IBM and Sun in Final Stages of Buyout Deal
One of the biggest talking points over the last few weeks has been the IBM buyout of Sun. It looks almost inevitable at this stage as reports are emerging that they are now in the final stages, where IBM would pay $9.55 a share for Sun, which is $1 less per share than Sun has expected. This would leave Sun at a price of about $7 billion. It's a good deal for IBM - in 2002 they were looking at spending $40 billion on Sun.
An interesting part of the story is that Sun wants assurances that IBM won't pull out of the deal, even if it faces regulatory scrutiny with respect to to anti-competition laws.
When we ran the story last month on JavaLobby, we had a very interesting mix of comments. Some were concerned about the future of Netbeans, considering that IBM are the ones behind Eclipse, the other main IDE for Java developers. But, even if Netbeans was canned, it seems that there's a huge number of developers out there willing to keep it going. There were also questions about what would happen to MySQL and GlassFish, when IBM backs WebSphere. One comment that caught my attention was the fact that IBM are more focussed on the commercial side of thing, while Sun has been taking open source quite seriously.
Others took a more positive spin, saying that IBM could spin off JavaSE as an non-profit foundation as they have done with Eclipse.
It seems like we'll find out if the hopes and fears of developers will come true sooner rather than later. Change is no bad thing, and maybe it's time for a bit of a shake-up for developers. I believe the Java community is strong enough that we can make things work, whatever happens as a result of this acquisition.
In other interesting buyout news, it looks like Google are setting their sights on Twitter.
Comments
Fabrizio Giudici replied on Fri, 2009/04/03 - 8:04am
Slim Ouertani replied on Fri, 2009/04/03 - 8:15am
Bad news, Microsoft on .net side and IBM for java one :(
Jim Bethancourt replied on Fri, 2009/04/03 - 9:50am
in response to:
Fabrizio Giudici
I could see how Geronimo might fade away -- you don't hear all that much fanfare about it, and every time I do read about it or come across it, it seems to be lagging in including more recent technologies.
Jacek Furmankiewicz replied on Fri, 2009/04/03 - 8:54am
Liyu Wang replied on Fri, 2009/04/03 - 9:10am
Osvaldo Doederlein replied on Fri, 2009/04/03 - 9:14am
Some thoughts:
1) Whatever happens, JavaOne 2009 is bound to be really "interesting".
2) I admit having a lot of fear of IBM, I'd prefer Sun to remain independent. But IBM is a complex organization, they are both great & evil depending on which products lines and practices you look at. I just cross my fingers so the Sun acquisition becomes part of the "good IBM". IBM, don't screw up with Java. Put you money where you mouth has been for many years (remember IBM's comments in every JCP ballot, complaining of Sun's licensing terms? Now let's see if IBM turns Java and the JCP into Apache's dream...). Keep other great tech like Solaris and Glassfish alive, you can only win in the long term.
3) Too bad than Sun invested the last few years completely reinventing the company around a open source software portfolio; IMO Sun's is currently in an inflection point, they're starting to reap what they sown - gaining recognision and market share for several great products - but this economic crysis is really the worst time to be in this position as the company needs confident and wealthy investors. Well, you may agree or not with this analysis, but the fact is that Sun has put their soul in a open source strategy and if they fail, this will be seen as a major failue of open source. Other companies with a large portfolio of proprietary software will think twice before following Sun's footsteps. This would suck.
Liyu Wang replied on Fri, 2009/04/03 - 9:30am
Jacek Furmankiewicz replied on Fri, 2009/04/03 - 9:41am
Yeah, they're greedy. But they got $$$ and resources.
Sun, it seems has neither. And hence we've had to watch Java getting neglected recently while some guy with a pony tail spent 1 billion $$$ of cash on a free open source database that barely generates a few million dollars of revenue per year.
Boggles the mind how such "business" decisions get made and justified to shareholders.
JeffS replied on Fri, 2009/04/03 - 12:53pm
"Yeah, they're greedy"
saying "greedy" and "company" in the same sentence is redundant. Companies have to be greedy, to meet their business needs, and their fiduciary committement to shareholders. Anything done by a company that appears "altruistic", or "compassionate", or "charitable", or "do no evil", or purely for the sake of great tech, is really only part of a larger, long term business strategy - to help support birnging in more $$$$. That's not good or bad, per se. It is what it is, and the way things need to work. So we all just need to look at it with a grain of salt.
So anybody worried about Java because "IBM is greedy" shouldn't worry. There might be good things that come to Java, their might be bad things. Same has always been true with Sun's stewardship of Java. A lot of what Sun has done with Java has been influenced by their need to monetize their crown jewel as much as possible (which they actually, for the most part, failed to do). And this has led to a lot of less than desirable stuff, from the standpoint of the larger Java community and developers, to occur.
I'm actually cautiously optimistic. IBM has no real need to monetize the Java runtime or libraries - they know full well that the real money lies much further up the stack. They may just create a foundation, a la Eclipse, and release it under the Apache or Eclipse licenses. And to go along with that, they can offer a dual licensed (free, open source, and for pay, proprietary) TCK - To put out a proprietary Java, pay for the TCK. To put out a free, open source Java, don't pay for the TCK.
Alessandro Santini replied on Fri, 2009/04/03 - 1:16pm
JeffS replied on Fri, 2009/04/03 - 1:42pm
Take the Oracle example.
Oracle has been on an acquisition binge the last 3 years. And, as far as I can tell, none of the acquired products have been EOL'd. Most telling is WebLogic is still alive and strong, and only compliments OAS. From Oracle's perspective, it doesn't really matter which one the customer goes with. It's still money pouring in for Oracle. Same is true of Peoplesoft and JDEdwards, and Oracle ERP. They're all still on offer, and supported. Oracle is trying to merge Peoplesoft and Oracle ERP into Fusion, but that's still very much a big work in progress.
In short, product overlap in one company does not necessarily mean death to one of the products. More often than not, the acquired, overlapping product just becomes another flavor of the acquiring company's product offerings.
And when you think about it, it makes sense. An acquiring company can only piss off existing users/customers if the acquired, overlapped with other offerings, product is EOL'd, sending said customers to the acquiring company's competitors.
And getting the existing customer base in an acquisition is where real gains can be had. Losing that customer base is flushing money down the toilet.
For this reason, I'm optimistic about the survival, and thriving, of stuff like Netbeans, Glassfish, MySQL and Solaris.
Just imagine IBM said:
"sorry mister customer, but we're EOL'ing Glassfish, because it has overlap with our WebSphere, so we're making you migrate to Websphere".
To which the "Mr Customer" says to IBM:
"F$%# you IBM, if I have to migrate, I'm migrating to your competitor who doesn't want o screw me, and to an app server that sucks less than WebSphere".
Well, that's assuming Sun has established Glassfish (paid support) customers (which I think they do). But you get the point. Just replace Glassfish and Websphere in that scenario with MySQL and DB2, or Solaris and AIX, or Netbeans and Eclipse. Same thing.
Anyway, no way in hell is IBM that stupid.
Besides, having more flavors on offer only means more integration services revenue for IBM Global Services.
Liam Knox replied on Fri, 2009/04/03 - 6:39pm
in response to:
Liyu Wang
Liam Knox replied on Fri, 2009/04/03 - 6:40pm
in response to:
Slim Ouertani
Guido Amabili replied on Sat, 2009/04/04 - 4:47am
in response to:
Jacek Furmankiewicz
If IBM buys Sun , it will kill Swing and JavaFx and you will be glad to use properties and databinding in a flex pane embedded in an ocx container for swt..... (and don't forget to port your os project to it)
;-)
Thierry Bodhuin replied on Sat, 2009/04/04 - 8:58am
cowwoc replied on Sat, 2009/04/04 - 11:52am
Greedy or not, give me a single recent example where IBM acquired a company and it didn't die horribly a few years later?
IBM will kill Sun because it doesn't know how to do anything other than charge for support. The worse the product, the more money they make. IBM has zero interest on the OS and programming language front.
Alessandro Santini replied on Sat, 2009/04/04 - 1:54pm
in response to:
cowwoc
If what you are saying was true, Eclipse would be a pay-only tool. IBM has a large offering of successful products - DB2, WebSphere MQ, WebSphere Message Broker to name a few.
I arbitrarily kept WebSphere App Server out of the list because I know it does not meet the taste of many people in this forum. However, I like to point out that the J9 JVM outperforms the Sun's JVM in many aspects, so I do not really get the meaning that IBM has no interest in the OS / Programming side.
OS/2 was a fantastic operating system, light years ahead of Windows. Unfortunately some unforgivable marketing choices made it sunset. I want to think IBM learnt from its mistakes.
Osvaldo Doederlein replied on Sat, 2009/04/04 - 4:21pm
in response to:
Alessandro Santini
Eclipse is an open source and free, but it's very far from being a great example of a community-driven FOSS project. It started as a closed/secret project, released as open source only at v1.0 so IBM could have absolute control of architecture decisions like the invention of SWT, and no competitors could derive a proprietary derivative before IBM. It was released in a time when no Java IDEs of that quality were open, basically to undercut the competition (the "OSS used as dumping" business pattern) as IBM's VisualAge product line languished. Up to this day, Eclipse is only alive because IBM and a few other big ESF partners fund near-100% of its development. When some project is abandoned by IBM and friends, like the Visual Editor project, that project dies miserably because nobody steps up to maintain and evolve the code. Not to mention other fiascos like the fact that Eclipse lacked any J2EE support for years, and they only introduced WTP after Sun started to offer massive J2EE supportin NetBeans. Eclipse still misses important functionality that IBM gives only to paying customers with the RAD/RSM/RSA line, stuff they'll also release as open source in the future only when such features are commoditized by other free IDEs. I love Eclipse as much as anybody, but if you compare, NetBeans is still much more feature-complete, well integrated etc., Eclipse is only really competitive in the high-end if you buy some of the commercial distros (even if it's a cheaper one like MyEclipse). You have to buy stuff like WindowBuilder for GUI development, it's ridiculous (while several free IDEs offer good to top-class visual GUI editors).
Even when I'm wearing my Eclipse User hat (I use it and also NB and RAD - the latter not my choice, mandatory in complex WebSphere-centric projects), I'm very worried that IBM's swallowing of Sun will certainly have the NetBeans project as one of its most certain victims. IBM will probably just abandon NB, so the community is left with no competition in the Java IDE space (JetBrains, CodeGear and others are not significant competition, and not open source). Without NetBeans, we'd be using garage-quality J2EE support like Lomboz up to this day. (Yeah there is JBoss-IDE, but it is vastly based on ESF projects. And RedHat is not giving away their full-featured Eclipse-based IDE either.)
In summary, IBM's support for open source - however large - can be described as part smoke&mirrors, part marketing. Their significant contributions to Apache projects or the Linux kernel serves mostly as carrots before IBM's product line, which is vastly dominated by proprietary projects (closed source, very high prices, poor interop and tracking of latest standards, totally closed development e.g. no public bugtrack or any kind of decent feedback cycle that doesn't require a support contract, etc.). Name a single significant IBM product that is fully open source; you can't - their real moneymakers (WAS, MQ, AIX, DB2, Rational, etc.) are basically in the same proprietary/closed status as Oracle's , Microsoft's etc.
People that believe that IBM's committment to open source and open standards is superior to Sun's - notice that I din't say that Sun is perfect, or even that Sun's behavior is altruistic - are probably bound to major disappointment when the dust settles in a couple years. I hope I'm grossly mistaken / pessimistic, IBM can also be great at times, but I'm just guessing the most likely future based on my averaging of IBM's past. We can hope that the future would be different, e.g. perhaps IBM is interested in becoming really serious in the OSS business and buying Sun would fast-track that plan - but right now this is just wishful thinking; if IBM was serious about that, they could easily start by publishing their J9 and WebSphere codebases under the ASL2, no need of complex strategies like contributing to Harmony and Geronimo, lobbying against Sun in the JCP, etc.
Jeroen Wenting replied on Sun, 2009/04/05 - 2:44am
IBM buys companies to remove competitors from the market and absorb people from those competitors. The products created by those companies are almost invariably abandoned, left to whither and die (no or very limited new development, little if any support, etc.).
For Java that means that whatever new development takes place will take place solely within the confines of the WebSphere platform, to the point where it will become quite pointless to use Java outside of the WebSphere platform. Effectively anyone using Java on other platforms will be stuck with unsupported and old JVMs unless and until they buy WebSphere and port everything they have to that. That's the real money for IBM in buying Sun, getting Java users to buy into WebSphere. The hardware business is also lucrative, we're likely going to see Sun customers getting "offers" to "upgrade" to IBM Blades or loose support.
Jacek Furmankiewicz replied on Sun, 2009/04/05 - 6:00am
in response to:
Jeroen Wenting
An Eclipse-style foundation for Java is the best scenario, and what you describe is the worst case.
Both are equally probable...we'll have to wait and see.
JeffS replied on Sun, 2009/04/05 - 8:41am
"For Java that means that whatever new development takes place will take place solely within the confines of the WebSphere platform, to the point where it will become quite pointless to use Java outside of the WebSphere platform."
If IBM makes Java Websphere specific, than they'd marginalize Java, which, in turn, would marginalize WebSphere.
Seriously, if IBM did that, the migration to .Net and other languages would be massive.
Or, there'd be a fork, most likely backed by Oracle, Google, and others, perhpas (ironically) based on Harmony.
Even if IBM were that stupid, I couldn't care less. I'd just use the other thing (not called Java).
And Java, as we know it, would become the next Coblol or RPG - IBM specific languages to support IBM specifc products (mainframes and AS400's, respectfully).
And the next Java, with real cross platform support, would be Harmony or IcedTea. And WebSphere sales would plummet, because it's based on a closed, proprietary, runtime.
Nos Doughty replied on Sun, 2009/04/05 - 12:59pm
That fact is, love them or hate them, IBM are the primary reason that we Java developers have a job. How many large projects in the government and banking sectors are built on IBM! Now if there was no support from IBM for Java and Linux etc, how seriously do you think these technologies would be accepted in the face of Microsoft propaganda?
I assume that if IBM gets control of Java it will do pretty much what they seem to do with most stuff, which is fap around all over the shop and send some checks to the main developers to keep the technology compatible with their broader strategies. This is a GOOD THING for us developers!
As painful as it will be to see IBM crap on some of our favorite tech, this really is the only way that I see the Java ecosystem surviving. I trust that companies like SpringSource will provide the counter balance to IBM's conservative tendencies, and if IBM knows what is good for them they will encourage them to do so.
I'm not sure we will ever see Java fixed, but at this point I think that the best hope is bringing other languages to the JVM and IBM would be the sort of company to encourage that.
Bring on the Blue Sun!
Bora Ertung replied on Mon, 2009/04/06 - 12:54am
I.B.M. Withdraws $7 Billion Offer for Sun Microsystems
Guido Amabili replied on Mon, 2009/04/06 - 1:53am
in response to:
Bora Ertung
Go Cisco Go ..... !
Jeroen Wenting replied on Mon, 2009/04/06 - 7:25am
If anything then, Sun listens too much to a community that's infested with people who (intentional or not) are trying to harm the language and platform.
Nos Doughty replied on Mon, 2009/04/06 - 5:20pm
I think that Sun can take the lion's share of blame for where Java is today. Generics are a mess because of the way they are implemented and the additional complexity of bounded wildcards. I never understood why Sun didn't just version the platform properly and allow breaking changes to the language. They could easily support two major versions at a time, and perhaps utilize the resources of the open source community to provide legacy support. Dotnet doesn't seem to have a problem with having several versions of the runtime and it has allowed them to vastly improve the language.
Closures are different, and help solve a very serious problem. On projects where I work, the quality of code these days from the majority of developers is shockingly bad. Closures allow senior developers to create APIs that are hard to use incorrectly. I can think of no other improvement to the language that would result in a greater increase in code quality over the long term.
The problem is not that Java is attempting to evolve, it's that as they are unwilling to adjust/break what they already have to make it work, they cannot implement them in a consistent and simple way!
Also, I'd like to point out that Java is *competing* for *developer* mindshare as well as business markets. Java is currently the most painful language I program in (It's still my main language). My own projects are now in Groovy and I am thinking that my next job will be a Rails position or a pure javascript one. If Java just sits there and refuses to evolve, then more and more developers will switch to languages that are more pleasant to use. To be honest, I already look at the recent dotnet releases with jealousy. LINQ looks awesome.
What I don't get, is if the JVM can handle multiple languages, then why can't it handle multiple different versions of Java? The only reason I can think of is the Sun is overly concerned with the Java *brand* and want's ONE TRUE VERSION of Java. THIS is what is causing Java to 'evolve' like a frankenstein monster with bolted on mismatched bits of other languages!
Whatever happens with IBM, it would mean that (a) IBM can't do any worse by applying their usual overly conservative approach. I think that IBM controlling Java would be be *more* to the liking of those that don't want it to ever evolve, (b) it'll be around for ever, and (c) It'll still be the premier choice for enterprise systems.
However, I have to say that for many of us that came into the industry in the early 90's, Sun's potential demise is a tragedy! When I was sitting at uni, all I ever wanted was a Sun workstation of my own (ok, and an SGI one too :) It was a very different world back then. Who would have thought that the 'crayon-os' (win 3.1) would one day rise to utter domination via piggybackingcheap intel pc's into the home!
At least for some of us we can enjoy NextOS in the form of OSX at get a glimpse of the future that might have been ours a decade ago without MS domination. If the cost of Sun going down is an increase in market share for MS and their coked up simian CEO, then it's a tragedy beyond words.