Submitted by We’ve been debating the merits of a possible IBM-Sun deal for a while now, and even Sun itself seemed to be in the dark as to if it would be a good idea to be bought by IBM. These debates are now all moot: in a surprise move (at least, I didn’t see any speculation about it) Oracle has bought Sun Microsystems, at USD 9.50 a share, which equates to a total of 7.4 billion USD. The news got out through a press release.
“The acquisition of Sun transforms the IT industry, combining best-in-class enterprise software and mission-critical computing systems,” said Oracle CEO Larry Ellison in the press release, “Oracle will be the only company that can engineer an integrated system – applications to disk – where all the pieces fit and work together so customers do not have to do it themselves. Our customers benefit as their systems integration costs go down while system performance, reliability and security go up.”
“This is a fantastic day for Sun’s customers, developers, partners and employees across the globe, joining forces with the global leader in enterprise software to drive innovation and value across every aspect of the technology marketplace,” said Jonathan Schwartz, Sun’s CEO, also in the press release, “From the Java platform touching nearly every business system on earth, powering billions of consumers on mobile handsets and consumer electronics, to the convergence of storage, networking and computing driven by the Solaris operating system and Sun’s SPARC and x64 systems. Together with Oracle, we’ll drive the innovation pipeline to create compelling value to our customer base and the marketplace.”
The deal has received unanimous approval of Sun’s board, and is likely to go down this summer, after shareholder approval.
I find this better than the IBM deal; Solaris and Sun have a good relationship anyway. On the other hand I’m wondering what will happen to MySQL. Forked away or killed?(unlikely)
Yes, I’m concerned about MySQL (vs Oracle) and Solaris (vs. Enterprise Linux)
Fortunately, the GPL allows us to fork it as a last resort.
Fortunately, the GPL allows us to fork it as a last resort.
Just to nitpick, any open source license would allow you to fork it.
The new BTRFS was launched by Oracle. And now Oracle has access to both BTRFS and ZFS.
Maybe Oracle will GPL the ZFS code? Merge some ZFS code with BTRFS? Everything might happen. Probably, it is good for the future of Linux filesystems.
You’re way ahead of Oracle there. Quite frankly, they’ve never been a company clever enough to realise what is going on within it.
Hmmm… Oracle doesn’t open-source much…
at least oracle won’t kill solaris anytime soon, as ibm might well have. sounds like a good solution to me.
It’s really only delaying the inevitable. The same people are going to be in charge that have presided over the very situation that has led to them needing to get bought out, and Oracle just aren’t clever enough to work out what’s required to arrest the slide. I still don’t see a future for SPARC whatever happens and I see Oracle maintaining the hardware parts of the business even less than IBM would have. It’s not what they do.
I think we’re going to end up with an even more horrendous Frankenstein’s monster and goodness only know what it means for the already horrific management and leadership of MySQL.
remember oracle made their ‘own’ linux. Maybe they will dump this and go theopen solaris route instead.
Certainly for their customer base it could prove a ‘better fit’
I think solaris as a standalone product will slowly be left with open solaris taking the lead
Yes, and look at how that’s turned out. Oracle Linux has got very little credibility at all amongst people who know that it’s better to get support from the upstream vendor. Those who don’t know that learn it pretty quickly.
I fail to see how things would turn out any different or how they would arrest the slide in Solaris’s usage over the past decade.
I don’t see how.
Unless Oracle can turn OpenSolaris into something that really is open rather than with ‘open’ in the name then the outcome is going to be exactly the same. They need a lot of contributors from outside of Sun and Oracle to reduce their maintenance costs and to increase the user base of it to drive reveue. That’s a tough thing to try and deal with when competing systems have already done that.
“They need a lot of contributors from outside of Sun and Oracle to reduce their maintenance costs and to increase the user base of it to drive reveue.”
I dont agree with that. Linux kernel is one of the buggiest piece of code there is out there. As Linux kernel Developer Andrew Morton says:
http://lwn.net/Articles/285088/
“I used to think [code quality] was in decline, and I think that I might think that it still is. I see so many regressions which we never fix.
…
it would help if people’s patches were less buggy.”
You see? I do NOT want Solaris kernel to be a piece of shit unable to scale as Linux. Way better if SUNs excellent R&D people do that. Many agree that SUNs tech is the best in the world. I want to keep it that way. Not let some amateurs contribute code to the superior Solaris kernel.
Hmms if your saying that linux has buggy code it could be nice to add some evidence to that.
The article that you linked to mostly deals with the exprimental tree that are used to try out new stuff.
Big surprise that there are many bugs i new code in the experimental tree’s. and it would not hurt to have the whole quote
“I used to think it was in decline, and I think that I might think that it still is. I see so many regressions which we never fix. Obviously we fix bugs as well as add them, but it is very hard to determine what the overall result of this is. ”
“Secondly: it would help if people’s patches were less buggy.”
And the last comment was only dealing with the exprimental tree.
If it’s only your opinon then you should say so but if you state it like a fact without backing it up well then it’s only trolling.
Edited 2009-04-20 15:38 UTC
You do know that when we talk about linux kernel bugs, we are talking actually for bugs on drivers mostly, right? And guess what, one of the reasons linux is in so widespread use is really because it damn works on most of the platforms out there. You also should know that all this talk about it not scaling is just rubbish, as seem on the many benchmarks that abound in the Internet.
That said, I think it seems a good solutions for Sun and, most important, for all us. I like OpenSolaris, Netbeans and Java and were very worried about getting all that eggs on only one basket (IBM). Hope they keep improving all them and all the good technologies Sun brought to market all these years.
It also make sense in the hardware side, as Oracle know can offer good products to be used with their databases and software. They got a lot of expertise on this front in just one swing. Consolidation just make sense on big market (storage and servers).
I´m a bit worried about MySQL and waiting for a Oracle general words about their intentions.
I’m not talking about making Solaris buggy or how buggy Linux is, which you provide no evidence and no figures for whatsoever, but it’s a simple fact of life. It is too expensive to produce your own OS, maintain it and add new features these days. Only Microsoft can really make that pay.
I’m not entirely sure what evidence you base that statement on because we’ve been through this before and there is ample evidence that has been provided on umpteen occasions that Linux ‘scales’ beyond anything that even Solaris has done. Repeating that will not make it true.
As for code quality, bugs and regressions, Andrew Morton does not tell us anything we didn’t already know about any piece of software – including Solaris. I can’t really fathom why you think that that statement can only apply to Linux.
The only reason why we get to hear these things is because Linux’s code is there for all to have a go at. You haven’t been able to see Solaris’s bugs, the decrepit device drivers that haven’t been touched for years or the glacial pace at which it has been developed for some time.
The main area for bugs within Linux is with device drivers, obviously, and it would be really nice if Solaris had some. You only need to look at how many device drivers Linux has that Sun would love to have for Solaris.
ROTFL. Well, I’m afraid the ‘amateurs’ have been handing the asses of the professionals back to them for the last ten years. That’s why Sun has just been sold. Maybe the great big penny will drop one of these days on that.
Edited 2009-04-20 16:11 UTC
Eh! Opensolaris.org. There is a source browser. Go read the source for your self.
Glacial pace? How far has sytemtap or kprobes come along? What about BTRFS? any of that crap out of alpha yet? How many users are using it in the data center?
Where is all the rapid development from the linux community?
You would love to believe that but then you would have to abandon your belief in Santa Claus and the tooth fairy too.
It was Intel doing that not linux alone. Without intel linux wouldn’t have done squat. IBM, HP, Apple still have their own OSes. Think about it linux s ionly sucessful on cheap intel hardware or embedded space… Desktops and Datacenters are still nicely mixed. With IBM peddling its mainframes and windows actually on the rise…
Oracle makes money on support contracts. Oracle can drop a box with a free OS and make a killing on the contracts and license for the rest of the stack. No one will be able to compete on price given that Sun’s hardware is already competitive with Dell and IBM. Never underestimate the power of subsidy. Works really well for cellphones…
Without the incentive to save a few bucks running linux most customers would just pay Oracle for the whole package…
Edited 2009-04-20 17:08 UTC
A source browser? Yer, and? After all these years where can I view the commits that Sun employees and contributors are making every week that will go into the next version of Sun’s Solaris and where others can make a bootstrapped alternative? How can people get patches into Solaris? I’ll tell you where. Nowhere.
What? Seriously. What? Intel is somehow single-handedly responsible for Linux’s success? Sunshine, run down the list of commits to Linux and you will see a minority of @intel.com addresses in there as opposed to @sun.com addresses with Solaris, if I could find Solaris’s repository that is. Sorry, but Linux gained momentum and went right through 2.22 and 2.24 before Intel touched it with a ten foot pole.
Mind you, you have touched on one thing. Intel are actually able to commit lots of device drivers to Linux ;-).
Oracle have been trying that ‘whole package’ thing for a long time, right from the laughable Network Computer over ten years ago to Oracle Linux today. There’s not really anything to suggest they’ll be any more successful.
Really! Nowhere?
http://www.opensolaris.org/os/project/onnv/
hg clone ssh://[email protected]/hg/onn/onnv-gate
This pulls the source.
Then run hg log on the cloned copy of the entire kernel source base that magically appeared in your local machine.
The log shows the commit history for the entire repository.
Nexenta, Belenix and others have no problem making distros based on OpenSolaris.
Got any more FUD to spread.
The rest of your FUD isn’t worth responding too.
Edited 2009-04-21 00:15 UTC
Yep, nowhere. Is there a repository for *everything* that people from Nexenta and elsewhere can commit to without signing away their children? As Ted Tso once said, Linus gets more contributions brushing his teeth than this approach will get in a week.
Like I said, where’s the repository? Maybe I didn’t make it clear what is meant by that in the open source world, but I figured people would be clever enough to know. Creating a Mercurial or a Git repository and accepting patches is not rocket science, but Sun won’t do it.
Ahhhh, Project Nevada eh? Sun’s blessed distribution that masquerades as OpenSolaris?
Yer. Read it ;-).
That’s because they have to build off the back of Nevada, which is Sun’s blessed and controlled ‘OpenSolaris’ distribution. They’re not building off the back of a free and open Indiana, which confusingly, was supposed to be OpenSolaris but became Sun’s pet offshoot. It’s confusing as hell. The notion that somehow you can pull the source down and build an independent and competing distribution without it is bollocks. As Roy Fielding found out, you can’t even call OpenSolaris anything else: 😉
http://roy.gbiv.com/untangled/2008/watching-the-ripples
If people are happy with that then fine, but it’s not an open source community we all know and have seen and pretending it is is just plain sad. I gave up looking at it a while ago. It’s dead until Sun gives up control.
Seriously, you don’t have to look far:
http://www.opensolaris.org/os/get/
AKA, Nevada, unless they’ve changed again. No cookies for you. I wouldn’t have a problem with this, but people try and pretend it is something it isn’t and that I do have a problem with.
I do wish people would understand what FUD stands for and means rather than throwing it around in conversations as a reply to things they don’t like. It would help if you read first.
Edited 2009-04-21 01:05 UTC
Where is a such a respository for GNU/Linux. Point me to where on kernel.org or anywhere there is a “repository” that all the other distros pull from and integrate too.
Didn’t think so. If you can’t shut your trap about your stupid defintion of Open Source..
You’ve got to be kidding me?! Seriously…………
You can pull code from all the projects that make up a ‘Linux system’ (kernel, userspace, whatever), bootstrap and roll your own distro or you can pull code from the many distributors out there who have repositories for their own development. The point is, you can fork any Linux distro in umpteen different ways but you can’t do that with Solaris. The freedom of the source code is everything.
You’re obviously upset that OpenSolaris isn’t what you thought it was, but I’m not. If Sun want to do what they’re doing then that is absolutely fine and I don’t have a problem with it at all, but pretending that they have something that is ‘open source’ with an OpenSolaris ‘community’ directing things is disingenuous and wide of the mark.
The basic acid test for an open source project is if people can create workable forks and work with the code 100% on a level playing field. In OpenSolaris’s case that means external people working on a level playing field with Sun, committing to the same repository and building a working Solaris system in the same way from source code with no intervention. Trademark and other issues can come into it, but that’s the acid test and OpenSolaris fails it. I don’t think Sun ever intended to have that, but they didn’t make it clear.
I really don’t have a problem with what Sun has done with OpenSolaris at all but let’s just admit it, let’s not pretend, let’s not get all upset and let’s not get pissy about redefining what ‘open source’ projects are when the reality has been pointed out.
SEGEDUNUM
SUN is opening Solaris as OpenSolaris. The Solaris 10 distro is not meant to open. All effort and focus goes into Solaris 11 (that is OpenSolaris). I myself prefer that SUN opens S11 instead of the soon legacy S10. If SUN committed all resources to open S10 instead of S11, then you would have complained; “who needs S10, why not open S11?”. Damned if you do, damned if you dont. SUN can never win, whatever they do. Whatever they do, they do it wrong.
As ZFS targets Enterprise servers big iron with lots of CPUs and lots of GB and 64bits CPUs and therefore is designed in a way to take advantage of lots of resources, then “ZFS requires 1GB RAM – ZFS sucks” as you say today about ZFS.
If ZFS targeted desktop with 64MB RAM, then you would say “why dont target servers? Anyone understands that all servers of today has at least 1GB RAM. ZFS is for desktop, it is a toy file system. ZFS sucks”.
You can download the source and build S11 (OpenSolaris). I very much prefer SUN to commit all resources to open source S11 than old, legacy versions of Solaris.
Your are contradicting your self. You said OpenSolaris isn’t open because there is no respository. I pointed you to mercurial respository for the Kernel and command line tools. You wanted one “repository” for everything which your are delusional to think makes something open source. Linux doesn’t follow that model either.
The same can be done for OpenSolaris. The kernel and most of the command line in the repository I linked.
Nexenta is the OpenSolaris kernel and GNU userland just like most linux distributions.
So what is the difference and why are you adamant that OpenSolaris isn’t open source?
You can with OpenSolaris. Belenix and Nexenta are proof. You just can’t call it OpenSolaris. Just like you can’t fork Fedora and call it Fedora.
The problem is symantics. OpenSolaris is everything. Unlike linux there is no name just for the Kernel. So if you say something is OpenSolaris people think it is like FreeBSD not like Linux. That is why you can’t a fork OpenSolaris, you have to call it something else. Just like you can’t with FreeBSD. There is only one FreeBSD. Apple uses FreeBSD code but calls it Darwin. Apple also uses OpenSolaris code in Darwin.
It is simple and true open source which means the source is open and others can use it freely and they are. The source is more free than with linux.
If you had a brain you would understand something that simple.
No I am upset that I have to waste my time explaining to morons like you what is very obvious with simple reading.
Doesn’t look like it. You have been harping senseless that OpenSolairs isn’t open source because it isn’t Linux.
The only person pretending OpenSolaris is something it is not, is you. OpenSolaris is Open Source whether your like it or not. You closing your ears and pretending to not hear when reality is pointed out against your insane notion is what is wrong here.
Take your head out of Linus’ ass and smell some fresh air for once. You might learn something.
If Linux wants to be GPL which doesn’t want to allow its code to be in Non-GPL projects then they should be fine with Non-GPL projects not allowing their code to be in linux. Anything otherwise is double standard which seems to be rampant in the linux community.
It is obvious you are jealous and upset that linux can’t leech stuff off of Solaris. So it seems are the Linux developers, their comments are so laughable. Sour Grapes really. Meanwhile Apple, FreeBSD, ONX are happily integrating Solaris code with no license issues. That is true Open Source whether you like it or not.
Edited 2009-04-21 15:22 UTC
No. You’re tying yourself in knots now because you’ve been shown what OpenSolaris really is and you don’t like it. You’re also trying to split hairs by saying “Oh, I pointed you at a Mercurial repository, it must be open source!” Errrrrr, no.
The fact that you pointed me to a repository that is read only to everyone but Sun unless you sign something convoluted is neither here nor there – as I had explained very, very clearly and carefully in what I wrote. There are is no way for external contributors to easily get patches into the repository, and even if you pull code from it you cannot build and bootstrap an ‘OpenSolaris’ system unless you do so from Nevada – Sun’s blessed and appointed OpenSolaris reference system.
If Sun wants to do that then it is absolutely fine. However, the fundamental conerstone of an open source project is that you can pull the code, compile and bootstrap a competitive system and get something that works. Without that it isn’t of too much use to anybody outside Sun because you get a case of Sun wanting your cake and eating it. I can’t say that any other way. If you don’t get that fundamental concept then I’m afraid you’re a moron.
Not only that but you can’t fork and bootstrap the resulting ‘OpenSolaris’ system without Nevada. Even Sun have made that crystal clear and you still don’t want to believe it.
Which can only exist bootstrapped with Sun’s official Nevada, as I have explained umpteen times. Stop trying to compare it with Linux. No Linux distribution can only exist with an ‘official’ Linux distro that it must be bootstrapped with. That would just be moronic, wouldn’t it?
1. There are still closed parts that you need.
2. You can only bootstrap your own competitive system with Nevada, and as Sun will probably want to prevent forks then that is unlikely to change. The fundamental principle of an open source project is the ability to fork. Sun don’t want you to do that. Great, but that’s not an open source project.
If that isn’t clear then you’re a moron.
Once again, you can only bootstrap your system with a provided Nevada system from Sun. I can only say that so many times, so I’m afraid Nexenta and Belenix are not examples. Without Nevada they disappear.
You’re a moron sunshine because OpenSolaris fails that definition you’ve laid down. You can’t just say it is more free than Linux and expect that to be magically true.
Nope, never said that and if you think that I said that then you have some severe mental issues regarding Linux.
I defined a set of principles and things you should be able to do within an open source project to ensure that the code stays free and you’re not relying on either closed parts or a reference system layed down by others. The fact that Linux distributions tend to meet those requirements is merely coincidental.
Hmmmm. Yet more mental issues regarding Linux again. Quite frankly, there’s not much you’d want to leech off Solaris apart from ZFS as it’s the only remotely interesting thing in there. Sun would love to leech Linux’s device drivers though. It looks as though neither will happen. No big loss.
I don’t believe this was ever a discussion about OpenSolaris’s usage of certain licenses, but it’s interesting that you seem to have issues over that and feel the need to get defensive about it as well.
ROTFLMAO. Come back when you have understood how things really work..I really can’t waste any more time. I can teach amoeba how to read english and understand this stuff with far less effort.
Read my original post again and again until it sinks in. It really isn’t that hard. Really! It isn’t.
Edited 2009-04-21 18:24 UTC
You think that if it gives you comfort. It’s been explained to you why you’re a moron and I’ve explained succinctly why OpenSolaris is not an open source project as we know one to be and what’s required to actually get an ‘OpenSolaris’ distribution working. You don’t want to discuss that. At all. OK.
Sun is OK with the situation, I’m OK with it, you’re not and many people around here have severe mental blocks and issues accepting it as well. It’s OK. We get it.
We didn’t need to have one massive thread where I ended up repeating my point five or six times because you didn’t want to go near it with a ten foot pole – as funny as it was to watch. 🙂
Blah! Blah! Blah!..You can keep telling your self whatever you want. You haven’t succinctly explained anything.
OpenSolaris is Open Source by any definition of the term. You are the only one that seems to have an issue with it.
Thanks for understanding my suffering. Next time don’t forget to take your meds…
I don’t want to go near it because like I said I could explain this stuff to a rock and it would understand it the first time.
Only you understand your stupid comments. Clearly others having a tough time comprehending your meaningless drivel as well.
There is and it is not different than how linux or any major open source process works. You have to have Linus and other Linux developers blessings to merge a change in if you want it to be in the main kernel. You need the same thing from OpenSolaris developers and must follow a process before a change gets accepted.
Nevada is the next realease of OpenSolaris that is where everything gets built from. The latest build is snv_111. nv stands for Nevada. Just like Linux 2.6.x is Linus’s reference build blessed and controlled by Linus.
That is exactly how it works. There is no sinister reason for having some closed parts other than Sun can’t open them up because they don’t have the right too.
Can I download the linux kernel, compile it and boot it up wihtout a Distro. Nevada only makes up part of the DIstro so just using it won’t get you a full working system. Nevada is the kernel + device drivers + libraries. It doesn’t include Grub for instance. You can’t boot linux without Grub. Neither can you Nevada.
Because Nevada is the main piece of it you idiot. Nevada is the kernel and main system. ON (OS/NET) is the linux kernel+ GNU equivalent. ONNV means ON nevada. It is the code name for the core operating system that will be next release. It is the base for the OpenSolaris distribution.
http://opensolaris.org/os/community/on/
“ON (pronounced “Oh En”) is the common abbreviation for the OS/Net consolidation which delivers the core Operating System and Networking components to Solaris. It contains the source for the kernel and all platforms (on all architectures), the bulk of the drivers, filesystems, core libraries, and basic commands that you’d expect to find on a Solaris system.”
It is like saying one can’t make a linux 2.6.x distro without Linux 2.6.x… which is utterly stupid!
Unlike Linux, Solaris has always been made of a core OS called SunOS. Which is what Nevada is the latest iteration of. The Solaris distribuition is built on SunOS or ONNV in this case.
Just like a Linux distribution is built on a Linux kernel.
If your main gripe is that there are some closed parts due to license issues that is pure Anti-Sun nitpicking. Sun has OpenSourced everything it can even stuff like ZFS and Dtrace. Sun even Open Sourced its latest SPARC processor. No other company has done that.
If you believe there is a sinister reason for having some closed parts or Sun is lying then there is no more argument. You are an anti-Sun troll plain and simple and must be pointed out as such. No amount of reasoning will change your mind.
There is nothing for me to believe other than that your a stubborn moron that doesn’t want to understand what it is you are arguing about and why you are completely and utterly wrong.
Nevada is the kernel and core OS you idiot. It is like saying fedora can’t exist without using the Linux kernel and GNU userland as the base.
That is how it works. You are pretending to understand something of which you have absolutely no clue what so ever.
Nevada is the kernel + base distro similar to a Linus’ main kernel release that every distro uses to get the latest and greatest. You can’t fork the linux mainline tree change it completely and call it Linux 2.6.14. Bad shit will happen if you do. You have to append a distro specific something to indicate it is modified version not the original.
If you say it is possible to fork Linux 2.6.x and call it Linux 2.6.x with the same name as Linus’s official release I want an example.
First understand what Nevada is then show me proof that Sun prevented a fork. Also show me that Linux has been forked with the same name as the official release.
You are wrong and if you don’t understand why after the explanation above you have proven once again you are an Anti-Sun troll.
Once again you are utterly and completely wrong. Without a linux kernel all Linux distros cease to be Linux too you idiot. Nevada is the base for OpenSolaris. It includes the main Kernel. You need it to get a working system you blithering idiot.
Please stop talking to your self and about your self in between a post it is very annoying. I know you are a moron and it is clear you know your are one too thanks for reaffirming it again and again.
Bullshit you defined no such principles. There is an industry standard for those principles and OpenSolaris meets those. Your own delusions might convince you otherwise but that isn’t reality. Stop posting when you are off your meds.
Edited 2009-04-22 13:59 UTC
Just wanted to clarify that Nevada is the base OS for the OpenSolaris distribution. It is not the Distribution it self but makes up a part of it. Like Linux is the main part of a Distro but not the distro in itself.
SEGEDUNUM
I would love to discuss with you if you had some substance in your claims. Please backup your claims and show us links that proves your point, especially when you state controversial claims like “ZFS need huge amounts of memory”, “Sparc is slower than Power CPU”, etc.
You know, negative criticism and negative articles are welcome in a debate. BUT DECLARATIVE CLAIMS SHOULD HAVE SOME SUBSTANCE AND PROOFS!
Otherwise, it is just your personal opinion, and if it is your personal opinions, you can NOT state things as “ZFS requires several GB RAM” – because it is not true. But you can say “In my opinion, ZFS is useless because it requires too much RAM”. That is a valid stand point, and I accept that. You dont have to back up anything about personal taste and opinions. That is entirely ok and nothing wrong.
I can say “I hate blue”. That is ok, and I dont have to back it up. But if I say “SPARC is slower than Power CPU” then I have to prove it otherwise I am lying. The correct would be to say “I like Power CPU better than SPARC” – that no one can argue against. It is a matter of personal preference. If I like Pizza or not, that is up to me.
Facts can be wrong. Opinions are never wrong. You have to back up facts. I can prove that Sparc has several performance world records, for instance.
You ask for a single repository for *everything* that makes up OpenSolaris, and then turn around and point at the bazillion separate repositories that make up a Linux distro as a counter-argument?
So … *not* having a single source repository for OpenSolaris means it’s *not* open-source … but *not* having a single source repository for a Linux distro means it *is* open-source?
Am I the only one who can’t figure out that statement?
No. That whole argument is bogus and he has been rambling on about it for years on OSnews and still hasn’t figured out he looks like a fool for making it.
Edited 2009-04-21 15:40 UTC
Seriously, is it possible to ban him as a Troll? I mean it. Can we vote or something like that? Ive proved my claims about Linux with links which we can discuss around. This uneducated LIAR SEGEDUNUM just states things again and again, despite us proving otherwise with links. He doesnt care. Is he paid to do this FUD and lying? Seriously, I wonder?
Everything would be fine if he could back up his comments, which Ive asked for numerous times without any reaction. I welcome critical arguments and a critical debate and discussion. But then you have to back up your claims, otherwise it is just FUD and lies. I am used to academia and discuss as grown people with rigorous arguments and proofs and counterproofs. Not false claims disguised as facts.
Pot, meet Kettle.
Thanks for introducing your self. Which one are you Pot or Kettle?
Oh dear. I think this thread needs to end here and now because we’ve obviously got some people with some severe issues who cannot discuss this in a rational manner. I’ve obviously hit some very raw and exposed nerves and even I’ve been educated in how far people will go to deny it to themselves.
I won’t reply to anything else because we’ve been over it many times and it has been proved to be deranged rambling as a result of not being able to accept reality. It’s been fun while its lasted.
I would appreciate if you could back up your false statements. But you never do. Instead you keep spreading FUD and lies. And when I ask for proof, you never give any proof. And when I give counterproofs to show your claims are false, you just ignore my links. How fair is that?
For instance, you keep on saying ZFS requires several GB of RAM. And then I prove you wrong. I point out that you have talked to FreeBSD people, and on Solaris ZFS runs on even 512MB RAM. What applies to FreeBSD does not necessary apply to Solaris. But still you keep spreading the lies about ZFS requiring several GB of RAM. I disproved you several times, but to that you never answer, and in the next post you say the say thing; ZFS requires several GB of RAM. This is very tiresome. Ive showed you this link before, but to no avail:
http://queue.acm.org/detail.cfm?id=1317400
Pawel Jakub Dawidek, a FreeBSD developer who successfully ported ZFS to FreeBSD says:
“The biggest problem [in porting ZFS to FreeBSD], which was recently fixed, was the very high ZFS memory consumption. Users were often running out of KVA (kernel virtual address space) on 32-bit systems. The memory consumption is still high, but the problem we had was related to a vnode leak.”
You say things, which are never true. And when I counterprove your false claims, you never give an answer. You just leave the thread, and in the next thread you state the SAME things. That is tiresome to go and watch your false claims, and disproving them. If you could prove your claims with links and articles, everything would be fine with me. Negative critism is welcome in a debate, but you should back controversial claims up. If you never do, despite being asked again and again – people can suspect you are lying. And you dont want that, do you?
If I state things about Linux AND prove all claims, then I am not FUDing, I am not lying. I have provided links that we can discuss around. It would be very bad if I stated “Linux scales bad” and never back that up. Then it could possibly be lies. And if I never prove any claims, despite being asked for, then it surely must be lies. But if I point to numerous links then there must be some substance in my claim. (Of course, when I say Linux scales bad, I compare to a real Unix. In fact, Linux scales very well compared to most OS. Of that is not question).
Regarding anyone can not contribute to Solaris source code. That is natural. There has been code accepted by non SUN employees. But, it is hard to get code accepted. That is a GOOD thing. Otherwise there is risk that the kernel degrades in quality, just as the Linux kernel, where a small upgrade breaks everything. Hardly stable.
Another point, is that SUN can be sued in court for lots of money if anything goes wrong, if SUN accepts bad code into the kernel. If a small upgrade breaks anything, hell will break loose. SUN wants total control of what they deliver, right?
But, you can download the source and recompile it. There are several OpenSolaris distros: Nexenta, Milax, Aurorax, Belenix, etc etc. The code is available and open for you. The code is open sourced under CDDL, Apple and FreeBSD and QNX are using the Solaris bits and pieces. Would that be possible if the source was not closed? No.
SUN will not accept any code changes without heavy scrutiny. Which is good for the quality. Otherwise, use Linux.
Now, will you leave this thread as usual? And in the next thread, you will go on and on again “ZFS requires lots of memory. SPARC Niagara is slower than Power CPU. OpenSolaris has not it’s code open sourced. etc etc etc”. Why dont you back up any of your claims instead of FUDing and lying? Leave this thread, will you, you LIAR?
I’ve been rambling on about single repositories for years…………….? Seriously, what planet have you just teleported from.
It might help if you knew what the argument was, and I’m afraid that wasn’t it ;-).
Hmmmmmmm, no. I don’t know why you and the other morons around here are trying to argue that route because it is totally unrelated to the point that was made. OK, maybe I do ;-).
Many Linux distros will pull code from many repositories, and most will have their own repositories where they will keep their own releases. OpenSolaris will pull code from many others, especially the userspace stuff, into their own repository. That’s neither here nor there.
The point being made is that I should be able to pull from a repository, or multiple repositories if I’m rolling my own as was stated, compile, bootstrap and get a working system. That is not the case with OpenSolaris unless you pull in the still binary parts and bootstrap from Nevada.
Point made. Get over it already.
That’s not what was pointed out at all, but then, you and the other morons around here know that ;-). I clearly stated why OpenSolaris was not an open source project as we would know it.
That’s because you’re trying to make an argument that isn’t there and you don’t want to see what the real point was. It’s amazing the lengths some people seem to want to go to to do that. Incredible.
Read this again and tell me what the difference is. Linux distros work the same as OpenSolaris is what this is telling me.
So, that is how it works. Nevada is the next Solaris releases codename. It is where everything gets integrated to. You don’t understand how this works clearly.
You don’t even understand the point your are making. There is nothing to get over you are WRONG. Plain and simple. Get over it.
Funny you call others moron when your own comments are beyond help.
You really need help. You seem to be talking about your self in third person. Seek some mental help. You are delusional beyond all hope.
SEGEDUNUM
You know, Solaris code is out there for me to see if I want to. On the mail list there are discussions about Solaris bugs. The bugs are not hidden to the public. Why do you imply that?
And for Linux having lots bugs in the drivers, but not in the kernel. Why do YOU ask me to back my statements up? Tell me, when did you back any statement up? You never did. Ive asked you numerous times.
You just state stupid things, such as a supercomputer and an server are used identically. Ive tried to explain to you that a supercomputer is differently constructed than a server, but you dont listen. Instead you keep reiterate it. Can you back that up, that Supercomputers are used just as an ordinary server? (Ive showed you wikipedia that says that supercomputers behave very differently from an ordinary server, but you ignored my links). So please, back any of your stupid links up. You can not blame me for not backing anything up, while never backing anything up yourself. Ive always provide links in my posts, you know that.
http://kerneltrap.org/Linux/Active_Merge_Windows
“The [linux source code] tree breaks every day, and it’s becomming an extremely non-fun environment to work in.
We need to slow down the merging, we need to review things more, we need people to test their f–king changes!”
And hackers below agree on this, whereas Linus Torvalds says in the discussion there will always be bugs in Linux. There is no testing. (In Solaris there are lots of testing).
Have you heard about the Linux shit list? The quality detoriated so much, drastic steps had to be taken. There is a shit-list for Linux kernel developers contributing bad code. Rafael Wysocki maintains the shit list.
The Linux code is really buggy and not tested. It is over 10 million lines of code. ONE KERNEL. You know, entire Windows NT was 10 million lines. You have seen my links, proving Linux scales bad, sucks as a file server, has bad uptime, buggy code, etc. Ive provided links for every claim. And still you ignore them, while you have never even once provided one link, despite me asking you many many many many many many times.
Linux zealot with blinders on… what do you expect? A well reasoned balanced argument..
You are right that he should provide links to backup his examples.
and that you have provided links to each one of your statements in the past.
the only problem is that i have yet to find one link of those you posted that actually are backing up your statements.
you make the statement Linux is bad for file servers.
And the article that you link to is “linux is bad for file servers if you wish to have a single volume larger than 100tb”
which are to completely different statements.
since you know not that many file servers run single volume file systems of 100tb+.
yes linux is not the best solution for everything, but guess what neither is solaris aix or anything else there is no golden ticket for achieving everything.
please read though the articles that you use to backup your statments since i have yet to find a article that you have linked to that have made the same statment that you have.
I dont agree with you. I think I have backed up my claims with relevant links.
If I state that Linux is unstable, what do I mean then? Do I mean that Linux is unstable for my desktop usage? Of course not. For desktop usage Linux is fine, Ive never seen Linux crash. I am talking about big installations.
http://www.computerworld.com/action/article.do?command=viewArticleB…
If I say Linux does not scale vertically, do I mean it does not scale on my quad core CPU? Of course not. Linux utilizes my 4 cores very well (I believe). I am talking about many hundreds of cpus.
If I say Linux sucks as a file server, do I mean for serving 1-2 client computers then? Of course not.
If you say Windows does not scale, does it not scale on your desktop? Of course it scales on your desktop PC. You talk about large computers.
Even Windows suffices for personal use. In fact, with SP3 Windows XP is quite stable. But I would lie if I said WinXP is good for large computers, because I myself has not crashed WinXP for a long time.
So if I provide links of WinXP crashes a lot on large computers, are those links not relevant then? Am I lying then??
I do not agree with you. I have provided relevant links. For personal desktop use, any OS will do including Windows. But that is not interesting, the large demanding installations are the interesting ones.
Edited 2009-04-20 19:32 UTC
Ok maby i am stupid or something.
but can you please explain the relevance
http://www.computerworld.com/action/article.do?command=viewArticleB…
in your last post.
it could be about “I am talking about big installations.
but the article just deals with about a unconfirmd rumor that google is trying solaris.
oh and it did also contain some cloud/hosting computing business that switched.
but we still dont know if the error is with bsd/linux or that they used a crappy support company, and sun is a great company for support.
but i did fail to see anything about linux not being able to run on big installations.
And yes in your last post you did make the statement that it was only big installation machines you are talking about. but not in the others i still think that you should think about what the article say’s and what you say.
That link I provided talks about Google trying Solaris. Since then, Ive changed the link to point to the third page in that same article. If you click on your link again, it says that Linux crashes while Solaris dont.
The point is, Linux is unstable compared to Solaris. There are several such links. Such as this one, where he changes to Linux on the same hardware and suddenly everything works:
http://lethargy.org/~jesus/archives/77-Choosing-Solaris-10-over-Lin…
Or Linux doesnt scale well:
http://searchenterpriselinux.techtarget.com/news/article/0,289142,s…
“The problems we encountered were because Linux doesn’t scale all that well,” Rand said.
…
The improvement is significant; with four compute nodes instead of five, Rand [SOLARIS] has more computing power and 99.99% uptime, compared with 97% uptime with RHEL, he said.”
http://www.computerworld.com/hardwaretopics/hardware/story/0,10801,…
Because Solaris 10 is “lighter weight,” meaning it’s less intrusive on applications, performance has improved, Greenwade said.
…
Philadelphia Stock Exchange Inc., has a new electronic options-trading system that runs on Solaris 10, which he said has improved trading capacity by 36%. Like Greenwade, he credited performance improvements to the TCP/IP stack, as well as improved multithreading support.
Thanks to Solaris, the stock exchange expects to reduce the amount of new hardware it needs to buy in order to scale up the trading system. “You can in effect scale within the same machine, as opposed to adding servers,” Morgan said.
The point is, the companies compare Linux to Solaris and then Solaris wins. Ive yet to see a link from you showing that Linux wins over Solaris. The links Ive seen from you guys, shows that Linux wins over Windows. So what, any OS win over Windows.
Edited 2009-04-21 10:13 UTC
You can find links proving about anything you want to prove.
There it goes: two links praising linux’ stability on servers.
http://www.computerworld.com.au/article/153880/manufacturer_moves_l…
http://www.eweek.com/c/a/Linux-and-Open-Source/LinuxWorld-Savings-S…
Thanks for your links. The thing is, those links compare Linux to Windows. And then Linux wins.
They do not compare Linux to Solaris (because Solaris wins). So please show us links where Linux wins compared to Solaris.
The links Ive showed, compare Linux to Solaris. There is on comparison of Solaris to Windows, because, any OS wins over Windows. It is pointless to compare against Windows, that is an easy win.
First link yes, second one talks about moving from the mainframe to linux while skipping over proprietary unix (doesn’t say which one).
I wont argue that Solaris can be a better choice for some, but the 100% uptime reports (as that of the previous link) and it’s use on mission critical systems(1) easily dismiss the idea of the linux kernel being a bug ridden POS.
(1) http://www.forrester.com/Research/Document/Excerpt/0,7211,34299,00….
I appreciate your statements as having some substance with links. Then we can discuss around your links.
Look, I dont mean that Linux is a piece of shit. In fact it is a good OS. But it does not compare to real Unix as Solaris/AIX/HP-UX. That is what I am trying to say.
One of the weaker point of Linux is it’s ABIs that are a moving target. How can you have something stable (relatively to Unix) when your upgrades break things? If you upgrade, it can break something working perfectly, which triggers an upgrade of that broken something, etc until you have upgraded everything. Device drivers can stop work in newer Linux versions. etc.
Not so with Solaris. The API and ABI has been frozen for many years. SUN guarantees binary compatibility back to solaris v2.6. Now Solaris is v5.10. No recompile, just copy your old software into the new Sparc Niagara, or to your Big Iron machine. THAT is enterprise.
Linux is a good stable OS, it just dont compare to top of the line. It is more stable and scales better than Windows (which many companies deploy also), that is no question. It is obvious it is far better than windows.
That’s not what I said, but then, you can’t read. I get that. I want to know where the central repository is for OpenSolaris that will form Sun’s next release and how people external to Sun can commit to it and how people can bootstrap an alternative. After all these years there is no such thing.
No I didn’t.
You made that statement yourself about ‘supercomputers’ because you were show up to be be a stupid twit. Apparently, Linux couldn’t scale beyond 32 or 64 CPU systems, you were shown to be totally and utterly wrong and then you made a ‘supercomputer’ distinction because you realised how wrong you were. Somehow, all of the Linux systems running 32, 64, 512 or 1024 CPUs were all supercomputers and didn’t count. Bollocks. The issue was scalability and you were wrong on all counts. Did I say the word ‘wrong’ enough?
You have some serious issues mate. Mind you, I’ve seen many people have the love-in that you do with Sun and it’s impossible for them to accept.
Meanwhile Linux still scales beyond Solaris.
Newsflash: Open source, bleeding edge development breaks every day! Come back and tell me when Solaris does development in the open like that.
There will be bugs in everything. I can tell you this much, there are bugs in Solaris and there is stuff that hasn’t been maintained for years. Tell us something we don’t know.
There is testing. It’s called iterative improvement and it’s why Linux has far wider hardware support than Solaris and it has equivalent device drivers that actually work, as opposed to learning that your IDE drivers are shit because ZFS detected some corruption.
Yep, that’s iterative development. It would be nice if we knew that Sun had something internal like that for Solaris because a lot of stuff hasn’t been touched since the nineties.
You mean the ones where you have been shown to be wrong and where you have mental issues where you are shown to be wrong?
If you pulled your head out of where ever it is right now, the answer is simple. I posted it below.
Of course being a linux zealot it is far easier for you to harp on about something that is entirely in your wrapped sense of reality. Instead of looking at the real world.
Belenix and Nexenta are already making alternative distributions.. I wonder how they are doing that… must be magic.
Edited 2009-04-21 00:23 UTC
In thread http://www.osnews.com/thread?354450 you wrote:
Here you state that there is no difference between a supercomputer and a server.
And then I quoted wikipedia, to show you that a Supercomputer and a server are very different:
To that you said:
Again you state that there is no difference between a supercomputer and a server. Why are you denying you said there is not difference between a supercomputer and a server? You stated it twice.
Regarding Linux scalability, you wrote
And you have not showed me the links yet. Why do you state things without being able to back them up? Are you deliberately lying or are you uneducated? Dont you know that if you state things you have prove your claims. Havent you been in academia?
To your claims I only say; LIAR. You should back up your claims, Ive requested this many many many many many times, and you have never done that. It must mean you are lying, yes?
So Mr LIAR, do us a favor and stop post here as you lie so much – otherwise why have you never backed up any claim? Ive always done that as Ive been in academia as a mathematician.
In addition, I should also point out that this is unstable development within the kernel between releases. There is yet another layer of testing and fixing beyond that that distributors do, so the chances of this rant affecting users is pretty small – unless they want to go on the bleeding edge.
Got news for you….Andrew and Linus idea of what is poor code is still about 100000x better code than is written by most closed source developers who don’t have the threat of many other eyes seeing their code.
Bah….What Linus and Andrew call ‘poor quality code’ is about 1000x better code than usually produced behind closed doors. Amazing what a decent motivator the idea that many different people will look at your code does for your quality, compared to being maybe the only guy who will see the code for maybe even decades.
How is OpenSolaris not open? Or do you mean that Linux can’t leech off it, and therefore it isn’t open?
Firstly, Solaris hasn’t all been completely open sourced yet. While I can understand the complexity of that it has been God knows how many years. Bootstrapping a system (a competitive one at least ;-)) has been damn near impossible, and at least it was the last I looked. I’m not inclined to look again.
Secondly, can you point me to a commit digest somewhere of everything that has been committed to Sun’s OpenSolaris repository in the last week that will end up in the next version of Sun’s Solaris and that everyone else can pull from? I’m sorry, what was that? You can’t and there is no such repository? Oh bugger.
The whole thing smelt like a cynical and desperate ploy so Sun’s consultants could go out and say “Look, we’re just like Linux!”
ROTFL. It’s the other way on actually. Sun would love to get their hands on Linux’s device drivers which is why they vaguely left the door open for that possibility but ensured that the *only* remotely interesting thing in ZFS would stay away. As Linus said on the matter two years ago:
I suppose that’s what you get for not kissing enough girls ;-).
Edited 2009-04-20 21:30 UTC
“Unless Oracle can turn OpenSolaris into something that really is open rather than with ‘open’ in the name then the outcome is going to be exactly the same”
Wait, what? What is not “open” about OpenSolaris? How can you make such a claim and provide zero backing?
GPL is not the only thing that defines wether or not a piece of software is “open.” OpenSolaris is pretty damn open, you get the source code an you get to do modifications to it, for crying out loud… what more do you need?
Considering Oracle’s interest in future-proofing linux from a filesystems point of view (see btrfs), will they now port ZFS officially?
Or will they try pushing their customers towards Solaris? Is this part of a strategy to compete with Red Hat?
…Sit back, I’ll grab the pop-corn.
Edit: 3 minutes to type a post is too long, now my post is redundant, and I look like a complete idiot. Oh well…
Edited 2009-04-20 12:22 UTC
OOOo — great abbreviation for an office suite. 😉
I hope they keep developing it, and keep that Open…
Sounds like an orgasm.
Gad.
Talk about a freakin Greek tragedy.
As we all know, Oracle is just so open with their products and really GETS open source.
-Hack
PS: Excuse me while I go throw up.
Huh? BDB hello?
http://www.infoworld.com/d/developer-world/what-if-oracle-bought-su…
So at least one person thought it was a possibility
Mate – this is hardly a shock announcement; the rumours have been going on for years and years about Oracle purchasing Sun. I’ll put money on it the only thing holding them back was waiting for the price to be right. In the past they were way over valued and of little strategic benefit to Oracle to plonk down the money required.
As for where Linux sits into this; I’ll put money on it that Oracle will scale back their contributions to Linux. The only reason why they contributed to heavily to Linux was so that they’re not dependent on a single vendor. Now they have an operating system of their own, its completely open source – why contribute more than necessary?
By the time Solaris hits 11, it will be 100% opensource, most of the hardware revenue will come from x86 sales, Oracle will take a ruthlessly pragmatic approach to hardware sales but at the same time investing into Solaris as to provide the customer with the complete widget (Hardware, Operating System, Database etc.).
What will be interesting is how this will stack up against IBM considering that now Oracle will probably have revenue of around $36billion – it’ll be interesting to see where SPARC will fit into the equation. Will we see a strong focus on Rock SPARC processor with Oracle being heavily pushed? lots of interesting things are going to be happening in the future and I have a feeling that Oracle want to leverage some of the ideas Sun has but needs the resources from Oracle to accomplish. The old story of having to have the funds to turn a drawing board idea into a real product that rolls off the production line.
Btw, this is a little off topic (that stupid 20 minute rule so I couldn’t add it to my original post). There are now OpenSolaris laptops; Toshiba laptops preloaded with Opensolaris:
http://www.shopopensolaris.com/suntoshiba/home.htm
It was only by accident I actually found it. From what I understand they are fully supported given that they are also the backbone for a lot of the Sun employee’s.
Ehh…. I’m not so sure about that. It still makes sense for Oracle to guarantee that its existing products work well in all environments, not just its own. Linux has much higher market penetration than Solaris; Oracle would want to maximize its potential customer base and not turn its back on an existing, and very profitable business model. Stuff like BTRFS may seem like unrelated cutting-edge advances, but it makes sense if it makes Oracle’s products more competitive.
Perhaps you’re right though, and Oracle will focus more of its efforts on developing its own OS, and instead of helping to push the envelope in Linux, will only marginally encourage projects that are directly relevant to its business model.
Please inform me at what point in my post did I state that Oracle would cease providing products for Linux or supporting their own distribution of Linux? I’d love to know where you came to that conclusion given that I clearly stated ‘scale back contributions’ and not ‘cease contributions’ or ‘cease supporting products for Linux customers’.
Why continue a model which ships a good portion of the profit stream off to Red Hat or Novell when one can bring in the revenue derived from operating system sales under one roof. Oracle is all about making as much money off the software stack – the more of the stack you control the more money you can make off the customer.
You’re right – why push the envelop on Linux when they have an operating system of their own, have the human capital which knows the system inside and out, and a model when coupled with their middleware can provide a complete stack to the end user which are optimised for Oracle software.
What Solaris needs is more hardware support – that is about the only thing I can think of that is letting it down. If they get the hardware support under control (maybe taking my advice to hire 1,000 programmers who sleep, eat and breath writing drivers – and get them writing drivers non-stop to catch up to Microsoft) then they’ll over take Linux – until they do that Solaris will not be on the radar for people. Developers want to run Solaris but if their hardware isn’t fully supported they are going to develop software on that platform. Customers don’t want to be locked into a single vendor who says, “we’re only going to support the hardware we sell” because ultimately that is the policy that has undermined Sun’s software efforts in the past. You need to decouple your hardware and software sales so that your software sales aren’t just used to prop up hardware sales.
Edited 2009-04-20 23:23 UTC
With the purchase from Oracle of Sun, I wonder what will become of Java and will it continue to be an open platform supported by Oracle/Sun? Sun has so much you have to beg the question What Oracle will prop up and support, what will be converted to internal use only and what stays marketable and what will go away? I beleive this will have far reaching impacts than anyone recognizes and companies that rely on Java like Novell and others could be impacted. Sun uses much of the Unix code that Novell once owned. Now what? and will this be a re-birth of Sun so to speak to the benefit of Orachle or will Oracle strip mine the company and toss it away.
Just a couple of points:
1) Sun’s implementation of Java is fully open sourced with the stuff that were not able to be opensourced are being replaced with opensource replacements.
2) Sun doesn’t control Java – that is the biggest lie made by people on osnews.com – JCP control Java and NOT Sun. The only thing that this purchase will give Oracle is an extra seat at the table which hopefully should mean that things will move quicker through the pipe line.
I wish people would stop lying – there is no need to falsify things; either stick to the truth or don’t post at all.
It’s difficult to know with Oracle. They’re a strange company sometimes, and they have been known to be ruthless with anything they feel is not pulling its weight. However, if they do decide to put their faith in Solaris then they’ll have exactly the same headaches as Sun has had with it – maintaining it, increasing the userbase, pushing in new features, increasing hardware support etc. That’s not a small thing to do for anyone considering Solaris’s decline in the last ten years.
One thing’s for sure, it’s completely new territory for Oracle. We’ll have to see how they handle it.
All of the problems of Sun comes down to two major points: A lack of resources and a lack of leadership. Both Sun lacks in truck loads. If you fix up the hardware support by going out, pinning hardware vendors against walls and bribing who needs to be bribed to get things moving along – you’ll find that people will be able to try it out, test it, install it and use it not only for the server but as a development platform.
I’d use OpenSolaris if they improved the hardware support and created a community maintained repository which allowed one to easily download and install patent riddle CODEC’s just like any other distribution; the ability to be able to choose my hardware based on what I want rather than the narrow selection that Sun has chosen to support (which is based on what they sell – thus forces you to buy their hardware).
The server and the developer workstation are closely linked together and for far too long Sun has dismissed making Solaris a pleasant experience in favour of this approach of UNIX as usual. What is required is to make Solaris easy to use and easy to setup – the ability for a hardcore UNIX guru to come in and install it or for the IT person who knows the fundamentals can configure without needing to become a UNIX guru – thus lowering the cost not only up front but also the on going cost of a paid up IT admin.
It has been suggested in other places, mostly because Oracle uses Java for a lot of there stuff. For Oracle to have more control over Java might be in their interest.
“The acquisition of Sun transforms the IT industry” said Oracle CEO Larry Ellison.
ibm just snapped its fingers.
“Oracle will be the only company that can engineer an integrated system – applications to disk – where all the pieces fit and work together so customers do not have to do it themselves.
oh snap.
Edited 2009-04-20 12:50 UTC
PR: “The Sun Solaris operating system is the leading platform for the Oracle database, Oracle’s largest business, and has been for a long time.”
I never heard oracle say that before. I guess that gives away some of their intentions.
Sun Chairman Scott McNealy: “This combination is a natural evolution of our relationship and will be an industry-defining event.”
I love it. he doesn’t like ibm
Sun and Oracle had already said something like that in 2007; there was a bit of ruckus because of that when Sun acquired MySQL…
I don’t think they’re gonna put out Solaris anytime soon, like what you said. If they have a bit of better vision (which I like to expect from Schwartz), maybe they’ll come up with something better than the “java desktop” gimmick.
If they put out Solaris it’s to propagate Unbreakable, and that’s pathetic.
also I totally missed this until now, but there is zero mention of opensolaris. The word solaris is used, but opensolaris is nowhere. This is probably on purpose.
“opensolaris” means two or three different things according the the clusterf–k that is “opensolaris” governance. it means the open solaris code. it also means the community surrounding it. it also means a consumer-oriented operating system built by sun using the open solaris code.
the opensolaris code, community, and distribution are all managed and governed and built and contributed to by SUN employees. there are only a handful of non-sun employees who work under sun employees to contribute.
the summary is that none of this has to have anything to do with the solaris operating system oracle wants to run databases on. will oracle continue to develop sun’s ubuntu clone? I doubt it.
all the things named opensolaris are probably going the way of the dodo in favor of enterprise development. solaris doesn’t even run java very well, so that is the sort of thing oracle will focus on, not improving the gnome user experience.
not to mention sun workstations. dodo!
“solaris doesn’t even run java very well,”
Ive read several SUN blogs that Java is fastest on Solaris when benchmarked. That sounds reasonable, dont you think?
the solaris jre is not the fastest jre. next time don’t let the lion tell you where to put your head
also, a translation of the schwartz email:
Oracle’s interest in Sun is very clear – they aspire to help customers simplify the development, deployment and operation of high value business systems, from applications all the way to datacenters. By acquiring Sun, Oracle will be well positioned to help customers solve the most complex technology problems related to running a business.
That needs no translation.
A combined Oracle/Sun will be capable of cultivating one of the world’s most vibrant and far reaching developer communities, (he means Java) accelerating the convergence of storage, networking and computing, (he means solaris and sun servers) and delivering one of the world’s most powerful and complete portfolios of business and technical software (he means oracle software).
That jives perfectly with the releases from oracle. Everything else oracle got with the sun purchase is bonus or chaff.
The target they should be going after is something like IBM’s iSeries. Something that’s a one-stop shopping. I think with the advent of blade servers it’s a good strategy. A tightly integrated OS + DB running on a custom blade, just plug into somebody’s chassis and point it to some LUNs let it be an IP “black box” of Oracle goodness. Using only Oracle parts inside the software would be really easy to optimize and build into clusters. Unfortunately both Oracle and Sun are companies that let the engineering branch have too much pull and build overly complex things that require and extra college degree to make them work well. They have a serious “forest for the trees problem” going on to miss something like this.
If thats the case, its sad!
Just about time. With Postgres completely free who wants to be tied to MySql uncertain future?
Talk about BSD vs GPL once more.
Nothing here seem too terrible a database company dipping its toes into Linux now has an open source unix of their own. Also a programming language, hardware business, an office suite, and another database… A database available for free that competes with Oracles databases oh wait this could be ugly.
@skeletor MySQL is gpl it won’t go bye bye, but there is definitely a conflict here.
Also way to eke in there a minute before me.
Edited 2009-04-20 13:02 UTC
MySQL can’t compete with Oracle. With Oracle Express maybe. MySQL is a piece of crap conceived as an interim solution that got all the press and became famous for using the GPL and being able to make money out of it. The fact that it’s most used with PHP should tell you much about it’s qualities. Does it implement stored procedures and triggers finally?
Yes, MySQL supports stored procedures and triggers since version 5.0 released sometime in 2004.
Yes it can. MySQL has a commercial distribution for companies that require the best of the best. Many companies use MySQL. Obviously, there will always be large companies that will stick to Oracle, but as MySQL matures, it competes more and more with Oracle, when comparing feature to feature.
You’re crazy.
Yes, the two most popular database and scription applications that power some of the largest web sites: Yahoo, Digg, Facebook, MySpace, etc…
Of course, it’s a full-fledged relational database system.
For Sun MySQL as a product had a business value. Even though they did not make much/any money from it, it had a tactical place in Sun’s offerings and as such could count on Sun’s investment and support. For Oracle MySQL is nothing more than a competitor turned into bait for a real sell. (Even 11g ‘Express Edition’ is hardly a light weight entry level database. It’s more like a neutered version of the real thing. Still as top heavy, just less powerfull.)
Seriously, I feel an ‘OpenMySQL’ fork on its way…
You mean something like MariaDB, which has been started by the lead developer of MySQL and a co-founder of MySQL AB due mostly to his disagreement with the disastrous way that Sun has been driving the latest release of MySQL and that led the said developer to terminate his relationship with Sun Microsystems?
http://askmonty.org/wiki/index.php/MariaDB
Thanks, interesting read! Must spend more time keeping up with those things 🙂
Or even Drizzle (https://launchpad.net/drizzle).
This is the worst news I ever heard about.
Why we don’t see Google buying SUN and releasing everything as GPL ?
Google is an internet services company. Not a hardware company, not a software company. All software it produces is totally geared to be used in the “cloud”. It may not have any interest in hardware or server OS or desktop suite. Maybe in MySQL.
But Google maybe interested in
MySQL,
ZFS,
Java (to improvePython ?)
OpenSolaris (to improve his Linux ?)
Netbeans (To remake it to GWT Developer)
There was many technology in SUN to wonder it
Google doesn’t need to buy anything. They already have access to opensource technology and as per the license they don’t have to distribute anything as long as the binaries stay in their machines. So how do you know they aren’t already using ZFS or opensolaris code in Linux? The fact is ZFS can’t be integrated in Linux because then you have to distribute linux source. But you can integrate it inhouse without the world ever knowing. They could be using ZFS right now.
http://www.computerworld.com/action/article.do?command=viewArticleB…
i wonder what will happen with netbeans. oracle has it’s jdeveloper which is swing based and pure java technology. they are moving towards eclipse with their latest moves (eclipse enterprise pack for oracle) and it will be very interesting to see how this will end up. i have very negative opinion on oracle developer tools .. in my company we work with oracle forms, designer, reports and jdeveloper. they are all immature, beta quality software. i haven’t seen so much instability anywhere: used Visual Studio, Visual Basic 6, Real Basic, Borland, Eclipse, Netbeans. Jdeveloper is the king of instability ! Oracle has probably the best database in the market, but developer tools are pure crap.
i hope opensolaris will profit on this, and java community.
i wish they just give up with jdeveloper, and merge everything with netbeans or eclipse.
I really wished it would, but that’s NOT going to happen, according to their own words (asked this same question directly to lead developer…)
The Eclipse Enterprise pack is an interim solution for customers who still use the BEA Workshop suite. Jdev will be the official Oracle development platform for all future releases (and that includes R 13, Oracle Fusion).
Besides, it’s still faaaar away in the future. Not even Oracle herself has a date. Latest Oracle Apps server was released 2 years ago, and BEA Weblogic DOESN’T work out-of-thebox (with ADF libs, that is…)
That’s why we went with a Webshpere – Custom Eclipse install for all our Java development: far cheaper and available… right now.
And I’ve yet to see any “big” J2EE oracle app… It’s still PL/SQL and Forms for pretty much everyone.
same thing here. everybody talks about jdev and how great it is, but all i can se are those ugly Forms and tons of pl/sql (which i prefer not to use).
i somehow cant imagine websphere being cheaper than oracle solutions.
that Sun went down, they were my fav tech company.
Some really good products. Now Oracle will transform them into overpriced crap.
Maybe now that Oracle is committed in some way to Linux I can get Oracle 11g on OSX server again. I mean OSX server will have ZFS which Oracle will take advantage of, The will clearly add some Unix focus. I know I am dreaming but man that would be great!
this could be a very very good thing, or a nutral thing. I can’t see it being bad for sun and it’s technologies. I can even see Oracle keeping MySQL as an alternative. I mean if your customers are from both camps, why not rake in the profit from both.
As for Solaris, I can see IBM’s linux disapearing in the face of Solaris. Solaris is an entprise grade OS and its going ot be a lot easier to market that to their customers.
I am looking forward to a technology press anouncement from Oracle on it’s “plan” (currently at plan version 0.1b, but it’s getting there
I hope they’ll fix MySQL. One way or the other.
Oracles DB is enterprise solution while MySQL is for private&small businesses..they in no way competite with each other so i don’t see MySQL being threatened by Oracles acquisition of Sun.
Why is it that nobody mentions Java? IMHO that is the biggest reason for Oracle buying Sun. The rest is just icing on the cake. Oracle has a mature database platform, reasonable mature middleware, kind of mature developer tools, solid management tools, and now with Java control of the glue that holds everything together. Java is more important for Oracle than .Net is for Microsoft. Java is what brings together their collection of more or less independent technologies. And, just like with Microsoft’s .Net technology, Oracle having control over Java means it has a structural advantage over any other competitor in the enterprise arena.
With Sun’s assets on board, Oracle doesnt need to interoperate with anybody or anything anymore. (Which was never one of their big selling points anyway.) They now own everything for a wall-to-wall enterprise solution, for any price they fancy asking.
Enlighten me a bit, does MySQL already support:
1. ACID transactions
2. ‘full’ SQL syntax support. all types of joins, subqueries etc etc
3. efficient work with indexes
4. efficient support of big db’s (terabytes)
Last time I was interested in MySQL was somewhere in 2004. Then it was in disastrous state, not supporting subqueries (!!!) and transactions.
I wonder, what has changed since then?
If 1-4 are still not thoroughly supported, it cannot compete with Oracle AT ALL.
Hope this answer your question and more.
http://en.wikipedia.org/wiki/Comparison_of_relational_database_mana…
I prefer Postgres and if I need the big guns its easy to migrate to Oracle.
Hey Oracle!….. If you don’t like what you bought. Don’t destroy it…. set it FREE!!!! (GPL it)
Thing is that a majority of MS SQL or DB2 or Oracle purchase really only use MySQL level database functions. Unless you’re a professional DBA (which most people rolling their own apps aren’t) most people just don’t need much past MySQL… hell, they’re using MS Access!!!
I think the world revolves around the small pieces. That’s what made Microsoft their big pile of cash. The majority of businesses are small businesses, and they just won’t ever use apps past a certain complexity. I think with too many key pieces of the Open Source stack in the hands of big companies they will continue to not be developed to the customers that would actually want them and fall in to obsolescence.
Oracle is about as BIG as big businesses come in the IT space, and I just can’t see them managing the small projects very well. Not to mention the power they’ll have over IBM via Java. Oracle’s leaders in the past have been more than willing to use dirty tactics to get a leg up over other companies and are more than willing to subject their customers to any collateral damage it takes to get there.
While Sun was pretty much a good steward of the Java language, their business stuggles could be a real albatross around the neck of Java, in terms of having the resources to make Java reach it’s full potential, and in terms of people’s perception of Java as an viable on-going concern.
But with Oracle controlling it now, all that goes away. Oracle have deep resources, and are spectalarly successful business, even in this horrible economy.
The question now is,l with Oracle being primarily concerned with enterprise middle ware, what happens with client Java. I’m sure JavaFX is dead, but what happens with Swing, or other client side stuff? Does Oracle see the value in also having compelling client side offerings (to help spur on server side middleware growth)? What happens to NetBeans?
We’ll see.
Swing is far from dead, the entire Oracle DB client side toolchain relies on it. In fact I would have been worried more about Swing if IBM had bought Sun!
JavaFX I am not sure but I assume also it will be dead, it is not Oracles business and it was too late to the table anyway and too slow, it might find its nieche in set top boxes however, Blue Ray for instance might be a new home for it…
JSF probably will have a bright future, the 2.0 spec is very good and JSF has been Oracles baby anway while Sun always treated it more like a stepchild dedicating only a few resources towards it because they had to.
My biggest concern (I am not the least concerned regarding MySQL, I always hated it) is NetBeans which is in direct collission with JDeveloper…
Netbeans has become a tremendously good ide over the last releases and has started to show its full potential with 6.5…
As for OpenOffice and Solaris, it probably will be business as usual probably Solaris will get more resources now…
I’ve tried netbeans multiple times over the last five or six years. Every time, it strikes me as one f-ing pile of crap slow-ass program.
The last time I ran it was 6.5 on my Pentium D system on my desk at work (with gobs of RAM). It was so painful to use that I went back to JEdit, and drew my own UML diagrams with dia. Doing that was faster than futzing with NetBeans.
“The acquisition of the sun transforms the IT industry, combining best-in-class enterprise software and mission-critical computing systems,” said Oracle CEO Larry Ellison. “Oracle will be the only company that can engineer an integrated system and power it indefinitely by harnessing the seemingly infinite power of our nearest star. Our customers benefit as their power bills and systems integration costs go down while system performance, reliability and security go up. We can’t imagine a better perimeter security appliance than one with a surface temperature of 5,778 degrees Kelvin. We are also on the verge of announcing a deal whereby our entire staff will be entitled to a free lifetime supply of sunscreen lotion.”
“On April 20, 2009, Oracle announced it has entered into an agreement to acquire Sun Microsystems (Sun). The proposed transaction is subject to Sun stockholder approval, certain regulatory approvals and customary closing conditions. Until the deal closes, each company will continue to operate independently, and it is business as usual.
The acquisition combines best-in-class enterprise software and mission-critical computing systems. Oracle plans to engineer and deliver an integrated system—applications to disk—where all the pieces fit and work together so customers do not have to do it themselves. Customers benefit as their system integration costs go down while system performance, reliability and security go up.”
If that is true then linux at Oracle is meaningless. Oracle develops linux because it needed something to compete with IBM. If Oracle is building integrated solutions now it already has enterprise class solution in Solaris.
Oracle will still continue Linux development but I doubt they would kill Solaris. The SPARC throughput chips do well for DB loads.. and the x86 stuff is the same industry standard kit that IBM, Dell and HP sell. Interesting times ahead.
For the last few years Linux has been the leading OS release for Oracle 11g and 10g updates in terms of what is made available by Oracle relative to timing(Linux releases first, others including Solaris later…
Current versions of Oracle have less problems, from an OS standpoint on Red Hat Linux than Solaris. I have Oracle 10g and 11g instances on both Solaris and RHEL. I would not be suprised to see a Solaris EOL statement by the end of the year with conversion assistance to Oracle Unbreakable Linux/RHEL..
Larry Ellison says in this press release that Solaris is more deployed on Solaris + SPARC than any other OS. He had also declared (years ago) that Solaris is the main prefered target OS for Oracle DB.
At Oracle the development and base platform where the first release occurs is Linux since 2005. Solaris is a secondary product line like Windows or AIX. That is the reality for the last several years.. Oracle 11G was first available for Linux months ahead of Windows followed later by Solaris. Even with the Sun acquisition I don’t see the trend of movement from Solaris to Linux in the customer base changing overnight. I would suspect the direction of Oracle on Solaris to be made clearer at OracleWorld in SF this Fall..
Does anyone know if Sun ever settled the lawsuit that they had with NetApp over the ZFS file system? I believe there was some controversy over who actually owned the rights to that piece of technology.
This could probably mean that Oracle will go after Microsoft?
I really hope Oracle integrates its products with Sun’s, and deliver a seamless experience that will enable them to challenge Microsoft in the enterprise market.
One thing the IT in my workplace love about going all Microsoft, is that all things work together.
What will happen to Sun’s (wasteful) open source effort?
I liked how a lot of Sun’s product were FOSS. Will Oracle keep them as is, spin them off?
Oracle needed servers. They found it cheaper to by Sun and let them sell a bunch of servers to Oracle at rock bottom prices. They then turn around and take a huge loss against Sun when they slowly trash it to pieces. They then sell off what is left.
I am an employee of Sun and now our local office dont even know if we will still have a job once the transaction is finalised. We are just finishing up RIFFS and still may lose more thanks to this acquisition.
Im in admin and do not want to lose my job as its the admins that ensure offices run day to day.
Spare a thought for all staff who are unsure where they stand
It’s really sad when big and significant corporations such as Sun Microsystems go this path. I guess all good things come to an end. Imagine all the innovations Sun brought us: Java, SPARC, Sun OS/Solaris to name a few…
Edited 2009-04-21 11:26 UTC