So, you go around claiming that switching from Linux to Solaris reduces memory usage from 200MB to 50MB, and then you admit that you can't even read the VM stats properly. That's pretty dishonest. Different tools on linux were giving me different numbers. I used the smallest one. It wasn't something I was really testing, just something that I was surprised to see and it was on my mind as the transfer was recently.
What difference does it make? That's not where Google spends their money. You claimed that OpenSolaris brings big performance gains, gains that would save Google hundreds of millions of dollars a year if they were true, yet Google sticks with Linux for their compute platform. Since Google isn't stupid, that's just another indication that performance claims for Solaris are bogus and that it does not perform significantly better than Linux. That may not be where they spend their money but it is where they make their money. I claimed I got better performance and memory usage in this one application. That has nothing to do with Google.
Anyway, I'm done debating the issue.
I think Sun was resting on it's laurels for a while but because of Linux, MS, x86, things have changed. With people like Andy Bechtolsheim and Richard Green coming back, Ian Murdock joining Sun (I guess Bill Joy is happy being a VC) it looks like things have been changing. Although not completely. How a server with 48 drive bays is not considered a storage product is weird but Bechtolsheim doesn't seem to care. He makes the products he thinks make sense and doesn't care what they are labeled.
Marketing is marketing anywhere. Sun's not alone. Even linux proponets promised more than they could deliver. Since 2.0 and 2.2 people were claiming it was as good as Unix and could be more than just an edge server. 2.2 the kernel was less than 2mil lines of code, 2.0 wasn't even a million. Now it's at around 10 million. You're going to tell me that's 8 million lines of device drivers and innovation and not catching up to provide the Unix features they were saying the already had? Not knocking Linux, it's come a long way.
I did not feel a particular need write a treatise intricately comparing the pros and cons of all plausible energy sources. Well if you're not part of the solution, you're part of the problem!:)
Except that you're missing one point: Novell claims that SCO never had the authority to make the deal that it did, and they so informed Sun last year once they got hold of the agreement between SCO and Sun. And you're missing a few big points as well. Except for one comment by McNeally, who is known to say weird things, Sun's position has always been that it's right to open Solaris source code wasn't what the deal was about. They had broad rights to SYSVR4 code for a long time. Remember, they worked on SYSVR4 with AT&T long before Novell purchased USL.
In fact, the Solaris code was opened under the SCSL in 1999 and I don't remember Novell saying anything.
What Sun was lacking was support on intel and amd hardware. UnixWare was the only Unix to have really good i386 support. To be able to offer Solaris on x86 they needed drivers from UnixWare and if you search most of the news stories I have seen regarding SCO/Sun's deal it talks about drivers. Solaris was heads and shoulders above UnixWare other than in it's x86 support so I doubt that anything in SYSVR4 code that Sun would want that they already didn't have.
What Novell is doing in regards to OpenSolaris, is basically what SCO was trying to do to Linux. They're just trying to make waves and scare people.
Really, lets look at SCO and Novell closely. SCO was a company that had a legacy (UnixWare) product and a linux product. They're legacy revenues were declining and their linux product wasn't generating much interest and revenue.
Now lets look at Novell. They had a popular set of products, NetWare and derivatives that are still their biggest source of their income. Quarter after quarter revenue from these products is steadily declining. They have a linux product, which they need to compete against RedHat but doesn't have anywhere near the market and really hasn't shown any signs of getting anywhere near the acceptance RedHat has. We're talking server deployments here, not the miniscule linux destop market. If I remember correctly they're making around $12mil per quarter or year from SuSE while RedHat is making 10x that. They wouldn't have even been able to buy SuSE since Sun was bidding for it as well until IBM threw them $50mil. I wish I could find that chart that showed RedHat vs SuSE deployments. If you didn't look close you couldn't even see that SuSE was on the chart.
Novell has no moral right, in addition to legal right, to demand anything from Sun. What the hell has Novell contributed to open source? Have they open sourced a lot of their products? No. They tried Hula but that died. They bought SuSE and Ximian but most people think they bought SuSE because IBM couldn't buy it and they didn't want Sun to buy it. Look at this rebort from the EU that shows Sun has the most contributions to Debian. Novell's not even on that table. Ximian is but it's minor
If it wasn't for the whole SCO thing, most people in the open source world wouldn't even know about Novell.
Ray Noorda was behind the Canopy Group, which funded SCO. The whole SCO, Santa Cruz Operations, Caldera thing can be confusing, but Noorda plays a role in Caldera and Canopy. Noorda was the founder of Novell and CEO for a long time. Noorda's companies have a history of purchasing IP from Novell and using it to sue mainly Microsoft. More details are on Wikipedia's page for Noorda. That seems to me like Novell is trying to keep it's hands clean while Noorda's company does the dirty work. So now, when Caldera/SCO/SantaCruz/Canopy tries to sue with Novell IP we're supposed to think nothing of it? And recently Novell has this deal with MS?
Though let me say, I mean no disrespect against Noorda (RIP), his target was mainly MS. I remember reading that he left Novell because they decided to pursue Unix as a way to save NetWare instead of Linux. When he left Noorda focussed on Linux.
I consider myself an interested outsider. When I first heard about OpenSolaris, what I expected was an Open Source distribution of Solaris. I think OpenSolaris makes sense for the distribution.
Maybe they should have named the kernel/userland tools something else. But it seems that can't be undone now.
And that doesn't make you the least bit suspicious? Are you naive enough to think that the use of Solaris gives you a fourfold reduction in memory usage in your web server because... it has a better kernel? Don't you think Google would switch over all their machines to OpenSolaris in a jiffy if they could shave off a few percent in memory and CPU usage, let alone a factor of 4? Doesn't really make me suspicious. For one, the Linux number may be inflated because of the way Linux treats threads as procs and some of the tools used to gauge memory usage don't know how to tell the difference. The memory usage on Windows isn't too far off from what I get in Solaris. Since Sun works on Java and Solaris they can work on making things run better between the two. Going from Solaris 9 to Solaris 10 showed improved Java performance for me. I think DTrace might have really helped them out there.
As far as I know, Google doesn't use Java for it's search platform. They do use Solaris and I believe Java for parts related to their billing for AdWords or something like that. Java is also used for some of their other applications. There were also some rumors a couple years ago that they were testing out OpenSolaris. Regardless, the app I'm running is using some specific libraries that I doubt Google, or anyone else uses that much anymore.
SPARC has it's performance problems, but Sun's SPARC systems offer some advantages that x86 hardware can't. Sun x86 line is pretty good and benchmarks well against similar hardware from HP, Dell, IBM etc from the reviews I've seen. In Some cases Solaris 10 is faster than RHEL and SuSE. In some cases it's not.
While Sun is putting new features into Solaris that get a lot of press, it doesn't mean that the core components, including the kernel are not getting improvements.
Sun's marketing is no different than Linux fans that were claiming it was better than commercial Unix kernels back at a time when that was far from the truth. I don't follow kernel development much but there was a time in the past where I was curious what was going on. And I remember a lot of comments in the lkml that basically something like Solaris seems to do this better, how can we do it like them.
NFS is widely used, I know that linux had some bugs related to it.
Java is also widely used and accepted, while it has it's criticisms, it has come a long way.
Architecturally, the Linux kernel isn't going to win any prizes. But at least the Linux developers concentrate on stuff that matters, as opposed to Sun engineers, who are trying to set themselves yet another monument in software. As for the kernel, linux developers concentrate on what matters to them and their users and sun developers on what they feel is important for theirs.
Why "rather than"? Because you can't have your cake and eat it too. The law does not allow you to be remedied twice for the same act.
Just like in the example where a salesperson sells something they shouldn't have sold, the store owner cannot get the product back and keep the money from the sale.
Sun already bought out their rights to Unix so if what Novell is saying is true, SCO coudldn't sell them anything. So if Novell's claim of SCO exceeding their authority is true and the deal is invalid, then Sun didn't get anything it didn't already have. If that's the case, how can Novell get something for nothing? Again, Sun along with AT&T developed SVR4.
Novell has said publicly that they will not be pursuing their Unix copyrights. I posted a link to that previously.
As an extreme example, if I sold you the Statue of Liberty (which I don't own) do you think you'd get to keep it? No. But if the state of New York tells me you are authorized to sell certain State properties such as landmarks... it's quite a different story.
If no-one had complained do you think Sun would have changed their mind? Uhm. Sun didn't change it's mind, Mickos SVP of databases former CEO of MySQL changed his mind. Or more likely Sun helped him change it to align with the corporate open source strategy.
Honestly, I hope Sun doesn't change the ZFS licence, because in another couple of years I expect we [zumastor.org] will have been able to achieve pretty much everything ZFS does without having to roll it all into one jinormous all singing+dancing filesystem that ate the OS. Good, hopefully the ZFS guys will read this and hold off their development for the next 2 years until you catch up.
You have to trade that off against the costs: licensing, porting, incompatibilities, potential bugs. For most people, the tradeoff never made sense. Sun's compiler tools are now available for free use on Solaris with registration so cost isn't a factor anymore. Porting I can't comment on. Incompatibilities and potential bugs are true for gcc as well. C and C++ have specifications behind them to help minimize these issues.
I don't develop compiled apps c/c++ apps that get distributed. All I ever compile is software that is installed and run on the servers I'm using. So maybe I'm not getting the difference. For me, having my local apps run faster is a good thing.
And it's the same with Solaris, DTrace and ZFS: no to small benefit for most people, some design problems, big switching costs, and big long-term risks. Design problems, big switching costs and big long term risks could be applied to people thinking of switching from Solaris to Linux. And there are still quite a few people out there that are still running Solaris. Solaris 10 has helped people stay with Solaris.
For me, the reason I like Solaris is that I've found I get less memory usage and better performance out of Apache/Tomcat. One webapp that was running well over 200MB on RedHat now runs in less then 50MB of memory on the Solaris setup under load. The performance benefit wasn't negligible either. A few years ago I remember Linux knocking the socks off more expensive Sparc hardware with the same java apps. Haven't tested any recent sparc systems but on x86 Solaris has been doing better than Linux for my apps.
I've only really been testing with Solaris 10 but have been playing with OpenSolaris more and more. I'm posting this right now from OpenSolaris running in VirtualBox.
I don't like Solaris because it has DTrace, ZFS, Crossbow, Zones, etc. I like it because it's developed by people that can come up with those technologies. They also have some pretty neat x86 hardware. I never really liked IBM servers, I always regarded the Proliant line well and I think Sun's x86 gear is on par with that.
One thing I noticed, is with other Linux Distros and even the most recent Ubuntu LiveCD, the fonts in Firefox have been horrible. What I'm seeing here now is nice and easy to read in comparison.
Anyway, we're not going to agree on this but as an aside, I never understood the animosity Linux proponents had towards Solaris. If you do a search of lkml you'll find more references to Solaris than AIX, IRIX or HPUX combined. And not all of them negative. The EU study that showed Sun was the biggest corporate contributor to Debian put Sun's Code in Debian to about 26%.
You should read my response above to the previous reply.
In addition let me add some history of SYSVR4 which explains why suing Sun over SYSVR4 is so funny.
In 1988 I believe, AT&T and Sun got together to work on SYSVR4. SunOS was primarily based on BSD at the time and a lot of BSD bits went into SYSVR4. Bill Joy, a founder of Sun was also a leading developer of BSD. AT&T and Sun handed off the rights to a seperate entity, Unix Laboratories, to handle licensing so that others can implement it. Sun always pushed for open standards, before open source was that popular.
Solaris is based on SYSVR4. You can see the relationship here. Solaris developed on it's own after it's SYSVR4 base. As did UnixWare. What Sun needed for Solaris 10 was better x86 support. SYSVR4 did not provide that. UnixWare did. At the time UnixWare had a large i386 deployment and good drivers to support that platform. So it makes sense that if Solaris got anything from SCO it related to UnixWare. Other than the x86 support, I don't think anyone can argue that UnixWare was better than Solaris.
Imagine if somewhere down the line, someone sued Linus over his use of Linux.
The problem the APA has is that it gives different treatment to UnixWare than it does to SYSVR4. Basically, SCO can do what it wants with UnixWare and Novell's lawyers are trying to separate the UnixWare parts from SYSVR4 which doesn't make much sense. It's like if you buy a source license for Windows 3.1 which is built on Windows 3.0 but you don't have rights to the code in Windows 3.1 that is from Windows 3.0.
Another interesting thing, the thing I think has Novell, IBM, and HP worried, is that in that simplified Unix History Tree. Solaris is the only Unix still on there. AIX, HP/UX, Irix are all gone. That indicates to me that Solaris still has value. UnixWare I believe is still on there because of the SCO trial. The original Unix History Tree is a mess and unreadable but you can google it and find it.
Novell and some others would like to get Solaris off the map because they think it would help Linux. To me that's a very short sighted view. I like what Jonathan Schwartz has to say on that issue.
I disagree and I'll leave it at that. Except to say that it would have made more sense for Novell to claim that SCO is not entitled to collect that money rather than SCO owes them the 95%.
Anyway, according to Schwartz in a 2003 interview he said:
eWEEK: Some critics are saying that its not just Microsoft funding SCO but also Sun, citing the fact that you acquired another license from them recently, received warrants to buy shares in SCO and are losing the most customers in the migration from Unix to Linux. It thus makes enormous sense for Sun to fund SCO, their logic goes. How do you respond to that?
Schwartz:We took a license from AT&T initially for $100 million as we didnt own the IP. The license we took also made clear that we had rights equivalent to ownership. When we did the deal with SCO earlier this year we bought a bunch of drivers and when we give money to a company oftentimes we get warrants, which is part of the negotiations. I have warrants in 100 different companies, we have a huge venture portfolio. I cant do anything about the perception thats out there and to be blunt, I dont care as those people arent going to drive our futureâ"customers are. Which makes sense considering Sun developed SYSVR4 with AT&T.
From this article:
"We're not interested in suing people over Unix," Novell spokesman Bruce Lowry said. "We're not even in the Unix business anymore." How would it look for Novell, to sue Sun over an open source project? Especially considering Sun is the biggest corporate open source donor. By the way, the article doesn't specify, but if you read the study, those numbers are from the source of Debian. They do not include OpenSolaris, OpenJDK or any of the other open source projects Sun participates in or has released.
While I do believe McNealy and Schwartz shared similar views, McNealy's mouth did tend to get in the way. As we've seen here Schwartz is sticking to his commitment to open source.
It looks to me like Novell is talking about OpenSolaris to bolster their case. Novell's CEO has also been spreading a lot of FUD about OpenSolaris.
I don't work for sun but I've been playing around with zones a bit and found a lot of useful information on zones. Haven't looked into dtrace much but I did find some good references online. Google for one line dtrace scripts as a start.
As for zones. I found this page useful for describing sparse root and whole root zones and how to implement them. Before I really understood what I was doing I had a zone running.
SuSE isn't that popular? In your personal little universe maybe. As far as I can see they are top three on distrowatch, above Fedora IOW, with the latest release 6 months old or so. And they have lots and lots of users, even though you don't hear much from them. German parliament being one of those. Distrowatch tracks hits to it's website to get those results. If you don't believe me then this might be a source you respect. http://www.groklaw.net/articlebasic.php?story=20070111043418436
Now, I'm not knocking Linux as a desktop but I think it is fair to say that the real value of Linux, at least where money is changing hands today, is on the server. And the number of server installs greatly outnumber the desktop installs. RedHat has always dominated server linux installs. I wish Netcraft kept freely publishing statistics like this as they are very interesting. When companies are talking about deploying Linux, they generally mean RedHat. At least here in the US.
And even if I'm wrong about SuSE's popularity, Novell hasn't monetized it the way RedHat has. Compare the two companie's quarterly reports to verify.
Novell said they have no interest in pursuing Unix copyrights.
Novell is trying to get their 95% portion of the license Sun paid to SCO. By saying the agreement between Sun and SCO was part of the APA between Novell and SCO they are affirming the deal between Sun and SCO. Sun actually helped write SYSVR4 with AT&T before Novell bought it. According to Schwartz, Sun paid AT&T about $100million for rights that basically gave them ownership. What was purchased from SCO were mainly device drivers since SCO's UnixWare had the best x86 support.
What is Novell's position going to be to the public? "We're an open source company but we're going to sue a company for releasing open source?" Nothing good can come to Novell if they challenge Sun.
This has all the hallmarks of a classic PR maneuver - Sun wants to figure out how they can extract more $$ from the high end users of MySQL. They need to find out how the market will react if they start selling closed source MySQL extensions without committing themselves if it goes horribly wrong. So they sprinkle some unsubtantiated vague rumours around and look for the reaction. The reaction was: PostgreSQL. So now they can kill the whole idea with minimal losses and try their next plan for how to "monetize" MySQL some more without pissing off their entire user base and killing the golden goose.
I don't believe for a second that things like this are an accident. These folks are far too smooth to just accidently let this kind of thing drop and run for a week.
I doubt that. When MySQL AB was a separate company and interested in an IPO, they had to find ways to boost their revenues.
While MySQL makes money on support and some of the "Enterprise" tools, their main source of revenue is from companies like Cisco that want to buy a non-GPL license to MySQL to embed it in one of their products. Since MySQL AB offered MySQL in dual licenses, it gives companies that want to include it in their projects, and possibly extend or modify it the ability to keep their changes to themselves and not have to GPL their own work.
Now that MySQL is part of Sun, it doesn't need to generate the same revenue from MySQL directly. Seems that MySQL's management didn't get this. But to be fair, they never specified that the source would be closed. They mentioned it could be available under the GPL but they haven't decided. What they said was that it would be in MySQL Enterprise and not MySQL Community.
No decision had been made. This announcement is the first actual decision on the subject. Baloney. The former CEO of MySQL even posted otherwise right here on slashdot:
The decision was made and then was reversed. Read your own link:
Additionally we will develop high-end add-ons (such as encryption, native storage engine-specific drivers) that we will deliver to customers in the MySQL Enterprise product only. We have not yet decided under what licence we will release those add-ons (GPL, some other FOSS licence, and/or commercial)
An open source company wants to close some of its new features. The "proprietary" software company that bought them wants them to keep everything open.
Somehow, everyone wants to paint the proprietary company in a bad light. The original blog post from the first story never even mentioned Sun but the title on Slashdot was about Sun closing MySQL.
Sun's management has MySQL change that decision and the headline is about MySQL reverting.
There's an obvious bias here that's laughable.
Mickos is the one that needs to adapt here. Though I personally don't think it's fair to give him a hard time about it. I mean the guy just closed a big deal, got a ton of money and needs to do a different job than he was doing for the last 7 or so years. If he does the same stuff in a couple more months, that's a different story.
And this wasn't a "public announcement" it was at a partner meeting. Which is a bit different. And nothing was actually released or finalized, it was just a roadmap to let partners know what to expect. Sometimes these things change but people made a big deal over it.
Most large acquisitions have their hiccups. That shouldn't come as any suprise.
"Pushing" against whom? MySQL ceased to exist as a separate entity once it was acquired. You know that companies are run by people right? Their not some big robot or computer program. People need time to adjust and get familiar with the new vision of their new company. You don't right click on MySQL AB, select Refactor and expect everything to just change.
Certain initiatives that were started pre buyout continued. When it was detected that those initiatives weren't inline with Sun's plans, it was corrected.
All the 400 or so employees that were with MySQL are now with Sun and they need to get used to how being part of Sun frees them from increasing direct revenues from MySQL.
Sun buys an open source company and doesn't force them to change their business practices. Doesn't sound so bad. And when it does get them to change their business practices, it results in being more open.
"Company forced to give up revenue stream due to open-source fanatics who refuse to acknowledge any boundary between open-source MySQL server APIs and closed-source enterprise utilities which call those APIs"
Despite the outcome, this is not a victory for the open-source movement. The original Slashdot story was inflammatory and designed to mislead, and now it has had the desired effect. MySQL AB needed to generate revenue directly from MySQL as that was pretty much their only product. They were looking for an IPO before Sun bought them so they needed to increase revenues.
Being part of Sun, MySQL doesn't have the same pressure to generate revenues directly from MySQL. Sun/Schwartz's plan is to drive revenue in Sun's other lines from MySQL. Hardware sales, support, etc.
Click the link and read more
* MySQL Server is and will always remain fully functional and open source,
* so will the MySQL Connectors, and
* so will the main storage engines we ship.
In addition:
* MySQL 6.0â(TM)s pending backup functionality will be open source,
* the MyISAM driver for MySQL Backup will be open source, and
* the encryption and compression backup features will be open source,
where the last item is a change of direction from what we were considering before.
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.
Slashdot Mirror
Anyway, I'm done debating the issue.
I think Sun was resting on it's laurels for a while but because of Linux, MS, x86, things have changed. With people like Andy Bechtolsheim and Richard Green coming back, Ian Murdock joining Sun (I guess Bill Joy is happy being a VC) it looks like things have been changing. Although not completely. How a server with 48 drive bays is not considered a storage product is weird but Bechtolsheim doesn't seem to care. He makes the products he thinks make sense and doesn't care what they are labeled.
Marketing is marketing anywhere. Sun's not alone. Even linux proponets promised more than they could deliver. Since 2.0 and 2.2 people were claiming it was as good as Unix and could be more than just an edge server. 2.2 the kernel was less than 2mil lines of code, 2.0 wasn't even a million. Now it's at around 10 million. You're going to tell me that's 8 million lines of device drivers and innovation and not catching up to provide the Unix features they were saying the already had? Not knocking Linux, it's come a long way.
You should read this thread, it has more information http://in.opensolaris.org/jive/thread.jspa?messageID=146731𣴫
In fact, the Solaris code was opened under the SCSL in 1999 and I don't remember Novell saying anything.
What Sun was lacking was support on intel and amd hardware. UnixWare was the only Unix to have really good i386 support. To be able to offer Solaris on x86 they needed drivers from UnixWare and if you search most of the news stories I have seen regarding SCO/Sun's deal it talks about drivers. Solaris was heads and shoulders above UnixWare other than in it's x86 support so I doubt that anything in SYSVR4 code that Sun would want that they already didn't have.
What Novell is doing in regards to OpenSolaris, is basically what SCO was trying to do to Linux. They're just trying to make waves and scare people.
Really, lets look at SCO and Novell closely. SCO was a company that had a legacy (UnixWare) product and a linux product. They're legacy revenues were declining and their linux product wasn't generating much interest and revenue.
Now lets look at Novell. They had a popular set of products, NetWare and derivatives that are still their biggest source of their income. Quarter after quarter revenue from these products is steadily declining. They have a linux product, which they need to compete against RedHat but doesn't have anywhere near the market and really hasn't shown any signs of getting anywhere near the acceptance RedHat has. We're talking server deployments here, not the miniscule linux destop market. If I remember correctly they're making around $12mil per quarter or year from SuSE while RedHat is making 10x that. They wouldn't have even been able to buy SuSE since Sun was bidding for it as well until IBM threw them $50mil. I wish I could find that chart that showed RedHat vs SuSE deployments. If you didn't look close you couldn't even see that SuSE was on the chart.
Novell has no moral right, in addition to legal right, to demand anything from Sun. What the hell has Novell contributed to open source? Have they open sourced a lot of their products? No. They tried Hula but that died. They bought SuSE and Ximian but most people think they bought SuSE because IBM couldn't buy it and they didn't want Sun to buy it. Look at this rebort from the EU that shows Sun has the most contributions to Debian. Novell's not even on that table. Ximian is but it's minor
If it wasn't for the whole SCO thing, most people in the open source world wouldn't even know about Novell.
Ray Noorda was behind the Canopy Group, which funded SCO. The whole SCO, Santa Cruz Operations, Caldera thing can be confusing, but Noorda plays a role in Caldera and Canopy. Noorda was the founder of Novell and CEO for a long time. Noorda's companies have a history of purchasing IP from Novell and using it to sue mainly Microsoft. More details are on Wikipedia's page for Noorda. That seems to me like Novell is trying to keep it's hands clean while Noorda's company does the dirty work. So now, when Caldera/SCO/SantaCruz/Canopy tries to sue with Novell IP we're supposed to think nothing of it? And recently Novell has this deal with MS?
Though let me say, I mean no disrespect against Noorda (RIP), his target was mainly MS. I remember reading that he left Novell because they decided to pursue Unix as a way to save NetWare instead of Linux. When he left Noorda focussed on Linux.
I consider myself an interested outsider. When I first heard about OpenSolaris, what I expected was an Open Source distribution of Solaris. I think OpenSolaris makes sense for the distribution.
Maybe they should have named the kernel/userland tools something else. But it seems that can't be undone now.
As far as I know, Google doesn't use Java for it's search platform. They do use Solaris and I believe Java for parts related to their billing for AdWords or something like that. Java is also used for some of their other applications. There were also some rumors a couple years ago that they were testing out OpenSolaris. Regardless, the app I'm running is using some specific libraries that I doubt Google, or anyone else uses that much anymore.
SPARC has it's performance problems, but Sun's SPARC systems offer some advantages that x86 hardware can't. Sun x86 line is pretty good and benchmarks well against similar hardware from HP, Dell, IBM etc from the reviews I've seen. In Some cases Solaris 10 is faster than RHEL and SuSE. In some cases it's not.
While Sun is putting new features into Solaris that get a lot of press, it doesn't mean that the core components, including the kernel are not getting improvements.
Sun's marketing is no different than Linux fans that were claiming it was better than commercial Unix kernels back at a time when that was far from the truth. I don't follow kernel development much but there was a time in the past where I was curious what was going on. And I remember a lot of comments in the lkml that basically something like Solaris seems to do this better, how can we do it like them.
NFS is widely used, I know that linux had some bugs related to it.
Java is also widely used and accepted, while it has it's criticisms, it has come a long way. Architecturally, the Linux kernel isn't going to win any prizes. But at least the Linux developers concentrate on stuff that matters, as opposed to Sun engineers, who are trying to set themselves yet another monument in software. As for the kernel, linux developers concentrate on what matters to them and their users and sun developers on what they feel is important for theirs.
Just like in the example where a salesperson sells something they shouldn't have sold, the store owner cannot get the product back and keep the money from the sale.
Sun already bought out their rights to Unix so if what Novell is saying is true, SCO coudldn't sell them anything. So if Novell's claim of SCO exceeding their authority is true and the deal is invalid, then Sun didn't get anything it didn't already have. If that's the case, how can Novell get something for nothing? Again, Sun along with AT&T developed SVR4.
Novell has said publicly that they will not be pursuing their Unix copyrights. I posted a link to that previously.
I don't develop compiled apps c/c++ apps that get distributed. All I ever compile is software that is installed and run on the servers I'm using. So maybe I'm not getting the difference. For me, having my local apps run faster is a good thing. And it's the same with Solaris, DTrace and ZFS: no to small benefit for most people, some design problems, big switching costs, and big long-term risks. Design problems, big switching costs and big long term risks could be applied to people thinking of switching from Solaris to Linux. And there are still quite a few people out there that are still running Solaris. Solaris 10 has helped people stay with Solaris.
For me, the reason I like Solaris is that I've found I get less memory usage and better performance out of Apache/Tomcat. One webapp that was running well over 200MB on RedHat now runs in less then 50MB of memory on the Solaris setup under load. The performance benefit wasn't negligible either. A few years ago I remember Linux knocking the socks off more expensive Sparc hardware with the same java apps. Haven't tested any recent sparc systems but on x86 Solaris has been doing better than Linux for my apps.
I've only really been testing with Solaris 10 but have been playing with OpenSolaris more and more. I'm posting this right now from OpenSolaris running in VirtualBox.
I don't like Solaris because it has DTrace, ZFS, Crossbow, Zones, etc. I like it because it's developed by people that can come up with those technologies. They also have some pretty neat x86 hardware. I never really liked IBM servers, I always regarded the Proliant line well and I think Sun's x86 gear is on par with that.
One thing I noticed, is with other Linux Distros and even the most recent Ubuntu LiveCD, the fonts in Firefox have been horrible. What I'm seeing here now is nice and easy to read in comparison.
Anyway, we're not going to agree on this but as an aside, I never understood the animosity Linux proponents had towards Solaris. If you do a search of lkml you'll find more references to Solaris than AIX, IRIX or HPUX combined. And not all of them negative. The EU study that showed Sun was the biggest corporate contributor to Debian put Sun's Code in Debian to about 26%.
You should read my response above to the previous reply.
In addition let me add some history of SYSVR4 which explains why suing Sun over SYSVR4 is so funny.
In 1988 I believe, AT&T and Sun got together to work on SYSVR4. SunOS was primarily based on BSD at the time and a lot of BSD bits went into SYSVR4. Bill Joy, a founder of Sun was also a leading developer of BSD. AT&T and Sun handed off the rights to a seperate entity, Unix Laboratories, to handle licensing so that others can implement it. Sun always pushed for open standards, before open source was that popular.
Solaris is based on SYSVR4. You can see the relationship here. Solaris developed on it's own after it's SYSVR4 base. As did UnixWare. What Sun needed for Solaris 10 was better x86 support. SYSVR4 did not provide that. UnixWare did. At the time UnixWare had a large i386 deployment and good drivers to support that platform. So it makes sense that if Solaris got anything from SCO it related to UnixWare. Other than the x86 support, I don't think anyone can argue that UnixWare was better than Solaris.
Imagine if somewhere down the line, someone sued Linus over his use of Linux.
The problem the APA has is that it gives different treatment to UnixWare than it does to SYSVR4. Basically, SCO can do what it wants with UnixWare and Novell's lawyers are trying to separate the UnixWare parts from SYSVR4 which doesn't make much sense. It's like if you buy a source license for Windows 3.1 which is built on Windows 3.0 but you don't have rights to the code in Windows 3.1 that is from Windows 3.0.
Another interesting thing, the thing I think has Novell, IBM, and HP worried, is that in that simplified Unix History Tree. Solaris is the only Unix still on there. AIX, HP/UX, Irix are all gone. That indicates to me that Solaris still has value. UnixWare I believe is still on there because of the SCO trial. The original Unix History Tree is a mess and unreadable but you can google it and find it.
Novell and some others would like to get Solaris off the map because they think it would help Linux. To me that's a very short sighted view. I like what Jonathan Schwartz has to say on that issue.
Anyway, according to Schwartz in a 2003 interview he said: eWEEK: Some critics are saying that its not just Microsoft funding SCO but also Sun, citing the fact that you acquired another license from them recently, received warrants to buy shares in SCO and are losing the most customers in the migration from Unix to Linux. It thus makes enormous sense for Sun to fund SCO, their logic goes. How do you respond to that?
Schwartz:We took a license from AT&T initially for $100 million as we didnt own the IP. The license we took also made clear that we had rights equivalent to ownership. When we did the deal with SCO earlier this year we bought a bunch of drivers and when we give money to a company oftentimes we get warrants, which is part of the negotiations. I have warrants in 100 different companies, we have a huge venture portfolio. I cant do anything about the perception thats out there and to be blunt, I dont care as those people arent going to drive our futureâ"customers are. Which makes sense considering Sun developed SYSVR4 with AT&T.
From this article: "We're not interested in suing people over Unix," Novell spokesman Bruce Lowry said. "We're not even in the Unix business anymore." How would it look for Novell, to sue Sun over an open source project? Especially considering Sun is the biggest corporate open source donor. By the way, the article doesn't specify, but if you read the study, those numbers are from the source of Debian. They do not include OpenSolaris, OpenJDK or any of the other open source projects Sun participates in or has released.
While I do believe McNealy and Schwartz shared similar views, McNealy's mouth did tend to get in the way. As we've seen here Schwartz is sticking to his commitment to open source.
It looks to me like Novell is talking about OpenSolaris to bolster their case. Novell's CEO has also been spreading a lot of FUD about OpenSolaris.
I don't work for sun but I've been playing around with zones a bit and found a lot of useful information on zones. Haven't looked into dtrace much but I did find some good references online. Google for one line dtrace scripts as a start.
As for zones. I found this page useful for describing sparse root and whole root zones and how to implement them. Before I really understood what I was doing I had a zone running.
http://www.solarisinternals.com/wiki/index.php/Zones
This was pretty interesting as well. http://www.google.com/trends?q=redhat%2C+suse%2C+ubuntu
Now, I'm not knocking Linux as a desktop but I think it is fair to say that the real value of Linux, at least where money is changing hands today, is on the server. And the number of server installs greatly outnumber the desktop installs. RedHat has always dominated server linux installs. I wish Netcraft kept freely publishing statistics like this as they are very interesting. When companies are talking about deploying Linux, they generally mean RedHat. At least here in the US.
And even if I'm wrong about SuSE's popularity, Novell hasn't monetized it the way RedHat has. Compare the two companie's quarterly reports to verify.
You're confusing ZFS with clusterfs GFS is more like Lustre
ZFS doesn't have anything to do with SYSVR4.
Novell said they have no interest in pursuing Unix copyrights.
Novell is trying to get their 95% portion of the license Sun paid to SCO. By saying the agreement between Sun and SCO was part of the APA between Novell and SCO they are affirming the deal between Sun and SCO. Sun actually helped write SYSVR4 with AT&T before Novell bought it. According to Schwartz, Sun paid AT&T about $100million for rights that basically gave them ownership. What was purchased from SCO were mainly device drivers since SCO's UnixWare had the best x86 support.
What is Novell's position going to be to the public? "We're an open source company but we're going to sue a company for releasing open source?" Nothing good can come to Novell if they challenge Sun.
You may have said something that makes sense. All I keep thinking is I'm hungry.
This has all the hallmarks of a classic PR maneuver - Sun wants to figure out how they can extract more $$ from the high end users of MySQL. They need to find out how the market will react if they start selling closed source MySQL extensions without committing themselves if it goes horribly wrong. So they sprinkle some unsubtantiated vague rumours around and look for the reaction. The reaction was: PostgreSQL. So now they can kill the whole idea with minimal losses and try their next plan for how to "monetize" MySQL some more without pissing off their entire user base and killing the golden goose.
I don't believe for a second that things like this are an accident. These folks are far too smooth to just accidently let this kind of thing drop and run for a week.
I doubt that. When MySQL AB was a separate company and interested in an IPO, they had to find ways to boost their revenues.While MySQL makes money on support and some of the "Enterprise" tools, their main source of revenue is from companies like Cisco that want to buy a non-GPL license to MySQL to embed it in one of their products. Since MySQL AB offered MySQL in dual licenses, it gives companies that want to include it in their projects, and possibly extend or modify it the ability to keep their changes to themselves and not have to GPL their own work.
Now that MySQL is part of Sun, it doesn't need to generate the same revenue from MySQL directly. Seems that MySQL's management didn't get this. But to be fair, they never specified that the source would be closed. They mentioned it could be available under the GPL but they haven't decided. What they said was that it would be in MySQL Enterprise and not MySQL Community.
The business decision on this was made by MySQL AB (by me as the then CEO)...
The decision was made and then was reversed. Read your own link: Additionally we will develop high-end add-ons (such as encryption, native storage engine-specific drivers) that we will deliver to customers in the MySQL Enterprise product only. We have not yet decided under what licence we will release those add-ons (GPL, some other FOSS licence, and/or commercial)
Are you for real?
An open source company wants to close some of its new features. The "proprietary" software company that bought them wants them to keep everything open.
Somehow, everyone wants to paint the proprietary company in a bad light. The original blog post from the first story never even mentioned Sun but the title on Slashdot was about Sun closing MySQL.
Sun's management has MySQL change that decision and the headline is about MySQL reverting.
There's an obvious bias here that's laughable.
Mickos is the one that needs to adapt here. Though I personally don't think it's fair to give him a hard time about it. I mean the guy just closed a big deal, got a ton of money and needs to do a different job than he was doing for the last 7 or so years. If he does the same stuff in a couple more months, that's a different story.
And this wasn't a "public announcement" it was at a partner meeting. Which is a bit different. And nothing was actually released or finalized, it was just a roadmap to let partners know what to expect. Sometimes these things change but people made a big deal over it.
Most large acquisitions have their hiccups. That shouldn't come as any suprise.
Certain initiatives that were started pre buyout continued. When it was detected that those initiatives weren't inline with Sun's plans, it was corrected.
All the 400 or so employees that were with MySQL are now with Sun and they need to get used to how being part of Sun frees them from increasing direct revenues from MySQL.
Sun buys an open source company and doesn't force them to change their business practices. Doesn't sound so bad. And when it does get them to change their business practices, it results in being more open.
Despite the outcome, this is not a victory for the open-source movement. The original Slashdot story was inflammatory and designed to mislead, and now it has had the desired effect. MySQL AB needed to generate revenue directly from MySQL as that was pretty much their only product. They were looking for an IPO before Sun bought them so they needed to increase revenues.
Being part of Sun, MySQL doesn't have the same pressure to generate revenues directly from MySQL. Sun/Schwartz's plan is to drive revenue in Sun's other lines from MySQL. Hardware sales, support, etc.
* so will the MySQL Connectors, and
* so will the main storage engines we ship.
In addition:
* MySQL 6.0â(TM)s pending backup functionality will be open source,
* the MyISAM driver for MySQL Backup will be open source, and
* the encryption and compression backup features will be open source,
where the last item is a change of direction from what we were considering before.
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.