Slashdot Mirror
IBM Derides OpenSolaris as Not-So-Open
MaverickFire writes "OpenSolaris isn't a true open-source project, but rather a "facade," because Sun Microsystems doesn't share control of it with outsiders, executives from rival IBM say.
"Sun holds it all behind the firewall. The community sees nothing," Dan Frye, the IBM vice president who runs the company's Linux Technology Center, said. Sun could do "simple things" to build a real OpenSolaris community if it were serious about doing so, Frye said. "They would push their design discussions out into the forums, so people can see what's going on," he suggested." I talked to one of the OpenSolaris developers at the project's LWCE booth in the "dot-org ghetto," and though it wasn't in response to this article, he pointed out that OpenSolaris takes contributions from all comers, has active public mailing lists, open IRC channels, and several online communities, so Frye's description seems at least overblown.
It's more open than AIX, that's for sure.
Game... blouses.
I think this has more to do with IBM feeling the heat over not doing *anything* to open-source AIX. Sure OpenSolaris isn't quite as open as some would like, but it's more than what IBM has done with AIX. C'mon IBM, open up AIX!
init 11 - for when you need that edge.
This is true. It is more open the AIX. IBM does not, however, claim that AIX is open. OpenSolaris is also more open than Windows, whatever software the NSA uses to crack codes, and a closed door, but of these things, only OpenSolaris claims to be open, and it is these claims Mr. Frye is addressing.
I don't know... IBM hasn't been out and about announcing how open-source AIX and OS/2 are (going to be) (any day now). The problem here is that Sun seems to want all the PR that a "leader of the FOSS community" deserves without actually dipping more than their big toe in the water.
My turnips listen for the soft cry of your love
Complaining about how Sun's OSS software isn't that opensource is merely a ploy by IBM in order to copycat Suns' projects and start making a profit. IBM is as ignorant as SCO. However, Sun has everyright to not be allow access to certain parts of code that might be pure proprietary. If they are funding the project than why should those that are not complain that they are not being 'that' open with their project?
This is why I think opensource is not that good for conglomerates such as IBM, SUN, etc. because they are all competitors to each other and would you share a secret to your enemy? That's suicide from a business standpoint!
Previewing comments are for sissies!
What else would you expect from IBM? Their entire Linux strategy is based on the idea of luring people in with Linux and then signing them up for ridiculously overpriced "consulting services' that usually results in a recommendation to purchase their own proprietary hardware running AIX and ever more extensive service contracts and recurring revenue for IBM. They are now seeing customers running Solaris 10 on IBM hardware and more and more requests for Solaris 10 instead of their own stuff and its not a pleasant prospect to see where the trends are heading for IBM.
When they open AIX and their Power chips like Sun has done with Solaris and SPARC, then they can compare and see how things stack up. For now, its just alot of sour grapes from an aging dinosaur to one that has recently been seen rising up again.
So this is just like OpenOffice.org then? I've read a lot of complains that OO.o is tightly controlled by Sun.
Sun should just do as AOL did and spin off their open source projects as a seperate company.
http://www.popularculturegaming.com -- my blog about the culture of videogame players
Whether IBM is right or not that OpenSolaris has a development community, OpenSolaris is true Free Software.
Free Software is not about a development method but about a way of licensing software. Free Software can build in a community process and in a in-house process as proprietary software can be developed in a community or in-house. It's not the development method which makes something Free Software it's the license.
Sad to see that even such a big company with such a big "linux-centre" like IBM doesn't really understand Free Software.
Support Free Software! Join FSFE's Fellowship: http://fellowship.fsfe.org
There are a variety of very good Open Solaris distros now:
o nid=6E46815A1C5CC33AC6470A9439DABAA6#all/
Belenix: http://belenix.sarovar.org/belenix_download.html/
Polaris, Solaris for PowerPC: http://www.blastware.org/
Nexenta, the Solaris/Ubuntu mix: http://www.gnusolaris.org/gswiki/Nexenta_OS/
And of course you can go straight to the official Open Solaris Communities page here: http://www.opensolaris.org/os/communities/;jsessi
Fight IBM FUD with Open Solaris Fact.
I wasn't trying to defend Sun, I find their actions quite ugly too. I just think it isn't IBM's place to point their finger at others and whine about their competitors while their own offerings suck such as bad or even worse.
I can't imagine how many licenses, agreements, contracts, and who knows what else would prevent IBM from sharing AIX or OS/2 code. Besides, what does "if IBM really cared about openness" mean? Of course IBM cares about open source. For the love of Pete, they have a vice president in charge of linux and open source. More importantly, can you think of any company ANYWHERE, for-profit or not, that's done more than IBM has for open source? How much of the modern linux kernel was written entirely or with significant help from IBM?
If IBM really cared about openness, they should open source AIX or OS/2 and shut up about Solaris.
IBMs donated some AIX features to linux and MS has some say in what happens to OS/2.
While I warmly thank Sun for their massive donations to free software, I wish they'd just STFU until they actually Open Source something. Most of the criticism they get is for flip-flopping on open source.
There are shills on slashdot. Apparently, I'm one of them.
IBM is just full of it. If OpenSolaris were not for real do you think they would have gone to the trouble of changing their source code control system from the in-house Teamware stuff to Mercurial (see this ).
No, that is the kind of wrenching and disruptive change that you do if you're really serious about pulling in developers outside the corporate WAN. If it were a facade they could have built a more impressive facade much more quickly.
Progress is slow on OpenSolaris because unlike Linux in 1991, Solaris is already a mission-critical operating system in many enterprises, and because they are trying to pull in non-employee contributions whilst maintaining quality. This is actually difficult.
Disclaimer: I was on the invite-only OpenSolaris pilot program and got some free t-shirts (none of which fit).
As you said, consumers have been clammering for IBM to OSS OS/2. (Tick, Tock, Tick, Tock) We're still waiting.
What would be interesting to keep an eye on is if OpenAIX or OpenOS/2 show up anytime soon. If they do, it could be indicitive that this FUD is all part of IBM's plan to promote their own OSS projects. Another thing to consider if this happens, is if they would have been released without Sun taking action first?
Javascript + Nintendo DSi = DSiCade
I am an outside contributor to OpenSolaris. I have several projects which are currently in the process of getting integrated into Solaris.
It is true that the development model at Sun is a bit more "Cathedral" than "Bazaar", and there are still some technical and administrative challenges to solve (for example they haven't figured out how to get folks to directly commit to OpenSolaris yet -- you have to hand off code to folks at Sun who integrate your code and walk it thru the process.)
Development of Solaris has always been a tricky thing, and historically has had huge amounts of "process" to get changes. This is because there are numerous quality safeguards, and committees that have been involved. There are famous questions that every project integrating has historically had to answer: (is it i18n safe, what interfaces does it expose? does it conform to various standards already established? is it portable to both intel and sparc? etc. etc.)
Part of the review process also has to uphold things like Sun's binary compatibility guarantee. In any respects, the _quality_ of Sun's Solaris product is much higher, I think, than what you find in say Linux, where churn is a lot higher and quality and oversight controls a bit less.
Anyway, it is possible to contribute to OpenSolaris now, though its a bit of a rough road right now. But they are making it better, and I expect it will be a lot easier in the next year or so.
The way I see it, the reason IBM is acting like this is because they refuse to open source their own major programs (like DB2 and AIX).
_ on_mac_os_x_ on_macos_x_at
So they can't say "Sun is doing a good job at open-sourcing their own software" because then they'd be asked "so why aren't you doing the same?" - and because nobody likes to admit a competitor is doing a good job.
So we get these mealy-mouthed attacks instead.
Given that DTrace has been integrated into MacOS X into Leopard:
http://blogs.sun.com/roller/page/bmc?entry=dtrace
http://blogs.sun.com/roller/page/mws?entry=dtrace
and that it's also being worked on for FreeBSD, isn't that proof enough?
Sun.
I wasn't so much saying the IBM guy is wrong, than I'm saying he's the wrong person to be pointing it out.
Between OpenOffice, OpenSolaris, and their work with GNOME, Sun has made plenty of solid contributions to OSS. Now they're supposedly opening the source for Java, which is the one thing everyone's been screaming about for the past five years and -- IMHO -- the only thing that keeps Sun relevant anymore.
I don't give a rat's ass about Sun, but they seem to be trying. Some douche from IBM doesn't need to be getting in their face because their OS code isn't open enough when IBM won't put AIX or OS/2 out there at all.
And the comment about there only being room for one open source OS is total bullshit. I hope his opinion doesn't represent the majority of IBM's staff.
Game... blouses.
Components of OpenSolaris are also showing up in other operating systems: DTrace will be in the next release of Mac OS X and FreeBSD. Speaking personally as one of the DTrace engineers at Sun, it's been quite a pleasure working with both the Apple and FreeBSD kernel engineers -- pretty decent community for a "facade".
It beats the hell out of OpenAIX. On acount of being somewhat more... existant.
But IBM never claimed to opensource their OS's.. so i don't see what their 'offence' is, sun however does tout the 'we are all about open source' horn, but in practise is not so much
;-)
Also IBM isn't such an offender, they've contributed a lot to the kernel, apache, and many many many oss projects; Which is something i personally value a lot more then opensourcing OS/2 forinstance
It's at least as Open as OpenVMS!
No folly is more costly than the folly of intolerant idealism. - Winston Churchill
This statement does not jive with your previous statement. Either Sun releases Free, OSS software (in which case they have a right to be heralded) or they don't (in which case they should STFU). Since I just rattled off three Sun OSS projects at the drop of a hat, I'm thinking that the former is the true case.
Javascript + Nintendo DSi = DSiCade
That doesn't really prove their commitment to open source in general beyond their commitment to making profit. Which is not a bad thing.
Yes, quite a few. Red Hat, SuSE, Novell, and even Sun, to name just a few.
How does IBM's contributing to the Linux kernel compare to Sun open sourcing an entire OS?
> "Sun holds it all behind the firewall."
Trans.: "I know a techie word and I'm going to use it."
Ade_
/
Big Bubbles (no troubles) - what sucks, who sucks and you suck
The poster claims
.Net is certainly open source. But it's not. I congratulate Sun on what they're doing, but that's still not true open source. Making the definition of open source muddy is really not a good idea.
"...he pointed out that OpenSolaris takes contributions from all comers, has active public mailing lists, open IRC channels, and several online communities, so Frye's description seems at least overblown."
With my apologies, if these things make something open source,
OS/2 will never be open sourced. AFAIK, Microsoft holds copyrights to some of that code.
But, maybe not in the way you mean.
Sun's own tools have driven more people to install GNU software on a Solaris machine than any other thing has caused people to migrate to Open Source.
Back in the day, a Sun which didn't have GNU tools was not very useful.
Cheers
Lost at C:>. Found at C.
I should have said:
I wasn't disputing Sun's contributions to Open Source, just saying perhaps they should hold off announcements until they're actually ready to like, you know open source something.
There are shills on slashdot. Apparently, I'm one of them.
I think that statement is relevant to this very day.
Like Eclipse, for instance!
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
Unfortunately, it has little to do with the issue at hand. OpenSolaris is fully released, and has several distros based on it. So this rant of one IBM executive is completely baseless and probably intended to promote IBM at Sun's expense.
Javascript + Nintendo DSi = DSiCade
So bugger off *BSD. Very open-minded of him
Most likely, AIX may also have some code in it that prevents them from open-sourcing it due to licensing.
Keep in mind they're already in enough legal battles over intellectual property licensing. While SCO's claims regarding IBM and Linux may be trollish, the impression I get is that SCO WOULD actually have significant legitimate claims against an "open source" AIX.
The end result is that rather than opensourcing AIX (which would be a rather pointless endeavor as the impression I get is that IBM is "sunsetting" it in favor of Linux), IBM is simply taking all of the Good Parts from AIX which they can and merging them into Linux.
Remember, open-sourcing a product isn't always a simple matter of taking a snapshot of your source tree, making it public, and adding a new license. Frequently, a company may not own all the code in a program and can't open source it without ripping out some of their code and either spending time replacing/rewriting it or releasing what is essentially open-source crippleware. In a situation where there is no even remotely competitive open-source alternative (see Quake and Mozilla), it makes sense to release crippleware and let the community fill in the holes over time, as even if it takes the community years (Mozilla/Firefox) to fix the holes, it still puts them way ahead. In the case of AIX, there would be utterly no point whatsoever in releasing it if IBM were required by licensing agreements to remove critical parts. Unlike Mozilla, with AIX there's a healthy and robust open-source competitor which would be dominant in developer and user mindshare even if it were open-sourced in complete form.
retrorocket.o not found, launch anyway?
I don't see openAIX floating around.
;)
Sure they've ported some of the technologies and added the opensource toolbox to AIX (imagine an RPM that can be installed on AIX and interfaces with the existing AIX package system).
Why is there no JFS2 for Linux? Why can't I mount a JFS2 filesystem on the SAN on my Linux machine? Why has the AIX lvm not been ported to Linux or why has IBM not contributed to the Linux LVM2 the ability to import AIX volume groups along with the requiste filesystem support on Linux? Why the hell don't I have lsdev, lscfg, lsattr for Linux? That alone would save me alot of effort.
Look the ODM is not the greatest thing since sliced bread but AIX has other good ideas that IBM should contribute instead of bitching about OpenSolaris. Shit they just want to sell more pSeries boxes anyway
"Fighting the underpants gnomes since 1998!" "Bruce Schneier knows the state of schroedinger's cat"
I suggest looking through the Linux kernel change histories sometime. There are a _lot_ of IBM email addresses in there.
And not just there. Have a look at most Apache projects too, for that matter.
There's a reason why SCO went after IBM. Well, ok, a second reason, beside the obvious "because SCO is on a pump and dump scheme." Like most lies, SCO's "IBM took our IP they had used in AIX and put it into making Linux enterprise-ready" is based on a small grain of truth, although in this case one irrelevant to the lawsuit. The truth is that IBM did donate that much code to Linux, and some which, indeed, is a part of why Linux is enterprise-ready OS instead of an academic toy. At any rate, a lot of that is either AIX code or it uses techniques developped for AIX.
If you read the RTFA, even there they spell it out repeatedly: "It prefers Linux and its own proprietary version of Unix, called AIX." ("It" being IBM.) Or even better: "IBM helped put Linux on the map, funding programmers to improve the operating system and offering early pledges of support that indicated it was safe for customers to use. The company has more than 600 programmers at its Linux Technology Center, but it's actively involved in many open-source projects besides Linux."
So basically IBM _does_ put a lot of money and work into a F/OSS OS. It's not AIX, but in hindsight, a lot of us actually prefer it that way. The great Unix fragmentation happened precisely because everyone wanted to make their own flavour deliberately incompatible to everyone else's, trying to lock their customers in. And that's how Unix lost back then, and why nowadays we have Windows instead on most computers. Does anyone (other than MS) want _that_ to repeat verbatim again? Not me, anyway. So thank goodness that IBM contributes to Linux this time, instead of trying to divide-and-conquer the F/OSS OS market with an OpenAIX.
I don't know exactly how "open" OpenSolaris is. Maybe it's really open, maybe it's one of Sun's usual smoke screens. No idea. I couldn't be bothered to care about it at that point.
But even OpenSolaris is a very new development. What I'm getting at is: IBM was putting its money where its mouth was, _long_ before Sun.
So excuse me if I find it outright funny to see someone claim that IBM isn't doing anything there.
A polar bear is a cartesian bear after a coordinate transform.
...I like some of the things IBM is doing now, but never forget they are a very, very big company whose agenda always directed at making money for their shareholders. They have a business motive behind everything they do.
IBM is a big champion of Linux now, but it wasn't all that long ago that they were issuing stern warnings to those who foresake the safety of proprietary software about the dangers of getting "locked into open source."
IBM would probably happily lock people into Linux... whatever, exactly, that would mean... if they can figure out how to do it and can see an advantage to IBM in doing it.
"How to Do Nothing," kids activities, back in print!
You do realise that IBM are in court right now, for the heinous crime of taking large gobs of its own AIX code and putting it in Linux, aren't you?
One good proof that Solaruis is really open is that DTrace is now in Mac OSX, or at least in the version of OSX released to Apple developers. I thinki that is a sign of an open project: When other projects can use your code in theirs. To be truely open code has to flow both ways to and from your project. Getting DTrace into OSX is a major contribution by Sun.
The problem is some of the OS/2 code is still owned by Microsoft. Stuff from back in the OS/2 1.x days when a lot of the code was still contributed by them.
-uso.
What you hear in the ear, preach from the rooftop Matthew 10.27b
How does IBM's contributing to the Linux kernel compare to Sun open sourcing an entire OS?
Not to mention having made such things as OpenOffice and NFS available. Oh, and that whole well-used language (which is now being open sourced) that a lot of projects are built on.
Everything I need to know I learned by killing smart people and eating their brains.
Last year I spoke with a group from Sun's Santa Clara offices at the OSBC East conference in Boston, and asked them about Sun's open source efforts. After drawing blank stares, and a bit of hemming and hawing after requesting that they actually call someone ("Umm...give us a card and we'll get back to you"), they finally relented and made a few phone calls. Got the name of someone in Austin. Came back to Texas, and to this day I've never heard back from said individual, despite several attempts to contact him. (I'm truly confounded as to why Sun would show up at an open source conference, only to disavow their open sourcedness.)
Sun appears to treat its OSS efforts as some sort of "dirty little secret," so much lip service paid to the OSS community so we'll just go away and stop hounding them. At this point, I don't believe Sun is sincere about OSS, at least from a corporate standpoint. IBM's position appears to be completely justified.
It's interesting to see IBM taking jabs at Sun, though
Typically, in recent years, IBM has been pretty classy about not disparaging competitors while Sun seemed to spend most of their waking hours trash-talking just about everyone.
When I see one company diss-ing another I tend to think less of them and I assume they are speaking from a position of weakness (which is what I have thought about Sun for years). So when I read this I tend to think like you are, is IBM worried about something?
Dan Frye also said that there was no Internet in 1991, so you'll forgive me if I laugh everything he says off.
Javascript + Nintendo DSi = DSiCade
Who sets the standard for what defines "Open Source"?
But who ever said you have to design in the open to be open source? I mean, thousands of smaller projects are generally designed by one guy in his den, without any public discussion. Are they not open?
I've read a number of open source licenses, and I don't remember any of them saying a thing about designing in the open. They all have different requirements, but they usually require that the source code be open (to different degrees). OpenSolaris is that. OpenSolaris is open source, and they are okay in my book calling themselves OpenSolaris.
In my opinion, I'm far happier with OpenSolaris being open source than I am about AIX and OS/2. I credit Sun for doing what they've done. The code was their property, and they were welcome to do what they wanted, how they wanted, and to what extent. I'm greatful they have made the source open, considering they didn't have to, and I think rediculous for IBM to criticize it.
It would be nice if we could have the rest, though. Even if half the OS needs to be rewritten, the other half won't, and maybe some project like osfree won't be such a lead balloon.
-uso.
What you hear in the ear, preach from the rooftop Matthew 10.27b
...on Sun to make the imminent open-sourcing of Java as free as possible.
I think this statement from an IBMer is much less about OpenSolaris than it is about Java.
And yes, Java is a huge matter for IBM (unlike Solaris).
Sounds like a pretty weak cheap shot on the side of IBM. There's not really a definition of a 'true open source project', the whole idea is that he who owns the project can call the shots how they see fit. Because it is Open Source, if IBM doesn't like how the project is progressing, they are free to fork the code and take it in their own direction. This guys whole gripe is that he thinks that IBM should have some right to hijack and redirect the project or get free benefits from the project without committing any significant resource to it. And at the end of the day, what does IBM care about openSolaris anyway? It's a competing product, which makes this guys comments amount to nothing but a whole lot of empty rhetoric
Andrew Cherry The Cherry Pit - www.andrewcherry.com
So this rant of one IBM executive is completely baseless and probably intended to promote IBM at Sun's expense.
Agreed - and of course it's intended to promote IBM at Sun's expense.
Isn't it crazy that IBM, who's contributions to F/OSS (whilst large & also warmly thanked for) are dwarfed by Sun's contributions are able to get away with this?
The reason I suspect is Sun's flip-floppiness & skittishness when it comes to F/OSS - they contribute much, but also help spread a litlle anti-F/OSS FUD, etc. IBM's stance hasn't changed for what? eight years now.
CDDL is part of that problem I think - as the article notes, linus had 10 times as many people contributing to linux in his first year than Sun - with all their resources - had contributing to opensolaris in its first year.... A pity.
There are shills on slashdot. Apparently, I'm one of them.
It is largely thanks to the collaboration of competing conglomerates such as IBM, Red Hat, Novell, (even Sun), Oracle, Intel, AMD, HP, SGI and others on the Linux kernel that has made it so fantastically capable. We wouldn't have the world-class portability, the performance, support for all the hardware under the sun, CPU/PCI/memory hotplug, multiple journalling filesystems, etc. without the above named companies realizing that they can get _more_ ROI via controlled cooperation than with proprietary engineering. These companies compete with each-other in the market place and then pay their engineers to collaborate on making the OS better. It is a hugely successful strategy. The fact that Sun doesn't get it (and sometimes tries to /fight it/) makes Sun stupid and obsolete.
Sun has at times recently engaged in actions that raise doubts about how much slack they should be cut. But please *do* remember OpenOffice. That was the first useable word processor for Linux (since MS bought out Corel and convinced them to drop WordPerfect). AbiWord just wasn't in the ballpark. KOffice was green and unready. Alpha quality.
And Linux NEEDED a word processor, not LaTex. And Sun provided one.
(This *doesn't* excuse their subsidizing SCO...but it *IS* a large contribution to Linux, even if I suspect their motives.)
I think we've pushed this "anyone can grow up to be president" thing too far.
I am SICK of using emulation layers. I want SmartSuite NATIVE or detoxed of enough ms spaghetti code so I don't have to limit my choices to the current options, NONE of which give me what SmartSuite does:
-Lotus WordPro, which is easier for me to use, neater, cleaner, and more visually aesthetic.
Lotus Approach, which by light years ahead delivers what most sane end-users would need or want out of a user-friendly, non-programmer, WYSIWYG database front end that produces forms, worksheets, charts, limited crosstabs, and fairly nice detail tables on the forms. Even the reports can have charts embedded.
NONE of this is available in Kexi, Knoda (is that a db front end?), Base, or anything else.
Please, IBM, let us have a dual-license/Open Source Lotus SmartSuite that is free of the crippling licensing that has to date made it untenable to release SmartSuite. If CodeWeavers can get ms orifice to install, then please sponsor them to make SmartSuite work. Mine WON'T, yet. They need your help. ***I*** need your help. Please be NICE: play ball. Sun, with all its sponsorhip and pocket change is NOT helping SO/OOo deliver to people like myself a database like Approach. They've HAD enough time, and haven't deliver. Why not now take the initiative, IBM? Geeks/devs would approach this just out of the sheer challenge and coolness factor of SmartSuite.
PLEASE!!!! PLEASE????
SPARC, Netbeans, OpenOffice and probably others that have heavy sun development.
Ever hear of, oh, NFS. No? How about RPC? These Sun contributions to open source predate IBM's involvement with FOSS by a long time.
Wow, that's some creative rewriting of history. In fact, NFS was proprietary for many years. I'm not sure at what point Sun did or did not release NFS source code, but it hasn't been relevant to the Linux world because (1) Linux already had its own NFS implementations by the time Sun released it, and (2) Sun's licenses were likely unacceptable.
Furthermore, both NFS and RPC were poor designs. The UNIX network file system world is still in shambles, and Sun and NFS are single-handedly responsible for that. NFS was so bad that with "NFSv4", Sun essentially started over from scratch, but it's been too little, too late.
When I set up two HPC clusters using IBM hardware, they had no intention of recommending AIX. It was in their scope of work that we use RedHat, as it was what we wanted. And they even through in a life sciences consultant to help with PBS setup.
True, the service contracts are recurring (which service contracts aren't?), but no more expensive than a Sun service contract would be. They were a good deal cheaper than Sun contracts, actually. If you have any Sun service contract history (and it sounds like you don't) you'd know that their hardware maintenance contracts are some of the most expensive in the industry.
Quoted from Sun's SEC filing on 2006-05-05 (pg. 40): Read [PDF] or HTML
"In particular, we are seeing increased competition and pricing pressures from competitors offering systems running Linux software and other open source software."
If you've seen/believe IDC market share reports of operating systems, you'd notice that the huge growth of Linux was not at the expense of Windows, but rather Unix (they seldom break Unix down into its flavors, but Solaris is the leader, so they'd be hurting from Linux). The IDC data showed that Linux was a competitive threat to Sun back in 2002, and it appears that Sun feels that way still.
See: An old IDC report
See: http://www.forbes.com/2002/07/15/0715linux.html Savio
I, along with others, complained about the "Joint Contributor's License" required to merge code into the primary OpenOffice.org tree. While other projects requires every copyright contributor to agree to a license change in the future, such as making the product close-source for future releases, the contributor's license already gives Sun the right to change the license at any time. In response to these complaints, Sun Microsystems claimed that there would be a non-profit formed and copyright controlled handed over to it. Unlike Sun that must answer it's board and investors to turn a profit which can sometimes run counter to being dedicated to keeping a project Open Source, the non-profit's purpose would be to ensure that OpenOffice will always be Open. It is now several years later and I'm still waiting for Sun to create the non-profit organization that will be handling the contributions.
Bottom line: Sun will claim anything to head off complaints but has no follow through. Only trust the Sun if you want to get burned.
IBM are being utter hypocrites.
NFS, ZFS, DTrace (coming to an Apple near you soon), T1 chip (GPL), ...
Zen tips: Pay attention. Don't take it personally. Believe nothing.
Certainly calling Sun names for being closed while they are making serious noise about opening Java seems ill timed from a tact perspective. However, NFS was never usefully open sourced. It was a well-specified protocol that was reimplemented by everyone. The historical Sun was generally not a creator of open code, but did believe in open interfaces as a tactical market weapon.
They may be changing their stripes, but the deliberate GPL incompatibility of the CDDL makes me wary. I'll believe it when I see it.
-josh
Ah, the mandatory straw man. How cute. I never said that IBM was "cool", or whatever straw man you feel like dismantling today. I just said that it supports Linux and F/OSS in general.
Sun, for better or worse, has been until _very_ recently just a sad case of corporate schizophrenia. It did a lot of smoke and mirrors shows of "we love F/OSS" and "we love Linux", followed by flipping sometimes even in the same fucking day to, basically, "Linux is teh suck! Die! Die! Die!" and "Proprietary software FTW!" Not to mention the whole deal with SCO at the apex of its anti-Linux campaign. If you look at where SCO's money came from, two companies stick out like a sore thumb: MS and Sun. If that's how they support their "we love Linux" theatre... I rest my cae.
Yes, it published some specs... same as everyone else, except maybe MS, did. Then proceeded to encumber the implementation with some "I own your ass if you even look at it" license.
But even there, let me clue you in: back in the days of Sun's "we love open specs and Unix interoperability" spiel, _everyone_ else put up the exact same "we love open specs and Unix interoperability" show. That makes Sun sooo special. Not. And everyone, Sun _and_ IBM included, actually sabotaged interoperability and deviated from those specs as far as they could, to lock in customers.
Except at some point, for better or worse, IBM started actually showing some support for actual F/OSS. You know, as in, you can actually take the sources, modify them, make something useful out of them. Plus, it's put its marketting and corporate weight behind Linux, which did a _lot_ more to get PHBs to accept it than ranting persecution-syndrom geeks did.
No, it doesn't make them "good" or "cool" across the board, but at least it did something useful in that particular domain.
Sun responded by... mainly putting up even more smoke and mirrors shows, and going even more schizophrenic. Mostly it just tried to muddy the waters about what F/OSS even means, and redefine some lame "well, you can do some free work for us, if you want to, but we'll sue your pants off if you as much as look at our sources the wrong way" offers as "open". See: Java.
No, IBM never was "cool" across the board, but from where I stand, Sun looks even worse. That's all. If you want to compare IBM with Hitler, so be it, but then it says something when Sun looks like an even bigger nutcase.
A polar bear is a cartesian bear after a coordinate transform.
Sure. Lets take that Linux box of yours and start removing Sun or Sun related code : /etc/shadow - Gone
OpenOffice - gone
Java - Gone
RPC - Gone
PAM - Gone
Much of Gnome - Gone
Not a very usable box anymore and this is only a very small set of examples.
IBM's business agenda, though, doesn't include lavishing praise on a rival operating system. It prefers Linux and its own proprietary version of Unix, called AIX. Solaris now runs on x86 computers such as IBM's System x servers as well as on Sun's own Sparc-based computers. OpenSolaris is designed to appeal to developers, who have the power to sneak software into companies the same way Linux snuck in during the 1990s.
That's hard to do when you cut out hardware documentation for entire SPARC platform - sun4cdm and limited bits of sun4u. The thing that they'd have to do is figure out how the GXT2000/3000/4x00/6x00 cards work and get them to work right - and they'd have something that IBM has so far not even documented in Linux. Otherwise I'd rather stick with IBM until they do that deed.
"OpenSolaris isn't a true open-source project, but rather a "facade," because Sun Microsystems doesn't share control of it with outsiders, executives from rival IBM say. "Sun holds it all behind the firewall"
Dunno about the facade, but their control, but they sure do hold some good stuff behind the firewall.
Yes, Sparcstations are "ancient hardware" - but at least IBM allowed machines as old as their POWER2 workstations and servers to have "Linux Affinity" on them for at least one release. That feature is as close as IBM may get to opening AIX itself directly (in a similar way as OpenSolaris), and is something that combines the best of both worlds for what you can do with their older hardware.
Why Sun seems to be adamant about keeping (outside of an ad hominem towards those who've objected - I saw that one bmc) sun4cdm and its hardware out of OpenSolaris (when their competition does well despite a longer hardware support cycle) speaks volumes of what they cant do.
Twitter supports and protects racists - by smearing their critics with the "Hate Speech" label.
No, it really isn't. DTrace functions on static trace points. The only thing that differentiates DTrace is that Sun put thousands of trace points throughout the kernel and userspace. All of the legwork to get DTrace to be worth a damn in OSX will be done by the core developers at Apple.
Do you say the same about Mozilla?
The CDDL is just the MPL rewritten to be more of a boilerplate license, rather than needing to be changed for each project. It may not be the best license around, but if you believe SUN is being disingenuous by using then CDDL, I think you need to hold all users of the MPL to the same standard.
Read more of this story at Slashdot.Read more of this story at Slashdot.Read more of this story at Slashdot.
The fbt and pid providers (which provide the overwhelming majority of probes) are not static probe points.
Tp.
OpenOffice - gone
/etc/shadow - Gone /etc/shadow isn't a Sun contribution, it's a convention.
As I was saying that's pretty much the only contribution Sun has made. (Of course, it wasn't actually written by Sun, Sun open sourced it for a specific business purpose and conflict with Microsoft, and they are still having trouble letting go).
Java - Gone
Sun didn't contribute Java to the open source world; in fact, they have been trying hard to prevent open source implementations. The Java implementations that run on my Debian box were painstakingly developed clean-room implementations, and there is no significant client or server code that relies on it.
RPC - Gone
First of all, Linux's implementation of RPC isn't based on Sun's. Second, RPC and NFS are horrible standards.
PAM - Gone
There are no Sun contributions in Linux-PAM that I can see.
Much of Gnome - Gone
"Much of Gnome"? Where is your evidence? I don't remember the last time I have seen a Sun copyright on a piece of Gnome software.
Not a very usable box anymore and this is only a very small set of examples.
Yes, and your small set of examples is typical. You keep listing things where Sun made decisions for their proprietary UNIX systems and open source then had to reimplement those decisions (sometimes against Sun's objections and usually without help from Sun) because that was the pragmatic thing to do.
If we went by your reasoning, that copying of a proprietary feature by an open source project contributes a "contribution" by the proprietary vendor, then Microsoft would be an even bigger contributor to Linux--after all, according to your reasoning, Microsoft "contributed" OpenOffice, Wine, Samba, and most of Gnome.
The real kicker is that that, not only did Sun fail to contribute most of the things you say they did, many of the standards they set for the UNIX world have been crap.
NFS, ZFS, DTrace (coming to an Apple near you soon), T1 chip (GPL), ...
The question is not whether Sun releases stuff under open source licenses, the question is whether they are making contributions.
Let's take NFS. NFS source code remained proprietary and Sun was getting licensing fees. The open source community eventually had to create their own, independent NFS implementation. I don't know whether NFS v1-3 ever was open sourced by Sun--it certainly isn't being used by most open source desktops or servers.
As for recent releases, NFSv4, ZFS, and DTrace, they all have a bunch of things in common: they are unproven designs, they are not established standards, they come under licenses that make incorporation into Solaris competitors difficult, and Sun controls them tightly. In short, they aren't "contributions", they are simply trial balloons.
Yes, your examples adequately sum up Sun's open source "contributions": too late, entirely self-serving, and useless.
The end result is that rather than opensourcing AIX (which would be a rather pointless endeavor as the impression I get is that IBM is "sunsetting" it in favor of Linux), IBM is simply taking all of the Good Parts from AIX which they can and merging them into Linux.
If you talk to IBM about their P series boxes, you get rather the opposite impression. They talk about all sorts of very cool technology - technology they have no intention of transitioning to Linux.
Alex
Then WTF are you doing posting here? You obviously haven't looked into it. Yes, OpenSolaris is mostly OpenSource (there are a few closed bits, but they are not necessarily critical bits anyway).
Why yes, sun4cdm support is quite significant - strange that they can get PPC32/PPC64 but nothing of their own platform. I guess the developers dont seem to like their own dogfood.
Just because Sun has control of OpenSolaris, doesn't mean you can't download the whole source tree and fork it and start your own project. (Some folks have already done this, check out the PPC port of Solaris, or the port of Debian userland to the Solaris kernel, for example.) That is what Open Source means.
Well, strange that if you try building for sun4m, a lot of pieces are missing(e.g. for that nicely performing, 32bit SS20 that could probably give some age-equivalent, OSOL supported PPC's the run for the money). Namely about everything sun4cdm is missing, save for a few lowend sbus modules.
bmc@sun.com, thank you for making me a solidly entrenched fan of your competitor, Big Blue. No wonder you had to have a selectionist part of time where not much was open short of The Thing that sun4m's Couldnt Run , dtrace.
P.S.
As for those who have suggested a *BSD - Sun comes short on documentation of their hardware again - from the sbus cards up to their Ultrasparc III processors.
Twitter supports and protects racists - by smearing their critics with the "Hate Speech" label.
Mozilla is tri-licensed, so your comments are wide of the mark.
The MPL itself was a new work, a creation. It was perhaps foolhardy in some respects, but it is difficult to presume that a group crafting a license will perceive the full longterm results of their creation. The CDDL was created years later with full understanding of the relevant problems. Moreover, the CDDL was specifically created to be used with the OpenSolaris release, which is a type of open UNIX release. The preeminent open unix implementation at this time is under the GPL. As a result, choosing to release OpenSolaris under the newly written, but long understood (see your point about the MPL) CDDL makes it clear that preventing source interoperability was a goal.
To sum up: Mozilla was in a very different position, the code is trilicened to be compatible. Some new MPL works may also be deliberately incompatible with GPL code, but may also not be in a field where there is an established GPL work with which they are deliberately incompatible. The Sun actions with the CDDL were the worst of this continuum, being deliberately incompatible when there is an obvious existing relevant work with which it would benefit everyone to be compatible, and moreover they minted a new license in text (though not substantially in function) which was aimed to create this incompatibility.
-josh