IBM in talks to buy Sun?

[Steven Herod]

http://online.wsj.com/article/SB123735970806267921.html

Personally, it's probably good for 'Java', bad for everything else you
might like about the Sun software and hardware ecosystem.

AIX v. Solaris
Power vs Sparc
Lotus Symphony vs OpenOffice
Websphere vs Glassfish
Eclipse vs Netbeans
DB2 vs MySQL

and so on....

[Jess Holle]

I'm not sure it is good for Java either.

IBM has a 2-tier approach to open source. All their products are closed
source and where these are based on open source they keep the
improvements closed wherever allowed by the license. Thus they've been
moving their Java implementation to Harmony so they can leverage what
others have done while keeping their improvements closed. This hurts
the underlying open source projects and makes their products hard to
troubleshoot.

When compared with Sun's JVM where you can get the full source code, the
contrast is stark.

Also IBM lags way behind most everyone else in implementing and
supporting new Java versions (and new J2EE versions and so on). It is
thus hard to see them pushing Java's evolution anywhere nearly as fast
as Sun does -- despite all the stones thrown at Sun in this category.

--
Jess Holle

[carljmosca]

I am not saying you're wrong about the possible conflicts (and
potential unfortunate outcomes) you list but I think it's better now
than it might have been 15-20 years ago.

IMHO IBM has changed a lot for the better in that time period with
open standards and open source. Remember when you couldn't use SCSI
drives on IBM midrange computers? I think they turned the corner with
the RS/6000, AIX, and Eclipse (just to pick some hardware, OS, and
software as quick examples). It's been a while since they bought
Informix and it's still around (not saying it will be indefinitely).
It would be shame to kill the friendly Eclipse/NetBeans
competition. :)

We shall see...

[phil swenson]

I am unimpressed with IBM's software. WebSphere, AIX, DB2.

The specialize in creating lousy, over-complicated software and make
money on professional services (the only people that can make their
stuff sort of work).

[Jess Holle]

If IBM buys Sun, I wonder if they'll orphan OpenJDK. They can't
un-open-source it, but they could stop contributing or supporting it in
any way.

OpenJDK seems too open for IBM's tastes and isn't based on J9 and thus
may fall afoul of "not invented here" knee jerk reactions.

Personally even apart from having source to Sun's JVM, I've much
preferred its behavior and reliability.

--
Jess Holle

[BoD]

Even Eclipse?

If anybody buys "Java" (maybe not Sun entirely, but just the Java part,
if that makes any sense (probably not)), I wish it was Google.

BoD

[James Ward]

Just be glad it's not Microsoft. :)

-James

[Ravi (Axolotl)]

What will software products stay? Predictions?

If they drop Netbeans I might have to start programming in .NET

[Jess Holle]

If Microsoft bought Java it would be an unambiguous call to action for the Java community to take full ownership of their own OpenJDK fork.

IBM could likely successfully waffle such that OpenJDK is effectively killed or crippled by a combination of their inactivity and the community's -- as the community could be convinced IBM will grow OpenJDK and thus be lulled into inaction, whereas they wouldn't buy that from Microsoft for a second.

[Derek Munneke]

BoD wrote:
> If anybody buys "Java" (maybe not Sun entirely, but just the Java part,
> if that makes any sense (probably not)), I wish it was Google.
>

Then we would all have to program "in the cloud", and anything code you
write, "you give Google a perpetual, irrevocable, worldwide,
royalty-free, and non-exclusive licence to reproduce, adapt, modify,
translate, publish, publicly perform, publicly display and distribute".

;)

/derek

[Frank Mathy]

Let's hope that they won't drop NetBeans for Eclipse. After using
IntelliJ for years I've recently started looking at it, being
surprised of all the goodies in it, good performance and nice UI. I
think that Eclipse is certainly ok, but horrible from a useability
perspective.

[Eric Winter]

Oh man I was dreading it all day and now that you opened my eyes that
they WILL kill Netbeans I am actually terrified. Netbeans is SO much
better than Eclipse and (ahg!) RAD that I am actually terrified. I
guess it is good for IntelliJ which would become my IDE. You can
argue that Netbeans is open source but I don't think something as
complicated as and IDE can thrive without strong backing.

[Christian Catchpole]

Would Duke's nose turn blue?

[Joe Data]

On Mar 18, 6:56 am, Steven Herod <steven.he...@gmail.com> wrote:

> http://online.wsj.com/article/SB123735970806267921.html
>
> Personally, it's probably good for 'Java', bad for everything else you
> might like about the Sun software and hardware ecosystem.

I think it's probably good for Java, too. Whether Sun does the right
thing with Java and it's other software projects or not, it's all in
vain if Sun, the company, can't survive. Looking at the financials,
Sun is the Dead Man Walking of the IT industry. Back on Oct 9, 2008,
long before the bottom fell out of the stock market, Sun supposedly
had close to $3.5 billion in the bank (http://bits.blogs.nytimes.com/
2008/10/09/sun-microsystems-a-lesson-in-failed-cosmetic-surgery/).
Today there were only around $1.6 billion left (http://
techpulse360.com/2009/03/18/ibm-to-buy-sun-microsystems/). In five
months, Sun burned through $1.9 billion in cash, meaning they could be
completely out of cash in four months, give or take a month or two
(yes, this calculation may be oversimplified, but I stand by the
trend).

So let's see, where does Sun revenue come from? 60% come from server
and storage hardware that go to financial institutions and telecoms,
mostly in the U.S., and that was down 14% year-to-year (http://
www.sun.com/aboutsun/investor/earnings_releases/Q209_SLD.pdf, slide
4). Sorry, but no short-term rescue in sight there, not in this
recession with a lot of banks expected to go under soon. Plus,
starting tomorrow, the phones in all Sun accounts will ring off the
hook because the Sun competitors will play the FUD card and try to get
customers to switch away from Sun.

Given the state of Sun, don't you think that big, conservative
customers feel safer when IBM backs Java, a company with deep pockets,
not a company on life support? And from all companies out there, IBM
as the second-biggest software company in the world has the most to
lose if Java goes under. Look at their five software brands (http://
www-01.ibm.com/software) - three of them run on Java: WebSphere is
Java + Eclipse OSGI kernel, Lotus (on the client) is Java + Eclipse
RCP + Open Office fork (and probably Java + Eclipse OSGI kernel on the
server), Rational is Java + Eclipse. Out of pure self interest, IBM
will maintain Java, given that (like Linux) it's the one thing that
runs across all their systems - mainframe, PowerPC, X86/64.

> AIX v. Solaris

Wrong. IBM declared a number of years ago officially that AIX will
eventually migrate to Linux, so it's "Solaris vs. Linux". But IBM
supports its products forever if that brings in money - heck, they
supported OS/2 ten years after it died (http://en.wikipedia.org/wiki/
Os/2#Fading_out). My guess: IBM supports Solaris for a long time but
offers migration to Linux.

> Power vs Sparc
No idea what would happen here.

> Lotus Symphony vs OpenOffice

Wrong: Lotus Symphony is "Eclipse RCP plus unknown OpenOffice
version" (http://blogs.zdnet.com/open-source/?p=2992), so I don#t see
a lot of conflict.

> Websphere vs Glassfish

Wrong: Either "WebSphere + Glassfish" or "Glassfish vs. Geronimo",
since Glassfish could be the low-end open source complement to the
high-end commercial WebSphere, just like Geronimo. My guess: IBM
would merge these two in one shape or another.

> Eclipse vs Netbeans

Yes, that's a tough nut - can't see how they get united, and Eclipse
underpins at least three of the five IBM software brands in one way or
another. My guess: Netbeans gets spun off into a foundation, gets an
initial check from IBM and is on its own from then as n open source
project without corporate backing.

> DB2 vs MySQL
Wrong: "DB2 + MySL" (see above - MySQL as low-end complement to DB2).

[Steven Herod]

Well, I like your picture more than I like mine, and in fact a number
of these comparisons give me some comfort.

I had no idea Symphony was Open Office. I have the CD on my desk but
I'm only installing it now.

Do you see the possibility that Solaris could supplant Linux in the
IBM suite?

On Mar 19, 3:16 pm, Joe Data <karsten.s...@gmail.com> wrote:
> On Mar 18, 6:56 am, Steven Herod <steven.he...@gmail.com> wrote:
>
> >http://online.wsj.com/article/SB123735970806267921.html
>
> > Personally, it's probably good for 'Java', bad for everything else you
> > might like about the Sun software and hardware ecosystem.
>
> I think it's probably good for Java, too.  Whether Sun does the right
> thing with Java and it's other software projects or not, it's all in
> vain if Sun, the company, can't survive.  Looking at the financials,
> Sun is the Dead Man Walking of the IT industry.  Back on Oct 9, 2008,
> long before the bottom fell out of the stock market, Sun supposedly
> had close to $3.5 billion in the bank (http://bits.blogs.nytimes.com/
> 2008/10/09/sun-microsystems-a-lesson-in-failed-cosmetic-surgery/).
> Today there were only around $1.6 billion left (http://
> techpulse360.com/2009/03/18/ibm-to-buy-sun-microsystems/).  In five
> months, Sun burned through $1.9 billion in cash, meaning they could be
> completely out of cash in four months, give or take a month or two
> (yes, this calculation may be oversimplified, but I stand by the
> trend).
>
> So let's see, where does Sun revenue come from?  60% come from server
> and storage hardware that go to financial institutions and telecoms,

> mostly in the U.S., and that was down 14% year-to-year (http://www.sun.com/aboutsun/investor/earnings_releases/Q209_SLD.pdf, slide

[Joe Data]

On Mar 18, 6:56 am, Steven Herod <steven.he...@gmail.com> wrote:

> http://online.wsj.com/article/SB123735970806267921.html
>
> Personally, it's probably good for 'Java', bad for everything else you
> might like about the Sun software and hardware ecosystem.

One more thing: I feel sorry for all Sun employees. I can remember
one episode (last December?) where Tor said how after the layoffs
everybody at Sun had laser-sharp focus on execution and wanted to
stick it to these Wallstreet guys (probably because of the low market
cap). Now it seems that these layoffs were just done to get the
company in "better shape" for a sell-off - the article mentions that
Sun has been shopping itself around for months.

[Steven Herod]

Yep, well, that's the constant perils of being an employee.

You can devote your life to something, and then someone up the food
chain can just erase your efforts with the stroke of a pen.

I know I'd be gutted.

[Peter Becker]

Eclipse used to be impressive, but nowadays I get the impression IBM has
no real interest in making it a decent Java IDE anymore. Compared to
NetBeans it still wins hands down in anything related to code hygiene
(NetBeans doesn't even bother to properly format the code it generates)
and it is still ahead in the refactoring category, but for writing code
I find NetBeans the much more pleasant experience. Apart from formatter
improvements I couldn't tell you what improved between Eclipse 3.2's JDT
and the one in 3.4. It certainly still has that really annoying bug that
clipboard operations fail sporadically on Linux -- you Ctrl-X something
and it is gone from your editor, but the clipboard still has whatever it
had before. No one seems to care enough to fix that and the Bugzilla
they use seems to be close to a one way communication system. If it is a
communication system at all.

I think the lack of interest in the OSS versions of their products is
sometimes quite obvious with IBM and I don't want to know what would
happen to Eclipse if NetBeans would be discontinued.

Peter

[Jess Holle]

Peter Becker wrote:

Eclipse used to be impressive, but nowadays I get the impression IBM has 
no real interest in making it a decent Java IDE anymore.

Eclipse methodology:

Step 1: Capture market share
Step 2: Let free products languish or simply keep it from providing a full solution and prod that market share into various commercial products built on top the free ones.

This methodology makes business sense to corporations involved in Eclipse and I believe is a major reason Eclipse has more corporate partners than NetBeans.  [Others being a lack of clarity on and good tooling for plug-in/module development in earlier releases and, of course, developer market share.]

NetBeans tends to try to provide a full solution.  It has arguably failed in cases (e.g. UML), but nothing is necessarily off limits.  [Well, the conspiracy theorist within me says CPU profiling is, as there are only a handful of things that need fixing in the CPU profiler that prevent it from being useful on large real world systems, but none of the issues I've filed in this area ever get addressed.]

Compared to 
NetBeans it still wins hands down in anything related to code hygiene 
(NetBeans doesn't even bother to properly format the code it generates)
  

Well it formats it in a default manner instead of that which you requested in the formatting options, which is quite egregious and needs fixing!

and it is still ahead in the refactoring category, but for writing code 
I find NetBeans the much more pleasant experience. Apart from formatter 
improvements I couldn't tell you what improved between Eclipse 3.2's JDT 
and the one in 3.4. It certainly still has that really annoying bug that 
clipboard operations fail sporadically on Linux -- you Ctrl-X something 
and it is gone from your editor, but the clipboard still has whatever it 
had before. No one seems to care enough to fix that and the Bugzilla 
they use seems to be close to a one way communication system. If it is a 
communication system at all.

I think the lack of interest in the OSS versions of their products is 
sometimes quite obvious with IBM and I don't want to know what would 
happen to Eclipse if NetBeans would be discontinued.
  

Eclipse is the one major open source product IBM has actually had prolonged interest in.  They contributed some XML libraries and such back in the day as well.

Overall, however, they're adamantly based on a closed source model.  That's fine, but (1) their bashing on Sun about supportinng open sourcing is disingenuous and (2) their products are sufficiently convoluted that I'm not sure how one is supposed to figure them out without a horde of IBM consultants or the source code.

--
Jess Holle

[Jess Holle]

Joe Data wrote:

Power vs Sparc
    

No idea what would happen here.
  

IBM is trying to avoid buying the hardware division according to most articles I've read.  Fujitsu might buy this slice of Sun, for instance.

Lotus Symphony vs OpenOffice
    

Wrong: Lotus Symphony is "Eclipse RCP plus unknown OpenOffice
version" (http://blogs.zdnet.com/open-source/?p=2992), so I don#t see
a lot of conflict.
  

IBM is likely to do as much work in Symphony as closed source rather than in OpenOffice as open source.

Websphere vs Glassfish
    

Wrong:  Either "WebSphere + Glassfish" or "Glassfish vs. Geronimo",
since Glassfish could be the low-end open source complement to the
high-end commercial WebSphere, just like Geronimo.  My guess: IBM
would merge these two in one shape or another.
  

Geronimo is a paper tiger.  Who uses it?  IBM uses it as WebSphere Community Edition just to have something in this space (though it has nothing in common with commercial WebSphere editions).

Glassfish has serious capability and momentum and only needs to be considered "low-end" in that it does not cost as much as WebSphere :-)

For this reason I'd bet IBM will kill Glassfish (or rather all support for it -- it is already open sourced) and keep supporting the paper tiger.

Eclipse vs Netbeans
    

Yes, that's a tough nut - can't see how they get united, and Eclipse
underpins at least three of the five IBM software brands in one way or
another.  My guess: Netbeans gets spun off into a foundation, gets an
initial check from IBM and is on its own from then as n open source
project without corporate backing.
  

I'd bet the same except with no initial check.

DB2 vs MySQL
    

Wrong: "DB2 + MySL" (see above - MySQL as low-end complement to DB2).
  

Except that MySQL may well kick DB2's rear :-)

[DB2 is no Oracle...]

--
Jess Holle

[Jess Holle]

I'm sure Sun execs told employees it was a focus thing and told them to be focused.

Meanwhile they focused on selling the farm...

[Vince O'Sullivan]

On Mar 18, 4:43 pm, James Ward <jaw...@adobe.com> wrote:
> Just be glad it's not Microsoft.  :)

mmm Java#.Net 2.0. I can just see it now.

[Jan Goyvaerts]

For what I am concerned, I would consider this a very sad thought...

I've known Java from the days it started and I've always associated Sun with it. If Sun would be bought, I think Java won't be Java any more. It will have been 'assimilated' by some commercial entity, just for the sake of profits.

And it would really be a shame to have all these great software go to waste. Especially if it's for the sole purpose of making even bigger profits. Never mind what will happen to the employees at Sun in case of such an event. I really wonder if the idea is that pleasing to them. I've been put out with trash once (for being too Java-minded btw) and that was no pleasant experience...

[Ruben Reusser]

Personally, I am wondering what it will do to JavaFX - it would be a
shame to loose it. It's just out there and IBM would have to commit to
the same path that Sun had in mind for it.

Ruben

--
Ruben Reusser
headwire.com, Inc
949 595 4365

[Joe Data]

On Mar 19, 12:29 am, Steven Herod <steven.he...@gmail.com> wrote:
> Do you see the possibility that Solaris could supplant Linux in the
> IBM suite?

I don't know. For all the talk by Sun about the fabulous Solaris
uptake in the market, it sure doesn't make a lot of money: $42 million
in the last quarter, -29% year-to-year. And that number includes
virtualization and management software (http://www.sun.com/aboutsun/
investor/earnings_releases/Q209_SLD.pdf, slide 6). Maybe IBM would
support Solaris if there's money in it, but they don't really need
another OS, they already have Linux. The Register thinks IBM would
just move the SPARC Solaris stuff to AIX using the emulating software
it recently bought (http://www.theregister.co.uk/2009/03/19/
ibm_sun_deal_comment/).

[Frederic Simon]

The main issues I have with IBM:

- They always proved that their product line don't take developer productivity has an input. Developers time is a big revenue stream for IBM global services and others and so they have negative incentive to actually reduce our suffering :(

- Sun has the worst sales department of any big company I had interaction with. But, IBM sales tactics (deception, black mail, pressure, ...) always disgusted me! I'm not naive and I know the sharks but IBM amazing ability to sell bad products is worrying me for the future of the amazing Sun technology offering.

Anyway, let's hope it ends smoothly,

Fred.

--
http://www.jfrog.org/
http://freddy33.blogspot.com/
http://nothingisinfinite.blogspot.com/

[Jan Goyvaerts]

that's one of my concerns too - I love the stuff ! I'm even going to teach it. Let's hope it won't feel like teaching latin - a dead language. ;-)

[Derek Munneke]

would be a pity to see zfs and dtrace dissappear - the linux guys can't
put into their kernel - need some space tourist to adopt opensolairs and
become it's SABDF ;)

[Joe Data]

On Mar 18, 6:56 am, Steven Herod <steven.he...@gmail.com> wrote:
> http://online.wsj.com/article/SB123735970806267921.html

A security filing from Intel reveals remarks from Intel chief Otellini
during a recent staff meeting (http://bits.blogs.nytimes.com/
2009/03/25/intel-boss-says-sun-was-shopped-all-over/):

"I can tell you that Sun was shopped around the Valley and around the
world in the last few months. A lot of companies got calls or visits
on buying some or all the assets of the company. It looks like I.B.M.
is in the hunt now."

This makes it more credible to me that Sun is being sold.

[Christian Catchpole]

There's been lost of talk about what IBM would remove from Sun upon
take over. Sure, there are conflicts in the product line, but would
IBM (or anyone for that matter) be stupid enough to throw away of the
assets they had just purchased? I think it's more likely they will
just run things in parallel. At least until one specific technology
has the writing on the wall.

Here's one I prepared earlier...

http://www.oracle.com/bea/index.html

[jharby]

I was wondering also what would become of the Java Community Process?
Would IBM just commandeer all the specifications and tailor everything
to meet their competitive needs? The agreement to join the JCP really
gives a good deal of power to Sun - http://jcp.org/aboutJava/communityprocess/JSPA2.pdf

On Mar 26, 5:54 am, Christian Catchpole <christ...@catchpole.net>
wrote:

[Van Riper]

On Thu, Mar 26, 2009 at 3:18 PM, jharby <jha...@gmail.com> wrote:
>The agreement to join the JCP really
> gives a good deal of power to Sun - http://jcp.org/aboutJava/communityprocess/JSPA2.pdf

I'm not disagreeing with assessment that Sun has more than its fair
share of influence in the activities of the JCP. However, I don't see
it being due to the structure of this legal agreement. I'm no lawyer,
but, this agreement seems to me to be mainly about making clear that
contributions to the JCP must be unencumbered IP. Not really seeing
where this particular legal document gives Sun any particular power
that IBM would inherit through acquisition. Please enlighten me.

Thanks, Van

--
| Michael "Van" Riper
| JUG-USA Interim President
| https://jug-usa.dev.java.net
----
| Silicon Valley Web JUG
| mailto:van_...@dev.java.net
| http://www.meetup.com/sv-web-jug/
----
| Silicon Valley Google Technology User Group
| mailto:van....@gmail.com
| http://www.meetup.com/sv-gtug/

[jharby]

I am not a lawyer either but FWIW, suppose IBM acquired Sun and IBM
wanted to get complete control of the Java Platform specs - this sort
of control could allow them to elevate WebSphere and other products in
the marketplace. The agreement I gave in that link is between the
participant and Sun (which would now be IBM), not the participant and
the JCP or some other open organization thus IBM could terminate those
agreements - of course it couldn't claim the IP already developed as
theirs but could prevent other vendors from having any influence in
the future spec directions.

On Mar 26, 4:01 pm, Van Riper <van.ri...@gmail.com> wrote:
> On Thu, Mar 26, 2009 at 3:18 PM, jharby <jha...@gmail.com> wrote:
> >The agreement to join the JCP really

> > gives a good deal of power to Sun -http://jcp.org/aboutJava/communityprocess/JSPA2.pdf

>
> I'm not disagreeing with assessment that Sun has more than its fair
> share of influence in the activities of the JCP. However, I don't see
> it being due to the structure of this legal agreement. I'm no lawyer,
> but, this agreement seems to me to be mainly about making clear that
> contributions to the JCP must be unencumbered IP. Not really seeing
> where this particular legal document gives Sun any particular power
> that IBM would inherit through acquisition. Please enlighten me.
>
> Thanks, Van
>
> --
> | Michael "Van" Riper
> | JUG-USA Interim President
> |https://jug-usa.dev.java.net
> ----
> | Silicon Valley Web JUG

> | mailto:van_ri...@dev.java.net

> |http://www.meetup.com/sv-web-jug/
> ----
> | Silicon Valley Google Technology User Group

> | mailto:van.g...@gmail.com
> |http://www.meetup.com/sv-gtug/

[Joe Data]

On Mar 18, 6:56 am, Steven Herod <steven.he...@gmail.com> wrote:
> http://online.wsj.com/article/SB123735970806267921.html
>

> Personally, it's probably good for 'Java', bad for everything else you
> might like about the Sun software and hardware ecosystem.

I still like IBM better than Oracle snatching up the software side of
Sun and HP the rest:
http://www.theregister.co.uk/2009/03/26/oracle_hp_joint_sun_deal/

[kirk]

How about this, GF is currently eating up IBM WS licenses. I know a few
large sites that are switching and it is saving the a bundle... huge
amounts of money. When these things happen (something I've seen it
happen in the past), IBM tends to sniff about but rarely do they buy.

Kirk