Submitted by Well, Oracle went from one of those big enterprise-serving companies most of us don’t deal with to one of the more hated companies in our little community. Not only did they just sue Google over Android and its use of Java-related technologies, they also just officially killed off OpenSolaris. Solaris will still be open source, but source code will only come after each major release – development will happen behind closed doors.
OpenSolaris kernel developer Steven Stallion was the one who broke the news by publishing an internal email sent to Oracle’s Solaris engineers. The email is extremely verbose and a little vague at times, but after a few thorough readings everything finally became clear.
Up until now, open source Solaris development took place out in the open, like the Linux kernel or GNOME. You could see exactly what was going on, and Solaris source code was updated continuously with nightly releases and all that stuff. This is going to end. Development will take place behind Oracle’s closed doors, and only after each major release will source code be released to the public.
“We will distribute updates to approved CDDL or other open source-licensed code following full releases of our enterprise Solaris operating system,” the email details, “In this manner, new technology innovations will show up in our releases before anywhere else. We will no longer distribute source code for the entirety of the Solaris operating system in real-time while it is developed, on a nightly basis.”
The reasoning behind this move seems a bit odd to me. Oracle cites they can’t do everything on their own due to a lack of manpower – and yet they cut off the entire open source community and all the work it could do. The second reason cited by Oracle is that by developing out in the open, competitors will have insight into what Oracle is doing.
Ben Rockwood’s assessment is pretty clear. Rockwood laments the role of the OpenSolaris Governance Board, which announced it would disband if Oracle didn’t take action on OpenSolaris, thereby playing right into Oracle’s hands. “As a community and governance, OpenSolaris has been a non-stop, end to end failure. Hands down. At every turn, it failed,” Rockwood writes, “As an open source project, it was luke warm at best.”
So, that’s that, then. In a single week, Oracle manages to upset the Android, Java, and OpenSolaris communities. Nice one, Larry.
Well no surprises there, still saddening though. But I guess we can all say we saw this coming.
What annoys me the most about all this stuff-which-we saw-coming-from-miles-away is that regulators in the EU were focused on the fate of MySQL. An insignificant piece of software in Sun’s portfolio compared to Java and OpenSolaris with plenty of (beter) alternatives (PostgreSQL, Ingres, FireBird to name a few).
Actually I’m surprised that they’ll still release the sources every now and then. I’d expected them to make Solaris closed source again which would have been a very bad thing.
I’m worried that they might decide to close-source java platform too on the same competitorship basis
I doubt the paying customers really care.
Sure, Solaris customers where already paying for the license. The OpenSolaris users are the one’s who’ll run the risk now if Oracle decides to close up the whole thing instead of doing a source burst after each major version release.
They will when the cost of development, and therefore the price, goes up even further to justify its existence.
After this http://www.osnews.com/story/23077/Solaris_No_Longer_Free
it looks like it will become impossible for me to use Solaris at all now.
Time to go back to Linux I guess.
Considering that Solaris stays FOSS, you may just as well use a 3rd party Solaris distribution like Nexenta OS.
Nexenta OS is even a commercially backed effort, so the guys at Nexenta Systems won’t shoot in their own foot.
The problem now is that 3rd party Solaris based distributions like Nexenta will forever lag behind the official versions of Solaris. There will be a lead time between Oracle releasing the source to a new release and Nexenta being able to rebase their distribution from it.
Weren’t they already behind anyway?
I tried Nexenta long long ago. There were just too many broken packages and inconsistencies between trying to do things the debian/linux and the solaris way. One example was SMF scripts for packages. Some had them, some didn’t.
Edited 2010-08-15 04:27 UTC
Well in Oracle point Nexenta is bit of leech. I think this move is targeted to get rid those and same time they can allow customers do code auditing easily (like you said forking is made lot harder). Whole OpenSolaris stuff reminds me Apple in 90’s when they tried to get more profit by allowing 3rd party Macs. I guess same happened here, Oracle lost profits to 3rd party companies without gaining much market share from Linux.
We don’t know when the Solaris sources will be released.
Even if Nexenta users have to wait two months. Who cares? Solaris isn’t targeted at the cutting edge audience anyway.
Don’t feel bad about it dude, everyone else already did.
Not unexpected. And the person who posted about Java… keep your eyes and ears open. I am sure we’ll see some changes in that arena as well.
Looks like Oracle thinks they can do a better job (financially and technically) without the “Open Source development model”. The Open Source way didn’t do much good to SUN, so hopefully the proprietary development model works better for Oracle.
Open Source is all about the way which software is developed. Open Source is about being pragmatic, it’s not about being political or ideological. The Free Software ideology, on the contrary, is all about doing the “right thing”(TM) — it’s about being ethical in your choices and thinking about users’ freedom.
If Oracle manages to make their business more profitable following the proprietary development model, then all the pragmatic Open Source supporters must applaud and agree that they made the right choice.
(Disclosure: I support FOSS in all of its variations, but I think FOSS is better for offering alternatives to the proprietary ways and for “scratching your itch” than it is for making money.)
I’m sorry, but this is an painfully false dichotomy. In fact, I would argue that the open source model is a greater source of profit than closed source (exhibit A: the web), because the concept of charging people for infrastructure software is outmoded, awkward, and expensive.
For example, a company I worked for has a supercomputing cluster with thousands of machines. If we installed proprietary databases, operating systems, and storage systems, the licensing costs alone would consume our entire margin. How is that business-friendly? Open source software was the only reason we could make money, in stark contrast to what you say.
But it’s not just niche exercises like clustered computing that benefit. Open source software and open standards allow for the creation of whole ecosystems, so that people can expend their creative energy building new and novel products, not reinventing the wheel. As I said before, the web is a great example of how open source software democratized infrastructure and created a whole new market, with a corresponding explosion of wealth.
The value of any app is not in what platform it runs on or which technologies it uses, but what utility it delivers to customers. You don’t visit a website because it was written in ASP or because it’s served by IIS. You visit it for its function. What does the nature of the source code’s licensing have to do with that? And if open source software is inherently cheaper than closed source, than isn’t it better from a financial perspective?
To label open source software as “unfriendly to business” is ignorant at best. Open source software is inherently business-friendly. The real culprit here is corporate hubris, where dictatorial control and short-term yield are the only metrics for success.
But, but, but … how are the developers of super-computer software supposed to make money?
Think of all those database and HPC devs that you just put out of business. Won’t someone think of their children? 
why is it all publicly traded companies tend to have the same mercenary mentality when it comes to OSS, doesnt help that sun essentially strung themselves up by their own bootstraps then jumped on a guillotine
Publicly Traded; the primary product is profit for the shareholders, it sells shares not products. The software or goods are simply the tool it happens to use to harvest wallets for the shareholders. It’s also a short term mentality; what produces a nice profit spike to make the current management look good before they move on to the next short term project.
Oracle produces shareholder equity, software just happens to be a tool it uses to do so. Microsoft produces shareholder equity, software just happens to be a tool it uses to do so. Sun was in the business of producing shareholder equity using hardware as it’s tool. And so on through the list of other publicly traded retail corporations.
Heck, look at Google, we are not the clients, we are the commodity. The clients are the companies that our information and page views are sold to in persuite of profits for the shareholders.
In this case, Oracle’s short sighted view is that they can do better without input from the open source developer community. They believe that they can do a better job of finding bugs and such with a budget and and head-count limited team of developers. We’ll see if it works out that way or not.
You nailed it there Sir. Good one. 🙂
I saw this coming…. EVERYONE saw this coming.
Really. I sad it months ago… But they laughed at me (The few, that didn’t saw this coming
)
openSolaris was one of the most advanced *NIX-like Systems. And now it is dead.Solaris will be crippled down to operating systems which does not confront Oracles own Linux System.
Releasing the source code with all major update is a joke. Really, just think of it: Solaris 10 was released 2005, Solaris 11 will be released 2011. 6 years of development. I think, it is impossible to fork it, while having a binary-compatibility…
(Thus i am not an expert in this)
Thank God there ist Illumos, you may think. But – honestly- even with the help from above, Illumos will not be able to attract important vendors of proprietary technology…
—————————–
OpenSolaris obituary
—————————–
Yesterday openSolaris, age 5, passed away. Born with an unhealable disease, its only hope has been that its inner values was a benefit for all others.
With the death of its parents, its end was sealed, by its stepparents decision to shut down its medical ventilator.
We will never forget its progressive genie.
——————————-
(Well, i think, that was too much
)
Edited 2010-08-14 01:18 UTC
“I can only maintain that the software we worked on was for the betterment of all, not for any one company’s bottom line. This is truly a perversion of the open source spirit.”
unfortunate. but some good has come of this. opensolaris meant freebsd got zfs. and zfs drove the development of btrfs faster than if there had been no competition.
solaris can drop off the earth and we’ll still have gotten some neat file systems.
“… zfs drove the development of btrfs faster than if there had been no competition …”
One dreads the prospects of linux/gnu work at oracle, OTOH, one of whose projects is … btrfs.
all bets are off there. I assume they will drop linux as much as possible and push solaris as much as possible. they do have a competitive advantage with solaris.
Eh? Oracle has been pushing LINUX (RHEL specifically) for years. I imagine they’ll demote / back-burner all OS related products – they are a database / platform company after all. RHEL is a nice low-R&D solution for them.
http://www.google.com/search?q=%22ported+from+opensolaris%2…
It looks like ZFS and DTrace, which are both huge contributions.
This is bad for OpenSolaris, but strengthens Solaris. Solaris 11 Express will be free for developers, as is Solaris 10 (for evaluation purposes, educational purposes, research, etc – just read the S10 EULA when you try to download it. Or you can buy S10 for 20USD). “There will be an optional support program” for Solaris 11 Express.
Solaris 11 Express will come at the end of this year. All source code will be released after Solaris 11 Express has been updated. With the new Oracle SPARC roadmap, with machines having 16.384 threads and 64TB RAM and brutal performance, Solaris has never been in a stronger position.
Earlier there was no business reasons to run Solaris, just technological reasons. Now Oracle will try to make all their Enterprise customers switch to Solaris 11 – Larry Ellison will make sure there are business reasons to switch. And the Enterprise customers are many. Solaris has never been in a stronger position.
But it is a sad day for the OpenSolaris distro. But a good day for the Solaris OS! The future is bright indeed! :o)
Perhaps.
The problem with your analysis, though, IMHO, is that Solaris has returned to a legal position of “Oracle’s OS”. Linux is “the world’s OS”.
I can buy Linux systems from IBM, HP, or a bazillion other companies, and my software will run on those systems with little to no effort. This isn’t true of Solaris, which runs only on Sun hardware, and without OpenSolaris will never be true.
You can argue that OpenSolaris failed and thus it would never have been true anyway, but that just confirms my point – Solaris’ lunch is already being eaten by Linux, and Oracle’s move hurts rather than helps its marketing case. For Solaris to reverse its decline in sales and market share, it needs to become multi-vendor, not more proprietary.
So I fail to see how “Solaris has never been in a stronger position”.
(Disclaimer: This is my personal opinion only, not speaking for my employer, etc. etc.)
hmmm…
not very impressed w/sun’s efforts of late (meanly years!)
it’s sort of a modern-day dinosaur lingering on old patents…
& the ellison’s lil larry’s been bad news for yonks — but — honestly can’t say that I’d ever thought “opensolaris” was all that viable for most ppl anyway — look at the track record: the last release was a year ago, & (what I thought was the better effort!) “belenix” hasn’t had an upgrade in nearly 2 years…
just input — imho — other’s ppl’s mileage, etc…
coeli
Yes, but Oracle has lots of Enterprise customers paying big money, 10x more than Sun ever had. If Oracle can make just a small percentage of them to switch, then Oracle has made a big win. The Oracle database is arguably the best many say, and a database is very important for any company. Larry will see that their big database plays best together with Solaris. For the big customer that prioritizes stability in big Enterprise server halls, they will be strongly encouraged to switch to Solaris.
From the mail this article is refering to:
“The growth opportunity for Solaris has never been greater. As one example, Solaris is used by about 40% of Oracle’s enterprise customers, which means we have a 60% growth opportunity in our top customers alone. In absolute numbers, there are 130,000 Oracle customers in North America alone who don’t use our servers and storage yet, and a global customer base of 350,000 (the prior Sun base was ~35,000). That’s a huge opportunity we can go attack as a combined company that will increase Solaris adoption and the overall Hardware server revenue. Our success will also increase the amount of effort ISVs exert optimizing their applications for Solaris.”
Again, the future is bright indeed for Solaris! Besides, you can buy Solaris from HP, Dell, etc
I dont get this statement? Solaris 10 runs on x86. As will Solaris 11.
Sun had 35.000 Solaris customers. Oracle has 350.000 Enterprise customers paying BIG money to Oracle. If Oracle can make a tiny percent of their customers to switch to Solaris, then Oracle has reversed the trend big time.
No disclaimer needed, you make it clear it is your opinion and you make it clear you are not spreading FUD, as the IBM camp loves to do: “I heard from an Oracle executive that Solaris 11 and SPARC is killed. Migrate to AIX now, or you will regret it”
Edited 2010-08-14 22:01 UTC
Yep, Oracle pushes Solaris to big enterprises.
Solaris will be the same as AIX and HPUX. Solaris systems from Oracle, AIX systems from IBM and HPUX systems from HP.
And if you don’t have the big bucks to get serious system, then there is always *BSD and Linux but don’t come knocking on the Oracle door then.
That’s a painfully optimistic and naive appraisal of the situation, especially from someone who was steadfast that this day would never come.
All that’s happened is that Solaris has gone back to being the completely proprietary Unix that was, and still is, leaking users and costing money. Oracle’s only hope is to push Solaris and SPARC right into the high end to try and compete with AIX and Power to try and get the kind of margins they need to stop them leaking money. That is Ellison’s goal, to have a go at IBM. Whether it will be successful considering how niche and crowded that market is I don’t know.
I dont agree. Sun had 35.000 customers. Oracle has 350.000 customers. If Oracle makes sure that their database plays best on Solaris (maybe a new killer functionality that is only available on Solaris) then many will consider to switch to Solaris – because it improves their business.
Sun always have been reknown for new innovative technology. They will find out new cool tech to apply on the combination Oracle DB + Solaris. Just like Sun did to ZFS and DTrace and got the new Fishworks storage servers 7000 where you can do uniquely new things that had no counterpart before.
That is the reason I do believe that Solaris never has been in a stronger position that today.
Sure, Oracle is bad for OpenSolaris, but for Solaris as a whole – this is good. Solaris 11 will continue to have open source code, so Illumos based distros will have a chance to get all the source code and synchronize every 3rd month.
Therefore I dont agree with you.
It be interesting to see when their firewall/paywall goes up for the ONNV gate. Still pushing code into it as of today as can be seen on the public mirror on genunix.
http://hg.genunix.org/onnv-gate.hg/shortlog
I use OpenSolaris at home and think it’s absolutely perfect for my limited needs (NFS shares with underlying ZFS, MySQL, and Squid). Sad as I am that OpenSolaris is being discontinued, I’m a little confused as to whether home users will still be able to use ZFS? Is Solaris 11 free as in beer, or will it be a product that you have to buy?
Also, can we continue using OpenSolaris with Solaris updates? This paragraph suggests that:
“Anyone who is consuming Solaris code using the CDDL, whether in pieces or as a part of the OpenSolaris source distribution or a derivative thereof, would therefore be able to consume any updates we release at that time, under the terms of the CDDL, LGPL, or whatever license applies.”
I’m not surprised. OpenSolaris pretty much failed as a tool to engage volunteer open source developers en masse, at least on a scale that would be noticeably beneficial. It was a losing proposition to try to get developers away from entrenched Linux. I said this from the beginning but some people still seemed to think that OpenSolaris technology would draw a ton of developers. Linux had already been “good enough” for quite some time when Sun finally decided to do this and by then it was at least 5 years too late.
Edited 2010-08-14 14:36 UTC
It is easy for many people to say “stop wasting your time and move to linux and BSD”, but many devs who are disappointed with Oracle have found alternative places to work that allow them to continue contributing to Solaris/Illumos. See the links in these blogs:
http://dtrace.org/blogs/bmc/
http://gdamore.blogspot.com/
Anyway, it is still bad news but I’d say it isn’t over for OpenSolaris except in name. Things will slow down while the remaining closed bits of Solaris are recoded as open source, and then we may even see a completely open source release by year end.
I wouldn’t think that a Solaris developer would leave Oracle and continue developing for Solaris. This ain’t Linux. As the release stated, Solaris software engineers are employed by Oracle so people who had the privilege and saw the source code of Solaris get released for OpenSolaris are the ones who will be in the dark.
I saw this coming 5 years ago, when I got my “First 5000” tshirt for the first 5000 who downloaded Opensolaris. I wondered who (I knew) was running the helm at Sun for doing something so idiotic as trying to start up a freelance development model. Repeat, this ain’t linux and customers who know the benefits don’t want it to be.
What will happen to VirtualBox then?
Another good product waiting to be closed to rot away?
That’s a good question. Its a really good product, but leaving it as is would ensure it rots away. VMWare owns the market when it comes to virtualization. They do quite well without an “OpenVMware”
With the open source freelance model, no one cares about a project per-se, but the technology behind it. That why you saw people jumping ship to the next best thing, leaving older projects to die. There was bochs, oh now, XEN is better, oh now there is KVM. Oh… Look shiney…
I think it is in the interest of Oracle to make it better. But not the OSS edition. Probably the Free (beer) but not Free (Libre) version. They don’t make money on virt, but it is part of the vertical market Oracle want.
leastways well always ahve qemu , if worst comes to worst fork
Was there a potential to fork OpenSolaris? If so, have they changed the licensing to block such a thing happening?
I mean, how sweet would it be to see a community BeyondSolaris project spring up and out code Oracle’s own development team with a truly open source, in spirit and code availability, project?
I guess the most effective way to kill a project is hand control over to Oracle.
As I read now, I see there are some OpenSolaris forks already but will be affected if they remain based on each latest Solaris release.
I think we all know what products are next in line for the hatchet: mysql and virtualbox.
Regarding the first one, I don’t know whether Larry would dare already to ruffle again EU-antitrust feathers.
I don’t think MySQL and VirtualBox are chewing through the money as Solaris is.
No, but a) mysql is actually a competing product and b) oracle already has their own virtualization product. Not that anyone is using it but still.
Who is this OpenSolaris you speak of?
Seriously – the OS itself can die a slow death and all the brouhaha over its demise won’t affect the majority of users who can simply move their stuff over to a Linux or BSD platform.
What WILL affect ME as a developer to a great deal, and something I’ve been kinda watching out of the corner of my eye, is the MySQL development. For me, that has MORE of an impact on my projects than does the underlying OS it runs on.
I can’t speak for anybody else, but yes, the status of MySQL does spark a concerning interest here.
Open source is a freight train that not even the giant Microsoft has been able to stop. This reminds me of SCO and Novell. SCO tried to stop the train and is a 15 mile skid mark now. Novell just looked at it wrong and they are still licking their wounds. Open source will suffer but Oracle has made a lot of enemies. I hope this is not what happens but that is what it looks like will happen.
I know the old phrase “Better the Devil you know than the one you don’t,” but looking at Larry Ellison’s behavior from being a World Cup jerk to Oracle’s attack on Red Hat, to killing OpenSolaris to suing Google, I’m starting to develop warm fuzzy feelings for Steve Balmer in contrast.
So, that’s that, then. In a single week, Oracle manages to upset the Android, Java, and OpenSolaris communities. Nice one, Larry.
Just look at Oracle’s altruistic track record.From belligerently taking over competitors to killing/embracing/extinguishing rival products,you name it,Oracle wont hesitate to lay your @## flat on the ground.Mysql would have suffered the same fate if it wasn’t for anti-trust.Whats baffling is that Java should have the main concern.It was obvious that Oracle was gunning for Java and/or Solaris and not Mysql since there are already plenty of DBs out there including FOSS ones.
ZFS fans,
How do you like them apples?
I think they love it! :o)
There was fear that Solaris and OpenSolaris would die a slow death as Sun got closer and closer to bankrupt. But now, Oracle has Solaris so it will not die. Instead, the Solaris future looks better than ever before. This means that ZFS will continue devlopment, in even more rapid pace. I think the ZFS fans likes the new big investment in ZFS, dont you think?
Solaris 11 source code will be released after the binary release. As of now, OpenSolaris releases the source code before the binary relase. Solaris releases upgrades something like every 3rd month. So, you have to wait 3 months to get the source code, instead as of now, where you get the source code immediately.
Yes, I agree that this is a problem for Solaris community distros that will be based on Illumos. But, in a couple of month, you will get source code – including ZFS. So FreeBSD will have to wait a couple of months before they can get the latest ZFS release. But, ZFS will live as Oracle is investing heavily in it. The best thing, you can migrate to any OS with ZFS – as it is platform agnostic. Try that with a vendor lockin hardware raid. :o)
“Solaris will still be open source, but source code will only come after each major release – development will happen behind closed doors.”
I couldn’t find where this was stated in the letter Steve blogged. Realize that in the Solaris language a major release may never happen. Solaris 10 (AKA 2.10) is major release 2.
It’s all so Steve-like.
will he kill mysql ?
Is it possible to maintain good products *and* stay competitive from a financial point of view ?
Jerkish moves like the ones from Oracle show how to make money by degrading the quality of the product. From the lifetime of modern cars and electronic devices to the bloated, uninteresting, and incompatible new releases of all major software, we can see that this way of making money is fairly common. Isn’t there another way of being profitable ?
Edited 2010-08-16 10:43 UTC
Um; is the new Oracle’s model for OpenSolaris (aka “death of open source”) different from Google’s model for Android (aka “hurrah for open source”)? I heard the latter gets also released only rarely in big dumps of code – did I hear wrongly?
ONLINE STORE——————-w w w.f a m a l e g o o d s.c o m
we accept PAYPAL/credit card payment and all the products are the FREE shipping .
Air jordan(1-24)shoes $33
UGG BOOT $50
Nike shox(R4,NZ,OZ,TL1,TL2,TL3) $33
Handbags(Coach lv fendi d&g) $33
Tshirts (Polo ,ed hardy,lacoste) $16
Jean(True Religion,ed hardy,coogi) $30
Sunglasses(Oakey,coach,gucci,Armaini) $12
New era cap $9
,
Bikini (Ed hardy,polo) $18
FREE SHIPPING
w w w.f a m a l e g o o d s.c o m
htru75
sells clothing,footwear,handbags,Sunglasses
Online Store,Get Name Brand Fashion From 12USD Now!
Our Website: http://www.famalegoods.com
Our main product list is as follows:
1. Fashion Sports shoes: $35
Shox,Dunk,Jordan, Air max, Air force,Adidas, Puma,Prada,D&G, Chanel,Gucci, Lacoste,
LouisVuiton,Evisu,Timberland,ugg,boots,Burberry,Dior,Greedy
2. Fashion T-Shirt $14 & Jeans $30 & Jacket $37:
AF,BBC,Evsiu,Juicy,GGG,Burberry,ED-Hardy,Chanel,G-Star,Red monkey,Christian
Audigie,sinful,lacoste,POLO,Armani,Smet,Baby,Levis,Justcavalli,Versace ,True Religion,Artful Dodger,Rock,Coogi,Crown
Holder,RMC,etc.
3.Fashion Handbags:$$35
LV , Prada , Chanel ,D&G, Fendi , EDhardy, Burberry , MIUMIU ,Gucci ,Chloe , Juicy and Chole,Burse,etc.
4. brand watches(rolex,Longines…)$50
belt(d&g, gucci, prada, chanel, burberry…. )$14,
hats $12
sunglasses $12 (Police,rayban,gucci,prada,LV)
Deliever: Nearly 5 days arrive your home or your friends’ adress by EMS,DHL,UPs !!!!!!!!!!!
Looking forward your visit.asdsf @#$R$$