer the darling of the IT industry. There is a new kid in town with UNIX pedigree – and the big boys are swooning around it like horny teenagers around the ‘hot chick’ at the pub.
Agreed.
The guys in Linux community are worried, that’s pretty obvious. They keep claiming “this is bad, that is bad” while they realize that such kid has a block to stand onto.
Even people like me who aren’t Unix/Linux users can appreciate how good it would be a free Solaris (because let’s confess it: it’s about price now) fully loaded with Enterprise-ready technologies and with a big guy like Sun behind it. That has a tremendous business value, expecially when you consider that even if Solaris is OS, Sun will always be able to bring a basic platform for “serious computing” while kids will keep releasing MyOwnLittleModifiedSolaris distro (though there are some distros which aren’t just about the new Gnome theme or the last MP3 player).
While I’m not interested in Linux for business reasons (at least until it will look so half-finished and dependent on untrustable pieces), I’m monitoring Solaris. It won’t be easy to drive us out of Windows field but if it will EVER be, it will be to Solaris, that’s for sure.
My concern is about Sun itself: will it be able to profit enough from all those open-sourced technologies? Sure, they have some advantages since they a h/w-maker too, but bringing Solaris back to compete with Windows and Linux means being able to profit enough to sell your hardware, keep R&D for that, continue development of Java, MySQL and Solaris itself and flood market with good apps / services to attract users. That means LOTS of resources.
Edited 2008-05-07 11:37 UTC
True. Lets put it this way; its like two beer companies; one is an expensive boutique beer and another is a cheaper one. The cheaper one gets most of the limelight because it is cheap – and good enough. The boutique beer comes out and makes their desirable beer cheaper; those who would have gone for the cheaper beer are now looking at the boutique beer. Guess who the cheaper beer is – its amazing when I see Linux advocates foam the mouth over Solaris when it should be a great thing there is more choice, more competition and more resources spent in the opensource world.
I’d say that Sun won’t create a desktop distribution in terms of the ‘unwashed masses’ but I do see Indiana becoming the operating system for workstations they sell – hopefully get people developing on OpenSolaris and then moving it beyond that to deploying it in enterprise desktops. I think that Sun has realised that Sun Ray product may be nice in theory, but business still want to keep traditional desktops – they’ve woken up and finally created a product to address that need.
Well, I think Sun is going to continue to improve – the problem is that wall street want instant results and simply ignore the reality that turning around the company takes time. Turning around HP was easy – it was bloated, over employment and bad direction. It was matter of slicing and dicing it down to size.
Sun’s problems are a hell of alot more complex than just a matter of firing a few people, bobs ya uncle, and things get back on track in an instance. When the whole business model has been changed, its going to take time, there will be a transition, but I think for the long term, profitability under the new model is going to be smoother, the margins are going to be better, and ultimately, you won’t see the dramatic swings in either extreme in terms of profitability/loss.
The problem is that the expensive boutique bear, that now have gone cheap is not ready for consumption just yet.
At least not for the big parties. ZFS and zones are nice, but even though already open sourced by Sun, software for clustering is, as far as I know, not yet integrated into the latest OpenSolaris release, and ZFS still doesn’t support concurrent access.
If you are going to do things like that, you are better off using Red Hat or CentOS using GFS, and on the Desktop Solaris is still Gnome 2.20, so you will miss the latest and greatest.
Once ZFS gets support for concurrent acccess, it will be great for datacenters, especially combined with its ability to easy export as iscsi devices.
Anyway, the new OpenSolaris feels snappy, and looks very good as a Desktop, even though it lacks the latest and greatest from Gnome. It is the first Solaris version in a very long time that I actually could consider for my desktop.
The performance on my laptop (Thinkpad t61p w/ Nvidia Quadro graphics card) works wonderfully; performance on my Dell Dimension 8400 is pretty spotty – basically the graphics card sucks (ATI X300) – I’m tempted to replace it with a low end Nvidia card some time in the future – I saw an elcheapo $80 one, I might get that.
Not really. There’s a sense of bemusement that Sun is trying to take on the one hand and not give anything back on the other. The license incompatibility is one thing, and the pseudo-MySQL development model that no one wants to talk about is the other. The rest of the open source community, and Linux, has been through all of this stuff.
Yer, and those resources have to come from somewhere. Sun don’t have them, otherwise they would have put them in by now. What they need is a community of kernel developers writing lots of drivers and people testing their software on Solaris systems along with Linux equally. In order to keep those developers, you have to give something back to them in terms of control and autonomy. That ship has sailed.
Can you please lay off the Anti-Sun FUD? Sun just completely open sourced MySQL.
Please share with us some examples of where nothing was given back. You can download the source from the website and people have come up with complete distributions like Nexenta and Bellenix from those sources.
Please don’t point to that one thread by Roy Fielding where you completely misread and misunderstood the issues that were being discussed.
On the other thread I asked you to provide examples to substantiate your claims but your disappeared and came here to spread more FUD. Now put up or shut up.
Edited 2008-05-07 16:18 UTC
If I may make a comment here… I have not followed this thread, but I assume that the objection is to the joint copyright policy on some or all of Sun’s projects. To that I would say (and referring specifically to MySQL here) that while GPL with joint copyright requirement is qualitatively a different thing than pure GPL, and it is useful to make a distinction between the two, the distinction is mainly relevant to contributors to the project and not to users. If it somehow results in a difference in volume of contributions to the project, that would only be relevant to users indirectly. The joint copyright agreement is “in your face” enough that the developer can hardly accidentally contribute without realizing what he is doing. All else being equal, my aesthetic preference is for a project not to require joint copyright. But it is really a decision which is up to the individual developers to evaluate.
Edit: And, of course, joint copyright on MySQL contributions is nothing new; MySQL AB did that, too. Also, I seem to recall that Sun’s agreement is or was not exactly joint copyright. I don’t know that it makes any fundamental difference.
Edited 2008-05-07 16:47 UTC
I was merely pointing out to the OP that just making blanket unqualified statements and blaming Sun is disingenuous.
OpenOffice.org also has a joint copyright requirement and that hasn’t stopped developers. Somehow the linux zealots never mention one of the only serious and successful Office suites on linux distros is a product Sun open sourced and manages.
Sun can effectively manage OpenOffice.org with nary a peep from the linux and GPL zealots but OpenSolaris and MySQL are somehow different, right?
If you have something relevant to what was said, be my guest.
MySQL was already open sourced sweetheart. What they’re not admitting to everyone is that they’re following a pseudo-MySQL development model of in-house development with some possible open sourcing, and they’re simply not admitting it to the people who are involved. That’s what the Roy Fielding thread was all about.
If they’re going to do the MySQL thing, then fair enough, that’s OK. But they’re trying to convince an awful lot of people that isn’t what’s happening.
I’m afraid I misunderstood nothing, no amount of jumping up and down on the spot will change that, and you proved beyond any doubt that you simply hadn’t read what was going on and didn’t want to face it – for whatever reason.
Project Indiana is now OpenSolaris (the trademark part of the issue), no one else is allowed to call their distro OpenSolaris, or even to use the name, and Indiana is an internal project headed by Ian Murdoch where code is thrown over the wall. Some people on that thread even talked about forking the code and calling it OpenUNIX.
That’s hardly the open development project model, where you would see Sun and other people committing code in a public repository, that was promised loudly by Sun’s executives and is continually being promised to everyone by various Sun employees. It’s been three and a half years now.
There’s little point in replying to any of that now (dead horses, flogging and all that) because you’re just desperate to deny what’s actually happening for whatever reason, and the points that I needed to make, I made. You can ask for examples until the cows come home, but they won’t change a thing about what I described. If you’re upset about them I suggest joining a gym or something.
Not it wasn’t!
http://blogs.mysql.com/kaj/
“The change comes from MySQL now being part of Sun Microsystems. Our initial plans were made for a company considering an IPO, but made less sense in the context of Sun, a large company with a whole family of complementary open source software and hardware products.”
No that thread was not about that and we have beaten that to death. You go on living is lala land.
Citation please.
You can close your ears and repeat that ad nauseum but i am still waiting for some hard evidence and all you can point me to is one misunderstood thread.
BTW I pointed you to Con Kolivas’ case in the linux development so linux is no different.
I have explained this to you already. No one is allowed to call any distribution OpenSuSe either because Novell owns the trademark for SUSE and OpenSUSE. Same with RedHat. Same with linux. Try forking the linux kernel and calling it some derivative of linux and see how Linus comes after you.
So? OpenSolaris is Sun’s distribution… That doesn’t change the fact that it is completely open any more than Fedora being owned by Redhat or OpenSUSE by Novell.
You dodged the OpenOffice.org question again! Classic troll tactic.
Cut the crap! It is no more closed than OpenSUSE or Fedora.
Really! So if you can’t respond you just put your tail between your legs and run!
You have proven nothing and once again we are back to the same place with you.
Your points are pretty merit less yet you make the same points on every Sun related thread. The bottom line is you are a troll nothing more nothing less.
If we could somehow convert the efforts expended bickering in threads like this one into OSS check ins, both Linux and OpenSolaris would be of Star Trek OS caliber and Microsoft would be history.
I really don’t see the point in all the negativity and conflict when there are so many constructive and positive places to put one’s efforts. In the end it is what get’s done that counts, and not which “side” is holiest.
Edited 2008-05-07 23:00 UTC
That will only happen if we ban sedgunum as a troll.
Go look at his comment history and you will find just how deep his anti-Sun trolling goes.
It takes two to tango. Sometimes it is most constructive, or at least less destructive, to simply agree to disagree. It’s obvious neither one of you is going to convince the other.
Edited 2008-05-07 23:06 UTC
Yeah the tango was fun but also a total waste of time. I concur.
You know GNU/Linux is not UNIX but one of it’s company own UNIX … Open Solaris advocate are copying the BSD tactics of lying about being the same as GNU/Linux and blaming all it’s problem on GNU/Linux instead of doing the work.
I higlighted the reality in your comment and offered copy paste of links , that are in the article that you claim don’t exist.
” it claims there is no opensource development community around it and yet provides no evidence to substantiate such claims.”
Linus gets more contributions while brushing his teeth than Open Solaris gets in a week. :
http://www.ratliff.net/blog/index.php/2008/02/14/not-with-a-bang-bu…
http://thunk.org/tytso/blog/2008/04/19/what-sun-was-trying-to-do-wi…
“So if you run into a Sun salescritter or a Sun CEO claiming that OpenSolaris is just like Linux, it’s not. Fundamentally, Open Solaris has been released under a Open Source license, but it is not an Open Source development community. Maybe it will be someday, as some Sun executives have claimed, but it’s definitely not a priority by Sun; if it was, it would have been done before now. And why not? After all, they are getting all of the marketing benefit of claiming that Solaris is “just like Linux”, without having to deal with any of the messy costs of working with an outside community. As a tactical measure, astroturfing is certainly a valid marketing trick. But after three years, the excuse of “just you wait a little longer, we’re just trying to figure this open source community stuff out”, is starting to wear a little thin.
Furthermore, if (as John Ploncher claims) this was about “empowering application programmers”, why was it that Sun’s first act was to trumpet how wonderful it was to release the Solaris source code under a Open Source license? This only seems to make sense if the Open Solaris initiative was really a cynical marketing tactic to try to save Solaris from being viewed as irrelevant. If that was Sun’s intention, I think it is fair to say that from a marketing point of view, the tactic has been at least partially successful – although as John has admitted, the goal of creating a full community with application developers, university students, and so on, hasn’t materialized for Open Solaris. Sun has the dream, the Linux community is living it.
However, from business standpoint, I wonder if Sun will really be able to sustain their Solaris engineering team if they will really be doing all of the work themselves, and outside contributions continue at the rate of 0.6 patches per day. After all, the margins when you are selling low-cost AMD servers are much lower than when you are selling ~A 1/4 ber-expensive SPARC servers. With Linux, we have a major advantage in that kernel improvements are coming from multiple companies in the ecosystem, instead of being paid for by a single company. And given that 70-80% of Sun’s AMD servers are running Linux, not Solaris, it’s not clear how Sun justifies their Solaris engineering costs to their shareholders. Furthermore, if Solaris on x86_64 were to actually take off, there’s nothing to stop competitors from selling Solaris support -except the competitors won’t have to pay the engineering costs to maintain and improve Solaris, so they would be able to provide the support much more cheaply than Sun could. So while Sun’s marketing tactics have kept Solaris alive in some verticals, I have to question how successful Sun will be in the long-term.”
Edited 2008-05-07 13:06 UTC
Dear God that is f–king pathetic; it truly is; you’ve gone from ‘no development community’ to ‘has less than Linux’ in terms of development community; wow, what a change, what next? if we’re going to use YOUR benchmark for it, we might as well claim that OpenOffice.org, Netbeans, Eclipse, KDE, GNOME and so forth are all monumental failures when compared to the almighty Linus “receives more contributions doing his teeth” Tovalds and his creation – Linux.
No, this is truly pathetic; when you move the goal posts, you make claims, and yet never substantiate them. Christ, is this the new trend now; instead of “*BSD is dying” we’re going to see pathetic gutter snipe such as you coming out of your troll bridges to beat you chest and spam forums with “opensolaris dying’ based on nothing more than what you ‘feel’. Pathetic, truly pathetic.
Edited 2008-05-07 13:03 UTC
No , you are the pathetic one here , leave GOD out of your lies and insanity too , try to follow , I copy pasted the linked text from a third individual that the writer of the article link to in it’s article to support is fact that there is no Open Solaris community. That you claimed don’t exist.
I don’t have any bechnmark , Those are your own creations and own words , and I don’t need to point out that I don’t support them or you in this case.
No , because WE don’t work in claims or make any claims we leave that to you. Those are again your own words and claims.
I don’t make claims and I proved that you where lying.
Insulting GOD sons name in vain too , leave that out too , I never said BSD is Dying , here , I said Open Solaris stupid advocate are copying the same failing attitudes.
You know , gutter snipe , spammers and trolls are removed from the site , It’s not because you call me one , because your definition is really anyone who disagree with me and prove me as wrong and that make me angry , that I am identified as one.
Yes , you are.
Edited 2008-05-07 13:50 UTC
You need to calm down a little, because you’re missing the essential truth:
1. Over the past eight or so years Linux has hurt Sun and Solaris, badly, and it has changed the rules of the game in how an OS overall is developed. If you’re hoping to maintain a large, proprietary kernel and OS by yourself, you want to port to lots of new and varied hardware and maintain lots of drivers, and you’re not Microsoft, you have no chance. Novell discovered that with Netware, and even that is a truly painful road.
2. Solaris needs more development investment and movement, particularly in the form of drivers.
3. Sun don’t have the resources to do that as Linux does, with various individuals and lots of companies being involved. They really do need a viable cost and effort sharing open source model for Solaris development to continue to be viable. When 70% – 80% of Sun’s x86 server customers are specifying Linux to be installed, that is a pretty difficult cost to justify.
4. In order to get external contributors and keep them you have to give something back to them, and that means letting go of Solaris altogether – ethics, trademarks, code and all and doing your development in a public repository.
5. A truly open source Solaris would mean that lots of companies could provide cheaper Solaris support than what Sun provide. Sun desperately don’t want that to happen, but they will have to grin and bare it if they want Solaris to have a financially viable future.
6. Rather than create that viable open source cost, code and resource sharing community, Sun seem to think they can quietly create a MySQL type open source model, keeping fairly tight control, but pretending it’s just like Linux really for customer impression, and hope that this whole Linux thing just goes away. Eventually.
7. If you’re getting 0.6 patches per day, and they’re taking ages to then actually be merged, you don’t have a community. Of any kind.
The above is what Theodore Ts’o’s blog entry is about.
There are plenty of people within Sun who still passionately believe that Linux is just an upgrade path to a proper Unix OS in Solaris, and that attitude has even been displayed around here. They also still believe that the commodity hardware, x86 growth that Linux rode off is just an aberration before people come back to using real SPARC hardware. They still haven’t learnt anything about why Linus wanted a Unix-like system to run on his 386.
It’s all very sad really, because they have everything they need to change and move on. Their current path will destroy them eventually, unless their customers continue to buy x86 and Linux from them.
I am calm (I am calm, I’m just passionate about certain things – to confuse the two says more about the reader than what is being spoken), the issue I have is hyperbole created by people on this forum who seem to create benchmarks for success on the fly.
I think that Linux is wildly successful – but that is based on a benchmark I created. To say that OpenSolaris is a failure, it has to compared to something, and for someones own subjective reasons. You might consider it a failure, but for Sun – their expectations and what has been achieved, it might be considered a great success.
I don’t doubt that. Linux is ultimately what has forced many big name UNIX vendors to start justifying what they’re charging; when a machine can do 95% of what a RISC UNIX machine can do – at half the price, the said company has to really justify the pricing – their whole business model in the case of Sun Microsystems.
Sun has a relationship with Intel, I just hope that the relationship with Intel will spur other companies on – hence the reason I purchased this machine (Thinkpad) 100% intel with an nvidia graphics card – all supported perfectly.
No doubt about that, but a the same time, I don’t want to see garbage half finished drivers being merged with the promise to ‘complete it later’. I’d sooner see finished and stable drivers merged into Solaris – and maintain the quality than seeing the half finished stuff that I seen enter the Linux kernel with a tag – “to be completed at a later date”.
I think what they *need* to do is find out *why* 70 – 80% of their customers want Linux – instead of simply throwing in the towel, they need to ask and address the issues; it might be as something as simple as offering a free cross grade for a given application. A given application might not be available which might require Sun to invest some cash into the given company producing the given application.
I’ve actually flagged the idea of a bounty project; I’m quite happy to throw $300 of my own money into it – I’m a student, I have limited resources, but I’m happy to put my money where my mouth is and start up a project to pay for the unsexy things that need to be done – to get done.
To bad I was ‘boo’ed’ off the irc and mailing lists when I said such heresy.
I don’t think that is the problem – the problem is simply, their marketing sucks. I’ve said it once and I’ll say it again, the current marketing manager needs to be tarred and feathered for the terrible job he has done.
What I also think they need to do is do something radical; announce their OpenSolaris 11 commercial product will be based 100% on opensource code; setup some prizes and generate some buzz – then again, it goes back to marketing. Something Sun is incredibly crap at.
It would be interesting to see on what basis it is taking so long to get things merged. They have a high code quality, but at the same time, I do think they need to open the cvs tree to more access by non-Sun employees. Get things moving along quicker and the contributors seeing that their contribution is being merged promptly – and as a community, we all benefiting from these improvements.
That’s exactly what has happened over the past eight years or so. x86 machines just have more raw processing power than RISC and SPARC machines as well, so it’s a tough sell to many people. Heck, many of Sun’s x86 machines compete with SPARC and they’re doing pretty well without too much prompting from Sun. There’s still a SPARC market out there, but what’s left is disappearing, and if you can’t put your Unix hardware right at the high-end (Power, Alpha, Itanic etc.) then it’s very, very tough.
Linux has drivers for devices and runs on hardware that Solaris and other systems just cannot. Sun would absolutely love to get their hands on Linux’s drivers.
The problem is, Solaris just doesn’t have the wide variety of drivers Linux does, and it doesn’t even have stuff being worked on to be completed at a later date.
Linux runs on all those little cheap ARM NAS boxes (I have three running Bacula), it has drivers for TV cards that form the basis of lots of PVRs, it has run well on SPARC for well over ten years without any help from SPARC’s major seller…… It’s actually shocking to think how far hardware support has come with Linux and how good most of it is.
I would hope they would throw in the towel slightly, do more sharing and then differentiate themselves with something better. The opportunities are all there for Sun, because their good stuff is very good (ALOM etc). It’s not inconceivable that if they got really rolling they could put Dell out of business. I’m not joking.
Not a bad idea. We’ve had them before with various degrees of success, but come on Sun, surely you can have a summer of code or something and get some code flowing? They desperately need a support system for Solaris in an open source form.
I have no problem with them doing the MySQL style open source method, neither would anyone else if they admitted it and they would still benefit a great deal. It’s the confusion that’s the problem.
From what I have been able to gather, Project Indiana is Sun’s internal OpenSolaris, of which code is then submitted publicly, and there just aren’t the mechanisms for outside patches to be reviewed and accepted. In effect, the public repository you see is not where the main action is happening and where Sun’s patches are going.
Again with the same bullshit. The Niagara based systems are making mince meat out of x86 cpus and Power in webservice related workloads that need to scale in throughput.
Itanic and Alpha are dead… Power and SPARC are the only game in town and SPARC is doing pretty well in terms of price/perfomance and Performance/Watt.
Intel still can’t do more than a 4 way Xeon effectively.
You know this how? Just from this thread we have seen that the graphics driver and wireless drivers are superior to linux. Its the quality that counts.
All of this is relevant how, again? How does ARM linux help linux on the desktop. It doesn’t. OpenSolaris is a desktop distro for x86 boxes. Why is esoteric hardware support in linux relevant to this thread?
Again like I said you are troll and all you can do is bring irrelevant points into a discussion thread.
Sun not sharing are you kidding. Sun is the only company to have completely opensourced their CPU RTL.
OpenSparc.net
No other company can claim that to have opensourced their own IP as much as Sun. IBM is closed and proprietary. So is HP.
Can you prove that there is no code flowing?
Backpedaling again?
Really? Can you provide proof or are you pull stuff out of your nether regions again?
Edited 2008-05-07 22:49 UTC
If you have a problem with the sound, just install OSS and you’ll even get a driver for Creative Labs X-FI
http://www.opensound.com/
It’s that easy.
ROFL. Simeon is what you get when you only have hear-say. Next time do some research.
LOL.
Perhaps they meant Ximian…..
Edited 2008-05-07 09:46 UTC
You guys. Why all this quarreling? You sound like some girls or something. (No disrepect to girls, but Ive heard some arguing in the bus just recently). Why dont we all be happy? Dont worry, be happy. OpenSolaris can only make Linux better. And Linux makes OpenSolaris better. Everyone benefits, even Windows. This competition is good. Or is it not? Lets sing “we shall overcome…” :o)
OpenSolaris, congrats for the release.
Didn’t RTFA, but from the various comments…
Try checking the SCOG v. Novell (er…Novell v. SCOG) trial transcripts over on Groklaw.Net and the various comments there. This isn’t a matter of FUD for OpenSolaris – it’s a reality that Sun has to face at the moment: the reality that Novell is contesting SCOG’s ability to have even entered into any kind of agreement with Sun of the nature that their agreement was, and that agreement looks like (from trial testimony) what Sun relied upon to open source Solaris – OpenSolaris.
So, then if SCOG didn’t have the legal standing to enter into that agreement with Sun, then does Sun have the ability to continue providing OpenSolaris? Not likely, unless Novell ratify’s the agreement as is.
So what’s the issue? The issue is that (a) Novell has not yet come out and said it would ratify the agreements, and (b) from the trial transcripts Novell seems to be hinting that it won’t ratify them as is but wants some modifications – likely just more money than what was paid to SCOG.
Now, I’m not partisan to any of the above in any manner – I’ve just been keeping track of the SCOG saga via Groklaw, and have really enjoyed keeping track of the trial last week through Groklaw as well. The above is just my impression from what I’ve read on Groklaw, and by no way tells what Novell will or is planning on doing – so take it with a grain of salt.
However, to say that it is simply FUD is wrong.
My final take: Novell is trying to hint to Sun that it wants changes to the agreement so get ready to see Novell and Sun sit down and talk about OpenSolaris; but I don’t think Novell will ultimately prohibit Sun from making/distributing OpenSolaris. More likely than not, Sun will just pay Novell more money and continue on its merry way – likely with some kind of re-occurring royalty stream going to Novell. In other words, Novell just wants a piece of the pie from any revenue OpenSolaris may generate, which I think makes business sense (that is the context that the comments were made in in the trail too – what would make business sense for Novell? – more on that below).
So, no – I don’t see OpenSolaris going away any time soon; just more money to Novell.
Again – I have not insights into what Novell is actually thinking/planning. This is just my take on the entire scenario brought up from the SCOG v. Novell litigation and last week’s trial. It’s also something that should be seriously considered.
Okay – so before I end this and go on to other reading: The context in the trial was about what would have made business sense for Novell. They have a business unit (SuSE) that sells Linux, so to have a competitor wouldn’t make business sense to allow the competitor when that competitor has to get a license from them to be a competitor, or rather to allow the competitor to exist without having some kind of royalty stream coming back to them since the competitor has to license…yeah a little confusing but the reality of it all. Thus, I think Novell really just wants Sun to be providing a non-ending Royalty stream for OpenSolaris back to them to justify letting Sun be a competitor to Novell’s SuSE unit – thus Novell wins regardless of who does better, and that just makes business sense for Novell. Sun would just out a little more money, but not likely enough to hurt OpenSolaris. So in other words, it would remain in Novell’s interest to continue allowing OpenSolaris be out there but only so long as they get the chunk of the pie they are owed.
Hopefully I didn’t lose anyone in that…
Actually, that’s wrong. Sun secured a majority of their rights from AT&T and Novell a long time ago. They beyond any question own SVR4. The only thing they licensed from SCO were some drivers and very few at that.
Read here for more:
http://mail.opensolaris.org/pipermail/opensolaris-discuss/2007-Augu…
So yes, it is fud, it is silly, and speculation is best left to the trolls.
Please quote a Sun source on that one. I’m not familiar with the person writing that particular e-mail.
My quotes go back to Novell’s lawyers. Please find an equally authoritative source before calling FUD.
In the end, Novell is going to taking Sun to task on what they got from SCO, and reviewing it. Now, it may be there is no issue, but from the looks of the Novell vs. SCO trial it looks like an issue for Sun and OpenSolaris. They also have the SCOG/Sun agreement in hand and are probably a lot more familiar with how all the licensing went than the author of the quoted e-mail.
Sorry, but please find a source that is at least equally authoritative.
He’s Joerg Schilling. Who’s been around longer than most versions of BSD, UNIX, etc. Not only that, he was part of the pilot program for the OpenSolaris project.
Also, here’s a Sun source if you want some comments:
http://blogs.sun.com/blu/entry/supreme_irony_sco_lost_the
Note that they state they were securing “rights to the remaining pieces of UNIX that we had not already secured”.
The judge also did not invalidate the rights that third parties secured through SCO itself.
In addition, Sun was one of the co-authors of UNIX with AT&T before Novell bought any rights at all way back in 1992:
http://wiki.linuxquestions.org/wiki/History_of_unix
Since Sun secured rights from AT&T, from their own authorship in UNIX, from Novell, and from SCO, I don’t see how any sane person could speculate against them.
Finally, Novell stated they had no interest in suing people over UNIX and that they’re not in the UNIX business anymore:
http://www.pcworld.com/article/id,135959-c,unix/article.html
It is FUD unless you can prove it. Trying to promote fear, uncertainty, and doubt about a product without being able to prove it via means other than a lot of interpretation, hand-waving, etc. is FUD. Pure and simple.
Edited 2008-05-08 03:39 UTC
I have a feeling that those unix copyrights are going to continue to bring trouble, if only for their FUD value, as long as they are held by anyone. I really think it would be best for all concerned if they journeyed to Mordor, climbed to the top of a volcano, and threw them in.
Edited 2008-05-08 03:51 UTC
Amen! Seriously.
Okay, thanks for the info on Schilling.
Please quote one that is actually attributable to an author, and someone that actually has some standing. I can’t find on there who the author is, thus I can’t determine whether or not it has more standing than Novell’s lawyers.
That said, it also seems to get the story quite wrong – as SCO/SCOG did not purchase the “rights to UNIX”. They purchased the right to collect money on behalf of Novell. (Check Groklaw, and check the courts. That’s basically what the APA gave SCO the right to.) So, please get a source that can actually get the story right at even the most basic level to start with.
That is one possibility, yes. But then they would have had to do due diligence to prove that SCOG actually had the right to license what they had. And that seems to been skipped. Of course, as Groklaw’s PJ put it – since the Sun/SCO deal was done, Sun’s management changed and Sun doesn’t seem to be coming to help SCO at all either.
Uh…it wasn’t for the judge to invalidate. The judge can only invalidate a contract if the contract is some how breaks the law, or the parties entered into it in a way that breaks the law or could show there was no real meeting of the minds. Neither was the case with the APA or its amendments; and I doubt (other than SCOGs lack of standing to enter into the agreements) that the SCOsource licenses would fail there either. Indeed – the law is such that if one enters into a contract without standing, then the person who has standing must decide whether to grant (ratify) to deny (revoke) the contract. In this case, SCOG entered into the contracts without standing, and Novell is the one who has standing and thus must decide.
Furthermore, the judge has only ruled thus far that Novell, not SCOG, owns UNIX. Which puts the ball in Novell’s court to decide first what to do about the contracts with Sun, Microsoft, and all the other SCOSource folks. Yes, the judge can say that SCOG didn’t have standing to enter into the contracts, but it is still left to Novell for what to do about them. If someone other than SCOG then contests Novell’s decision, it then goes back to court to decide that. Until then, it’s in Novell’s realm of things to decide. That’s what the APA gave them the right to do.
As does Novell, IBM, and Microsoft. However, they still have to license parts from AT&T/Novell – SYSV/SRVX. Just because they have rights to SVR4 doesn’t give them rights to SVR5.
Also, this isn’t a matter of speculating against Sun. Certainly they have a right to most of the Solaris code-base – just as IBM does with AIX. But (again) looking at the Novell vs. Sun trial transcripts, it seems that Sun may have relied upon SCOG to provide something SCOG couldn’t provide – something that only Novell could authorize – and Novell is saying they’re not so sure about it.
Oh – and I bet you’d love this link:
http://www.groklaw.net/article.php?story=20080506120544368
I have referenced, nor said anything saying they wanted to sue. Only that they may want Sun to sit down and renegotiate the deal. I’ve even said it would (in my opinion) likely go through, just with more money going to Novell.
I still ask you to call an equally authoritative source to counter. And if you noticed – in everything I’ve said that this is my interpretation.
FUD, btw, is saying things that cannot be backed up and referencing it as fact not ones interpretation or ones opinion. And (again) for the record – I’ll continue to state that I have no insight into Novell. This is just what Novell has said – via its own attorney’s and the testimony of SCOG executives in response to questions from those attorneys in court, under oath.
Again, please go to Groklaw and read the transcripts. I’m sure you’ll reach the same conclusions I have.
Oh – and per quoting Sun – I’ll wait until I hear something from Jonathan Schwartz, their attorney’s, or an official press release from Sun – hey, I’ll even accept it from Schwartz’s own blog (http://blogs.sun.com/jonathan/). I think that would be about the only thing from Sun that could counter the court transcripts – don’t you? Or something from Novell.
Finally – just to note – I really do like the fact that OpenSolaris is around, and I want it to stay that way. But denial is not going to get Sun out of this either. All I am saying is that this is a real possibility because of how backly SCOG has messed things up. But the fact of the matter is that there are only two groups that can really say if there is an issue: Sun and Novell. So until one of them speaks further we can neither deny nor confirm whether it is an issue beyond what I’ve said and referenced.
For your reference:
Trail Day 1:
http://www.groklaw.net/article.php?story=20080429182910157
Trial Day 2:
http://www.groklaw.net/article.php?story=20080430193035760
And you can find the transcripts here:
http://www.groklaw.net/article.php?story=20080501194602283
The PDFs provided are of the official documents from the court. While the quotes above are eye-witness summaries, you’ll still find it all in the transcripts.
….blah blah blah. Seriously.
The point is this:
* The court is treating SCO as a legal reseller for the purposes of third parties which is what allowed Novell to collect funds collected on “their behalf”
* Sun and Novell already have business agreements, licensing, parternships, etc.
* Sun has a massive patent portfolio to protect itself
* Sun acted in clear, good faith to secure rights
While your research is appreciated, it is still interpretive and selective and remains FUD since its purpose is to create fear, uncertainty, and doubt.
Edited 2008-05-08 04:44 UTC
Patents won’t protect Sun on this one – patent portfolios only protect against patent attacks, not bad contracts or copyright infringement or anything else, just other patent attacks; but again – everything you said backs up my thought that Sun and Novell will reach an agreement that leaves OpenSolaris as it is but in a way that Novell finds makes business sense for Novell. (Why would they do anything else?!)
BTW – what I am saying is not FUD because of (a) how I am stating it, (b) my disclaimers, and (c) the very fact that I am not doing so for your stated FUD purpose – I am writing what I am writing b/c it is a highly probably possibility based on what is on-record in an authoritative source.
You call my quotes selective – show me how they are out of context or other materials that invalidate what I am saying? Please do. I beg you. I’ll retract what I’ve said if you can show it – even edit out the old posts if you so desire. But please use an authoritative source to prove it – such as the transcripts I referenced. That’s all I ask – nothing more, nothing less – and I’ll cede you victory on the issue.
Patents won’t protect Sun on this one – patent portfolios only protect against patent attacks, not bad contracts or copyright infringement or anything else, just other patent attacks; but again – everything you said backs up my thought that Sun and Novell will reach an agreement that leaves OpenSolaris as it is but in a way that Novell finds makes business sense for Novell. (Why would they do anything else?!) [/q]
Assuming that Sun even *needs* an agreement. Which is a big assumption.
Furthermore, patent portfolios do matter since if Novell decided to be nasty, I’m certain that Sun has more than a few patents that Novell might want to consider licensing
No, it really is FUD. I can see no purpose to your writing other than creating a negative view of the matter. Instead of just stating a simple thing like, “I wonder if Sun needs to secure further rights from Novell” you have to create your own legal theory based on the situation. That’s the problem I have.
There is no victory here. For there is no war to win — Sun already did
I wouldn’t normally reply to my own post, but ran out of space and wanted to draw special attention to the following quote (also available in the parent):
So, even though Sun had all the rights to Solaris, they didn’t necessarily have the right to distribute it the way OpenSolaris is distributed, and they may have needed to acquire such a right. SCOG conveyed they had the ability to give it to Sun – even though they didn’t – and that is what this whole issue is about.
Again, until Novell or Sun comes out and says otherwise, the issue will be there for OpenSolaris. Of course, I fully expect (and again will state I have no knowledge beyond what I quoted in the parent) that Sun and Novell will sit down and reach an agreement that will leave OpenSolaris in tact as it is now.
I’d really like to be able to run Solaris instead of Linux for sake of the classic pros like ZFS, Zones and DTrace, but I’d also lie it to take advantage of the open souce goodies that are lying around there, ready to be picked up and used in respect to any available open source unix-like OS, be that Open Solaris, Linux or BSD …
I just don’t see what makes Linux users make such a big deal out of the fact that some people choose otherwise, like e.g. *BSD or Open Solaris, when all they tend to be able to base their distrust on is the fact that ‘Linux does the same things that *BSD or Open Solaris does’ with the occational ‘we don’t need ZFS at all and SystemTrace basically does the same job as DTrace’ attitudes …
Why not just let people make their choices – after all, open source is all about free choice and ditto access to software. It’s all about what makes the user able to put together his or hers favorite set of software offerings and giving that person freedom to customize everything down to the last itty bitty detail….
So be joyful be happy, that Open Solaris, *BSD and Linux are all here to make the concept of freedom of development and deployment shine in all its glory!
Only a moron will make a claim that Solaris does not have a community. If you look at how Solaris was developed in the past and compare to how it is developed today, you will be shocked at the differences. Just look at what Solaris is turning into today. Xorg is used for their Xwindow system. Gnome for their desktop environment… and it is no longer even heavily modified. And let’s not forget Bash as their default shell. Just about everything in Solaris but the kernel is now being developed by the community. What does this do for Sun? It lets them focus on what is truly important while drastically driving down costs. It lets the kernel be the only thing differentiating it and Linux and IMHO, with DTrace, ZFS, etc., Sun is in a good spot for that comparison.
My impression is that OpenSolaris has a different kind of community than Linux. Solaris has traditionally been a different kind of OS than Linux. As such, one should expect that a Sun led community effort around Solaris would have both similarities and differences with the community around Linux.
I would advise that OpenSolaris advocates not take these things too seriously, out of respect for their own blood pressure and general mental health.
As OpenSolaris expands into new territory, as it obviously is, judging from my experiences with this live CD, it is going to spend some time in the limelight. As a long time advocate of Linux, let me tell you something about what it is like to be in the limelight. Along with all the roses being thrown your way, you are also a perfect target for the rotten vegetables. And the best thing to do is simply to resign ones self to that fact, accept the roses gracefully, wipe off the tomato pumice, smile, take another bow, and get back to work.
An endless stream of rebuttals, and rebuttals to rebuttals will only get you a reputation for being thin skinned. Letting these articles get to you… taking them too seriously… is a recipe for strife. I should know.
It will be interesting to see how OpenSolaris’ community evolves. It will be interesting to see how Sun’s community-building technique evolves.
Edited 2008-05-07 20:29 UTC
Hi All,
I installed OpenSolaris 200805 the other night and here are my opinions for what they are worth (since many people here may not have tried it yet):
Pros:
* As an earlier poster mentioned, the desktop is *very* responsive. Enough to notice, and I don’t usually care about such things. Better than Ubuntu, and it seems better than XP (which is saying something). I’m guessing it is the the Solaris kernel (which can be used for true real-time stuff), the Xsun display server, and the proprietary Nvidia driver (although Xorg is available as well). It installed all the Compiz goodness without tweaking anything. Nice.
* OpenSolaris is now far more ‘friendly’ to Linux users than it once was. The Gnome user interface will immediately familiar to Ubuntu users (and others).
* The default GTK+ theme seems to be based on Nimbus (which is a very recent vector-based theme for Java’s Swing). Looks much better than clunky CDE or Motif style themes.
* You don’t have to let Solaris do the boot-loading, which is a problem I found with Solaris Developer. So OpenSolaris cohabits nicely with Linux and Windows.
* Barring a single mouse issue, all my other hardware worked flawlessly. This is on a four month old custom built desktop (with new/recent hardware). Device support is getting better (although not perfect, see below).
* It’s Solaris (and associated goodies such as ZFS, Dtrace). Which, in my experience, means the kernel will never crash unless there is a hardware failure (such as dead RAM).
* It has a LiveCD, which is better than Solaris has had in the past.
* A package manager in the vein of Synaptic (although not quite as good in my opinion).
Cons:
* Didn’t detect a Microsoft wireless mouse (although the wireless keyboard that came with it was found). Tried another microsoft mouse and it worked instantly. So the device support isn’t as extensive as Linux, but it is better than what Solaris has in the past.
* The installer died the first time round as it couldn’t change a spare partition marked as type Linux to Solaris. A quick reboot to Ubuntu, use fdisk to mark the partition as Solaris, restart the OpenSolaris LiveCD and the install worked. A show-stopper for ‘Aunt Nelly’, but pretty minor for anyone with a modicum of sysadmin experience.
* The package manager didn’t easily show which packages were installed (unlike Synaptic which makes it blindingly obvious).
* Using the package manager categories didn’t show packages that were available. I had to manually search for “gcc” to find it. Likewise for java and netbeans. Although, the nice thing is, they are available without having to recompile anything.
* The packages don’t seem to be very highly compressed (if at all, although perhaps it is done invisibly by cpio). Downloading things seemed to take a fair bit longer compared to getting things on Ubuntu (even though the bits-per-second download rate was similar).
* Politics. Ubuntu will probably always be around in some form. Who knows before corporate policy changes and OpenSolaris is no longer of interest (sure it’s open sourced so can remain, but without Sun pushing it then progress will be glacial).
Conclusions:
OpenSolaris improves on Solaris a lot, making it an easy fit for those familiar with Linux (Ubuntu specifically). It would definitely make sense on the corporate desktop, especially for companies nervous about a product that is maintained entirely through community support (although most of us know this is ridiculous). I’m sure the Solaris pedigree will make deployments run rock solid over extended periods.
The OpenSolaris team have done a great job, although probably not enough to replace long-time Linux user’s desktops, yet. Give it a year or two or improvements at the current rate and OpenSolaris could be a real contender.
The Linux community should benefit from the cross-pollination from an added Linux-friendly Unix. Lets face it, Sun knows engineering (it’s just a shame they often get the same engineers to do design).
Audio- and video-philes might have an interest, since Solaris has better real-time responsiveness than Linux (since though Solaris’ evolution it also targetted embedded devices).
The bulk of Windows users will still hate it as much as Linux since they still can’t find anything as the file browser still doesn’t list a “C: drive” and it doesn’t work exactly “the same as my PC” – that can’t be helped I’m afraid.
If you like playing around with operating systems (you are reading OS-news aren’t you) and have an open mind, then an hour downloading and installing OpenSolaris wouldn’t be wasted if you are curious.
I’m not going to pick up a quote for it, since it’s pages ago, but there’s enough “Linux” applications that boast with portability, POSIX or even go as far as claiming to compile on Solaris often simply don’t. There’s always some fiddling required. I’ve seen it too often. Apart from the kernel and libc, the recently released OpenSolaris distro ships with everything you’ll find in a default Linux environment, including headers and libraries. Yet things still fail to compile. It’s one thing if an application relies on specific kernel APIs that aren’t shared across the different *nix variants, but standard userland stuff shouldn’t fail if it claims portability.
It’s not portable if it requires patches. Being able to compile in different Linux distros isn’t portability.
Oh hey, segedunum is also again hard at work talking crap. Sigh.
….
Edited 2008-05-08 00:01 UTC
I believe that only applies to the binary release. You cannot mess with the binary cd and redistribute it. The source, which is what matters, is under CDDL. At least that is my understanding. If there are binary drivers or whatever on the CD for which source is not available, well… those are closed. Like Nvidia and fglrx under Linux.
This would be a major break for OpenSolaris and shake the Linux world.
And how would that help OpenSolaris? drivers? well, they can easily get that from OpenBSD (and other *BSD’s for that matter). Drivers is about the only thing I can think of which is holding it back.
One thing I would love to see, however, is opensourcing the compiler stack – now that would be cool; a complete end to end opensource operating system and development tools.
I need a good NX client.
But as to drivers, I would think that the driver infrustructures of the two kernels would differ so much that there would be little to be gained by direct sharing of code. And developers can already look to the Linux drivers for information on how the device works, and how to talk to it, as long as they don’t copy code.
I would like to see a GPL2/3 OpenSolaris. But I’m not sure how much difference it would actually make. It might make some difference regarding perception among potential developers.
To some extent, I think that would just shift the focus to the joint copyright issue, because I believe that some people in the community I hail from, the Linux community, are just plain distrustful of Sun and their OpenSolaris efforts. In my opinion, there is really no need. A rising tide lifts all boats. And the success of OpenSolaris, as with the success of the *BSDs is, in the grand scope of things, a success for Linux, as well.
Edited 2008-05-08 16:53 UTC
I installed this incomplete Solaris cut, opensolaris 2008.05. I was shocked at how much solaris was regressed and how many things didnt work.
Solaris 10u5 + companion software is really very good.
SXDE (Nevada) is also quite usable.
I wouldnt touch solaris 2008.05 with a 10 ft. pole.
When they do things like this , its like running a really old incomplete version of linux.
Screw Solaris AND Linux
Im holding out for OpenAIX!
Edited 2008-05-09 03:37 UTC
developers, developers, developers. Right now OSOL is a project that is – what – 90-95% in the hands of SN developers. But that’s simply not enough to have a successful open source product.
The product itself? Yeah, nice, but some technical advantages (ZFS, dtrace etc.) will make few people switch horses. On the contrary, one realises how well crafted Linux is. In my case there were no drivers for ethernet (!), wifi, sound, and SATA …