Sun Spokesman Says "We Screwed Up On Open Source"

An anonymous reader sends along a video from Builder AU, in which Sun's chief open source officer Simon Phipps describes 2001-2002 as 'a period where Sun 'screwed up' in their dealings with the open source community. Phipps says that Sun is trying to remedy the situation with the open sourcing of Java, Solaris, and the rest of Sun's software."

Comments

Never too late

[thammoud]

Thank you Sun for all the great products that you have open sourced. Unlike your competitors, you have outsourced your crown jewels.

Re:Never too late

[dintech]

And I really, truly hope it works out for them. I hate for it to go the wrong way...

Re:Never too late

[pegdhcp]

Probably because they are an UNIX company at the end, their track record is better than that of Novell. SUN is closer to the core of FOSS community. Also this is not the first time they admit a mistake, which takes some balls to do in IT industry. It was really appreciated (by me at least) when they switched from SunOS to Solaris and it was not just the name that was changed. I hope Novell would take the clue one day...

Re:Never too late

[somersault]

you're saying that has a high percentage of Latter Day Saints among its employees? Or are you being ironic?

Re:Never too late

[jdfox]

Actually, he was being irmonic.

Re:Never too late

[BadOPCode]

I'm confused. What did Novell do? I'm not talking 8 years ago. What did they do today? Signed a sweetheart deal with Microsoft to get some inside poop on Windows tech to deliver some compatibility to Linux. Wow that is horrible. Ya they were trying to do what the EU forced Microsoft to do. Sun is rotten too... because back in 'Nam ... for christ sakes people ... get over it. These companies are delivering massive amounts of resources to the OSS community. DO NOT WHINE ABOUT IT! I understand if they do something you feel is morally and/or religiously wrong, by all means do not participate in the matter. But call me a troll but I think its ridiculous and stupid to look a gift horse in the mouth. Also note: Fanatical fanboys make no difference to the scheme of anything other than everyone's annoyance levels.

Re:Never too late

[somersault]

Is that some kind of mnemonic?

Re:Never too late

[teknopurge]

Sun is the modern-day Bell Labs.

Re:Never too late

[halcyon1234]

As stupid as the parent was being, this is actually kinda true. One of Novell's key locations is in Provo, Utah. Due to demographics alone, with about 61% of the population being Mormon, Novell most likely has a higher-than-average employ of Mormons. (Howard Tayler, of Schlock Mercenary fame, and a Mormon, used to work there)

Re:Never too late

[PalmKiller]

Bell labs is still kicking under Alcatel-Lucent, so its more like sun just resembles Bell Labs.

Re:Never too late

[houghi]

Probably because they are an UNIX company at the end, their track record is better than that of Novell.

Perhaps you have heard of things like openSUSE, the build service, KDE, GNOME, the Linux kernel and several other things.
http://en.opensuse.org/Novell_Supported_Projects for a list

Re:Never too late

[jdfox]

I prefer to think of it as a mnemnonic.

Re:Never too late

[pegr]

Sun is still Evil (capital E):

"The simple fact is, I gather from Jones' testimony, Sun could have prevented the harm SCO sought to cause by simply telling us what rights it had negotiated and received from SCO prior to SCO launching its assault on Linux. Yet it remained silent. When I consider all folks were put through, all the unnecessary litigation, and all the fear and the threats and the harmful smears, including of me at the hands of SCO and all the dark little helper dwarves in SCO's workshop, I feel an intense indignation like a tsunami toward Sun for remaining silent."

Pamela Jones, Groklaw.net (Linky.)

Sun is Evil. Google is Evil. Microsoft is Evil, but getting tired and slow.

Re:Never too late

[pegdhcp]

Yes I heard them all. Also I read PJ's recent post in Groklaw about SUN's silence on their deal with SCO for right to open source OpenSolaris. I am not saying SUN is perfect. I spent so much time on SPARC workstations and hate it most of the time, that I even today prefer to use AIX, if I must to use a branded UNIX.

What I tried to (maybe poorly) say is, SUN, being a UNIX -however bad- producer, is closer to FOSS community. Also while not saying Novell is totally evil, even after spending (wasting?) a part of my younger years writing programs on Netware, I cannot believe that they are honest, especially after wasting (this time with no question mark) some time while trying to install mpeg libraries to SUSE 10.x or after reading (thanks to PJ again) about their deal with M$.

Re:Never too late

[somersault]

All the memory aiding goodness but with added tasty mnemniness?

Re:Never too late

[Anonymous Coward]

If the Novel/MS deal was strictly between the two of them, i would have migrated from Suse and left it at that. However, Novells covenant with MS potentially gives MS ammunition against the rest of the 'linux community' (whether it be FUD or if they make good with their IP threats). That's what i have an issue with. The deal exposes the rest of us, non-Novell linux users

Re:Never too late

[horatiocain]

I'm guessing it was the 'Google is Evil' bit.

Re:Never too late

[Belial6]

No, you want to look a gift horse in the belly.

Re:Never too late

[fm6]

I don't think the parent was concerned with simple demographics. He's dredging up this obsession many people have with the LDS church: Is it really Christian? Do they really think that heaven is another planet? Have they really given up polygamy?

As a serious unbeliever, I look at Mormons as slightly stodgy but basically likable people who do a lot of home canning. I could care less about their theological quirks, but passionately believe in their right to worship as they choose. I have little use for people who drag "Mormonism" into this kind of conversation, where it's relevant only to the terminally paranoid.

I do wonder, though, if Bill's going to marry that Serbian lady. But that's just TV.

Re:Never too late

[Anonymous Coward]

It's not the same thing. Those are all open source products they've purchased. Novell Groupwise, Netware, Zenwork, eDirectory, etc ... all their core products, they haven't open sourced. Whereas Sun has opened every software product they've ever produced. It's great Novell is supporting the open source community, but Sun is in a completely different league.

Re:Never too late

[rtb61]

Well, dang, you just gotta remember slashdot is a forum for the good old computer geek/nerd network and not a forum for corporate marketdroids. Yup, one thing to remember about computer geeks/nerds they have really good memories and things that happened one or five or ten years ago are all just like yesterday and they will take the piss out of misbehaving corporations as often as it tickles their fancy to do so.

Personally I think Sun had a few rough patches with open source as the reward from open sourcing a lot of great software, might not have been as great for the bottom line as they had expected and it them a bit anxious and pushed them into making some skewed decisions. Of course now the rewards for pushing open source are coming into to sight and it is becoming apparent that the beast of redmond has been neutered and open source is where you need to be today to open up future opportunities.

Re:Never too late

[mounthood]

And I really, truly hope it works out for them. I hate for it to go the wrong way... You can put your money in JAVA for the cheapest price in ~5 years. http://finance.google.com/finance?q=NASDAQ:JAVA

Re:Never too late

[pegr]

>I'm guessing it was the 'Google is Evil' bit.

google-analytics.com

Ever parse that jscript they push down? Yeah, capital E.

Re:Never too late

[Anarke_Incarnate]

Have you considered the possibility that Novell cannot open some of those products due to conflicting license arrangements with companies that they collaborated with and licensed technology for themselves? As it were, Novell is one of the top contributors to open source. Only recently has Red Hat made mention of opening their RHN tools like Satellite server.

Re:Never too late

[GuyverDH]

Sun was in the same boat, however, they took the effort, time and expense to make arrangements to allow them to opensource it, regardless.

They are even taking the time to re-write things that they cannot get agreements with.

Novell, on the other hand, has not done this - or at least, not to this extent.

Re:Never too late

[houghi]

Installing MPlayer under 11 (10.3 will also work if you look it up) is done with one click install: http://api.opensuse-community.org/searchservice//YMPs/openSUSE_110/e208863ffe1f342b2a9eed35603427dcac9fa27c
http://en.opensuse.org/One_Click_Install for more info
They have bought SuSE and have opend up not only YaST but the whole process. You can even make your own openSUSE based distribution where they activaly helped solving on how to do this in several ways.
They are working on how to open source other things as well. However that is not as easy as they would like.

The fact that they have gotten a lot of money from MS so that MS can give away SUSE is realy nice and except for all the emotional reactions to it, I am still waiting for the negative things that were going to happen what people were promising. Instead I see no change, except the fact that Novell got a bunch of dollars for SLES licences.

And being closed to FOSS is something I disagree with if you compare it to Novell. Novell is not closer to FOSS. They are standing in it and are part off it.

Perhaps it was true for Novell a few years ago. It certainly isn't at this moment.

Re:Never too late

[crotherm]

Pamela Jones has a bug up her arse about Sun. Her extreme stances is what drove me from visiting her web. It is one thing to have ideological differences with others, it is quite another to act rabidly about it.

Re:Never too late

[omnipresentbob]

I would look a gift horse in the mouth.

Re:Never too late

[Max Littlemore]

The way the whole SCO thing played out was really good for Linux.

If SCO had been shut up earlier, their loss would not have been as convincing. To me, the fact that so many people from non technical backgrounds backed SCO so strongly and lost so hard not only makes Linux in particular look stronger, but it also will make others less confident about similar actions against large open source projects. Combine that with the fact that most small open source projects stay under the radar until they get large, the whole situation has helped consolidate FOSS and provide reasonable protection from spurious claims.

Not saying it wasn't painful for a lot of people, wars generally are, just saying that it was probably worth taking the hard road for a larger and more emphatic victory.

Not that any of this is specifically Sun's doing. According to you, Sun's great sin was silence. If I was the great dictator of Sun through that time, I would have done the same, especially if I wanted SCO to lose. Anyone with real knowledge of the case and the code knew that it would be difficult for SCO to win. For Sun to remain silent instead of weighing in has reinforced the validity of the victory and the respectability of Sun, who is now open sourcing everything. Win win win

I actually feel they did FOSS a great service, perhaps unintentional, but certainly a blessing in disguise.

Re:Never too late

[mrbluze]

Wrath is a sin as well. God is a sinner. And to many people, the Sun is God. So beware of Sun's wrath.

Re:Never too late

[drinkypoo]

Have you forgotten all the cases in which Sun has kissed Microsoft's ass, taken their money and all that? Sun is Microsoft's hand puppet. Never forget that. Microsoft bought SCO to mount an assault on Linux, and bought Sun's silence around the same time. Those who forget history...

Re:Never too late

[Skrapion]

Reply before you mod, jerks...

How can I mod after I reply?

Re:Never too late

[houghi]

... yet. They are working on it.

Re:Never too late

[JeffSchwab]

For good reasons. Sun may be the single worst-managed large company in America.

Re:Never too late

[crotherm]

I have been a user and an administrator of Sun products for close to twenty years. Seeing how many of their products ended up in linux and BSD amazes me. How many other industries does thing like that happen? Sun is no more evil than any commercial company. Saying Sun is MS hand puppet is just fan boyish rhetoric that does not belong in any serious debate of issues.

Oh yeah... this is /.

Re:Never too late

[turgid]

Have you forgotten all the cases in which Sun has kissed Microsoft's ass, taken their money and all that? Sun is Microsoft's hand puppet. Never forget that. Microsoft bought SCO to mount an assault on Linux, and bought Sun's silence around the same time. Those who forget history...

Are you nuts? The less Microsoft Windows there is in the world, the easier things would be for Sun. Well, for everyone, actually. Don't forget that the Microsofts and SCOs (Caldera) of this world screwed Sun. And IBM screwed Sun with their patents. (IBM only supports Linux because AIX sucks c.f. Solaris and they want to compete with Microsoft).

Sun is only guilty of being a bit too laid-back, and not being viscous enough in business. Sun could have bought Apple, AMD, nVidia and SuSE. Maybe it's being Californian? Who knows.

Oh, and who was it who started the anti-trust proceedings against Microsoft in the first place?

Re:Never too late

[drinkypoo]

Oh, and who was it who started the anti-trust proceedings against Microsoft in the first place?

There's a certain matter of timing, which could have been better for Microsoft, but not by much. They were found guilty so the geeks were appeased, but then they got a handslap instead of actual punishment...

GPL zfs

[C_Kode]

Phipps says that Sun is trying to remedy the situation with the open sourcing of Java, Solaris, and the rest of Sun's software."

GPLing ZFS would go a long way with me!

Re:GPL zfs

[FauxPasIII]

> GPLing ZFS would go a long way with me!

Prepare to be surprised.

Re:GPL zfs

[IAR80]

ZFS is already implemented on FreeBSD.

Re:GPL zfs

[An dochasac]

I think anyone who has attempted to legally link proprietary drivers, video codecs with the Linux kernel would understand some of GPL's limitations. The fact that GPL can take from so many licenses (without necessarily exporting) doesn't automagically make it the best license.

CDDL is based on MPL which has an explicit patent protection clause (bring a patent suit against another CDDL licensee and you lose all CDDL rights)

The fact that (besides Java), Sun hasn't released much GPL code should not cause us to ignore significant contributions by Sun to the opensource community. According to a E.U. study on The economic impacts of free and opensource software, Sun contributed 312 million Euro's worth of FOSS which amounts to over 51000 person months. This was 44% of all corporate contributions to FOSS. The next highest contributor was IBM with 13% then Red Hat with 8%. The rest SuSE, Netscape, AT&T... don't even add up to Sun's contribution. And this study came out before Java was GPL'd.

Re:GPL zfs

[IAR80]

The day they will open source ZFS will be when their next file system will be ready for sale. And from the looks of their recent performance that will be never.

Re:GPL zfs

[squiggleslash]

The GP was asking for ZFS to be released under the GPL. It already is open-sourced, just under the CDDL which is, unfortunately, incompatible with the de-facto standard copyleft license, the GPL.

So, er, what file system is Sun selling then? ;)

Re:GPL zfs

[larry bagina]

ZFS is Open Source and has been since November 2005.

Re:GPL zfs

[lolocaust]

What's wrong with the GPL "taking" from other licenses? If you don't want your code taken and modified without being given the changes you should pick another license that doesn't allow that (I can name one that comes to mind). I don't see anyone bitching about modified BSD code in Windows or OSX, although I'm sure some changes were given back by Apple. This isn't really directed at you as much as it is to the driver developers who got their panties in a bunch a few months ago, so please don't think I'm taking one word you wrote out of context for the purpose of this rant.

Re:GPL zfs

[An dochasac]

ZFS is already opensourced and available to all Operating systems with CDDL compatible licences. It is already available in BSD, OSX, OpenSolaris (All distributions including Nexenta,OpenSolaris2008.05,Nevada,Belenix,Schillix,Martux,Milax)

ZFS read is already in OSX 10.5 and I've installed the beta ZFS write in Mac OSX 10.5, created a pool on a USB keychain, imported that pool into OpenSolaris2008.05 (which automagically mounted it). Put stuff on it, snapshotted it, exported it and reimported it into OSX. This is the filesystem of the future. The fact that GPL isn't compatible with ZFS is Linus's problem. Good luck with that FUSE module.

Re:GPL zfs

[C_Kode]

You can have $3 billion dollars, you just have to cut your arms and legs off before you can spend a single penny.

The point is; Linux is the most used open source operating system and Sun licensing it under the CDDL is like dangling a carrot out, but saying only one in ever ten people can have a bite.

If the GPL is good enough for Java why isn't it good enough for ZFS? I'll tell you why. Sun's Java was on a road it couldnt' correct so it gave to the GPL community to keep it afloat. ZFS doesn't actually have any real competition so they use the CDDL to thumb their noses at Linux and the GPL.

Sun has always had a sore spot for Linux. They only grudgingly accept Linux at times because their very existence depends on it.

Re:GPL zfs

[Anonymous Coward]

The numbers in that study seem to be *way off* reality. Really, RH and only 400kloc? I suggest you look at the OSDL study or any of the recent "Who wrote linux 2.6.x" statistics on LWN for more realistic numbers.

Re:GPL zfs

[JohnFluxx]

Meh, it was clearly on purpose that ZFS was put under a GPL-incompatible license.

If the GPL was compatible with CDDL, then Sun would have just chosen another license

Re:GPL zfs

[Anonymous Coward]

I think that GPL-fanatics' attitude is all the time quite unwelcoming towards people outside the GPL realm. Does it make them feel good? I think not, but old habits die hard.

Re:GPL zfs

[drinkypoo]

The fact that GPL isn't compatible with ZFS is Linus's problem. Good luck with that FUSE module.

FUSE gets faster all the time, but it appears that ZFS may go GPL here soon anyway. It would be kind of surprising if it didn't, what with Sun moving other projects that direction.

Compared to Linux, OSX is not even in the running for servers, except for budget SANs.

Re:GPL zfs

[FishWithAHammer]

Because the GPL is manifestly less free than the CDDL and contributions licensed under the GPL could not be folded back into the CDDL version?

And don't try to say "well, they could stipulate that all submissions have to be dual-licensed"--you and I both know we'd see some stupid little gnuZFS the same day as ZFS was GPL'd, just to get around that.

Re:GPL zfs

[samkass]

licensing it under the CDDL is like dangling a carrot out, but saying only one in ever ten people can have a bite.

Actually, it's like saying that only 9.5 out of every 10 people can have a bite... among all the OSes out there, I think only Linux has problems, and that's a tiny fraction of the desktop OSes out there.

Re:GPL zfs

[mrjatsun]

> And?

    Mac OS X (http://zfs.macosforge.org/trac/wiki/)
    FUSE (http://www.wizy.org/wiki/ZFS_on_FUSE)

Re:GPL zfs

[y86]

Sun contributed 312 million Euro's worth of FOSS which amounts to over 51000 person months. This was 44% of all corporate contributions to FOSS. That's why I bought 50 shares of "JAVA". It's 10$ and it's a good company overall.

Re:GPL zfs

[teknopurge]

Hi zealot hater. Your smug attitude is quite unbecoming. Does it make you feel good?

It made me feel good. =)

Re:GPL zfs

[Bralkein]

And don't try to say "well, they could stipulate that all submissions have to be dual-licensed"--you and I both know we'd see some stupid little gnuZFS the same day as ZFS was GPL'd, just to get around that. While the GNU people certainly can be rather over-zealous when it comes to the subject of licensing, inclusion of ZFS is surely a matter of kernel development, and the Linux kernel does not (AFAIK) fall under the aegis of the GNU project.

Linus Torvalds always struck me as a pragmatist, and Linux kernel development is backed by a number of groups with a genuine commercial interest in improving Linux with the inclusion of good technologies like ZFS. In the light of these facts, I would be very surprised (not to mention disappointed) if a dual license prevented the inclusion of ZFS in Linux.

Re:GPL zfs

[jedidiah]

Ultimately it doesn't matter how many lines of code Sun has contributed.

Their big mistake has been ignoring commodity hardware, not their policies
towards an particular operating system or 3rd party codebase. Open standards
go a long way to deliver what you would otherwise need Free Software to do.

Sun simply chose to ignore us. That lack of vision cost them their position
in the marketplace. This is typically what happens to dinosaurs. This is
especially true in a free market where artificial exit barriers aren't a
problem.

They can "liberate" whatever they like. It doesn't matter as the boat sailed
over 10 years ago.

Re:GPL zfs

[Hatta]

ZFS and Sun need Linux more than Linux needs ZFS or Sun.

Re:GPL zfs

[street struttin']

among all the OSes out there, I think only Linux has problems, and that's a tiny fraction of the desktop OSes out there. This is quite telling. OSX hasn't had too many problems adding it, and neither has freebsd. It's the GPL that has issues, not CDDL.

The fact that GPL needs to have everything that touches it be opened makes it very difficult to use it in proprietary environments. By using CDDL and allowing ZFS to be in freebsd, I could now use freebsd to create a proprietary network storage device using freebsd as the OS, zfs as the file system, and not have to release any source if I don't want to. That's pretty powerful.

Re:GPL zfs

[tm2b]

You can have $3 billion dollars,

Would you get it from an ATM machine?

Re:GPL zfs

[Hatta]

Man, the only thing worse than his anti-zealot zealotry is your anti-anti-zealot zealotry.

Re:GPL zfs

[larry bagina]

The CDDL is like LGPL but on a per-file basis. Any changes made to a CDDL file will remain CDDL, but any other files or code is not affected. It's 100% usable by FreeBSD or Apple.

Re:GPL zfs

[MrFlapjacks]

You can have $3 billion dollars, you just have to cut your arms and legs off before you can spend a single penny.

It's a deal. PM me and I'll give you my paypal info.

I'm sure with that much cashola I'd be able to build myself some pretty awesome legs and arms.

Re:GPL zfs

[Anonymous Coward]

The fact that GPL isn't compatible with ZFS is Linus's problem. Yeah... bad, bad Linus. The only reason a Finnish CS student chose the GPL back in 1991 or so was that he knew that 15 years later, a US-American company would come up with a filesystem that they'd want to integrate into his kernel, and obviously, he could have none of that - so he chose the GPL deliberately to ensure incompatibility with ZFS's license and then proceeded to gloat and stroak his goatee for the next 15 years. And there was nothing Sun could do, since it was impossible for them to use the GPL, too - everybody knows that this can't be done.

Yes, that's how it happened - it's all Linus's fault. Sun really wanted to contribute ZFS to Linux so much, but the bad, bad Linus cunningly made it impossible for them. What a world.

Re:GPL zfs

[AKAImBatman]

Their big mistake has been ignoring commodity hardware

Not really sure what you mean here. I was rather surprised when I decided to do some Solaris development that the primary focus has moved from Solaris/SPARC to Solaris/x86. Half the cool stuff in OpenSolaris is designed around the x86 platform.

Similarly, the primary focus of the Java codebase is the x86 platform first, remaining platforms later.

Sun is also a massive seller of AMD64 and Intel Xeon based servers and workstations. Amazingly, Sun's prices have even come out of the stratosphere and are extremely competitive with other manufacturers like Dell.

Sun is even working to virtualize these "commodity platforms" with their surprisingly good OpenxVM project. I actually passed on a free copy of Parallels because Sun's VirtualBox was working so well for me.

I know Sun has the stigma of selling only overpriced iron, but the truth is that they're fairly well in tune with their customers and are working hard to provide them with the products and services they need. Along the way, the Open Source community is benefiting greatly.

Re:GPL zfs

[EvilRyry]

Unless you modify ZFS which is CDDL licensed.

Although it is true that FreeBSD and OSX have/will be getting ZFS, its also true that Linux is much bigger than either of these in the server room. So it is still true that most people on their current platform will not be able to benefit from ZFS under its current license.

Re:GPL zfs

[bill_kress]

Strange... I thought that users could add in anything they wanted under the GPL... Is there something in CDDL that prevents a user from adding a ZFS package?

Personally I like the GPL, but I also support anyone who creates a product releasing it under whatever license they damn well please (or not at all, if that's what they want).

Also, aren't there quite a few distros that include proprietary software anyway? I believe it's only a few distros that get all anal about not including anything that's not GPL, but I could totally be wrong about that.

Re:GPL zfs

[mzs]

No, here is why GPL was not chosen:

If you wanted a copyleft license, why didn't you just use the GPL or LGPL?

We needed an open source license that allowed files released under the license to be linked with files released under other licenses. While a license like LGPL would allow this for dynamically-linked code, we also needed to be able to release software that statically links source files available under different licenses. In addition, we wanted to allow others to add externsions to OpenSolaris with different license terms. This was only possible under a license like the MPL; however, we could not use the MPL because it is not a "template" license allowing reuse by others. Consequently, we crafted a variant of the MPL, taking the opportunity to make it a template license as a step towards reducing license proliferation for others finding themselves in a similar position.

The other reason for the creation of the CDDL has to do with software patents:

What does the CDDL say about patents?

The CDDL provides an explicit patent license for code released under the license. This means that you can use, modify, and redistribute code released under CDDL without worrying about any patents that the contributors of the code (including Sun) might have on the contributed technology. The license also includes a provision to discourage patent litigation against developers by revoking the rights to the code for anyone initiating a patent claim against a developer regarding code they have contributed.

The reasons that the GPL is incompatible with the CDDL are very complicated and nuanced but in large part have to do with the patent clauses. It is clear why a company such as Sun needs such clauses. So if it were not for the stupidity of the existence of software patents a CDDL-like license could have been created that would have been compatible with the GPL minus some other niggles. Because of the existence of software patents and the need for such clauses the CDDL is incompatible with the GPL. It is too bad about the GPL being so restrictive about adding clauses that protect the copyright holders.

Really the spirit should have been that if CDDL source is used in another project that this project needs to be open, but then all sorts of real world complications get in the way. That is basically the spirit of the GPL as well. Sun was against anything BSD-like where another company could take their source and create a closed source product.

Re:GPL zfs

[spitzak]

I will indeed say: well, they could stipulate that all submissions have to be dual-licensed.

You appear to have discovered the answer to all your problems while you were typing your response. That is exactly what they could do. You then came up with this weird and rather convoluted excuse as to why that very simple solution would not work:

you and I both know we'd see some stupid little gnuZFS the same day as ZFS was GPL'd, just to get around that.

Do you have ANY examples of this happening, ever? Please name the project where such a fork survived in any useable form.

Re:GPL zfs

[An dochasac]

Sun already has most of the useful stuff in the upper part of the GNU/Linux stack because most of it is licensed under BSD/MPL/Apache... licenses which are all compatible with CDDL and run well on OpenSolaris (see openSolaris2008.05 and Nexenta for examples.)

The tiny part of "Linux" which can't be easily used on OpenSolaris, the real Linux almost no one thinks about when they're talking about RedHat, Novell... is Linus's kernel, the filesystem, drivers and a few other bits. What the #^~@% would Sun do with another kernel? The kernel in OpenSolaris has scalability, security and observability features that are only being dreamed about in Linus's kernel. But more importantly, the OpenSolaris kernel has stable APIs and ABIs so you won't have to rebuild and requalify all of your business logic the next time Linus adds a kernel module to support this week's latest X86 (&@?ware.

Re:GPL zfs

[Jah-Wren Ryel]

OSX hasn't had too many problems adding it, Try getting the source for the OSX-specific changes.

The fact that GPL needs to have everything that touches it be opened makes it very difficult to use it in proprietary environments. No shit sherlock. The GPL is all about ensuring that the end user has full access to the source code for the software he uses. That concept is completely alien to 'proprietary environments.'

By using CDDL and allowing ZFS to be in freebsd, I could now use freebsd to create a proprietary network storage device using freebsd as the OS, zfs as the file system, and not have to release any source if I don't want to. That's pretty powerful. Powerful for you.
Suck-ass for your customers.

Re:GPL zfs

[Hatta]

What Linux has that Sun needs is a thriving community.

Re:GPL zfs

[LaserLine]

For those that are confused I think the parent might be referring to this.

Re:GPL zfs

[Anonymous Coward]

Try getting the source for the OSX-specific changes. http://zfs.macosforge.org/trac/wiki/downloads

Google is hard... let's go shopping!

Re:GPL zfs

[larry bagina]

"Everybody should share. That's why I won't share with you".

Re:GPL zfs

[PipsqueakOnAP133]

As long as it's not exclusively GPL'd.
If ZFS was exclusively GPL, it's essentially a dead end technology since nobody else aside from Linux would adopt it. It'd be like NTFS. A filesystem for one OS.

Re:GPL zfs

[larry bagina]

$24,000. What kind of dirt is that?

Re:GPL zfs

[jedidiah]

...like I said: this all happened about a DECADE too late.

Linux already blazed the path that Sun refused to go down for whatever reason.

Sun is a follower. The Linux Community might benefit if it suits Sun. The Linux Community primarily benefits by not waiting on dinosaurs like Sun.

Re:GPL zfs

[larry bagina]

Consider ... BSD code. There are plenty of instances where BSD code was relicensed as GPL and changes weren't given back under the BSD license.

Re:GPL zfs

[spitzak]

Name one, please. One where the non-BSD branch actually contains some useful code that is not in the BSD branch.

Re:GPL zfs

[larry bagina]

BSD yacc vs GNU Bison.

Re:GPL zfs

[spitzak]

BZZZT! Not based on Yacc's source code:

from http://www.gnu.org/software/bison/manual/html_mono/bison.html:

Bison was written primarily by Robert Corbett; Richard Stallman made it Yacc-compatible. Wilfred Hansen of Carnegie Mellon University added multi-character string literals and other features.

Sorry try again.

Re:GPL zfs

[larry bagina]

Must be a total coincidence that he also wrote berkeley yacc. (And a total coincidence that he works at Sun.)

Re:GPL zfs

[Gazzonyx]

Yeah, it seems like overpriced iron when compared to x86, especially when you look at it through the eyes of 'MHz Myth', but Sun does have great hardware. I don't know about you, but I spend a good deal of time each and every day waiting on bottlenecked busses.

I start feeling it when an X86(-64) box hits a load average of about 3; while the box I SSH to for school (SPARC) chugs along smoothly at a load of 15. Then again, I think Linux and Solaris compute load averages differently, so take it with a grain of salt moreso than most anecdotal stories. I'm thinking about investing in an old SPARC box. They're cheap second hand at Anysys.

Re:GPL zfs

[Guspaz]

I don't particularly care which is more free, or less free, or which license has which clauses and provisions. An enduser shouldn't have to care about any of that. And as an enduser, I'd like to use ZFS natively without resorting to FUSE.

Then again, Sun has already said that anybody is free to re-implement ZFS themselves under a GPL license if they feel so inclined. Unfortunate that nobody has bothered.

Re:GPL zfs

[Scullywag]

They came out six months after we bought a netapp filer. Thumper was half the price for twice the storage. That's the sort of dirt cheap he's talking about.

Re:GPL zfs

[tolkienfan]

That's exactly the scenario (well, one of the many) the GPL is supposed to prevent.

Re:GPL zfs

[drinkypoo]

Strange... I thought that users could add in anything they wanted under the GPL... Is there something in CDDL that prevents a user from adding a ZFS package?

No, there's something in reality, it's called the necessary work, which is not going to be done if people can't benefit. You think integrating a new filesystem (especially one as complicated as ZFS) is a trivial job?

Also, aren't there quite a few distros that include proprietary software anyway? I believe it's only a few distros that get all anal about not including anything that's not GPL, but I could totally be wrong about that.

Well, you're pretty wrong. Most distributions only ship Free Software by default. Many or even most include non-Free media and other resources, although some are anal about that too (Debian, and the new all-Free distribution of Ubuntu come to mind.) The license doesn't require it, but some users do.

Re:GPL zfs

[drinkypoo]

Linux already blazed the path that Sun refused to go down for whatever reason. Sun is a follower. The Linux Community might benefit if it suits Sun. The Linux Community primarily benefits by not waiting on dinosaurs like Sun.

How is Sun different from IBM in this regard? IBM is a latecomer to Linux like Sun, IBM has been making massive contributions to Open Source and Free Software like Sun... okay, so IBM was a litle earlier/faster, but Sun is not unique in this regard.

Everyone can't be redhate... thank god :D

Re:GPL zfs

[drinkypoo]

What the #^~@% would Sun do with another kernel?

Perhaps they would have support for, you know, some hardware. Sun has always had poor hardware support, back in the early days of Slowlaris x86 I bought some intel/mach64 boards to serve as fancy X-Terminals basically, for some engineers at SEI who needed faster, cheaper workstations to display magic on. Everything in the system was on their compatibility list, and none of it worked worth a shit. Today, Linux supports everything, and Solaris supports nothing.

Re:GPL zfs

[drinkypoo]

Because the GPL is manifestly less free than the CDDL and contributions licensed under the GPL could not be folded back into the CDDL version?

The CDDL and GPL are licenses and cannot themselves be free or not.

The CDDL delivers more freedom to the developer. This was Sun's immediate concern. The GPL delivers more freedom to the user. This is not Sun's immediate concern. Sun is in business to make money, not to give nerds presents. If they give something out there is an ulterior motive. Not placing ZFS under the GPL is an attack on the GPL, plain and simple. That's Sun's prerogative, mind you, but I don't have to be amused.

Re:GPL zfs

[JohnFluxx]

Meh, I think you're naive if you believe their official reasons for why they chose CDDL. If CDDL was compatible was the GPL then they would have just chosen another license and come up with some reason for that license instead.

Re:GPL zfs

[spitzak]

Still it's a rewrite.

Of course this dates back to when there was the BSD advertising clause, so it was not legally possible to copy the code to GPL. Thus the CDDL would not have stopped this from happening either. However the modern BSD does not have the advertising clause and thus can be converted into a GPL project. Still waiting for you to name an example where such a GPL version exists and contains useful functional additions over the BSD version.

The problem with your premise is that anybody who did this would be blasted with negative publicity and thus their code would never be accepted.

Re:GPL zfs

[PipsqueakOnAP133]

Considering you posted this twice, I think you're just trolling, paranoid, or ignorant.

The argument the parent made makes reasonable sense to me.

If you can suggest a GPL-compatible license which accommodates the reasons stated, let us know.

2001-2002?

[Rosco P. Coltrane]

How about the other years? or is 2001-2002 the period they screwed up the worst ?

Re:2001-2002?

[cpuh0g]

Oh, I'd say 2008 is shaping up quite nicely to be one of the worst years in their history. The huge revenue miss in Q3, combined with the total lack of organic revenue growth, the continued exodus of top execs, and the pending layoff of up to 3000 employees doesn't exactly bode well for the future of Sun.

Ponytail-guy and his pals have basically given away the crown jewels and have not been able to "monetize" any sort of decent return for their efforts. The company can't sell servers, gives away software, and keeps purging the budget of the services and marketing teams that are the only pathway left for revenue growth. Geniuses.

I love Sun, I love their software, and they even have some really impressive hardware, but the management team there is absolutely clueless about how or where to take the company. The bets they made on open source and other areas are clearly not paying off. The stock is as low as it has ever been over the past 8 years and is showing no signs of life.

It is utterly depressing to see such a great company go down like this.

Re:2001-2002?

[IAR80]

the continued exodus of top execs

This might prove to be a very good thing!

Re:2001-2002?

[IAR80]

and they even have some really impressive hardware,

Have you ever worked with the Sun Blades?

Re:2001-2002?

[cpuh0g]

Yes, agreed. I think the clueless execs are a big part of the problem.

Re:2001-2002?

[IAR80]

I think THEY are the problem.

Re:2001-2002?

[Anonymous Coward]

Sun has tens of thousands of employees, many exclusively engaged in software development. They've had many rounds of layoffs before and since 2001. They looked at Linux vendors such as Red Hat and saw they were much smaller. Have you ever had to meet a payroll? It's easy to sit back and say, this company should do this, should've done that, they should get a new business model.

The wrong decision can sink a company. Look at Sybase - they were one of the hottest RDBMS vendors in the late '80s. Then they ran into a cash shortfall and had to make a source code licensing deal with Microsoft. Now Microsoft has the majority of the SQL Server business that Sybase once had, even though Sybase still has joint ownership of the source code. Yes, there are plenty of nice people out there willing to roll up their sleeves and help, but there are also plenty of un-nice people who will take what you've got and use it to push you aside.

Re:2001-2002?

[raddan]

Even if Sun as a corporate entity disappears, though, I'm sure that their influence will continue to be felt. Look at Netscape. Really, they only had two significant products (which came in one package): Navigator and JavaScript. Now, Netscape is no more, but Navigator lives on as Mozilla/Gecko and JavaScript as ECMAScript, and both of those technologies have been essential to the "2.0"ing of the web.

Sun created Java, which (love it or hate it) is still being taught as part of the core curriculum in many computer science programs. And SunOS/Solaris and its many associated technologies are being integrated into many places (PAM, DTrace, ZFS, and so on). If you have experience with any of Sun's technologies, you know they're not perfect, but they're damn well thought out, and they make many parts of your daily work easier.

I hope Sun weathers these changes-- they're one example of a company that saw a coming shift in the business of selling computers and software, and instead of lobbying the government to prop up their failing business model, instead changed their business model. There's plenty left for Sun to fix in their company-- e.g., have they opened up their hardware documentation yet? (we would probably buy Sun hardware if we could run other OSes, fully-supported on it). But it would be a shame to see such an innovator go the way of Xerox PARC, Bell Labs, etc, etc, etc...

Re:2001-2002?

[Attila Dimedici]

Oh, I'd say 2008 is shaping up quite nicely to be one of the worst years in their history. The huge revenue miss in Q3, combined with the total lack of organic revenue growth, the continued exodus of top execs, and the pending layoff of up to 3000 employees doesn't exactly bode well for the future of Sun.

I'm curious, how do you have the Q3 revenue numbers when Q2 isn't even over yet?

Re:2001-2002?

[cpuh0g]

Uh, the fiscal year runs from July - July. Fiscal 2008 Q3 ended at the end of March. The numbers were reported at the end of April. They were horrible. Heads should roll from the top down, not the bottom as they appear to be scheduled. Head of sales left of his own accord, and got a nice payout from Sun and fat new paycheck from his new employer as a reward for the horrendous job he did growing sales over the past several years for Sun. Nice work when you can get it.

Re:2001-2002?

[Attila Dimedici]

Sorry, I'm not an employee or stockholder of Sun, so I didn't know what their fiscal year was. Not all companies have the same fiscal year.

Re:2001-2002?

[Cal Paterson]

Well, it appears that most of them do have the same fiscal year. Supposedly it is most normal in the USA to run from October to October.

Re:2001-2002?

[Attila Dimedici]

What you linked to is the US government fiscal year, which runs October to October. I have worked for several companies that had different fiscal years. One was April to April, one was July to July. The oddball was a large privately held company that ran May to May. When I see a Q? listed I assume it means the calendar year unless otherwise listed. For example when the U.S. government releases quarterly economic figures they are usually based on the calendar year not the government's fiscal year.

Re:2001-2002?

[dedazo]

Sybase has no ownership of the code they licensed to Microsoft, nor does Microsoft have any stake or ownership in Sybase's current products. The purchase of the old Sybase code was a one-time deal with no future involvement or collaboration clauses whatsoever. Microsoft just took a snapshot of the Sybase code along with technical documentation, and that was it.

And I'd challenge your claim that Sybase is doing badly. I'm not seeing that out there, especially in the OLAP niche where they are cleaning Microsoft's clock left and right.

Re:2001-2002?

[Pinback]

Milestones for Sun:

1) They stopped shipping a C compiler with the SunOS. Frustrated people gradually moved to GCC.
2) They moved from BSD to SYS-V. Hey, all the other kids are doing it.
3) Linus wrote his kernel while reading the SunOS kernel API.
4) They spent years championing everything but PCI.
5) They spent millions trying to make IDE and NIC chipsets that sucked. Ooh, remember the Ultra5, and HME?
6) The spent years playng a peekaboo game of "Look NFS is standard, except only we have a lock daemon".
7) They spent years giving Veritas and Oracle blowjobs. If Solaris was a car, it would come without wheels.
8) They spent years smoking mental crack, and planning to dominate the desktop by making Java the dominant paradigm. (It looks painful, being thoroughly outmaneuvered my Microsoft.)
9) Cache RAM NDA fiasco?
10) ZFS is so great, we're giving it away. Wow excellent plan. XFS, JFS, ReiserFS, yup we need one more FS.
11) During all this dick fumbling time, *IBM* has gradually turned AIX/RT into something that eats a portion of Sun's lunch.

How do you guys keep morale up over there? Do you console yourself by saying, "Hey, at least we're more relevant than Novell?" or "Hey, Solaris is in better shape than OS/2?" or "Hey, HP completely dumped PA-RISC, and now their an Intel Junkie. ROFL"

How can so many smart people make so many stupid decisions, and keep from running a company completely in the ground? Are they flying by throwing themselves at the ground and missing?

The UNIX market, and arguably all modern OS business, was Sun's to loose. At this point, throwing one more thing out of the gondola isn't going to help. If you opensource everything tomorrow, is anyone really interested in taking you back?

Quick buy some Sun memorabilia before they become the next Intergraph/SGI/Sequent.

Re:2001-2002?

[James Carnley]

10) ZFS is so great, we're giving it away. Wow excellent plan. XFS, JFS, ReiserFS, yup we need one more FS.

ZFS is a paradigm shift in file systems whereas the others are simply evolutions of the basic file system.

Don't be too quick to discount it as another piece of shiny tech that won't make any money. Someone has to provide support for it and support brings in cash.

Re:2001-2002?

[cpuh0g]

I agree with almost all of your points except for ZFS. ZFS is truly revolutionary and amazing. XFS, Reiser FS (i.e. murder-your-wife-FS) pale in comparison. ZFS renders Veritas and NetApp moot and useless in one fell swoop. No wonder NetApp is suing.

However, unfortunately, I'm pretty sure Ponytail guy and the rest of the C-level nerds will f* things up as usual and fail to capitalize on the technology. Sun = great engineers, shitty management. Someone needs a haircut. Its 2008 for crissakes, get with the decade.

Re:2001-2002?

[Binary Ninja]

Hmm, sounds like the bullshit stories people told about Oak. A paradigm shift. You mean like digital compact cassette, minidisc, DivX, or HD-DVD? The NAS and block storage companies are not going to just roll over and concede their market to Sun.

On the flip side, Sun has deep coffers. They should be around for a while.

I've seen people playing with ZFS, but everyone I know who is using ZFS paid zero dollars for it. Sun is still sending me "please buy a copy" email.

FWIW: After reading the ZFS, Nexenta and Thumper whitepapers, I'm more interested in Hammer than ZFS.

Re:2001-2002?

[IntlHarvester]

> 2) They moved from BSD to SYS-V. Hey, all the other kids are doing it.

Considering Sun's bread-and-butter for many years was multiprocessor systems, this was a very wise move.

still skeptical

[IAR80]

I am still skeptical of Sun. They are coerced into behaving nicely by the huge open source community which is not that much interested in what they have to offer anymore and have a lot of influence in the market. And let's face it, they opened up Java after IcedTea is out for quite a while now.

Re:still skeptical

[/ASCII]

IcedTea is based on OpenJDK, released by Sun.

Re:still skeptical

[Kentaree]

They open sourced their compiler, virtual machine and most of their libraries before IcedTea was started according to wikipedia. And how would you say they can be coerced? It's not like their entire revenue is based off open-source, so I don't see any distinct advantage open-sourcing would give Sun

Re:still skeptical

[julesh]

They are coerced into behaving nicely by the huge open source community which is not that much interested in what they have to offer anymore and have a lot of influence in the market

Not really. Java still dominates the enterprise application market (the only place it ever made any money for Sun), and its open source status is likely to have little effect on this. Even without ZFS being open-sourced, Solaris would still have a world leading file system. And I don't see where any pressure at all came from for them to open source the design of their UltraSparc T1 and T2 processors. Sun have been progressively opening more and more of their key business IP, and as far as I can see the only reason they have done so is that they really believe in the benefits of open source.

Re:still skeptical

[Woy]

Your shell is my shell of choice. Thank you.

Re:still skeptical

[jedidiah]

Sun doesn't "dominate".

It shares the market with IBM, HP and Linux.

Depending on your "enterprise" app of choice, Sun may appear to be nothing but a has-been.

Re:still skeptical

[Cal Paterson]

I like it, have just switched. Thank you for pointing it out.

Re:still skeptical

[gwniobombux]

They are coerced into behaving nicely by the huge open source community which is not that much interested in what they have to offer anymore and have a lot of influence in the market. You mean coerced into open sourcing, like IBM, Microsoft and pretty much everybody else.
Sorry, speaking of coercion seems to detract from what Sun's doing here. I, for one, am thankful.

Re:still skeptical

[Jah-Wren Ryel]

Sun have been progressively opening more and more of their key business IP, and as far as I can see the only reason they have done so is that they really believe in the benefits of open source. I see it the other way around. They are being forced to confront the value of open source and are fighting it every step of the way. They are losing the fight, precisely because open source is so powerful. The results of each lost battle has been them open-sourcing a piece of the pie.

If the company whole-heartedly believed what Schwartz has been saying, they would GPL everything they've got - proprietary apologists be damned.

FWIW, I'm waiting on the sidelines for that event. The day it happens, assuming it does happen before they go tango-uniform, is they day I move a third of my retirement savings into JAVA shares.

Re:still skeptical

[pdusen]

Do you even know what "The cat is out of the bag" means? You're thinking of "That ship has sailed" or something like that.

Re:still skeptical

[bill_mcgonigle]

I see it the other way around. They are being forced to confront the value of open source and are fighting it every step of the way. They are losing the fight, precisely because open source is so powerful. The results of each lost battle has been them open-sourcing a piece of the pie.

That's a nice aspect of Open Source, isn't it? Even if they come kicking and screaming, they'll still benefit.

If the company whole-heartedly believed what Schwartz has been saying, they would GPL everything they've got - proprietary apologists be damned.

Some of those apologists have recently resigned in protest. That's a good sign, I think.

FWIW, I'm waiting on the sidelines for that event. The day it happens, assuming it does happen before they go tango-uniform, is they day I move a third of my retirement savings into JAVA shares.

Well put.

Re:still skeptical

[HiThere]

Java was relatively open from the beginning. I remember how surprised I was at finding the Java 1.0 libraries as source code. (Well, not ALL of them, but most.)

What Java wasn't was free. It as $0.00 for the entire set, but though it was Open Source (mostly) it wasn't Free (libre).

It's reasonable to be skeptical of Sun, but not dismissive. Sun has never been as closed as MS, and it's never been as open (or as free) as the FSF. In the last several years it's been edging towards the FSF position. I don't expect it to ever get there.

OTOH, it's important to remember that Sun is a corporation, and corporations have management, and top management changes. There's no guarantee that Sun will continue to be as open, and whatever is released under a license that doesn't provide openness into the future under a different management is untrustworthy. I don't know about the CSSL (or whatever it's called), except that it's incompatible with the GPL. OTOH, Java is being released under the GPL. And OpenOffice is being released under the GPL. Those are two very significant offerings. Don't discount them. (Note, also, that OpenOffice had no significant competition at the time it was released. And Sun still chose GPL. Unfortunately, about that same time it subsidized the SCOX lawsuit.)

Re:still skeptical

[julesh]

Sun doesn't "dominate".

It shares the market with IBM, HP and Linux.

Depending on your "enterprise" app of choice, Sun may appear to be nothing but a has-been.

I said Java dominates, not Sun. I'll agree that only a small proportion of those applications are running on Sun hardware these days, but that wasn't my point. Java is still a successful technology in the only market it was ever profitable in, and I therefore don't see any reason to say that Sun were "forced" to open source it. It wouldn't have gone away if they hadn't.

You know how you can help, Sun?

[MikeRT]

Just finish open sourcing Java, and then bring JRuby and Jython up to par with the latest versions of the C-based versions. Oh and throw in there a spec to allow easy integration of those languages into JSP and some other areas to make them peers to Java itself in the enterprise world.

Re:You know how you can help, Sun?

[BestNicksRTaken]

you might be asking the wrong company there - as far as i recall, the main jython developer also wrote ironpython, and now works for microsoft who seem to not really take python seriously as its a bit of a bolt-on hack and not nicely integrated into visual studio like c++ etc; they're not exactly the kings of opensource either....

plus, even though the jython library version is out-of-date, it still makes c-python look like a snail - and i never thought i'd say that java is faster than c!

hopefully python3000 will bring us speed if not compatibility.

Re:You know how you can help, Sun?

[julesh]

plus, even though the jython library version is out-of-date, it still makes c-python look like a snail - and i never thought i'd say that java is faster than c!

C-Python is a bytecode interpreter written in C (Apples). Jython is a bytecode->bytecode translater that operates on top of the Java runtime (Oranges).

Re:You know how you can help, Sun?

[Tim C]

Oh and throw in there a spec to allow easy integration of those languages into JSP and some other areas to make them peers to Java itself in the enterprise world.

Now I don't doubt that you'll see a fair amount of Java code in JSPs, but that doesn't mean that it's right or desirable. The correct thing to do is to write a tag, which presumably can be written in any language that can be compiled to JVM bytecode.

To be honest I think that the ability to use scriptlets in JSPs is a bad thing that I'd rather see removed than extended to include other languages.

Re:You know how you can help, Sun?

[drewness]

plus, even though the jython library version is out-of-date, it still makes c-python look like a snail - and i never thought i'd say that java is faster than c!

Bwuh? Here's a mini benchmark I did a few months ago out of curiosity. Take with as large a grain of salt as needed:

Timing test of creating a sequence of 10,000 items by for-loop, list comprehension, mapping a function, and generator comprehension. Test machine is 1.83GHz Core 2 Duo iMac with 2GB 667MHz DDR2 SDRAM running MacOS 10.4.11.

awatts@lab1:~$ python2.4 timerseqs.py
2.4.4 (#1, Oct 18 2006, 10:34:39)
[GCC 4.0.1 (Apple Computer, Inc. build 5341)]
forStatement => 4.00885510445
listComprehension => 2.35688495636
mapFunction => 1.81449890137
generatorExpression => 2.70679593086

awatts@lab1:~$ python2.5 timerseqs.py
2.5.1 (r251:54869, Apr 18 2007, 22:08:04)
[GCC 4.0.1 (Apple Computer, Inc. build 5367)]
forStatement => 3.87032794952
listComprehension => 2.28621602058
mapFunction => 1.88603901863
generatorExpression => 2.94665503502

awatts@lab1:/Applications/IronPython-2.0A6$ mono ipy.exe ~/timerseqs.py
2.5.0 (IronPython 2.0A6 (2.0.11102.00) on .NET 2.0.50727.42)
forStatement => 10.0953903198
listComprehension => 7.43706512451
mapFunction => 5.51852416992
generatorExpression => 8.20207214355

Note: Jython doesn't support generator comprehensions
awatts@lab1:~/jython2.2.1$ ./jython ~/timerseqs.py
2.2.1
forStatement => 20.13700008392334
listComprehension => 15.575999975204468
mapFunction => 10.496999979019165

Jython is *by far* the slowest. I tried to rerun it just now with the newest version of all of them on my new machine, but I'm having java issues with Jython.

Re:You know how you can help, Sun?

[totally bogus dude]

hopefully python3000 will bring us speed if not compatibility.

Maybe, but 992 years is an awfully long time to wait to find out.

Kudos to Sun

[cryptodan]

They restored some respect that they lost from me. Lets see how it goes from here.

To be fair

[Frogbert]

To be fair they wanted to open source the code in Q4 1999, however their Java ftp client just finished loading a few months ago so they couldn't upload the code until just now.

Took long enough...

[joe$007]

You have failed me for the last time!

Sun is still screwing up, albeit not as much

[speedtux]

I appreciate how much effort it must have taken for Sun to move this far on open source. Nevertheless, I think Sun is still screwing up.

Solaris, for example, is being positioned as an alternative to Linux: it's "pick us or pick Linux". From an open source point of view, it would be better if Sun picked a license that allowed the best parts of Solaris and Linux to be combined, and for end users to decide what those best parts are.

For Java, Sun still has most of the control, they have torpedoed attempts to certify Apache-licensed implementations as Java compliant, and their dual licensing scheme for Sun Java means that the project just isn't run the way an open source project ought to be run.

In the short term, Sun's behavior is disruptive for open source, but sadly not in the positive newspeak sense, but in the sense of merely annoying a lot of people for no good reason.

In the long term, Sun is going to lose with Solaris and Java if they persist in their take-it-or-leave-it approach to open source. If they want the technologies to survive in some form, they need to allow a mix-and-match approach; that's what open source is really all about.

Re:Sun is still screwing up, albeit not as much

[IAR80]

Mysql 6 will be the perfect example of Sun screwing up even more than before.

Re:Sun is still screwing up, albeit not as much

[pirhana]

> Solaris, for example, is being positioned as an alternative to Linux: it's "pick us or pick Linux". From an open source point of view, it would be better if Sun picked a license that allowed the best parts of Solaris and Linux to be combined, and for end users to decide what those best parts are.

This is a very important point. Regardless of any so called technical merit Solaris kernel has over Linux, its NOT going to catch up with Linux in adoption or momentum. At least not anytime in the near future. I am telling this as I have managed to get Solaris(intel version) installed on a machine after about half a dozen failed attempts. Mostly due to hardware incompatibility. The tried hardwares include even the very common ones like DL-385. Just to manage it from my laptop(Kubuntu) I installed OpenSSH on the solaris box. It took almost 30 minutes to get it installed where as in linux it would take less than 30 sec. Solaris is no where near to Linux in hardware compatibility , ease of installation, availability of applications ..... But it DOES have some cool technologies like Dtrace and ZFS. So what best SUN can do is to integrate these technologies with Linux and try to get maximum hardware sale and service contract on Linux platform. The problem with SUN has been that they are late in everything. They do things after much resistance. That is what has happened with Java and now going to happen with Solaris. I really wonder why its so hard for the SUN execs to understand such simple things.

Re:Sun is still screwing up, albeit not as much

[cpghost]

But it DOES have some cool technologies like Dtrace and ZFS. So what best SUN can do is to integrate these technologies with Linux and try to get maximum hardware sale and service contract on Linux platform.

Meanwhile, this is exactly what FreeBSD developers are doing w.r.t. ZFS and Dtrace... On the one hand, as a long-time FreeBSD user (since 2.0) I'm glad to see this happen, yet on the other hand, as an even longer time SunOS/Solaris admin (since SunOS 4, way before Solaris), I'm kind of sad to see Sun's gems being canibalized this way, even though Solaris would definitely deserve more exposure. And don't get me started w.r.t. Java on officially non-supported platforms: the hoops we FreeBSD admins had to jump through to get Java support were just... well, totally insane. Thankfully, things have improved a little bit now, and let's hope Sun sees the light, though it may already be too late for them.

Re:Sun is still screwing up, albeit not as much

[speedtux]

So please explain how A) it's awful that they're offering an alternative to Linux, alongside of Linux;

It's not awful, merely stupid of them because it's not going to work.

and B) how this is so much worse than what any other systems company is doing? IBM offers AIX, z/OS and "i" as an alternative to Linux. HP offers HP-UX, OpenVMS, and Tru64 UNIX as an alternative to Linux.

IBM and HP have released plenty of kernel code, and they have done so under a license that is Linux compatible, Sun has not.

The difference seems to be that Sun is open sourcing their alternative to Linux. The bastards.

No, the difference is that Sun is being dishonest. If Sun just had remained a closed source vendor, nobody would care.

rock and hard place

[speedtux]

Ponytail-guy and his pals have basically given away the crown jewels and have not been able to "monetize" any sort of decent return for their efforts.

And what brilliant strategy would you have recommended? What should they have "monetized" and how?

meh

[n3tcat]

So Sun decides that they can't make money by pushing the open source community around, so now they come on their knees and beg forgiveness before they have anymore problems.

Hmm...
/me wraps his arm around Sun's shoulder

It's alright man. Good to have you with us.

Re:meh

[WingedHorse]

Sun has been... Pushing the open source community around? Eh? They have given the OS community as much as all the other companies counted together and it just hasn't given them any money back.

Re:meh

[xSauronx]

reminds me of an arrested devlopment bit for some reason

Michael: Is this about the money?

Gob: No.

Michael: What do you want?

Gob: I mean, it's not about money in the sense that I'm coming here saying, "Here, Michael. Take some money." It's just more of a "may I have some" kind of visit.

Sun Doesn't Have Much of a Future

[segedunum]

Sun has, and has had, some great products in the past, and some of their hardware is still pretty excellent, but the problem with the company is that they still have a deep rooted protectionist attitude towards SPARC and Solaris. Why do you think it took so long to get Solaris on x86, why it took so long for Sun to accept that x86 servers was where the growth was, why most of Sun's customers still get Linux pre-installed on Sun's systems and why Sun paid a couple of billion for an excellent business opportunity in Cobalt, and then promptly destroyed it?

If they could make Solaris and SPARC stand out and pay off then fine, but they can't hence the half-hearted and pretty sad move to 'open source' Solaris just so all their consultants and execs can run around trying to tell us that it's 'just like Linux'. However, in the cold hard light of day, Linux ate Solaris's lunch, and SPARC just competes too closely with x86 based servers without the comparable performance. SPARC is so inferior to x86 in terms of raw performance it's so laughable. Solaris also suffers from the fact that Sun just don't have the resources to push development to where Linux and other operating systems are, and these days it is increasingly expensive to try and maintain an entire OS yourself.

In terms of open source, Sun's problem is that the vast majority of open source software is written for Linux and the BSDs first. No one thinks of Solaris as their first platform of consideration, and it's difficult to see why they should do so now. It's still like that now, and it was still like that a few years ago when a former employer scratched its head trying to work out why Zope and Python performance was so terrible on Solaris and an UltraSPARC. A Sun guy even recompiled Python in Forte. The bottom line answer we got from the Python devs was "We use open source systems, and possibly Windows, first and foremost on x86 systems, x86 and Linux performs better anyway, and while we'd like to help, we just don't care about your corner case problem on an OS and hardware we don't have access to and can't reproduce. Just use Linux and x86". That's not literal, but it's the general gist, and I couldn't say I blamed them.

The solution? They moved to a far cheaper x86 system with Linux, they had no installation problems with Python as it came within the package management system itself, things were far easier to manage, performance increased exponentially which pleased everyone and Python and Zope ran with no issues whatsoever. That still holds true today.

Re:Sun Doesn't Have Much of a Future

[wprowe]

Is it possible that Sun finally decided to open-source Solaris in hopes the OSS community would help them port it to more platforms? The user gets the "stability" of a commercial operating system with the "compatibility" of an open source operating system. Sun may also brag about the enhanced security that Open Solaris provides because of all the eyes allowed to poke into it, discover the holes, and close them.

Just a thought.

Re:Sun Doesn't Have Much of a Future

[drinkypoo]

Sun OBVIOUSLY gets it which is why they are concentrating on Linux on x86 today. On the other hand, it's hard to beat Solaris on SPARC for the big tasks that need a single machine. On the gripping hand, there's only so many RDBMSes with Oracle or Java Application Servers out there (even if the numbers are considerable) and if you don't have more reasons to exist than that, you're not going to exist long.

You really have to hand it to Linus for his work and his choice of license. Today AIX, Solaris, and HP-UX are all being destroyed by Linux; The commercial x86 players (BSDi and SCO) are already done. UNIX is dead, long live Unix!

Re:Sun Doesn't Have Much of a Future

[segedunum]

Sun OBVIOUSLY gets it which is why they are concentrating on Linux on x86 today.

I'm not entirely sure what you think is obvious, and the capital letters aren't going to make it true. The only thing obvious to me is that they've basically been forced into that by market forces, but they still continue to waste R & D time, money and resources on Solaris and SPARC that are not growing and are producing ever decreasing returns. They either need to do something spectacular with them or get rid. They'd be better off cutting costs that way rather than having yet another round of 'restructuring' where they fire thousands of employees and totally demotivate those remaining, wondering when they are next.

On the other hand, it's hard to beat Solaris on SPARC for the big tasks that need a single machine.

SPARC is dead, and it can't handle 'big tasks' in the way that x86 has been doing for years because it just doesn't have the raw performance - as I'd pointed out. Sun have tried to back SPARC into a niche with Coolthreads, but that is a very, very specific niche that few will see any benefit from. SPARC just overlaps too much with the work scenarios of x86 servers, a hangover from the dot com era, and it loses. At least IBM saw sense and put Power into a market where it made sense.

Re:Sun Doesn't Have Much of a Future

[gclef]

I'd love to say that you're right, but you know what? Compared to x86, SPARC performance still sucks for some tasks...especially serial (non-threaded) ones.

Example: we had a perl script processing high-volume syslog data & inserting it into a database. On sparc hardware, we had trouble getting the script to read any faster than about 600 messages per second. Switching to a similarly-priced x86 box, we max out somewhere around 4,000 messages per second.

Sun's heavy emphasis on threading makes *some* tasks work better. But, if your task is very serial (like reading & parsing syslog), SPARC just doesn't compete.

Re:Sun Doesn't Have Much of a Future

[drinkypoo]

they still continue to waste R & D time, money and resources on Solaris and SPARC that are not growing and are producing ever decreasing returns. They either need to do something spectacular with them or get rid.

While I agree, without SPARC and Solaris, Sun is just another Linux system integrator, plus Java.

Sun have tried to back SPARC into a niche with Coolthreads, but that is a very, very specific niche that few will see any benefit from.

Yeah well, I agree all around. Maybe they're waiting for a major rev of AMD or intel hardware so they can claim "this is what finally makes SPARC irrelevant". I previously thought Hammer might have done it, but it didn't, so I'm probably wrong again.

Re:Sun Doesn't Have Much of a Future

[segedunum]

I'd love to say that you're right, but you know what? Compared to x86, SPARC performance still sucks for some tasks...especially serial (non-threaded) ones.

That's my point. In terms of raw performance per task SPARC is just far too inferior to x86. To get anywhere with something like Coolthreads your workloads have to be unbelievably parallel with very little depending on each other. There also comes a tipping point where you can just get so much throughput on one thread that it outweighs the advantages of doing them in parallel.

Rather than improve SPARC Sun are just backing it further and further into a niche.

Re:Sun Doesn't Have Much of a Future

[segedunum]

While I agree, without SPARC and Solaris, Sun is just another Linux system integrator, plus Java.

The problem is, that's not any reason to hold on to SPARC and Solaris and it makes an assmption that there is some differentiation people are really going for. They're not differentiating Sun in any tangible way, and are just adding to the costs.

Yeah well, I agree all around. Maybe they're waiting for a major rev of AMD or intel hardware so they can claim "this is what finally makes SPARC irrelevant". I previously thought Hammer might have done it, but it didn't, so I'm probably wrong again.

Love the sarcasm in there ;-), but that's the attitude that is killing Sun's financial performance. The performance of SPARC is just far too inadequate, Sun are not able to put the effort in to keep up, they're trying to differentiate SPARC with things like Coolthreads in the same market as x86 and all that's happening is that SPARC's market share and share of Sun's total revenue are being eroded away.

Re:Sun Doesn't Have Much of a Future

[Jah-Wren Ryel]

It sounds like this coolthreads (is that the latest marketing name for their niagra stuff?) is like Itanium. It needs a lot more support from the software tools than is feasible. Itanium kicks x86 ass on certain, very specific (math), workloads. But as a general purpose cpu its way over-priced for the performance. Itanium had the benefit (to Intel) of scaring off most other non-x86 development (MIPS, gone. Alpha, gone. Power fading into obscurity, etc). Sun really doesn't gain any significant technical or market advantages by clinging to their current cpus.

Re:Sun Doesn't Have Much of a Future

[drinkypoo]

Sun's heavy emphasis on threading makes *some* tasks work better. But, if your task is very serial (like reading & parsing syslog), SPARC just doesn't compete.

That's why we have SPARC and x86.

Why we don't have both in the same box yet, I have no damned idea. Maybe it's just Sun's insistence on having their own extra special bus.

PJ calls Sun out for "abetting" SCOundrels

[Gimble]

Look at this post here from Groklaw, reviewing the testimony from the SCO v Novell trial.

PJ notes that SCO enacted a license, illegally according to Novell, with Sun in 2003 that allowed Sun to open source Unix Sys V. Knowing they had that, Sun still allowed SCO to embark on their SCOSource campaign against IBM and Linux users for allegedly putting Sys V code into Linux.

Re:PJ calls Sun out for "abetting" SCOundrels

[HiThere]

PJ *MAY* be overstating the case. Perhaps.

If this were a criminal case, Sun should be given benefit of the doubt. Possibly they were contractually bound not to say anything. We haven't seen the agreement.

OTOH, Sun hasn't claimed that they couldn't say anything. They've pretty much acted like their attitude was "No skin of *my* nose." But this is inferred, not deduced.

It certainly doesn't make me want to trust Sun...OTOH, it's short of enough evidence for total condemnation. (Then again, the only Sun products I use are OpenOffice and Java. And not much Java. They wouldn't miss my custom.)

Not Open Source

[Anonymous Coward]

I think they screwed themselves. I am happy to see they are changing their minds and hope that is not too late.

Re:Should have been from the Start

[julesh]

[Java]'s a great learning language, like Turing was back in the 80's-90's or so, and it can be used to teach the basics of OOP and OOP compiler design.

Java is a useful language with a well-designed and extensive set of libraries that make it one of the best languages available for implementing most kinds of business applications.

What it isn't, in fact, is a good learning language.

How they EVER planned to make money off java, is beyond me, [well, maybe the embedded versions]. This should have been open-source from the start.

1. The original plan for Java was as a language for embedded systems. The other applications were added as an afterthought, effectively.
2. It's hard to see how they would have been better off had it been free software[1] from the beginning. The language was adopted at a phenomenal rate, has had a huge amount of community input and is now effectively one of the three most important languages for commercial programming today (along with C++ and C#).

Sun's lack of profit from Java stems primarily from the fact that they never developed the kind of leading support software for it that the development community required, leaving things like IDEs and application servers for others to successfully commercialise. If Sun had produced an IDE even a fraction as good as eclipse, or a server environment as robust as IBM WebLogic, they could have made a fortune from it. But they failed to achieve that.

[1]: I use the term to avoid confusion: Java has always been a "source code available" product, but simply did not have the redistributability of modified versions that we expect from free software. Many people understand "open source" to include the former.

Re:Should have been from the Start

[julesh]

a server environment as robust as IBM WebLogic

Errm... s/WebLogic/Websphere/. Sorry, was thinking about something else...

Only 2001-2002?

[Mathness]

Reading this story seem to indicate that 2003 and forward should have been part of the "screw up" period.

It is good to see Sun throw their weight in to support open source.

Sun has a long way to go..

[Anonymous Coward]

Yeah, they screwed up big time. Groklaw has a nice article up on their involvement with SCO:
http://www.groklaw.net/article.php?story=20080625020853732

And they are still screwing up. Just have a look at the legal agreement you need to agree to when downloading even the *specification* from any of the Java Community Process groups where the project lead works for Sun. Evil and completely unacceptable terms for open-source developers...

And in those projects where they have released the source under a free license, they still keep an iron hand on the development process. So unless you work for sun, you need to beg to get your changes in (and sign all sorts of agreements). Closed bug-reporting systems. Version-control repositories that you need to apply to get read-only access to. Closed mailing lists. Design meetings held in person (Sun employees only of course).

This is a company that has a *long* way to go before they understand what Open Source is about.

Or, less charitably, this is a company that does indeed understand what open-source is about and is manipulating the system. Yes, once the source is released a fork is then possible, but for a large project inertia and an existing pool of developers all from one company make that something that takes real anger to do. So the changes Sun has made so far don't achieve a whole lot; they still completely control the direction their open-source projects take.

Real OSS companies are different; they contribute upstream, allow derivatives downstream, and are open in their process. A whole world of difference. See RedHat for a good example.

Re:Sun has a long way to go..

[JerkBoB]

Or, less charitably, this is a company that does indeed understand what open-source is about and is manipulating the system.

Never attribute to malice what can be explained by simple incompetence. Or ignorance, in this case.

Sun is a company comprised of over 30k employees. That's a small city's worth of people. Many of those people have been with Sun for a long time, from times before OSS really came on the scene.

People at the top may get it. People at the bottom (i.e. new, younger hires) may get it. The problem is that there are many people in-between who have been doing things the Sun way (indeed, the standard corporate way) for so long that OSS is just alien and bizarre.

There is indeed a lot of internal hostility toward Linux. A lot of it is just sour grapes, but there is also quite a bit of feeling that Solaris is the superior solution, and people are downright baffled that anyone would knowingly choose inferior technology. "If we just showed them the light, they'd use Solaris instead of that Linux crap!"

As with most huge multinationals, the company is made up of several distinct business units. Hardware, Software, Sales, Services, IT, etc. Sales people make money on software sales and support contracts. They also make money on high-margin government and finance sales. What they don't make much money on is bare hardware sales, especially if the customer wants Linux. Unfortunately, what this all means is that the people who use revenue streams to try and shape corporate focus are in a battle with the senior executives who are trying to shift the company away from relying on those high-margins-but-shrinking-buyer-pool revenue sources.

There is also the problem that for many people, a job is a job. They're not particularly interested in keeping up with things outside of their sphere of influence. Change means having to learn new things, and sadly, there is a lot of resistance to change (not just at Sun, of course!)

It will certainly be interesting to see what Sun looks like in 5-10 years, if it still exists as a distinct corporate entity.

I think that Sun is doing Open Source fairly well

[MarkWatson]

Sure, Simon Phipps's quote makes a good headline, but between OpenOffice.org, Netbeans, Glassfish, slowly but surely Java, etc. I would personally give Sun a good grade.

Open Source can be good for business, huge, large, and small. A bit off topic, but: while I earn most of my living consulting on (unfortunately) closed source projects, I almost always try to initially talk my customers into at least considering Open Sourcing all or parts of development projects. I believe that software development should be done in the least expensive and highest quality way possible: better for almost everyone to drive down the cost of software development; I argue that the less expensive that useful projects are, then more projects get funded. Also, about an hour ago, I received a small grant from someone in Europe to convert one of my LGPL projects from Java to Pascal/Delphi :-)

PR problem for Sun...

[mkcmkc]

The fact that GPL isn't compatible with ZFS is Linus's problem. Actually, it's more of a PR problem for Sun than anything else. The fact that Sun chose an explicit "poke in the eye" to the Linux community when they could easily have chosen otherwise has been duly noted (and is one more reason not to bother looking back at Solaris, IMO).

Re:PR problem for Sun...

[An dochasac]

From the CDDL FAQ at OpenSolaris.org:

If you wanted a copyleft license, why didn't you just use the GPL or LGPL? We needed an open source license that allowed files released under the license to be linked with files released under other licenses. While a license like LGPL would allow this for dynamically-linked code, we also needed to be able to release software that statically links source files available under different licenses. In addition, we wanted to allow others to add externsions to OpenSolaris with different license terms. This was only possible under a license like the MPL; however, we could not use the MPL because it is not a "template" license allowing reuse by others. Consequently, we crafted a variant of the MPL, taking the opportunity to make it a template license as a step towards reducing license proliferation for others finding themselves in a similar position.

If Sun were "out to get Linux", it would relicense StarOffice, Java and MySQL in a way which would prevent it from running on GNU/Linux.

There is a really nice document showing the redlined diffs from the MPL license. GPL is a great license in theory but it hasn't been tested in court and does run into problems in the real world where it hurts Linux's long-term hardware compatibility and stability, prevents many multimedia codecs (e.g. DVD+MPEG2+CSS) from ever being legally supported in a GNU/Linux distribution.

Let them come truely clean

[WindBourne]

and speak about the deal with MS to fund SCO. When you pay 10 million for software that is worth at most 200K, and you obtain a big chunck of stock, well, that is not about a set of drivers. And let not just the OSS world know about that, but some of the regulatory groups.

Re:I think that Sun is doing Open Source fairly we

[jedidiah]

I give Star Division all the credit for Star Office.

Sure, Sun decided to come along after it was pretty much finished
and but it out and then give it away. However, it was the Germans
that actually built it from the ground up and gave us something
useful. They had the vision and the interest in creating the thing
in the first place. They chose to support us.

It wasn't 'ideal' in terms of free software zealotry but they were
there at the table with us at least showing us enough respect to
acknowledge us.

Giving Sun the main credit for Star Office is like saying that Gates invented the internet.

Re:Still find Java hard

[Tim C]

I write decent Visual Basic code and I am quite adept with OOP, but still find writing Java code quite hard. Is it the case for Java folks out there too?

Well I can only speak for myself, but no, I don't find Java hard. Then again I've used a variety of different languages casually, academically and professionally over the last 25 years and I've never had much trouble picking up a new one (that sounds rather grander than it really is - for much of the first 11 years of that I was only programming very occasionally indeed; my professional work began a little over 9 years ago)

Also as others have said VB is no more dying than C or C++ are; there will be plenty of work at least maintaining existing code for a good few years. You can still find COBOL jobs if you look hard enough (and Fujitsu has released COBOL.NET!)

I'am thinking WABI circa 1993

[jabuzz]

I would say this is the single biggest screw up that Sun have made on the open source front. Had they put WABI on an open source license then the effort that has gone into Wine would have started with a mostly functioning version, and would have given people a real OS choice over a decade before it became one for most people.

I would also cite the failure to opensource the XView desktop as another mistake.

Re:I'am thinking WABI circa 1993

[boner]

WABI was not Sun's IP to give away.

Re:Counterpoint to your Solaris x86 install issues

[pirhana]

> I have installed Solaris x86 8/07 on DL385 systems; what exactly was your problem? HP provides the SmartArray (cciss) to use during the installation (and Broadcom NetXtreme (BNX) drivers to install afterwards).

This is exactly my problem. You need to load the drivers at the time of installation where as in Linux it was supported from day one for DL-385. This is for servers and for Desktop/laptops its even worse.

> but a *guess* would be that for current, popular server systems sold by major vendors (Sun, HP, Dell, IBM), Solaris x86 support shouldn't be that bad.

But people don't deploy things on the servers first. They deploy it on laptops/desktops first and test/learn from it. So support for all type of hardware is of utmost important for widespread adoption.

possible correction

[drewness]

I got Jython working on my new machine, and it's considerably better. Like, almost on par with CPython - modulo the fact that it's still quite behind feature-wise. It might be the JVM (1.5 before vs 1.6 now) helping. Also, IronPython has gotten worse at generators between Alpha6 and Beta2 somehow.

Test machine is 2GHz Core 2 Duo iMac with 4GB SDRAM running MacOS 10.5.3.

awatts@platypus:~$ python2.5 timerseqs.py
2.5.1 (r251:54863, Jan 17 2008, 19:35:17)
[GCC 4.0.1 (Apple Inc. build 5465)]
forStatement => 4.59940099716
listComprehension => 2.38550591469
mapFunction => 1.92037510872
generatorExpression => 3.14438390732

awatts@platypus:/Applications/IronPython-2.0B2$ mono ipy.exe ~/timerseqs.py
2.5.0 (IronPython 2.0 Beta (2.0.0.2000) on .NET 2.0.50727.42)
forStatement => 9.48001098633
listComprehension => 6.89910125732
mapFunction => 5.57528686523
generatorExpression => 19.7973022461

Note: Jython doesn't support generator comprehensions
(With Java 1.6.0)
awatts@platypus:~/jython2.2.1$ ./jython ~/timerseqs.py
2.2.1
forStatement => 6.0929999351501465
listComprehension => 3.7769999504089355
mapFunction => 1.6570000648498535

yeah they are mormonic

[commodoresloat]

I hear the upper management takes a lot of LDS. Some of them have been dropping it since the early 80s.

Re:yeah they are mormonic

[neomunk]

Well, ya can't blame them, two hits of double-dipped polygamy can really boost your outlook on life!

Re:Should have been from the Start

[CastrTroy]

What's wrong with NetBeans. I know a lot of people prefer Eclipse, but personally I like Netbeans better. Netbeans definitely isn't a bad product. And if you count Forte, it's been around quite a bit longer than Eclipse.

Our Fiendish Plan

[fm6]

Speaking as a Sun employee: you're welcome.

But do remember that there's an element of self-interest in this open-sourcing strategy. It's all part of our fiendish plot to sell people hardware and services.

Take Solaris, for example. By opening it up, we do lose the income we would have had from selling it to people. But that's been dwindling anyway, as Solaris loses ground against Linux and Windows. By opening up the OS, we make it a better product through user contributions, and encourage its spread. More Solaris users means more people who will seriously consider out products and services.

Of course, even Linux and Windows people should be looking at us anyway, since we are now serious about products that run those OSs. (I work on documenting several of them.) But if you're already a Solaris user, then your options go beyond our x64 systems to the systems that are still the core of our business: the SPARC machines.

There are many reasons SPARC systems have been losing ground. But a big one is that they don't run "standard" operating systems. Promoting Solaris through open-sourcing (and through other means, such as supporting it on other vendor's hardware) drastically changes that particular equation.

Re:Our Fiendish Plan

[hherb]

I believe the very best situation arises when a company actually sees the benefits of Open Source and can reap some profit from doing it.

There is nothing wrong if both customer as well as vendor benefit and both sides can be happy about a deal without one side ripping the other one off - it is the Nirvana of commerce

Good on Sun for doing what they did and are still doing.

Re:Should have been from the Start

[rapiddescent]

1. The original plan for Java was as a language for embedded systems. The other applications were added as an afterthought, effectively.

this is not entirely correct; I met with Arthur van Hoff (he of java.lang.String fame) back in the late 90's when a large chunk of the original java team went to Marimba to develop Castanet - they really thought that the future of Java would be centralised software as a service style apps that would download to your thin 'pc' 'workstation' etc as you needed them and would always be up to date. I have a Java Kettle workstation in the loft that had the Java VM-hardware in it and ran only Java.

at that time, large organisations like reuters were spending a tonne of money using this paradigm to reinvent online applications. reuters had a huge team working on a java version of RT that sucked down java classes as it needed them - after all, the Internet/WWW hadn't quite taken off yet - sure web browsers were around, but in 1995/1996 functional web pages were rare and web sites not taken seriously. (ask Bill Gates).

the original java team were wrong of course, Java has taken off in an unimagineable way in enterprise middleware. large enterprise middleware such as trading systems, healthcare systems, enterprise messaging, ERP, etc etc totally rely on J2EE and J2EE implemented Service Oriented Architecture. RedHat/JBoss, Apache Jakarta, BEA/Oracle and IBM make a living selling enterprise Java systems.

Do you know how you can help Sun!

[boner]

Everybody out there complaining about Sun needing to open-source this or that, should take some time to consider what they would do in return!

Next time you need a new, heavy duty workstation or server, checkout Sun's offerings - rocksolid engineering, excellent price performance.

We all know Sun made mistakes, and like in all human endeavours they will likely make some more, at least they are trying to innovate and share - it's up to you to participate.

Re:Should have been from the Start

[Captain Spam]

What it isn't, in fact, is a good learning language. To be honest, I disagree. The concepts of object-oriented design didn't make much sense to me at all (tried to learn it in C++ in college classes) until I started messing about with Java. And, more specifically, once I started leafing through the standard Java libraries. Looking at how the API was organized suddenly made OO make sense to me. I think the exact point of realization was "Wait, I get it... all these GUI objects ARE JComponents, and that's why I can treat them all the same!".

So I'd say it's a decent learning language as well. Ironically, learning Java made C++ make a lot more sense to me than learning C++ ever did. :-)

not that bad

[tatermonkey]

As I sit here at work reading this, I have a spreadsheet minimized in my nice and free ( as in beer) Open Office. Need I say more.

Re:not that bad

[HiThere]

OpenOffice may be "free as in beer". It's also GPL.

Re:I think that Sun is doing Open Source fairly we

[MarkWatson]

I thought that Sun paid a lot of money for StarOffice, and then Sun open sourced it.

Isn't this right?

Overpriced

[DesScorp]

"I know Sun has the stigma of selling only overpriced iron"

They're considered overpriced now, but the reason Sun (and DEC as well) hit the big time in the first place was because their products were far cheaper, but just as reliable, as Big Iron products from IBM/Sperry/Burroughs/etc. It's not so much that Sun got greedy as X86, Windows Servers, and Linux came and changed the marketplace the same way SunOS and Sparc servers did.

That's quite a TROLL...

[mkcmkc]

First of all, the question is not why they didn't use the GPL, but why they didn't use one of the many GPL-compatible licenses.

Second, a license that prevented programs from running on GNU/Linux would (by definition) not be an Open Source license.

Third, I suspect the GPL is the Open Source license *most* court-tested.

Fourth, Linux's GPL license does not prevent any codec from running on it. It's the authors of the codecs and patent holders that do that.

And finally, the GPL hurts Linux's stability? Truly it is a powerful license, but I never imagined that it had such capabilities...

Re:That's quite a TROLL...

[An dochasac]

First of all, the question is not why they didn't use the GPL, but why they didn't use one of the many GPL-compatible licenses.

That question is answered in the link, at least as far as the most common OSI approved licences, GPL, LGPL, MPL...

Second, a license that prevented programs from running on GNU/Linux would (by definition) not be an Open Source license.

Neither GNU nor Linux is mentioned in OSI's definition of Open Source.

Third, I suspect the GPL is the Open Source license *most* court-tested.

You may be correct here, but don't be surprised if it does eventually break apart, businesses (especially U.S. based businesses) are accustomed to dealing with extremely complex laws and if GPL gets in the way of the likes of Microsoft/GM/ADM... look out! Many companies have higher revenue than the GNPs of all but the G8 and they don't even like abiding by U.S. tax law, much less Kyoto, WTC etc. GPL is hardly a bit of road-kill, much less an obstacle to the wishes of these mega corprations.

Fourth, Linux's GPL license does not prevent any codec from running on it. It's the authors of the codecs and patent holders that do that.

Then you'd better spend your time trying to lobby Sony, Apple, Microsoft, Fraunhoffer, the MPAA... instead of worrying about what license Sun frees its software under. You can blame the bad people who wrote these CODECs or you can look into the real incompatibilities between GPL and codecs, problems which LGPL, BSD, MPL and CDDL don't have:

This is taken from the fluendo website http://www.fluendo.com/resources/fluendo_mp3.php) "If you are living in a country where the mp3 patents don't apply you can of course use the source code provided by Fluendo (or anyone else) to get legal mp3 support onto your Unix/GNU/Linux desktop. On the other hand, if you live in a country where the patents apply, or if you are a distribution maker who sells your distribution in countries where the patents apply, then you need the licensed binary from Fluendo. This of course is no problem, but be aware that even if our binary is made from MIT licensed source code the resulting binary combined with our license is not free software, at least not GPL-compatible. This means that if you ship GStreamer with our binary mp3 plug-in, you need to be sure that you don't ship any GPL-licensed plug-ins that could end up being used together with the mp3 plug-in, as this would violate the GPL. And you don't want to violate the GPL. You also need to make sure you don't ship any GPL-licensed players which would use this plug-in.

And finally, the GPL hurts Linux's stability? Truly it is a powerful license, but I never imagined that it had such capabilities...

It hurts Linux's stability by allowing Stallman and thousands of others to live in a fantasy world where GPL's viral nature overcomes Microsoft/Sony/Apple and other IP holders will bow to a non-paying 1% marketshare and make things convenient for this minority. It allows Linus to "not care about ABI stability" because he too lives this fantasy.

What if it does?

[DesScorp]

And I really, truly hope it works out for them. I hate for it to go the wrong way... What if it does? What effect would that have on other companies considering opening up their code? Sun took a big gamble. They didn't GPL some lame side project... they surrendered their bread and butter. And the fiscal results thus far have been, shall we say, not encouraging.

I think they made a mistake. IBM doesn't open source everything. Unless I missed it, their mainframe OS and AIX are still closed source, and it hasn't hurt them. Face it, most really profitable IT products involving software have at least a degree of exclusivity. If you give away the store, what's left to sell? Thus far only Red Hat has managed to make significant money selling services, and even they accomplished that by introducing a level of exclusivity, namely by making it very very hard for the average user to get a whole copy of RHEL without paying for it; they make you go through a song and dance to get the code, and then you have to compile it all. So technically they're free, but not practically.

It's a shame too, because Sun's latest line of Sparc processors are simply outstanding. They don't have record breaking clock speed, but in work-per-clock-cycle, and in work-per-watt, the new Sparcs are really outstanding. They also had some really good product ideas in their emergency-mobile computing areas... putting instant server farms in truck containers, for instance.

While "giving away the store" has hurt them, I also think their current leadership has hurt them as well. Paying 1 billion dollars for MySQL? Are you kidding? When Jonathan Schwartz was bumped to the top office, it was said that there was a collective "what the???" at Sun. I think his performance since then has only reinforced that.

Sun complicit in SCO lawsuit?

[Tracy Reed]

According to this Sun may still not be honestly supportive of FOSS:

http://www.groklaw.net/article.php?story=20080625020853732

This says that Sun already had the rights to open source SYSV even when SCO sued IBM for open sourcing SYSV technology (which didn't actually happen) into Linux. Yet Sun stood by and said nothing.

Perhaps they are secretly hoping Linux will get stomped so that Solaris can make a comeback?

It seems more and more that you just can't trust anyone and that Sun isn't and probably never will be as FOSS friendly as they would have us believe.

What went wrong

[guacamole]

In 2000-2001

1. They screwed up by announcing the end of line for Solaris on X86
2. They screwed up by refusing to offer X86 hardware.
3. They screwed up by not offering Linux on any of their hardware
4. They screwer up by not open sourcing Java, Solaris, and other goodies.

In the end, they are trying to correct all those errors, but I wonder whether doing that 7-8 years later means that they missed a golden opportunity to become a leader in the Linux and Unix software and hardware market (including on X86).

Apology is spelled sun4m

[sethstorm]

In order to remedy the alienation, Phipps said Sun is "leading by changing behaviour rather than by just saying good words". Fine. Start by ending the stonewalling of sun4m code. Then start looking at opening hardware such as SunPCI cards (that is, hardware that has no non-Sun code equivalent).

Open that code up and consider that a significant part of a complete, sincere apology.

Re:How to really help out SUN.

[sethstorm]

Then discover that they yank support and documentation well before about anyone else. No thanks, as IBM gets that one right.

Re:Should have been from the Start

[julesh]

1. The original plan for Java was as a language for embedded systems. The other applications were added as an afterthought, effectively.

this is not entirely correct; I met with Arthur van Hoff (he of java.lang.String fame) back in the late 90's when a large chunk of the original java team went to Marimba to develop Castanet - they really thought that the future of Java would be centralised software as a service style apps that would download to your thin 'pc' 'workstation' etc as you needed them and would always be up to date.

When I say originally, I really do mean originally. Van Hoff didn't join the Java team until they'd been working on it for two years, with the primary goal being a language to enable more advanced applications for 'digitally controlled consumer devices'. The focus was already shifting by the time he joined the team, so I'm not surprised he felt differently about it.

Re:In other words...

[HJED]

Java is a very good programming language, which is not buggy

Re:Should have been from the Start

[julesh]

To be honest, I disagree. The concepts of object-oriented design didn't make much sense to me at all (tried to learn it in C++ in college classes) until I started messing about with Java. And, more specifically, once I started leafing through the standard Java libraries. Looking at how the API was organized suddenly made OO make sense to me. I think the exact point of realization was "Wait, I get it... all these GUI objects ARE JComponents, and that's why I can treat them all the same!".

So I'd say it's a decent learning language as well. Ironically, learning Java made C++ make a lot more sense to me than learning C++ ever did. :-)

I suppose Java isn't an absolutely terrible language to learn OOP in. C++ is definitely worse. But there's a problem with Java in that it encourages a "string together these prepackaged solutions to make a new one" approach to programming. While this is great for actually achieving stuff, it's not ideal for learning with. Which is why I'd recommend a simpler language for learning.

I picked up OOP using Borland Pascal with Objects (the forerunner to Delphi), which was a good experience. It came with a standard library ('TurboVision') that allowed the same intuitive understanding that you're talking about picking up from Swing, but was substantially simpler and could be easily learned in its entirety. Of course, using a DOS environment to learn might not be entirely wise these days. Delphi is more complex, but still simpler than Java.

I'm generally impressed with the skills of those whose first OO experience was with Smalltalk: that's a language that really encourages you to think about what objects and classes are and how to use them effectively. I imagine that the gap left by Smalltalk's waning popularity has been filled by Ruby, which is a similarly expressive language at least. Python, also, is a rather expressive language, although its standard library lacks the strict OO basis I'd consider ideal.

These are all much better languages than Java for a learner, IMO.

Don't agree.

[Stu Charlton]

For a while, I would agree with the above, but I'm seeing a lot of signs of life out of Sun lately. They really get open source software, and are making money off of it. Simon Phipps this week at Jazoon '08 noted they're making more money off OpenSolaris these past couple years than the past 8 years combined.

It's very easy to pick at a company's decisions -- but it's really hard to turn around a huge company with an entrenched culture; other UNIX players weren't pure plays and are so diversified that it's both easy to hide their own problems (how's HP-UX doing?) or to entice hardware purchases because of broader relationships & bundling (IBM is classic at this).

Sun still has a lot of runway ahead of it. $3B in cash. $13.8b in revenues a year, which is UP $2.8b from 2005. The recent quarter problems are concerning but in context were a 0.5% drop in revenue year-over-year. Yes, it's very bad that they're not very profitable, but let's put it in context -- their big losses were 5 years ago or more. Apple was in much, much worse shape in the mid 90's. Motorola just lost $2.1 billion in revenue from their mobile phone division.

Sun is drifting slowly towards death, but, as they say, the green mile is sometimes quite long.

Cunning Marketing Ploy

[turgid]

Announcing the end of Solaris x86 was a cunning marketing ploy to test the waters as to the demand for Solaris on commodity hardware. Not long after they resurrected it a port to x86-64 was done...

Powerful my ass....

[jotaeleemeese]

Now if I want to buy your appliance I have to ponder that I would be locking myself in.

With somebody selling me a similar appliance with a GPLed core, I would not have such concern.

For consumers GPLed stuff is the way to go, any companies that do not heed this may struggle in the future.

Sun open sourced it.

[jotaeleemeese]

SO credit where credit is due.