Oracle and Sun Team Up to Provide .NET Alternative

segphault writes "Ars Technica has an article about the new partnership between Sun and Oracle, designed to provide an alternative to .NET." From the article: "According to Ellison and McNealy, their mutual goal is the production of a complete Java-centric enterprise datacenter architecture that leverages Solaris 10 and Oracle's Fusion middleware. Designed specifically as an alternative to Microsoft's .NET technology stack, the new platform is competitively priced and based on robust frameworks."

Comments

Team work

[Beuno]

Big names just keep on teaming up to beat down Microsoft.
How long will they be able to resist this?

Isn't this what EJB was supposed to be

[Anonymous Coward]

i thought EJB was supposed to be the .NET competitor

Re:Isn't this what EJB was supposed to be

[slashk]

yes, but EJB was designed by a committee and turned out to be a complete misfire.
that hasn't stopped people from using EJB, though, and for some even liking it - remember that ignorance is bliss

people have used it because they were told that it was the right thing to do
however, in doing so, they have suffered serious productivity losses

if you notice, .net does not have an equivalent to EJB - just doesn't exist
why is this? IT REALLY IS AN UNNECESSARY TECHNOLOGY! for many reasons.

and if you look at EJB 3.0, it is so completely different than EJB 2.0, it would be hard to compare them

why, you may ask - EJB was done by a committee lead by IBM and Sun, with less than knowledgable engineers.
this is NOT a troll - i know this for a fact, have spoken to them,
and have heard them admit it was a mistake.

as you can tell, i have an issue with EJB or any crap technology 'standard' that is delivered to the general public as the right thing to do.

Re:Isn't this what EJB was supposed to be

[sbenj]

Right, except a bit harsh, I think. EJB has lots of flaws, things that people have pointed out over the years, and it certainly wasn't all that simple or transparent. While better things have come along over the last few years, I think that some of those, at least, reflect the lessons people learned from EJB, and the reactions to it (too heavyweight, too obtrusive, too much of a hassle to configure, to slow....) .I think banging on them too much is a bit of monday-morning quarterbacking. What would you have designed for a persistance framework in 1999?

I don't know if I'd go so far as to call it unnecessary, more like, it set out to solve problems that were bigger, or at least different, than the problems people had to actually solve most of the time.

As for the knowledgeability of the engineers, I've found that it's an easy temptation to fall into as a developer to think that you're smarter than everybody else (can't they see how much better your idea is than theirs?)

Re:Isn't this what EJB was supposed to be

[grrrgrrr]

As I understand it EJB does some things very nice and transparent like distributed transactions.. I see programmers were i work that do not know wat distibuded transactions are and probebly just hope for the best but some aplications need things like that and all aplications are probebly better of doing the right thing.

Re:Isn't this what EJB was supposed to be

[slashk]

a bit harsh, probably.

but, they had this pie in the sky idea that EJB would become an enterprise component model for distributed computing.
Session beans were designed to be kind of a modern day CORBA implementation, in fact using IIOP as their wire level protocol.
Entity beans were designed to be a kind of coarse grained persistent component model.
And for 1999, it was a novell concept.

What people ended up trying to do with them is create web applications.
Entity beans were used, often poorly, as a general OR mapping system, which is a tough way to go.
Session beans were used occasionally for remoting, but mostly for either state tracking or state sharing.
Both Entity and Session beans are almost always used locally, hence their introduction of the Home interface.

EJB as an enterprise component model, where applications achieve this SOA style architecture never happened.
Internally, IBM product devisions agreed on EJB as a communications platforms for integrating their applications. This never happened.

IBM's push for this made the EJB specification process very political.
For example, IIOP was pushed as the wire level protocol so it would support legacy C++ CORBA implementations. However, I don't know of any J2EE application that communicates with a C++ CORBA app over IIOP. I'd love to hear if there are some out there.

I'm not saying I had a better solution at the time, but when it did come out (and I knew several people on the original EJB committee), I felt it would not achieve its goals.
My take on it then was XML on the wire, XML as an IDL, with pluggable transports. Yes, even in 1999, some of use were doing this!
But, this is basically what we see with SOAP/WSDL.
This has turned out to have it's own limiting issues, though.

Personally, I would have provided very minimalist interfaces for a lot of this. Then, I would have allowed someone else to take the arrows.
Heck, .NET is only now planning to release a persistence framework, after literally thiking about it for 2 years, and it hasn't seemed to affect their market share.
And they took 6 years so far to build WCF (indigo).

In any case, 1 more interesting note.
I had the opportunity once to corner some of the J2EE leads and architects at day long private meeting at Sun.
Their response was basically apologetic, although the architects were really hung up on JDO. Marketing told us that they have devoted 99% of their efforts to Web Services.
Furthermore, we were told that the Java group is being put under the manager who really pushed Solaris to where it is now, and that in time Java/J2EE should being to improve.

I have a lot of interesting Sun/Microsoft stories, actually, but those are for another day :-)

Re:Isn't this what EJB was supposed to be

[slashk]

point is that .NET and many other technologies can do distributed transactions without and EJBish thing. actually, it is JTA/JTS that makes distributed transactions possible. EJB just allows you to remote your transaction with remote invocations, which is very very rare.

Re:Isn't this what EJB was supposed to be

[Anonymous Coward]

Great Summary -- EJB was pretty obviously designed to solve IBM's (and their customer's) legacy systems issues. What never made sense is why Sun got behind it and pushed it as a non-legacy application architecture.

Re:Isn't this what EJB was supposed to be

[slashk]

harsh, perhaps.

EJB had this pie in the sky concept of becoming a general purpose distributed enterprise component architecture, where applications could be integrated SOA style (before SOA was a term).
Internal IBM divisions saw this as a way to integrate their various applications.
None of this ever actually happened.

EJB Entity beans were meant to be coarse grained persistent distributed components.
Session beans were meant to be a means of remoting.
IIOP was adopted to integrate with legacy CORBA apps.
Hardly anyone uses EJB this way.

People use EJB these days mostly to build web applications.
Entity beans are used to do OR mapping, which is a tough way to go.
Session beans are used for state tracking and state sharing.
Most EJB invocations are local, hence the Home interface introduced in EJB 2.0.

In other words, its kind of like buying a house because you need lumber.

It was clear when the spec first came out that complexity and complex interdependencies would prevent the grand vision from happening.

My original take on it was that it wouldn't achieve the desired goal (and it hasn't).
In fact, back then, some of us were advocating XML on the wire and XML as an IDL. This is kind of what SOAP/WSDL became. But that has it's own issues.

In any case, I would have implemented a very minimal interface for J2EE and let someone else take the arrows.
MS took 2 years just to think about LINQ in C# 3.0 and over 6 years so far to build WCF!
And it hasn't seemed to hurt its market share so far.

I was at an all day private meeting once with Sun on J2EE.
I would summarize it as this - they promised to fix it over time, and to prove it told us that they put their Solaris dev manager in charge of Java.
This caused some rifts in Sun, but they were getting beat up too much over Java/J2EE's immaturity.
So, they gave it to one of their most accomplished dev managers.

Re:Isn't this what EJB was supposed to be

[grrrgrrr]

But with ejb you do not need to know.I must admit i like spring a lot more than ejb. Often real inovation does not come from the real big guys the are to bussy inventing acronyms.

Re:Isn't this what EJB was supposed to be

[slashk]

to some degree, there was some cross over of personnel between the divison of IBM spearheading EJB and Sun Microsystems.
some of the people on either team went way back
i think that had at least something to do with it

IBM was also threatening to start up their own version of Java at the same time.
Sun didn't want to see Java move in 2 different directions, and J2EE became a way of gluing it together for a longer haul.

however, you are beginning to now see another push to split java into an open source version and a sun version.
so, sun will have to do something again.

Re:Isn't this what EJB was supposed to be

[slashk]

you still have to demarkate your transactions, just as with .net

but, yes, spring and hibernate started with a focus on OR mapping.
EJB entity beans were kind of meant for something else originally, that's why they don't quite do the trick.

I was told that Entity beans were meant to be a 'coarse grained distributed persistent component' with few, if any, relationships.
This becomes apparent when you try to use them.

EJB 3.0 is actually sort of based on hibernate.

Re:Isn't this what EJB was supposed to be

[revengance]

I think it is more likely that EJB is oversold by companies that are making EJB products. EJB is useful or even necessary for some applications. However, I would say that using EJB in most applications is not appropriate and could kill the applications faster and better than anything else. Another problem is that nowadays, the IT industry field is dominated by people who don't know anything about technology. A lot of people who I came into contact with does not understand what is EJB or J2EE. I am not saying that I understand it very well, but at least I will not use EJB as a de facto persistent technology like what a lot of other people did.

Re:Isn't this what EJB was supposed to be

[Decaff]

yes, but EJB was designed by a committee and turned out to be a complete misfire ...
people have used it because they were told that it was the right thing to do
however, in doing so, they have suffered serious productivity losses

Yes, like the most successful site on the net - EBay - which is based almost entirely on J2EE/EJB.

if you notice, .net does not have an equivalent to EJB - just doesn't exist
why is this? IT REALLY IS AN UNNECESSARY TECHNOLOGY! for many reasons ....
and if you look at EJB 3.0, it is so completely different than EJB 2.0, it would be hard to compare them

You are confusing two things. EJB (Enterprise Java Beans) and Entity Beans (or to be very specific - container managed persistence of Entity Beans).

Most of EJB has been very successful and popular - Message Beans, Stateless Session Beans.

Contrary to what you say, .NET DOES have equivalents to these.

What has been a problem, both in terms of performance and complexity is Entity Beans - the object relational mapping (ORM). .NET has no equivalent - this is one of the most difficult things to implement.

EJB 3.0 differs mainly from previous versions in the way it handles ORM. There is a new, very simple, version based on well-established Java technologies used successfully by hundreds of thousands of developers - Hibernate, TopLink and JDO.

this is NOT a troll - i know this for a fact, have spoken to them,
and have heard them admit it was a mistake.

Entity Beans as implemented up to EJB 2.1 were probably a mistake - they are the only part of EJB that EBay doesn't use.

as you can tell, i have an issue with EJB or any crap technology 'standard' that is delivered to the general public as the right thing to do.

Fortunately, Java is about choices. EJB 3.0 ORM is based on the choices that the public (the developers) have already made.

However, to claim that all of EJB is flawed is to misunderstand and misrepresent things - it has in general been highly successful.

Re:Isn't this what EJB was supposed to be

[Decaff]

however, you are beginning to now see another push to split java into an open source version and a sun version.
so, sun will have to do something again.

Why? It does not matter whether or not there are open source versions of Java. What matters is whether they pass the compatibility tests, so they can be labelled 'Java'. Sun have stated that they are entirely happy with this. There are many open source implementations of Java APIs that have passed such tests - indeed many of the reference versions of these APIs are open source!

Re:Isn't this what EJB was supposed to be

[ad0gg]

Thats funny because i remember Ebay blaming sun for their outtages a while back which costs ebay millions of dollars.

Re:Isn't this what EJB was supposed to be

[Decaff]

Thats funny because i remember Ebay blaming sun for their outtages a while back which costs ebay millions of dollars.

Firstly, that was in 1999! You are trying to make a point about modern J2EE with an example six years old!

Secondly, that has not stopped them continuing with J2EE/EJB and making billions ever since.

Re:Isn't this what EJB was supposed to be

[Darth Daver]

"as you can tell, i have an issue with EJB or any crap technology 'standard' that is delivered to the general public as the right thing to do."

    Then you must have an issue with the Windows family of OS, Internet Explorer, ActiveX, .Net, Visual Basic, and many other crap technologies.

    Your post does not explain why EJB is crap. It only states that you think so and that you claim some IBM and Sun employees have admitted it was a mistake. I honestly have no experience or opinion either way, but your "Score:4, Insightful" post does not help me form one.

Re:Isn't this what EJB was supposed to be

[mabinogi]

No.

EJB is just one part of J2EE which is the closest thing to a .NET competitor.
You can do J2EE without EJB (and in most cases, probably should).

Re:Isn't this what EJB was supposed to be

[slashk]

i'll restate this:

harsh, perhaps.

EJB had this pie in the sky concept of becoming a general purpose distributed enterprise component architecture, where applications could be integrated SOA style (before SOA was a term).
Internal IBM divisions saw this as a way to integrate their various applications.
None of this ever actually happened.

EJB Entity beans were meant to be coarse grained persistent distributed components.
Session beans were meant to be a means of remoting.
IIOP was adopted to integrate with legacy CORBA apps.
Hardly anyone uses EJB this way.

People use EJB these days mostly to build web applications.
Entity beans are used to do OR mapping, which is a tough way to go.
Session beans are used for state tracking and state sharing.
Most EJB invocations are local, hence the Home interface introduced in EJB 2.0.

In other words, its kind of like buying a house because you need lumber.

It was clear when the spec first came out that complexity and complex interdependencies would prevent the grand vision from happening.

My original take on it was that it wouldn't achieve the desired goal (and it hasn't).
In fact, back then, some of us were advocating XML on the wire and XML as an IDL. This is kind of what SOAP/WSDL became. But that has it's own issues.

In any case, I would have implemented a very minimal interface for J2EE and let someone else take the arrows.
MS took 2 years just to think about LINQ in C# 3.0 and over 6 years so far to build WCF!
And it hasn't seemed to hurt its market share so far.

I was at an all day private meeting once with Sun on J2EE.
I would summarize it as this - they promised to fix it over time, and to prove it told us that they put their Solaris dev manager in charge of Java.
This caused some rifts in Sun, but they were getting beat up too much over Java/J2EE's immaturity.
So, they gave it to one of their most accomplished dev managers.

Re:Isn't this what EJB was supposed to be

[slashk]

ActiveX, COM, VB were pretty hokey, and never lived up to their expectations
in fact, COM was mostly used to communicate between languages, rather than a component architecture
and DCOM/MTS - don't even get me started.

however, .net has been different.
MS is doing something they never really did before - take their sweet old time to produce quality.

check out C# 3.0 and LINQ* - it's pretty amazing.

and if you think .net is flawed, then why is Java changing its original 'formula' to look like C# to such a great extent.
really - generics, attributes, first class treatment of iterators, etc.

now, i'd like to see what happens to EJB and J2EE when WCF comes out.
EJB was thrown together quickly with little or no feedback based on usage.
WCF represents nearly 7 years of R&D at Microsoft, with iterations of usage in applications, feedback, etc.
and it's based on lessons learned from DCOM.

In any case, the .net stuff at MS is doing things much differently than the stuff you saw at of MS before.

I for one like where they are going, and I think they are generally producing higher quality products than most other vendors right now.

Geeez

[matr0x_x]

It's like the fall of the Roman empire - when everyone is fighting the same corporation it's tough even for the biggest corporations to hold ground

Re:Geeez

[smittyoneeach]

What an interesting comparison.
Does that imply that the Empire will be divided, say, an OS in the West, and an Office suite in the East, and the East hang around for another thousand years after the FOSSian hordes have crossed the Columbia to re-partition the buggy carcass of the West?

hilarity

[BitterAndDrunk]

Since Oracle Applications is still driven by ActiveX controls.

As is their AIM methodology.

In fact, Oracle Apps downloads are unsigned, untrusted. You have to open the browser (and it must be IE) pretty dern wide to use it.

Re:hilarity

[Anonymous Coward]

Are you running version 8 or something? That is not my experience on version 10 or whatever version my company is running.

Re:hilarity

[Threni]

More hilarity:

---
"SAP believes that they can modernize their applications without changing them. They keep writing programs in a language called ABAP [Advanced Business Application Programming], which is a 25-year-old proprietary language not related to Java.
---

It's 25 years old, for gods sake! And it's not related to Java! You say it like it's a bad thing! Guess I'd better stop using C then.

What's flavour of the month right now?

Re:hilarity

[Anonymous Coward]

It's 25 years old, for gods sake! And it's not related to Java! You say it like it's a bad thing! Guess I'd better stop using C then.

True, but the difference is that ABAP is garbage and C isn't. After all, how many other 25+ yr old languages are still in heavy use? A few, sure, but most have fallen by the wayside. What was the last app you saw written in JOVIAL? BLISS? PL/I? SNOBOL?

Ok, so *I* still think SNOBOL is a million times better than Perl, but whatever.

Pricing...

[Anonymous Coward]

"the new platform is competitively priced"

What!? I remember when Oracle and Sun charging was based on how much money fell out your pockets when they turned you upside down and shook you.

Seriously though, an alternative is nice, but isn't that alternative already here and called Java? I suppose a nice end-to-end branding a-la .NET wouldn't go amiss, but what's the point? Sufficient technologies already exist out there to do what they're trumpeting as new...

Re:Pricing...

[Debiant]

I think It's not just about technology, but about user end and development support.

If I compare Java and .NET, I must say I think it's right now much easier to do things with .NET .

I'm not talking about being platform independent, robustness or things directly related to merits of some programming language or enviroment, but more about how many potential people have access .NET technology.

For example, VB programmer may with some training be able to move his old VB code's business logic to .NET server. Same goes to C++ programmers. Even Java developers may find C# much more intresting than Java, because it pretty close but still diffrent(and not with a negative way). In a way .NET is a culmination of many programming languages, and that way looking far ahead of Java where you can only 'plug in' with Java only.

Besides Microsoft with it's traditional method, is trying to support .NET much as it is possible.

So I can understand why Ellison is trying to do what he is. as he sees that .NET has much synergy. More I look .Net, more I've started to wonder why it has been so overlooked.

It's coming fast, where I'm looking at it.

Re:Pricing...

[NemoX]

I agree. Java IS the alternative. They just need to improve a few areas:
1. Web applications - why would someone pay $20,000 for a java app program to run on top of a web server (i.e. websphere, jrun, etc.). They need an easy, cost efficient web application part that is identical to their client GUI.
2. They need to improve the performance of the GUI, and the creation thereof (i.e. easier creation in an IDE).

Re:Pricing...

[malraid]

I wouldn't say that .Net has been overlooked. .Net is mostly popular with MS centric developers, people that mostly used VB o VC, not so much with Java people. Some Java developers might be tempted to look to what C# has to offer, but at the end of the day, both are only tools. You can build great or crappy programs in any language. Large enterprises that have lot's of J2EE code WILL NOT swtich to .net, simply because J2EE has delievered, while .Net still has to prove to be as robust as Java. Java used to be sexy back in the 90s, now it's the new COBOL.

Re:Pricing...

[Asprin]

Java?! Why can't Sun come up with a java compiler that can target freakin' Mono?

Re:Pricing...

[OpenServe]

So I can understand why Ellison is trying to do what he is. as he sees that .NET has much synergy. More I look .Net, more I've started to wonder why it has been so overlooked.

What is truly mind boggling is the apparent conclusion that Java's correct "answer" to .NET is anything but a heavy embrace of open source. (Apache/LGPL style) The only way to match the growing "synergy" of .NET is to work from the grassroots up and not look only at the "big iron" markets. While lucrative, they can not yield sufficient growth for the Java platform. .NET is not just an enterprise datacenter solution -- it's a unified desktop/server/mobile/web platform and as such it is very much "embrace and extend." The areas where Java is weakest are on the client side and this is where .NET has / will have it beat hands down unless drastic changes are made. Java Desktop was a good initiative but it hasn't gone nearly far enough. At this point, I'd have to put my bets on open source Java projects as the true saviors of the language. (even as pioneers of fresh approaches not designed by committee.. think EJB2 -> Spring/Hibernate -> EJB3) This is not to say Sun and Oracle lack a significant market in the high-end, but theirs is also a niche compared to the whole landscape .NET is going after. If they don't expand their horizons and embrace open source in the low/mid-range markets, the high-end niche could shrink as .NET marches forward on both hype and shiny new technologies.

If anyone from Sun happens to be reading this, please don't overlook where .NET is going with UI technology. Swing/SWT/JSF came too late and are inadequate in comparison to upcoming .NET XAML/WPF. Nevertheless, XAML is not perfect and there is still an opportunity to leap-frog ahead of it before it gains too much traction post-Vista-launch. Now would be the perfect time to tear down the popular image that Java is good for the server but awful for client-side.

Re:Pricing...

[Anonymous Coward]

In a way .NET is a culmination of many programming languages, and that way looking far ahead of Java where you can only 'plug in' with Java only.

Are you seriously saying the best thing about .net is that you can use another language with it? That the best thing about using it is that you can not use it? Ok, I'll not use it.

Re:Pricing...

[Decaff]

For example, VB programmer may with some training be able to move his old VB code's business logic to .NET server.

It is not just a matter of training - VB.NET has many differences from VB.

and that way looking far ahead of Java where you can only 'plug in' with Java only.

I really don't know how people come up with statements like this. The facts could not be more different. There are more than 200 different languages than run on the JVM. A large proportion of them integrate well with Java, and can used Java classes and libraries. There are implementations of LISP, Ruby, Python, Basic, Modula, Pascal, Fortran and even COBOL. There are currently far more languages implemented on the JVM than on .NET.

Re:Pricing...

[Zigurd]

More I look .Net, more I've started to wonder why it has been so overlooked. .NET sure is nice. C# is the only language that can go toe to toe with Java and win most aspects of the comparison. .NET confers the abilty to mix in most other languages that are/have been used in various Microsoft products. .NET Compact Framework puts J2ME in the shade, and has since it first came out several years ago, and will continue to at least until a replacement fot MIDP 2.0 ships in quantity - at least 18 months, and perhaps twice as long.

So why can't .NET get any respect? Because the world has changed, and only a minority of customers will settle for a really nice closed-source solution that runs on one server OS, with one supplier controlling the pricing model of the tool chain, platform, and OS.

In mobile platforms, Microsoft insists on tightly coupling .NET with their mobile OS, and its pricing model and sole-supplier disadvantages. Handset makers have no desire to add another royalty burden while becoming further commoditized. Mobile network operators have no desire to propagate Microsoft branding among their customers.

And then there is the fact there is no compelling user-visible use of .NET in the everyday Windows XP experience. While Sun is at least as pathetic as Microsoft in marketing their technologies, most Windows users have seen a Java logo in the course of running an applet or a Java application. Tens of millions of mobile phone users - a number that could plausibly grow tenfold in the next couple of years - know their phones run Java applications and have bought at least one of those applications in a mobile commerce transaction. Do you know any end-users that know they bought anything that depends on .NET?

Re:Pricing...

[jb.hl.com]

I stopped reading CNN a while ago, now I read "1984".

I tried that once, but got a little ticked off by the lack of a sports section. Or news.

Re:Pricing...

[Anonymous Coward]

And what about mixing and matching data structures between those languages? Thought not... that's why .net is vastly superior. It's not just about a shitty byte code that was designed only to run with one langauge... it's a much richer environment. Making other languages run on the JVM is like jogging in high heels - you can do it, but you'd be better off with a pair of trainers.

Re:Pricing...

[nxtw]

While it'd be nice to say that all of those languages are equally supported on Java, they aren't. Java is the primary language on the JVM; most documentation is for Java only; and functionality, performance, etc. with non-Java languages varies.

At the very least, Microsoft offers VB.NET and C# which are both equally supported by Microsoft and most vendors; they also have J# and C++, which have varying degrees of support (I can create web projects using C#, J#, or VB.NET; .NET Compact Framework projects with C# or VB.NET; .NET Windows Forms applications with any language.) All of these languages are supported out of the box with Visual Studio 2005.

Microsoft also has JScript.NET, which seems to be the bastard child of Microsoft's .NET languages -- probably intended for simple scripting use only.

Still, when someone says a product works with .NET, it's safe to assume that it's supported with at least VB.NET and C#, whereas most Java products are supported with Java as the language only.

Re:Pricing...

[naden]

For example, VB programmer may with some training be able to move his old VB code's business logic to .NET server.

Wrong. VB.Net is very, very different to VB6 to the point where much of the code will have to be rewritten to either run or run well. Not to mention that there will be extensive training required for VB6 developers to learn the more OO ways of VB.Net.

and that way looking far ahead of Java where you can only 'plug in' with Java only.

Wrong. There are many languages that will run on the JVM e.g. Python, Groovy, COBOL. See here for a full list. Not to mention that having multiple languages is a nightmare in itself for support, maintenance but thats just IMHO.

Besides Microsoft with it's traditional method, is trying to support .NET much as it is possible.

Great. One company supporting .Net. For Java we have: Sun, IBM, Oracle, BEA, Apache, CA, HP, Novell, Apple etc etc. Just look at the list of supporters for the JSR-168 (Portal) and JSR-170 (Content). The JSR process whilst not perfect is a great part of Java and why the platform is doing so well.

More I look .Net, more I've started to wonder why it has been so overlooked.

Easy. (a) Because Microsoft is the ONLY company involved. And that company just happens to have a long history with anti-competitive behaviour (b) Because I want a choice of which servers/OSs to developer and deploy on (c) I want competition so the quality of all the technologies will increase. .Net provides none of these. Java provides all of these.

Re:Pricing...

[Decaff]

And what about mixing and matching data structures between those languages? Thought not... that's why .net is vastly superior.

Of course you can mix and match data structures between these languages.

It's not just about a shitty byte code that was designed only to run with one langauge... it's a much richer environment. Making other languages run on the JVM is like jogging in high heels - you can do it, but you'd be better off with a pair of trainers.

Utter nonsense. .NET CLR is not that much different from the JVM - it designed for procedural/OOP languages like Java/C#/VB.NET. Other languages can struggle to get good performance on it.

Re:Pricing...

[Decaff]

While it'd be nice to say that all of those languages are equally supported on Java, they aren't. Java is the primary language on the JVM; most documentation is for Java only; and functionality, performance, etc. with non-Java languages varies.

The support depends on the vendor - there are many vendors for non-Java languages on the JVM. Naturally, support will vary.

Still, when someone says a product works with .NET, it's safe to assume that it's supported with at least VB.NET and C#, whereas most Java products are supported with Java as the language only.

Java is a different culture than .NET. You can get support for a range of languages on the JVM - but not from Sun. You can get them from a variety of source. For example, language support in different Java IDEs is broad (especially Eclipse).

Re:Pricing...

[Heembo]

Seriously though, an alternative is nice, but isn't that alternative already here and called Java?

No, .NET is a microsoft specific but language-neutral standard. You can write all of your .NET aplications in java. Network app frameworks like this (language neutral) are absolutely fantastic - but who can blame anyone for wanting to see a non MS entity control the standard?

Re:Pricing...

[Fulcrum of Evil]

t the very least, Microsoft offers VB.NET and C# which are both equally supported by Microsoft and most vendors; they also have J# and C++, which have varying degrees of support (I can create web projects using C#, J#, or VB.NET; .NET Compact Framework projects with C# or VB.NET; .NET Windows Forms applications with any language.) All of these languages are supported out of the box with Visual Studio 2005.

Tell me again how VB.NET differs from C#, aside from the syntax candy. Last I checked, they supported almost exactly the same feature set, while the other languages that MS advertised as being supported were cut down to fit in a C# shaped hole.

Re:Pricing...

[electroniceric]

Sure, C# as an individual language is better than Java. And certainly for any UI applications, C#/.NET rocks Java's world. Absolutely, Java should have a linker and "registry" of sorts - certainly this classpath nonsense is way too much work as it stands.

On the other hand, Java is still waaaay more modular for almost any server application than .NET. In Java, you call components which are add-on libraries (hence the incredible aggravation of 10 million directories, XML config files and classpath entries), any of which can be removed, re-implemented, locked down, etc. In fact, many of the enterprise (or close) grade solutions are open source (Tomcat, JBoss, Hibernate, Struts and everything else Jakarta), so you have complete implementation details to boot. With .NET, as with all Microsoft products, everything is "integrated", and all the documentation points you at the Microsoft solution. And typically you have little information about the architecture and internals of the .NET libraries provided by MS: no UML diagrams, no published interface specs, just tech docs, albeit thorough and well-written ones.

Finally, the only thing that will make VB.NET programmers as good at making robust architectures as Java programmers is disciplined OO design of the type Java essentially mandates. .NET will probably catch up in this department, but it will be up to Microsoft to push that kind of development in its tech docs, rather than "here's how build a form with all the logic inside widget event code". .NET is definitely a step in the right direction, and as a language C# definitely improves on many aspects of Java. But the overall .NET development ecosystem and the houses that do it have yet to get the kind of robust that you find in Java. Where Java was a giant leap forward in development languages because it brought OO, managed code compile-time safety to the masses, .NET is merely an evolutionary improvement on those concepts. Many of Java's defects could be remedied with a little standardization and simplification. It will be interesting to see how these two languages mature.

Re:Pricing...

[nxtw]

VB.NET has these where C# doesn't:
Automatic type casting (where appropriate);
Late binding; Functions like IsNumeric, IsNothing, etc.;
the My namespace (in 2.0);
other things I can't remember right now

There are less things an inexperienced programmer has to worry about in VB.NET than in C# -- they can simply ignore what type a variable is and just use it, most of the time (e.g. SomeInt = SomeString will compile and hopefully work, if SomeString can be converted properly.) Many of the changes to VB are necessary in order to have full .NET object oriented functionality. There's a lot of only VB-specific features that make things easier and require less programming.

As for other languages, I can't really comment. There are other languages that have been ported to the CLR as well, but I have no interest in using them, and if I did, it'd probably be with a native/non-CLR implementation.

Re:Pricing...

[Doctor Faustus]

The areas where Java is weakest are on the client side and this is where .NET has / will have it beat hands down unless drastic changes are made.
But so what? The client side is rapidly being taken over by JavaScript/HTML and to a lesser extent, Flash.

Re:Pricing...

[Doctor Faustus]

Are you seriously saying the best thing about .net is that you can use another language with it? That the best thing about using it is that you can not use it?
No, that .Net is not just one language.

Re:Pricing...

[Ravatar]

The new ATi display control panel is powered by .NET, and many other applications are taking this path too.

Re:Pricing...

[petermgreen]

well ms used the .net brand on passport so anyone who uses hotmail or msn messenger will be somewhat familiar with the .net brand.

Re:Pricing...

[misleb]

Ha! You said "synergy!" For a second there I thought you might actually be technical.

-matthew

Re:Pricing...

[petermgreen]

doesn't ms have one of those by the name of j#?

Re:Pricing...

[misleb]

Still, when someone says a product works with .NET, it's safe to assume that it's supported with at least VB.NET and C#, whereas most Java products are supported with Java as the language only.

Who cares? What serious programmer actually uses VB.NET anyway? It is just there to attract the VB code monkeys. Might as well just be C#.. which is so much like Java anyway, the only real difference is the frameworks.

-matthew

Re:Pricing...

[Anonymous Coward]

Well, hell will have frozen over when Steve Gibson (slammed yesterday) gives up writing apps directly in 80x86 assembly and starts hand-writing them instead in MSIL...

Re:Pricing...

[nxtw]

Who cares? What serious programmer actually uses VB.NET anyway?

What makes a "serious programmer"? Given VB.NET's alleged closeness to C#, why wouldn't a "serious programmer" use VB.NET if he or she preferred it?

VB.NET is designed to be easy. The features I listed in this post basically let the programmer write less code. Automatic casting/etc. can be disabled with Option Strict. VB.NET provides additional functionality that programmers can choose to use. Or not. With Option Strict turned on, the language is very similar to C# from the programmer's point of view. Without it, programmers need to think about a lot less and can get away with doing things they normally wouldn't be able to -- which, as long as they're careful (checking that strings have valid numeric data before changing to a number format, not abusing late binding, etc.) isn't too bad. It definitely saves a bit of time, although it does make it a bit easier to make mistakes.

I can't think of anything that C# can do that VB.NET can't. VB.NET provides additional functionality over C# in My and Microsoft.VisualBasic -- but ignoring those, and with Option Stirct on, VB.NET and C# are very, very similar minus the syntax differences. At that point, the functionality is very similar and there's little reason to use one over the other beyond syntax preference.

Re:Pricing...

[Anonymous Coward]

If .net is so great then why does the new.net always bork my computer?

Re:Pricing...

[Debiant]

Damn!

They're on to me!

Re:Pricing...

[Anonymous Coward]

Of course you can mix and match data structures between these languages.

Liar... certainly not in the way that .net's CLI/CLR allows you to do.

Utter nonsense. .NET CLR is not that much different from the JVM - it designed for procedural/OOP languages like Java/C#/VB.NET. Other languages can struggle to get good performance on it.

The words "byte code" apparently shot right over your head, just like many of the crappy things about Java. The JVM was designed to run Java -- and it shows. .net's CLI/CLR may have its flaws, but it fixes much of the brain damage of the original JVM design. Microsoft had a clean slate... Sun is stuck with years worth of back compatiblity from a crappy design in the first place.

Re:Pricing...

[Decaff]

"Of course you can mix and match data structures between these languages.

Liar... certainly not in the way that .net's CLI/CLR allows you to do."

You can use any class or data structure written in one language on the JVM in any other, providing it is a compiled class. There are no restrictions. In OOP languages you can inherit from any other class, no matter what the language the original was written in. Just like .NET.

The words "byte code" apparently shot right over your head, just like many of the crappy things about Java.

Why can't we have a polite debate?

The JVM was designed to run Java -- and it shows.

and the CLR was designed to run languages like C#, J# and VB.NET - and it shows! Perhaps you ought to read some of the development blogs of various language implementors. .NET is certainly a better platform for some languages in some ways than Java, but to assume there are no problems is false. Have a look at the journal of the developers of SmallScript on this matter - it is illuminating!

Re:Pricing...

[DrXym]

I'm not talking about being platform independent, robustness or things directly related to merits of some programming language or enviroment, but more about how many potential people have access .NET technology.

I interviewed 5 people for a .NET position recently. Some of them had put VB.NET or ASP.NET on their resumes and (surprise) turned out to be utterly clueless when asked simple questions such as "what is reference counting", "what is garbage collection" etc. I find asking such questions a good way to gauge someone's competence.

For example, VB programmer may with some training be able to move his old VB code's business logic to .NET server.

I wouldn't let a VB programmer within 20 yards of a server. I'm sure you could train them to VB.NET but I don't believe they have the suitable background discipline to be a natural or even easy fit. In my experience, the best programmers you can get are those already grounded in Java or C++ (with Win32 / COM) experience and therefore fully versed with most of the principles of .NET. I think I'd even pick PHP, Python or Ruby experience over VB.

Re:Pricing...

[GotenXiao]

POINTY HAIRED BOSS ALERT!

Stop buzzwording :P

So I can understand why Ellison is trying to do what he is. as he sees that .NET has much synergy. More I look .Net, more I've started to wonder why it has been so overlooked.

Why has it been overlooked? Maybe because to run .NET the primary path would be to run IIS, unless you want to try using the ASP.NET Apache2 modules or Mono.

Besides the fact that .NET is nowhere near as portable as Perl, PHP or Java, it's Microsoft - not exactly good for peace of mind if you're working on a secure product >.>

Re:Pricing...

[Anonymous Coward]

"but to assume there are no problems is false."

Urrm... having trouble reading, are you? From the message you just replied to: ".net's CLI/CLR may have its flaws, but it fixes much of the brain damage of the original JVM design."

And you got modded up too... slashdot eh? You are either stupid, or doing it on purpose.

Re:Pricing...

[Decaff]

Urrm... having trouble reading, are you?From the message you just replied to: ".net's CLI/CLR may have its flaws, but it fixes much of the brain damage of the original JVM design."

The problem with this statement is that it assumes that the JVM design was 'brain damaged'. It wasn't. There were very specific design considerations which it met well: To be easily JITted/translated to native code, to be a compact code and the well suit procedural/OOP language Java.

The .NET/CLR is very similar. It may have a few more features that better suit other languages, but they are few. Although there have been successful implementations of dynamic languages on the CLR (IronPython), many dynamic languages (like Smalltalk) have struggled to make use of the CLR, and have had to make compromises.

The idea that the JVM is 'brain damaged' is just ranting. The sheer number of languages successfully implemented on the JVM provide the clearest possible evidence against this statement. Things could certainly be better, and it is likely that new opcodes will be added to future versions to make implementation of dynamic languages easier.

But anyway, one developer's 'brain damage' is another developer's sensible choice, so using such terms is meaningless.

Re:Pricing...

[llefler]

(a) Because Microsoft is the ONLY company involved.

Sorry, nope. Borland has been along for the whole ride, and doing a good job getting useful features added. Borland supports .NET with their own C# and Delphi. They support Win32 with C++ and Delphi. And they support JVM with JBuilder. If only they would look to their roots and support JVM with Delphi too.

I'm sure there are other companies supporting .NET as well, I seem to remember hearing about COBOL for .NET, maybe from Microfocus, but I'm too lazy to look it up.

Re:Pricing...

[llefler]

Java?! Why can't Sun come up with a java compiler that can target freakin' Mono?

How many reasons would you like? It's not in their best interest. JVM supports more platforms that .NET. What motivation would they have to do so? Why doesn't Microsoft support WINE?

Re:Pricing...

[Fulcrum of Evil]

VB.NET has these where C# doesn't:
Automatic type casting (where appropriate);
Late binding; Functions like IsNumeric, IsNothing, etc.;
the My namespace (in 2.0);

You can get the automatic typecasting by using variants all over the place like VB does, which will also get you the functionsyou mentioned. Late binding seems to be implemented via IDispatch, so that's supported just as well in C# or C++. What I was asking about was language features, not library stuff. For example, does VB.net allow for closures?

Re:Pricing...

[nxtw]

You can get the automatic typecasting by using variants all over the place like VB does

there are no Variants in VB.NET. automatic type casting/conversion is done with strongly typed variables (e.g. Dim x As Integer: Dim y As String = "3": x = y makes x = 3.)

Late binding seems to be implemented via IDispatch, so that's supported just as well in C# or C++.

Late binding is done by the VB.NET compiler automatically; this isn't done at all in C#. (Dim x As Object = SqlCommand.ExecuteReader(): x.Close() in VB.NET is valid -- object x = SqlCommand.ExecuteReader(); x.Close(); in C# is not.) Of course, this is slow and bad.

IDispatch? Only if you're using COM. For .NET objects, late-binding uses .NET reflection.

What I was asking about was language features, not library stuff.

VB.NET has "library stuff" that's for use in VB.NET only -- specifically, the "convenience" functions like IsNumeric, and new "My" features. Those features, along with late-binding and automatic type casting/conversion, are what really differentiate VB.NET from C# -- not any significant non-syntax language features.

For example, does VB.net allow for closures?

No.

Re:Pricing...

[Fulcrum of Evil]

Late binding is done by the VB.NET compiler automatically; this isn't done at all in C#. (Dim x As Object = SqlCommand.ExecuteReader(): x.Close() in VB.NET is valid -- object x = SqlCommand.ExecuteReader(); x.Close(); in C# is not.) Of course, this is slow and bad.

Looks more like it's done at runtime via IDispatch or similar mechanisms.

IDispatch? Only if you're using COM. For .NET objects, late-binding uses .NET reflection.

so it's the same as C#, more or less?

VB.NET has "library stuff" that's for use in VB.NET only -- specifically, the "convenience" functions like IsNumeric, and new "My" features.

Well, you probably could use the my namespace in C#, but it'd likely require the Vb runtime and be a pain.

Those features, along with late-binding and automatic type casting/conversion, are what really differentiate VB.NET from C# -- not any significant non-syntax language features.

Which is really my point - VB and C# are basically the same, so trotting that out as an advantage over the JVM, where only Java is first class, doesn't make sense.

Re:Pricing...

[nxtw]

Looks more like it's done at runtime via IDispatch or similar mechanisms.

Did you completely neglect to read what I said? The VB.NET compiler does the required late binding -- in other words, generates the late-binding code. Furthermore, as I said before, IDispatch is NOT USED -- .NET Reflection is -- unless you're using COM, and in my example, you aren't.

so it's the same as C#, more or less?

No. C# does not do late binding for you. You have to write the code yourself if you need to use late binding, or cast object types to what they're supposed to be.

Well, you probably could use the my namespace in C#, but it'd likely require the Vb runtime and be a pain.

You can do something like that with the aptly titled "That". "My" is in VB.NET, however, to make things easier for VB programmers, which tend to be beginners.

Which is really my point - VB and C# are basically the same, so trotting that out as an advantage over the JVM, where only Java is first class, doesn't make sense.

It is definitely an advantage. VB.NET and C#, while similar, provide choice. VB.NET has a lot of features for beginners/old VB developers -- some of which can't be done in C# the way they are done so easily in VB.NET (late binding and automatic type casting). Without them, it's more or less C# with different syntax. And some people prefer End statements and newlines to brackets and semicolons, so that alone is an advantage, at least to some.

Re:Pricing...

[Fulcrum of Evil]

And some people prefer End statements and newlines to brackets and semicolons, so that alone is an advantage, at least to some.

No accounting for taste :P

Re:Pricing...

[cfuse]

I really don't know how people come up with statements like this. The facts could not be more different. There are more than 200 different languages than run on the JVM. A large proportion of them integrate well with Java, and can used Java classes and libraries. There are implementations of LISP, Ruby, Python, Basic, Modula, Pascal, Fortran and even COBOL. There are currently far more languages implemented on the JVM than on .NET.

Then maybe they should spend their money on making this fact more widely known. I didn't know this, and I doubt many others did either.

Still, even with 200 implementations it's not quite visual studio, is it?

Re:Pricing...

[wheresjbob]

"Alternative to .NET"?? .NET is a cheap imitation of Java, yet ironically cost M$ about $1B. I like to think of .NET as a virus-enabled and memory leak-enabled version of Java.

It's funny how a copy becomes the original, somehow. Like Microsoft actually invents things themselves! HA!

Re:Pricing...

[Decaff]

Then maybe they should spend their money on making this fact more widely known. I didn't know this, and I doubt many others did either.

The JVM/JRE is a platform that many develop on. There is no central control over what languages are developed, so who would make these languages widely known? What you get is individual language vendors promoting their product.

This emphasises a difference between Java and .NET - Java has a wider culture.

Still, even with 200 implementations it's not quite visual studio, is it?

I think that it is better. There are very high quality cross-platform, multi-language IDEs like Eclipse and NetBeans. This is a far better situation than with Visual Studio.

Re:Pricing...

[finkster]

While it is entirely true that .NET does have more of a "Production Ready" mix of languages, which boil everything down to MSIL (Microsoft Intermediate Language) - more so than Bytecodes (Java's IL), I have to ask why is this such an issue?

Every project that I have worked on in the past 18 years has been completed using a single-language. The exception for this is web-based applications where you have a mix of languages in each tier:
        Front-end: HTML, JavaScript
        Middle Tier: VB, Java, VB.net, or c#
        Data Tier: SQL, Stored Procedures (PL/SQL, TSQL, Java SPs)

Would it NOT be more productive if you had a single language that was common amongst all of the tiers in an application? This would provide maximum coverage for all of the developers and would extend the knowledge base as new aspects of the application are added.

While I am not saying Java is the BEST tool for all problems, it is a good choice for solving problems on many different tiers of an application.

If it were YOUR money running an IT shop, wouldn't you want to leverage all of your developers on all of the different parts of an application?

Re:Pricing...

[OpenServe]

But so what? The client side is rapidly being taken over by JavaScript/HTML and to a lesser extent, Flash.

In some ways, I wish this prognosis was as true to reality as the current hype suggests. The lightweight web makes everything neat and tidy since you don't have to worry about complex client-side platforms. (beyond a modern browser.. which will eventually mean Mozilla or Safari if IE continues to ignore W3C) Unfortunately, there are many things you still cannot do with JS/DOM/XHTML/etc. How important those shortcomings are has yet to be determined. I believe there is still a future in the myriad of W3C standards (especially with upcoming SVG, Canvas, CSS3, etc.), but I'm not sure how far these will extend beyond the document-centric web we know today.

One thing that could significantly change the picture would be a first-class integration of Java by the Mozilla project. This would provide a managed extension beyond the browser itself and the inherent limitations of JavaScript. (local hardware access, data caching, offline support, standalone web-managed apps, more efficient libraries for working with the DOM, etc.) This can all be done today with applets, Java Web Start, and Eclipse Rich Client Platform, but these are not bundled and integrated with Mozilla by default. If they were, Java / Mozilla / modern W3C standards would become an easily targetable platform for next generation web apps, while providing a seamless integration with today's web. It wouldn't be the end of the story, but it'd be a major step forward for open standards.

This misses the point

[Anonymous Coward]

.NET is not just about enterprise datacenter, but mainly about the next generation of client software. .NET is at the core of the windows vista API. So in order for sun to compete with .NET, they would have to improve the client side support, e.g. Swing.

Re:This misses the point

[Anonymous Coward]

Swing is terribly ugly and slow. Try wx4j - a Java Binding for wxWidgets - instead and enjoy the native widgets with their normal speed.

Re:This misses the point

[Decaff]

.NET is not just about enterprise datacenter, but mainly about the next generation of client software. .NET has had little impact in the 'enterprise datacenter'. The only place where it is truly successful is client software. .NET is at the core of the windows vista API. So in order for sun to compete with .NET, they would have to improve the client side support, e.g. Swing.

Sun aren't just competing with .NET - in terms of client side GUI, Swing is now more widely used than WinForms - check any job site.

Re:This misses the point

[Decaff]

Swing is terribly ugly and slow.

Just FUD. Swing is whatever you want it to look like - the look and feel is pluggable. As for slow, it is OpenGL and DirectX accelerated.

Re:This misses the point

[Anonymous Coward]

So then it looks like no one ever wanted Swing to look unugly and yeah, I know Java is supposed to be fast but reality is that it sucks on that front.

Re:This misses the point

[Decaff]

I know Java is supposed to be fast but reality is that it sucks on that front.

Why not try a recent version? Things have changed hugely. It really is worth looking at an application like say, JEdit, under Java 1.5 before commenting.

Re:This misses the point

[Anonymous Coward]

Correction, Swing requires MANY MORE PEOPLE to build than WinForms. Or GTK. Or TCK. Or just about anything else for that matter. I work in a mainly Java shop, and believe me, if long term effectiveness and efficiency are what you're looking for, Java isn't it, and Swing is definitely out.

Simply using job site stats misses a very important indicator, the number of people required per project of a given size. Swing is practically king here, requiring far more people than any other GUI toolkit. As a matter of fact, the number of Swing postings is largely because employers have discovered they can't build Swing UIs with as few people as WinForms, thus generating an artificial demand.

Job demand can increase in 2 ways:
1) Genuine demand for the technology
2) Artificial demand because of the sheer number of people required.

Swing and Java falls into 2.

Re:This misses the point

[man_of_mr_e]

You want to know why .NET is growing at a fast pace? I can download VS.NET Express for free, and inside of 10 minutes, have my first web app (hello world, or whatever) running right there on the box, with all of about 10 clicks total from download to running app.

Now, how do you do that with Java? It's a bunch of disparate technogies with all their own quirks that need to be all working in unison with magic incantations to get everything to work. You need to get and configure Apache and Tomcat. You need to install the JVM. You need to download and install Eclipse or something similar. You need to build your skeleton and get it all going. You really have to be an ubergeek to get started in this environment. VS.NET makes it so ANYONE can get started if they want to.

Re:This misses the point

[VGPowerlord]

Swing is whatever you want it to look like - the look and feel is pluggable.

That's part of the problem. Consistency between applications in a windowing system is a great boon to the casual user... and to users who don't like

Swing is pluggable, yet I can't choose to use my own OSes widgets. Why is that? Instead, I have to use

UIManager.setLookAndFeel( UIManager.getSystemLookAndFeelClassName() );

, which, instead of using the library already in memory to render them, loads its own library.

Just FUD.

Here, I'll even help spread some FUD. I was looking at using the Java Media Framework (JMF) recently to play some video files. However, I was using Swing to write the GUI. Here's what Sun has to say about Swing and JMF:

JFC/Swing components are light weight components and JMF by default uses heavy weight components. Heavy weight components are used to JMF's advantage as they permit using native rendering methods for higher frame rate video.

Got that? Sun doesn't use Swing components because they don't perform as well as native components. If Sun doesn't use them, why should anyone else?

Re:This misses the point

[VGPowerlord]

Sorry, ended a sentence without completing a thought. The first paragraph in the preceding post should read "That's part of the problem. Consistency between applications in a windowing system is a great boon to the casual user... and to users who don't like skinnable applications."

Re:This misses the point

[ZenShadow]

IIRC Swing has been around way longer than .NET, and that's probably why you see so many more postings. Or maybe it's because you don't need a "WinForms developer" to write for Winforms. YMMV. But that's not what I wanted to post about.

I wanted to speak to Enterprise data centers, since I work in such an environment. Said environment is, in fact, one of the top 10 most visited web sites in the US according to some analysts (lies, damn lies, and statistics as always, so I won't bother being more specific).

We use .NET for over 50% of our services, and it's growing rapidly because .NET has proven that it scales better than most other solutions (Java included). .NET is just now creeping onto the radar in the enterprise. My prediction: 2006 or perhaps 2007 will be the year that .NET overtakes Java in terms of new development. It won't take much longer than that, though. What you're seeing is the after-effects of Microsoft's early marketing attempts, which didn't really tell anyone what .NET was (why would I want it? Microsoft didn't answer that question well). Now it's starting to get traction as more and more people understand just what .NET is, and it will probably be exponential.

--S

Re:This misses the point

[Decaff]

Correction, Swing requires MANY MORE PEOPLE to build than WinForms

Don't be silly. A Swing interface can be done in a few seconds using a GUI designer like NetBeans. This has to be one of the worst attempt to justify an argument I have ever read on Slashdot!

Re:This misses the point

[Decaff]

IIRC Swing has been around way longer than .NET, and that's probably why you see so many more postings.

My prediction: 2006 or perhaps 2007 will be the year that .NET overtakes Java in terms of new development.

These show a misunderstanding of what is happening. Swing has only just taken off on the desktop because it was truly awful in terms of speed and looks more than a couple of years ago: No-one was seriously using Swing before that.

Secondly, there has been mention of .NET overtaking Java for a long time. But this is meaningless as they are largely used for different things - .NET as an upgrade for the existing (and pretty saturated) VB6 and VC++ Windows development and Java for Server Side development. Most large companies use .NET and Java together.

Re:This misses the point

[Decaff]

That's part of the problem. Consistency between applications in a windowing system is a great boon to the casual user...

I have never believed this. It is just a developer's myth. Users usually haven't the slightest problem determining what a button or menu is, no matter what the GUI.

Anyway, if consistency is a major advantage, why to Microsoft keep changing looks and feels so much>

Re:This misses the point

[Anonymous Coward]

It's even easier to do with Sun's Java Studio Creator. Drag and drop web apps which will run on a multitude of servers... not just one.

Re:This misses the point

[VGPowerlord]

Anyway, if consistency is a major advantage, why does Microsoft keep changing looks and feels so much?

I don't know, but I'd really like to slap the Office team.

Re:This misses the point

[man_of_mr_e]

You're still missing the point. All of those "multitude" of servers are complex to set up in the first place.

Mono?

[Anonymous Coward]

I thought Mono is (will be?) the alternative to .NET.

Re:Mono?

[Anonymous Coward]

Mono IS .Net, n00bface.

I'd give it a shot

[TubeSteak]

If it isn't a resource hog.

Hopefully they'd put some effort into making sure it is at least as secure as .Net

Re:I'd give it a shot

[Anonymous Coward]

You'd "give it a shot if it isn't a resource hog"? I guess your job must involve making purchase decisions for enterprise-grade software at a large corporation...or you don't know what the fuck you're talking about, one of the two.

Re:I'd give it a shot

[Anonymous Coward]

If it isn't a resource hog.

Dude... it's Java. Fat fucking chance.

That's funny...

[ltwally]

"...the new partnership between Sun and Oracle, designed to provide an alternative to .NET. ... Designed specifically as an alternative to Microsoft's .NET technology stack..."

That's kinda funny, 'cause here I was thinking that .NET (which is only a couple of years old) was the alternative to Java (which is 10+ years old).

Re:That's funny...

[timeOday]

You have to wonder where Java might be right now if they'd gotten EJB and AWT right the first time!

Re:That's funny...

[Lauritz]

Have you ever tried?

Re:That's funny...

[hackstraw]

Another insightful post mismodded as funny.

The real funny thing is that the article is a cross between an anti-Microsoft flame and another newsvertisement.

TFA also says, "Both companies have lost a lot of relevance in the modern world, where cost-effective open source software and disposable commodity hardware reign supreme."

Oops. That is the meat of the article. Its hard to keep stock prices up when you are "loosing relevance" in your area of doing business. Stock plot here.

Offtopic, but I have to know. I am not a stock or money person. Is it normal for stock graphs to have a logarithmic y axis in terms of dollar amount and percentage by default? I thought that both money and percentages were linear. I understand where log transforms of data are important for things like perception in hearing and many other things, but does this make sense to some money guru out there?

Re:That's funny...

[BarryNorton]

That's kinda funny, 'cause here I was thinking that .NET (which is only a couple of years old) was the alternative to Java (which is 10+ years old).

Come off it, look at the core .Net technologies (before they were re-branded): COM (implementing the multiple interfaces per object idea without multiple inheritance) predates Java, ODBC predates JDBC etc. etc.

Re:That's funny...

[skraps]

Sorry, I'm as big a Microsoft fanboy as any, but this is just wrong. .NET and COM are completely different. COM is basically a convention for interoperating between C/C++ programs. .NET is its own virtual machine and set of languages. ODBC may be similar to JDBC, but that has nothing to do with .NET! .NET uses something called ADO.NET, which is nothing at all like ODBC.

Re:That's funny...

[msloan]

Yeah, you're just plain wrong. A wide selection of apps will run on mac OS/linux.

Same binary code, and as long as you stay within the System namespace you should be fine. True if you use some external dll that pinvokes things (only supplying methods for windows), or you pinvoke things yourself, its not cross platform. However it's generally bad practice to pinvoke things yourself, and many libs that use PInvoke provide cross platform solutions.

No cross platform solution can really be perfect, especially when the platforms are made by seperate organizations.

Re:That's funny...

[CableModemSniper]

I have. It works for at least "Hello, world!".

Re:That's funny...

[NewbieProgrammerMan]

IANAMG but exponentials pop up in some financial applications, like the value of an investment with reasonably constant annual return over a long period of time. On a linear plot, all but the most recent behavior is crushed into a nearly flat line because the value rises (roughly) exponentially; however, on a log plot it's easier to see the long-term behavior.

For example here's the Dow Jones Industrial Average for ~80 years:
Log: http://finance.yahoo.com/q/bc?s=%5EDJI&t=my&l=on&z =l&q=l&c=
Linear: http://finance.yahoo.com/q/bc?s=%5EDJI&t=my&l=off& z=l&q=l&c=

The ugliness of the crash in the 30's is essentially invisible in the linear plot, but clearly visible in the log plot.

Re:That's funny...

[BarryNorton]

You were there when COM was originally released? You were there when VBX was replaced with ActiveX? You were there when ODBC was wrapped in an ActiveX interface forming ADO? You were there when COM was the means of interfacing with Visual J++? You were programming Visual J++ when Sun started legal challenges eventually leading to .Net? You've been programming Microsoft more than a few months?... seems not! Thanks for the lecture though.

Re:That's funny...

[CyricZ]

What exactly are you suggesting, Barry? That .NET is an implementation of COM? I would hope that you're not, because anyone in the know understands that COM and .NET are completely different beasts.

Re:That's funny...

[Anonymous Coward]

Can you read? The very last thing he said was 'implementation of COM'. You're as retarded as the grandparent to your post, with his 'COM is basically a convention for interoperating between C/C++ programs'.

Re:That's funny...

[killjoe]

Too much comittee thinking on both of those. They tried to solve a problem that so few people had. They listened to the corporate CxOs instead the developers and got J2EE. This is a good lesson for all.

I think they should have renamed EJB3 to something else, it is going to suffer from it's name being associated with the madness that was EJB.

Re:That's funny...

[ZenShadow]

I can't speak to any similarities/interoperability between modern ADO and ODBC, but I can speak to the rest. For ADO.NET, I wouldn't be surprised to learn that it's a completely new environment that happens to be capable of interoperating with ODBC drivers -- but I haven't done the research to support that statement.

COM is the basis of most older Microsoft technologies (OLE, DirectX, DirectShow, DirectSound, ActiveX, ...), and .NET can interoperate with COM -- but pure .NET doesn't use COM or DCOM at all. The core of .NET is the CLR, which is Microsoft's answer to the Java VM. By design, it supports mutliple languages (C#, VB.NET, J#, ...).

The CLR leaves any modern JVM in the dust in terms of both performance and usability. A pure .NET app uses .NET-specific remoting protocols, not COM. A .NET app can invoke native system calls and other "unmanaged" code running outside the VM, without any ugly hacked-up proxies (and without using COM-related technologies). And C# is a far more versatile language than Java ever will be.

In short, .NET blows Java out of the water for real-world applications. It's not as portable as Java, but it can be when using pure managed code (hence Mono, which I use daily in a production environment). ASP.NET provides Microsoft's answer to JSP and friends, and it works reasonably well. But most of all, a .NET executable has the same appearance as any other executable on the system, unlike Java (which is a pain in the arse outside of an app server environment).

Being a Unix company, you'd think Sun would've at least supported #! syntax, but noooo...

I am not ordinarily a Microsoft fan. I run Linux on the desktop and on many of my servers, and on the server I vastly prefer it to Microsoft products. But I will defend Microsoft in that .NET, while not revolutionary in concept, is definitely revolutionary in execution. I have built complex production systems in C, C++, perl, Java, .NET, and more. .NET is my favorite environment yet, because it has nearly all the benefits of Java while managing to have nearly none of the drawbacks. And C# is simply a kick-ass language.

And yes, I *was* around when COM hit the market.

--S

Re:That's funny...

[ZenShadow]

I have with far more complex stuff than Hello World. .NET 2.0 stuff is choppy (the Mono team isn't supporting it officially yet, and it isn't nearly ready), but .NET 1.1 seems to work pretty well.

And even with .NET 2.0, I had no problem writing a fairly nice monitoring suite for one of our server farms. It uses the network libs, the XML DOM libs, and other similar stuff. I compile under Visual Studio (sorry, MonoDevelop just ain't there yet), push the code, and it just runs.

--S

Re:That's funny...

[dnoyeb]

I dont know what the point is there? .NET was not 'rebranded' from COM. COM is completely different and does not compare to Java. COM or better DCOM compares with RMI and CORBA. I was into DCOM long ago and my take was this

DCOM: Windows only
CORBA: between languages on the same platform (may have changed since)
RMI: Java only, but cross platform.

COM is rebranded OLE.

Re:That's funny...

[skraps]

Actually, I was programming Windows when COM was originally developed. Back then it was called OLE. The first version of VB I used was in the 2.x series, IIRC. So, yes, I watched as VBX was replaced with ActiveX.

I was there when ADO was first released. But you are wrong about how that came about. ADO is not a wrapper for ODBC. ADO is built on top of OLEDB, and OLEDB has a compatibility provider that interoperates with ODBC. Unless you had to interoperate with an old ODBC provider, there was no reason to use the ODBC compatibility provider in OLEDB. There was a native SQL Server provider made specifically for OLEDB, that did not rely on ODBC.

ADO.NET is another from-scratch reimplementation. The ADO.NET object model only vaguly resembles the ADO object model. Neither ADO or ADO.NET are based on ODBC! They have compatibility providers that can talk to ODBC if you have a legacy data source that is only available as an ODBC provider.

If you don't understand that basic fact about ADO and ADO.NET, then you obviously haven't been around long.

And yes, I do remember when they came out with Visual J++, but I can't say I've ever had the displeasure of using it.

Re:That's funny...

[skraps]

You're as retarded as the grandparent to your post, with his 'COM is basically a convention for interoperating between C/C++ programs'.

My post. Care to back up your use of "retarded" here? What exactly do you disagree with?

COM, at its simplest level, defines:

• a mechanism for locating DLLs using registry information
• loading those DLLs and obtaining an instance of the class factory
• how to call methods on an interface pointer
• the IUnknown interface

At its core, that is all that makes up COM. All of the rest came later. Interprocess marshalling, which is the foundation of DCOM, came later.

All of the technologies commonly associated with COM are just services built using COM. ActiveX, DirectX, etc, are all things built using COM. They are not part of COM.

Re:That's funny...

[Ulrich Hobelmann]

I don't want to spend time trying. I want a platform that's guaranteed to run on different systems. I want to be able to use an API that WILL work everywhere, not one that's patented, or that's at least not existent on Mono.

But the moderators think I'm only a poor, stupid troll, so be it...

Re:That's funny...

[Ulrich Hobelmann]

Yes, and the separate organization part and the patented APIs part is what's bugging me. Without a guarantee, why should I use .NET, even though C# is a bit better than Java, as a language?

No thanks, I'd rather have some safety for my planning. No lawsuits, and a platform that DOES run.

Re:That's funny...

[BarryNorton]

So, yes, I watched as VBX was replaced with ActiveX

Actually it wasn't. Since you're being picky about names, I thought you would have pointed out that OLE controls replaced VBX and that ActiveX was a rebranding. Not so important in itself but...

I was there when ADO was first released. But you are wrong about how that came about. ADO is not a wrapper for ODBC. ADO is built on top of OLEDB

OK, you're right about ADO being the interface to OLEDB, but again I was simplifying and skipping generations. Jet was wrapped in DAO, ODBC was wrapped in RDO and then ADO was a simplification, the underlying layers were 'thinned' (with ODBC being just one option - but the one everyone used for years!) and the object interface was rationalised. Still, if we stop with the willy-waving and look at the actual point I was making, I don't see how you contradicted it. Somehow there seems to be an inability here to distinguish between evolution of pre-existing, pre-Java (and, dare I say, innovative) technologies, and exact code bases/interfaces.

And yes, I do remember when they came out with Visual J++, but I can't say I've ever had the displeasure of using it.

Then maybe, like that troll CyricZ whom I refuse to address directly, you're missing the point - that a common object-based interface between heterogenous languages (not just C/C++, which skraps also seems to be missing) already existed and was applied to Java in J++, showing what Microsoft's inclusive vision for a Java like language was back in 97 (I think it was - that's when I used it). Not only does this strategy form the basis for interoperability in .Net (whether it's based on the COM/COM+ libraries, or slimmed down and more efficient), but it also allowed the provision of libraries like ADO and ActiveX controls, which were summarily ripped off in an non-inclusive language-specific way in Sun's Java.

Re:That's funny...

[BarryNorton]

Now we're really off at a tangent I've no interest in pursuing (these are arguments for 1997, not 2006), but to set the record straight

DCOM: Windows only

DCOM is not Windows only, there was a Linux and Unix implementation even back in the 90s.

COM is rebranded OLE

COM was not rebranded as OLE, they were released at the same time, OLE implemented over COM - see Brockschmidt's book (from which I was inseparable in 95). You're perhaps confusing that OLE Controls were rebranded ActiveX Controls, and the OLE documents/Office stuff as ActiveX Documents.

Re:That's funny...

[BarryNorton]

pure .NET doesn't use COM or DCOM at all. The core of .NET is the CLR, which is Microsoft's answer to the Java VM. By design, it supports mutliple languages (C#, VB.NET, J#, ...).

As I said on another branch, .NET might separate the full and developed capabilities as COM+, a heavier-weight library, but if you want to know how the multi-language interoperability is based around the COM approach, you had only to look at their 'first cut' (before having to design a separate wholly non-Java language) in Visual J++. This is how interoperability, and the provision of libraries, across languages, and with an object-based interface, is achieved in .NET.

Re:That's funny...

[Knetzar]

Log graphs show more of a percent change, where linear graphs show an actual change. When looking at these graphs I don't care if the stock went up $5 (linear) in 1 year, what I do care about is if it went up an order of magnitude (log) or just a small percentage.

Re:That's funny...

[skraps]

Then maybe, like that troll CyricZ whom I refuse to address directly, you're missing the point - that a common object-based interface between heterogenous languages (not just C/C++, which skraps also seems to be missing) already existed and was applied to Java in J++, showing what Microsoft's inclusive vision for a Java like language was back in 97 (I think it was - that's when I used it).

OK, I agree that many languages eventually connected to COM, and Visual J++ was one of them. I don't think that invalidates my point that COM started as a convention for interoperating between C/C++ programs. If you look at how COM is actually implemented under the hood, it is very C++-like. You have what amounts to a vtable at the center of everything. Sure, several other languages use what amounts to a vtable, but the COM vtable is defined exactly the same as the C++ vtable. If you look at the rules for the ordering of the vtable, it is the same as C++. The ordering rules define how to deal with base class members, multiple inheritance, etc. In C++ those rules are actually defined. In Java and most other OOP languages, those decisions are implementation-dependent.

Another issue is the variety of data types. COM can handle anything C and C++ can handle. The basic inproc method invocation is just the thiscall convention. When you look at out of process method invocation, you'll see that IDL and the standard marshaller can do just about everything C and C++ can do.

When other languages began to interoperate with COM, there was a big problem with supported data types. Few languages are expressive enough to use all of the marshalling options of IDL, and fewer still if you include all possible C/C++ data types.

Some languages had such limited type systems that Microsoft went and intented the "automation" subset (using the OLEAUTOMATION attribute in IDL). This would limit you to the types that a VARIANT could hold. Quite a restriction!

After that, they had the brilliant idea of IDispatch for dynamically typed languages. This thing was just a nightmare.

It's not at all fair to say that you could write a C/C++ program using COM, and "seamlessly" interoperate with, say, a VBScript using COM. Yes, by some definition, they both use COM. But there was no solid abstration. To make it work, you had to know that you were going to be talking down to the automation subset.

You can't look at all of that and say Microsoft planned it all from the beginning. IDispatch, OLEAUTOMATION, and everything? If they planned that from the beginning, it was a horrible plan.

.NET isn't perfect with language interoperability, but it goes a heck of a lot further than COM.

I have never used JDBC, so I can't comment on how similar it may be to ADO. That stuff may very well have been ripped off.

Re:That's funny...

[BarryNorton]

If you look at how COM is actually implemented under the hood, it is very C++-like

Yes, it's like Objective C or C++ with multiple inheritance, but:

1) it is not like C (the predecessor, RPC with an unextended C-based IDL, is precisely what this is not - there is object identity here);

2) it was not exclusively, or even initially, for C-like languages;

3) it does not require multiple inheritance (which is 'coincidentally' what Java also claimed credit for some time later).

Some languages had such limited type systems that Microsoft went and intented the "automation" subset (using the OLEAUTOMATION attribute in IDL). This would limit you to the types that a VARIANT could hold

Actually - drifting off point again - Visual Basic (and more importantly VBA) does have a core system of strong types, but it also allows dynamic typing... something that Java also has!

You can't look at all of that and say Microsoft planned it all from the beginning. IDispatch, OLEAUTOMATION, and everything?

Read Brockschmidt...

Re:That's funny...

[ZenShadow]

Visual J++ != Visual J#.

Yes, COM was used in pre-.NET languages. That has no bearing on the present .NET infrastructure.

--S

Re:That's funny...

[BarryNorton]

Visual J++ != Visual J#

Gee, really? You're going to tell me that?

Yes, COM was used in pre-.NET languages. That has no bearing on the present .NET infrastructure.

No bearing, huh?

Well, thanks for your in-depth analysis of what the difference in the object-based interoperability and library model is. What a dull discussion this would have been if the crux of your argument was 'they have different names'.

Good night...

Re:That's funny...

[ZenShadow]

If you wish to continue believing that .NET was designed around COM, do feel free. Reality shouldn't stand in your way. :-)

--S

Re:That's funny...

[BarryNorton]

And if you wish to continue believe that Enterprise Java is 'built around' a database interface that owes nothing to ODBC in its concepts and design, you should likewise feel free. I didn't think that was either of our positions, but reality shouldn't stand in the way of your believing you've won an argument.

Re:That's funny...

[ZenShadow]

And I mentioned what Java was built around exactly... where?

Dude, when you have to resort to inventing "my" beliefs, then I *have* won the argument.

--S

Re:That's funny...

[BarryNorton]

And I mentioned what Java was built around exactly... where?

Nowhere... nor did I, that's the point!

Dude, when you have to resort to inventing "my" beliefs, then I *have* won the argument.

Same right back at you - I was only doing it back to demonstrate the point. This is what you'd done, therefore thanks for (implicitly) conceding (which you're now forced to do), talk to you some other time...

Some odd reason

[hsmith]

I don't see the Oracle solution being cheap... But who knows!

err wtf?!!

[Anonymous Coward]

I mean HELLL-O?!!

J2EE??!!!!

How does Eclipse fit in with this?

[kt0157]

Err.. Am I missing something? Can't this be done with the Eclipse platform?

K.

Re:How does Eclipse fit in with this?

[Dan-DAFC]

Maybe it would fit in on the development tool side of things if it weren't for the fact that Sun has nothing to do with Eclipse, it's IBM's baby. Even Sun's notoriously stupid marketing department wouldn't call a Sun product 'Eclipse'. And given that name, Sun are unlikely to ever get involved even if they wanted to (which they don't). Expect to see Sun pushing the improving NetBeans platform as part of this offering.

Re:How does Eclipse fit in with this?

[kahei]

I actually laughed out loud, but after about a second I realized you were serious.

I admit, it is a little unclear how a product can be a 'datacenter platform' and also a '.NET competitor', given that the two things hardly overlap at all.

I wouldn't be surprised if this was really a set of Java wrappers for Oracle Fusion, bundled up for Solaris 10 to create a one-step solution for getting a datacenter going. The weakness of such a product would, of course, be that it has Fusion :)

Re:How does Eclipse fit in with this?

[grimJester]

I admit, it is a little unclear how a product can be a 'datacenter platform' and also a '.NET competitor', given that the two things hardly overlap at all.

As I see it, they are simply going to bundle existing stuff to porovide a standardized platform. Sun hardware running Solaris, with Oracle databases and Oracle's Fusion app server running Java web apps.

.NET is likely mentioned as the competitor to Java. Oracle and Solaris are going to do their damndest to get big business to pick Java. Currently they don't mention any new products at all; all this is is a complete hardware / software package for heavy-duty web applications.

Proprietary

[trollable]

"According to Ellison, this is all about providing users and developers with technology based on standards. But what standards is he talking about, and are those the standards that consumers care about? The availability of an open source .NET implementation based on ECMA standards certainly makes Java look more proprietary."

The whole JDK1.5 API is public and totaly available to be implemented by anyone (www.jcp.org). Also there is already a 98%-complete implementation of it (www.classpath.org). OTOH, only a small part of .NET has been proposed to the ECMA, which is not even a standard organization. Mono provides only a small subset of .NET.

(that said, the most used Java Platform (Sun) is still proprietary)

Re:Proprietary

[slashk]

not a standard organization?

what is oasis?

i know a few people on the oasis committees and they never seemed very standard. in fact, they seemed a bit sub-standard.

Re:Proprietary

[Anonymous Coward]

You think JCP is a standard organization? JCP is owned and operated by Sun.

ECMA is a standards body. ISO is also a standards body. The C# language and the CLR are both standards under both standards bodies. Granted, the standardized CLR is not as extended as the .NET framework that is available from Microsoft, but it includes everything required to construct the foundation of the framework. Unstandardized portions include Windows Forms, Windows Management, COM+/MTS, ADO.NET and ASP.NET, which are fairly proprietary in general. Mono has their own extension libraries, including GTK#.

It is absolutely hilarious to see Oracle and Sun have to band together like this. When JDK 1.5 was released, while it was painfully obvious, Sun never bothered to mention that they were attempting to catch up to .NET (and if you think they weren't, you have some serious fucking denial problems.) If anything this is Sun and Oracle noticing that their babies have competitors with bite, enough so that the existing offerings aren't enough to compete directly.

Well, good luck to 'em both. Sun and Oracle both do create some good products, but largely they are little more than producers of rhetoric. This is especially true of Sun who has been sitting on the gold-mine that is Java for over a decade without a fucking clue what to do with it. Neither of these companies "get it," and I doubt they'll learn anything important from this soon-to-be-failed venture.

Re:Proprietary

[ichin4]

Mono provides only a small subset of .NET.

You are welcome to go one all you like about how you like the Java standards story better than the .NET standards story, but you really need to stop making this demonstrably false claim about Mono's API coverage.

As you can see here, Mono covers about 98% of the v 1.1 (Everett) framework, which is what most shops still use. This is comparable to the JDK 1.5 implementation you just touted!

And as you can see here, Mono already covers about 90% of the just-released v 2.0 (Whidbey) framework.

And of course, these statistics leave out all the LDAP, GTK, CORBA, and other Unix-centric APIs that are in Mono but not in the .NET Framework.

Re:Proprietary

[trollable]

I apologize. It seems they have progressed a lot since I checked. At that time, a lot was missing but more important, there was no intention to implement the part out of the standard. Things have changed and it is great. And you're right that the situations are comparable, in terms of implementation (according to the percentages reported, that may have a different signification).

Re:Proprietary

[trollable]

You think JCP is a standard organization? JCP is owned and operated by Sun. ECMA is a standards body. ISO is also a standards body.

No I don't think the JCP is a standard organization. But ISO is. There is a big difference between an auto-proclaimed association of companies and an official organization supported by governements.

As you wrote, a lot of .NET APIs are not "standardized". OTOH, the whole Java APIs are open -- but not their implementation and not their tests :(.
Concerning the changes in 1.5, I dislike most of them (in their current form). But you're probably right about the motivation

Concerning your last paragraph, I totaly agree.

Re:Proprietary

[trollable]

[ECMA] not a standard organization?

No, it is an association of companies. Same for OASIS (that also includes non-for-profit organizations). OTOH, ISO is the standard organization. By definition, standards are supported by governments, nationaly (national standard organization) or internationaly (by delegation).

Re:Proprietary

[finnif]

> OTOH, only a small part of .NET has been proposed to the ECMA, > which is not even a standard organization. Mono provides only a > small subset of .NET. ECMA is the organization that standardized Javascript, and is the organization that Firefox looks to for that standard. I'm not sure what else you have to do to be a "standards organization". Technicalities aside, why does this matter? Python is not a standardized language; nor is Ruby a standardized language. Yet both are mentioned on this site daily and are being used widely in production. With C# and .NET, you have the future direction of the #1 software company in the world. You have the adopted language and framework of a very large number of Fortune 500 companies and Wall Street firms. Microsoft/.NET is the new version of "you can't get fired by going with IBM," and IT'S ACTUALLY GOOD. That's the part that terrifies Oracle and Sun. In contrast, Java is a terribly disorganized. Developers rely on tons of third party tools to get anything done. The development environments aren't very good compared to VS 2005 (Eclipse, Netbeans -- though much improved with 5.0 -- fall flat in comparison). Bottom line is, if anyone really worries about .NET not being standardized, they're not in the .NET target market anyway, and they're not part of what Sun and Oracle are at all worried about. Good luck to Sun and Oracle, they slacked off on Java to the point where they have a long hard road ahead of them. ASP.NET/C#/CLR 2.0 and VS 2005 are awesome.

Re:Proprietary

[alexo]

> By definition, standards are supported by governments, nationaly (national standard organization) or internationaly (by delegation).

Which government "supports" TCP/IP?

Re:Proprietary

[ad0gg]

So when microsoft doesn't follow W3 specs, its because they aren't a standard organization?

Re:Proprietary

[ichin4]

Wow, I'm so impressed to actually have an intelligent give-and-take discussion on slashdot. Thanks!

As a sysadmin...

[Anonymous Coward]

As a sysadmin, I work with a plethora of applications, systems, integrators and vendors. We run everything: AS400, PHP, J2EE, linux, windows, perl, oracle, db2, postgres, mysql...I could go on, and on. Windows bashing aside, Java is the only technology that's "advanced" enough to break itself. I can literally run some of my perl scripts over and over until the cows come home...or leave my cisco routers up for 700 days...or reboot linux til I'm blue in the face and it's always predictable. When they fail, there's some reason: Disk space, upgraded software, user error, low memory, gamma rays, etc. Java is not that way - java has a mind of its own doesn't need an excuse to not work 1/1000 times.

My point here is that I feel for the people who will be administering this system - all of those sleepless nights troubleshooting transient failures with no fixes or even causes. Oh well, they made their bed, I suppose.

Re:As a sysadmin...

[SimplyBen]

I'd have to 100% disagree with you here. "java has a mind of its own" is an extremely ignorant statement. While the quality of many java applications is below acceptable, critizing the virtual machine and its related frameworks and apis from the perspective of a systems adminstrators is doing nothing but spreading FUD. Java has several advantages that the majority of other technology stacks lack. That advantage is choice. This being said it is a double edged sword. Don't like writing SQL? Use hibernate, toplink, iBatis, torque, OJB, castor or about 20 other functionaly similar technologies. The main problem imho with java is to write an end to end application you have to be proficient in a breadth of technology stacks. With this choice comes responsibility which in many cases leads to failure.

Re:As a sysadmin...

[Anonymous Coward]

Java is not that way - java has a mind of its own doesn't need an excuse to not work 1/1000 times.

I guess that's why so many financial institutions and government software systems [where failure is not an option] are Java EE-based...

Then again, as a systems administrator who likes writing Perl scripts, I'm sure your opinion on Java's stability is 100% valid.

Re:As a sysadmin...

[Anonymous Coward]

Could you give us an example of this?

Re:As a sysadmin...

[twiddlingbits]

Java by itself doesn't "BREAK". Applications that are poorly written BREAK. Some application crashes CAN crash the JVM and you lose all your Java apps but if you need 100% uptime there are ways to configure the JVM to deal better with the errors. You could also look into some of the newer operating systems such as Solaris 10(Containers, Predictive Self-Healing, DTrace tool), or virtualization. Modern Java servers such as Websphere/WebLogic can be setup where one flaky program only kills one instance of the JVM, all the other JVMs and thier programs continue to run just fine.

Re:As a sysadmin...

[Anonymous Coward]

I hear you brother. The crashes are a monster PITA too because its never too clear if its the app, the universe of frameworks, the VM or the OS. Complexity has a price.

Re:As a sysadmin...

[Anonymous Coward]

Some application crashes CAN crash the JVM and you lose all your Java apps

Wow, I didn't know that. That's piss poor.

Re:As a sysadmin...

[Anonymous Coward]

Some application crashes CAN crash the JVM and you lose all your Java apps...

I hope you're not implying that this is acceptable, since it would be roughly analogous to saying that some operating system (Win*, for example) being taken down by an application is acceptable. And we all know that's not acceptable here on /. ;)

Re:As a sysadmin...

[Anonymous Coward]

This is pretty much only going to happen if something goes nutty in JNI, at the OS level. .NET and everything else is just is much if not more vulnerable.

Re:As a sysadmin...

[Mindwarp]

It's not at all analogous. It's like saying "an extremely bad piece of Java code can cause the sandbox instance it's running in to quit. All the other sandboxes keep going just fine."

I've never once seen a piece of Java code running on an approved JVM do anything approaching the travesty of Win* being taken out by a piece of code.

Re:As a sysadmin...

[ZenShadow]

As an administrator who has run many Java apps, and a programmer who has written many Java apps, I will say this:

Java is a PITA in a production environment.

Can it be stable? Yes, it can. Or it can break in completely opaque ways that take weeks to troubleshoot. The problem is that which one you get is luck of the draw. 99% of this problem, IMO, is the garbage collection mechanism. Java's GC is horrible. .NET's GC is excellent. The difference is amazing.

Add to that the lack of simple invocation of java apps from a command line (eg. without OS-level hacks you can't just add them to your PATH), the fact that most professional Java applications seem to require reams of XML/Greek for configuration (ever used a Java app server? I've used several, and they all suck!), and other such issues, and you get an environment that most sysadmins absolutely DESPISE. .NET, on the other hand, behaves remarkably like a normal native executable, even under Linux (though you still need the OS hack, it's easier than Java's). It's perceptibly faster than Java for many applications. And there is no equivalent of JNI when you need to do things that the VM can't do -- you pretty much just do them.

The list goes on.

While I agree that many instances of poor performance / VM crashes / out of memory errors tend to be caused by the programmers writing bad Java, it says something that I've yet to see those same issues with C# -- and the quality of the programmers isn't that much higher.

--S

Re:As a sysadmin...

[Anonymous Coward]

"I guess that's why so many financial institutions and government software systems [where failure is not an option] are Java EE-based..."

Sure, like most government operations, government software systems never fail.

Re:As a sysadmin...

[Anonymous Coward]

The .NET GC is so poor that Microsoft server teams keep state in big arrays of structs rather than relying on it.

the truth finally comes out....

[Anonymous Coward]

that java is inferior to .NET.... Only took sun 5 years to acknowlage it...

imitation...

[Swamii]

...is the sincerest form of flattery.

Rather than teaming with Larry Elliscum, a better move for Sun would be to open Java up to the ECMA/ISO for standardization.

Re:imitation...

[nacturation]

Rather than teaming with Larry Elliscum, a better move for Sun would be to open Java up to the ECMA/ISO for standardization.

I'm sure that's on ISO's to-do list, but they're waiting to receive the standards documentation for PHP. :p

Re:imitation...

[Blakey Rat]

And an even better move would be to help the Java VM use native controls for its host OS, so Java programs on Macintosh stop looking like ugly hideous mutants that nobody wants to use.

Run a .net application alongside a non-.net application on Windows. Can you notice the difference, visually/behaviorally between the two? No? That's the POINT, and that's something Java totally missed.

Re:imitation...

[Anonymous Coward]

It's not that they missed, they simply didn't care about it. (Sun that is).

It's not a problem with the Java language. If you run a Java-GNOME app and a Mono/GTK# app in GNOME you'll get the same look and feel in both. :)

fix java or give it up to the community

[slashk]

i for one am sick of dealing with classpaths and 250 jars inside of jar files inside of war files inside of ear files - catch my drift.

i'm also sick of J2EE containers with class loaders schemes that are more complicated than my senior year algebraic structures course.

build a linker into java just like .net has, and something like a GAC.
than allow versioning of libraries.

then get rid of checked exceptions so i don't have to do try/catch/wrap/rethrows(or do nothing) in 90% of my J2EE code.

then get rid of stateful, local session beans - how redudant is that???

then find a way to get rid of the 14 million defines i need in my server.xml to specify which implementation of each 'open, standard' interface i need

so, java as a language - it's ok
java as a platform - SUCKS!!!!!!!!!!!!!!!!!!!!!!

left java for .net after 6 years of dealing with Sun's bullcrap and i have never looked back.

Re:fix java or give it up to the community

[Anonymous Coward]

Sorry you got modded a troll. I happen to agree with everything you said.

Re:fix java or give it up to the community

[slashk]

thanks. i was kind of surpised by that.
kept it mostly factual.

how does one get marked a troll anyway - somebody at slashdot or other readers?

Re:fix java or give it up to the community

[masklinn]

Once in a while you may get 5 mod points, you may use these mod points on any post in any thread (any article) with the only limitation that you may NOT participate (e.g. post non anymously) in a discussion where you used your mod points, nor may you moderate in a discussion where you moderated.

You also get mod points faster if you meta-mod a lot, I think.

While you're supposed to act fairly and impartially when moderating, not everyone does it and most /. users use the moderation system to push forward their take on the subject without considering the actual and factual value of the moderated post.

Since you were bashing java with actual arguments, you got modded as a troll by a few java fanbois passing by.

Re:fix java or give it up to the community

[ttfkam]

i for one am sick of dealing with classpaths and 250 jars inside of jar files inside of war files inside of ear files - catch my drift.

So unpack the jar files into a common directory and re-archive them all together. They're basically just renamed zip files.

War files are intended to be independent: that's what they're for. They are meant to be a drop-in web application piece.

i'm also sick of J2EE containers with class loaders schemes that are more complicated than my senior year algebraic structures course.

Fair enough. They are indeed complicated. I happen to believe that their raison d'etre, preventing App-A from redefining classes used by App-B, is a good thing. But you're right. It can be extremely difficult to wrap your head around.

build a linker into java just like .net has, and something like a GAC.

Nah, I'd prefer they standardized the native interfaces on something like GCJ's CNI. But we're in total agreement that JNI is not sufficient.

then allow versioning of libraries.

Can you be more specific? I think I know what you are intending, but I don't think it's a good idea. Principle of charity, I'll assume that I'm misunderstanding what you intend. Can you be more specific as to what you're intending here?

then get rid of checked exceptions so i don't have to do try/catch/wrap/rethrows(or do nothing) in 90% of my J2EE code.

Just extend from RuntimeException. Problem solved. Checked exceptions are good if and only if the calling code can actually do something about the situation. If that's not the case, extending from RuntimeException will give you the fast track out that you are looking for.

Just don't overdo it.

then get rid of stateful, local session beans - how redudant is that???

Not redundant at all. J2EE/EJB components work from interface definitions, not the bean code itself. Without this, you are calling the implementation code directly. Probably not the end of the world in most cases, but a distinct and non-redundant case nonetheless.

then find a way to get rid of the 14 million defines i need in my server.xml to specify which implementation of each 'open, standard' interface i need

Now you're talking app server (Tomcat, right?) configuration. This is hardly the fault of the language, the VM, or the spec.

so, java as a language - it's ok
java as a platform - SUCKS!!!!!!!!!!!!!!!!!!!!!!

left java for .net after 6 years of dealing with Sun's bullcrap and i have never looked back.

Just like capitalism, it's the worst choice. Well, except for all the others. .Net has its own warts.

You really should take a look at the EJB 3.0 spec. It really is a lot nicer.

Re:fix java or give it up to the community

[Doctor Faustus]

Just extend from RuntimeException. Problem solved.
Except that all the libraries throw regular, checked, Exceptions. You can catch them and rethrow a runtime exception, but it's still a nuisance.

Re:fix java or give it up to the community

[ttfkam]

Fair enough, but blame the lib author, not the language.

Re:fix java or give it up to the community

[Orlando]

so, java as a language - it's ok
java as a platform - SUCKS!!!!!!!!!!!!!!!!!!!!!!

I so agree. Java was designed and developed with programmers in mind, NOT sysadmins. It may be a great language to code, but my heart sinks every time I come across it wearing my sysadmin hat.

Re:fix java or give it up to the community

[Anonymous Coward]

Since you were bashing java with actual arguments, you got modded as a troll by a few java fanbois passing by.

Yeah, then he got moded up by an army of java-haters

Re:fix java or give it up to the community

[p3t0r]

There are a lot of nice solutions to the 'classpath' issues you describe. For instance take a look at maven2 with its transitive dependency management. You just tell maven 'hey, my project uses hibernate 3.1'. It will download the requested jars, and add them to the classpath. How much easier should it be?

Re:fix java or give it up to the community

[Anonymous Coward]

Sun is the author of both the language and the libraries, so just blame them. In any case, one can argue that it was a mistake to put checked exceptions in the language in the first place. No other language before or since Java has included that feature.

Re:fix java or give it up to the community

[Anonymous Coward]

It is a FANTASTIC feature and the main reason Java libraries are as widespread and robust as they are. If you are having problems with checked exceptions it means you have structured your code wrong.

Re:fix java or give it up to the community

[Anonymous Coward]

i for one am sick of dealing with classpaths and 250 jars inside of jar files inside of war files inside of ear files - catch my drift.
i'm also sick of J2EE containers with class loaders schemes that are more complicated than my senior year algebraic structures course.

They do what they're supposed to: allow you to deploy/undeploy independant apps on your appserver without a hitch. And they do that well. The problems are:

• it's needlessly complicated for appservers running just one app - which happens to be 95% of them.
• it's a pain to admin - a few tools allowing me to find/grep/cp in and out of the hierarchy and to tell me which class was loaded by what classloader and what else it loaded at runtime, without having to change code would go a long way.

build a linker into java just like .net has, and something like a GAC. than allow versioning of libraries.

Well, on unix you just wrap it in a shell script anyway. On mac you wrap gui apps in a bundle. So that's pretty much a windows-only problem. But you can buy commercial ones :-(

then get rid of checked exceptions so i don't have to do try/catch/wrap/rethrows(or do nothing) in 90% of my J2EE code.

NO f*cking way. I like my checked exceptions the way they are.
I like that if joe blow decides he needs to throw another exception, javac will tell me.
I like that it strongly encourages people to deal will error conditions instead of waiting for something to die later - or at least log something.
I like the way eclipse can generate all the boiler plate code for me anyway.
I like that I don't have to dig in the doc to know what can go wrong with a method.
Granted it's a bit of a pain, but we're not writing 50 lines shell scripts here.
And if one guy here thought about starting throwing RuntimeException, or it lesser cousin, some generic superclass, he'd better have earplugs.

then get rid of stateful, local session beans - how redudant is that???

Don't need them? Don't use them! Usually the problem is not that J2EE is badly designed (although there ARE issues), but that devs, more often than not pushed by management and so-called "architects" or "designers" start making everything an EJB.

then find a way to get rid of the 14 million defines i need in my server.xml to specify which implementation of each 'open, standard' interface i need

No argument there :-) I will point out that usually the best way to deal with all that XML is to generate it. Xdoclet, some XSLT...

so, java as a language - it's ok java as a platform - SUCKS!

It doesn't. What sucks it the way, willingly or not, people sometimes use it.

Re:fix java or give it up to the community

[Anonymous Coward]

Ada and C++ do. How else do you know what to try and recover from, instead of just coughing and dying?

Re:fix java or give it up to the community

[ttfkam]

When an error occurs, I personally would like to know if it was (a) a malformed URL, (b) a security (sandbox) violation, (c) a data I/O error, or (d) a null reference. But all of these are possibilities if I'm trying to programmatically grab a file from a server.

In this case, you have a few possibilities:

1) Check status codes returned from each function

2) Catch relevant exceptions

With unchecked exceptions, all you know programmatically is that an error occurred. Sure, you can take a look at the stack trace after the fact, but while the program is running, there's no way to differentiate and take appropriate action. For example, a security violation would mean that I'm connecting to a server I shouldn't be. A MalformedURLException means that the user entered invalid data. An I/O exception implies that there was a problem tranferring the data. A SocketTimeoutException means that the server never responded. How my program reacts to each of these situations can be completely different.

If, however, upstream of the raw I/O layer, you decide that any error should be treated the same, that's your business.

catch (Throwable t)
  - or -
catch (Exception e)

And you're done. Alternatively you could wrap it in a RuntimeException.

catch (Throwable t) { throw new RuntimeException(t); }

It's up to you. The important point here is that if the info simply isn't important to you, you can easily discard it. On the other hand, if that info isn't there in the first place -- if all exceptions are unchecked -- you can't retroactively enforce them; you can't react to info that wasn't already there.

The beauty of Java is that one way or another you must acknowledge possible error conditions. By wrapping them in a RuntimeException, you are still clearly aware that the errors are possible. You just don't think they're important in a particular case.

This is in stark contrast to C where folks regularly ignore error response codes, compiling cleanly at 4am, sometimes with disastrous results down the road at a user's workstation.

Given that humans are inherently flawed creatures that have been know to write inherently flawed code, better to err on the side of caution.

Re:fix java or give it up to the community

[Doctor Faustus]

Yes, I understand the logic behind checked exceptions, and I think it was an experiment that was worth trying. However, it just hasn't worked out. Checked exceptions force programmers to handle error correction when they're not yet ready to, so the code that's written tends to swallow errors invisibly. "catch (Exception e){}" has become as widespread in Java as "On Error Resume Next" was in Visual Basic.

Re:fix java or give it up to the community

[ttfkam]

That has not been my experience. Teams I've worked with both on a regular basis and only occasionally have at least one member that knows what he/she is doing, i.e., kicks the crap out the catch(Exception e) {} junior programmer.

Other than that, I've seen it in some open source projects. That failed to convince me because 80% of all open source project in any language suck. Most are 19-year-old first-timers learning the language.

From my point of view, checked exceptions were a qualified success. It wasn't a panacea, but I don't agree with you that we should throw the baby out with the bathwater. They simply became yet another programming case of "use when needed, don't when not."

Re:fix java or give it up to the community

[Doctor Faustus]

Change the checks to warnings, maybe?

Re:fix java or give it up to the community

[slashk]

BTW - EJB 3.0 now uses runtime exceptions!

Re:fix java or give it up to the community

[slashk]

something like that would make sense
i think this is what Bill Joy was trying to achieve when he insisted on putting checked exceptions into Java, AT THE LAST MINUTE.

you can argue that they can be used when makes sense, but if you are calling into a library that declares them, you have to deal with it, even though you might not be able to at the time.

the result, as pointed out earlier, is that 90% of programmers tend to do nothing - catch (Exception exc) {exc.printStackTrace();}

and as far as rethrowing it, they might not be in a position to rethrow it properly - e.g. JSP page

it was a neat idea, but one that has many, many issues.

Um...

[msimm]

No? Why would I be interested in another .NET lock-in project. Open would be news, but this just sounds like more crap to tag onto my tech budget that could be done with any number of existing technologies.

Re:Um...

[NewbieProgrammerMan]

I expect this really is semi-big news for those that may need extra ammo to get management to buy off on using Oracle & Sun products where they're appropriate. But I do agree that Sun and Oracle are definitely hoping to tag something onto a lot of people's tech budgets. :)

J2EE??!!!!

[Schnapple]

Oh but you see J2EE, Java, Eclipse, etc. - they're not obliterating .NET and Microsoft like Sun would have hoped. So instead of beefing up their offerings and maybe fixing whatever is keeping them from "taking down" Microsoft and .NET they're going to do something "new" - because otherwise, they'd have to explain why J2EE didn't do it.

Re:J2EE??!!!!

[jcaldwel]

J2EE only addresses server-side applications. Their biggest flaw was that the early specifications were needlessly complex. EJB 3.0 addresses that flaw perfectly, IMO.

No thanks

[Anonymous Coward]

So I'm supposed to trade a solution written by a company with a maniacal leader for a solution written by TWO companies with maniacal leaders? No thanks.

Which version?

[MosesJones]

The reason I ask is that most people seem to use the Java front-end or the HTML front-end. I haven't seen people use ActiveX at all with eBusiness Suite or Oracle database....

Just checking the install that I have... yup that just uses Java as well.

I can't find ActiveX anywhere on the various Oracle products I've got installed at the moment.

Re:Which version?

[BitterAndDrunk]

Yes, it's all java. I should have been more clear - the JRE 1.1.18 that Oracle Applications will download is triggered by an ActiveX control.

As is the entire AIM (Application Implementation Methodology) suite.
The eCommerce suite (CRM, iStore, iSupplier Portal, et al) avoids this issue entirely, as would an alternate method to download the JRE. But the "standard" implementation of Oracle Apps wil require opening the security settings wide like I said.

.NET in the data-centre....

[MosesJones]

So the world's largest database vendor is paring up with the world's largest big server provider as competition to Windows and .NET?

Sounds like Microsoft joining up with Dell to compete with Apple on the desktop.

Re:.NET in the data-centre....

[Redwin]

Sounds like Microsoft joining up with Dell to compete with Apple on the desktop.

Don't they already?

Re:.NET in the data-centre....

[jsight]

So the world's largest database vendor is paring up with the world's largest big server provider as competition to Windows and .NET?

Sun is not the "world's largest big server provider".

Re:Predictions

[twiddlingbits]

ROTLMAO..The last time Sun took on MS they WON. Or have you not been around long enough to remember the Java battle? The part I don't get is the Oracle Middleware, Sun has a whole set of Java Services that can function as middleware and even integrate with .NET to provide Web Services. This seems to be taking money away from Sun's software group.

There allways has to be balance in everything

[Anonymous Coward]

Even in software industry. If SUN feils somebody will do it somewhere, look what happeneds with Ruby on Rails? They remind me on famous swordman Miyamoto Musashi that is cutting great software gigants with its light framework.

To be honest, I expect that Sun and Oracle will do great thing but their strategy has to be very focused to practical solutions and usable product, not to their sales.

Re:There allways has to be balance in everything

[Decaff]

Even in software industry. If SUN feils somebody will do it somewhere, look what happeneds with Ruby on Rails?

Yes, and what has happened with Ruby on Rails? Actually, very little so far. A lot of hype and noise.

read my mind

[bcrowell]

Y'know, I was just saying to myself, "Self," I said to myself, "you really need an enterprise datacenter architecture that leverages middleware based on robust frameworks." Wow, they must have been reading my mind!

Re:read my mind

[g2devi]

I thought so too!

But then I realized that it would be useless without a solution that integrated the various business silos with a results driven, customer directed, strategy that maximized ROI. Few businesses have the right paradigm or the Google Juice to effect such re-engineering.

It is my recommendation that Sun and Google set up a task force to appoint a counsil that would set up a committee to scan for low hanging fruit that can act as a change catalyst to empower businesses to achieve TQM.

Oh yeah, BINGO!

OK

[Dunbal]

the production of a complete Java-centric enterprise datacenter architecture that leverages Solaris 10 and Oracle's Fusion middleware.

      I realize this is offtopic, but I wonder if anyone would have understood the meaning of the above sentence as little as 50 years ago... :)

Re:OK

[cyber-vandal]

Or even 20 years ago. In fact I barely understand it now.

Re:OK

[Anonymous Coward]

No, but if they tried to they would think that someone was trying to design a building where a bunch of accounting for a coffee based company would be done, and it leverages the power of the sun and a magic artifact made by an oracle.

J2EE... we need a change

[Gunark]

Lets hope this means they're going to do something about J2EE. Between Enterprise Java Beans and Java Server Faces, J2EE is a sordid mess right now.

Correction to above post

[BitterAndDrunk]

As respondents have posted, it's the JRE that uses this control scheme. The actual application is java based.

But AIM is still ActiveX so the original point halfway stands.

Competitive pricing?

[Anonymous Coward]

You mean free? Because .NET is free (and so are the tools to develop with it).

That's the whole problem many of these "competitors" have. They price themselves out of the game (*cough* QT *cough* Opera *cough*).

Re:Predictions

[slashk]

i wouldn't root for sun.

they've attempted man times with much fanfare to unleash some new technology onto the market - usually with a resounding dud (java chips, jini, jxta, java desktop to unseat MS)

their latest concept of utility computing is based on the assumption that the evolution of computing will mirror the evolution of power grids in civilized countries.

their open sourcing of their flagship product, solaris, smacks of desperation, when confronted with the linux/dell threat.

the new java enterprise system seems pretty weak - how many developers do you know who have downloaded the newly openned source code yet?

java as a platform is moving slower than a snail.
i met with once with one of the j2ee managers, and his explanation for their slow evolution of java was its 'industrial inertia'.
they shipped the J2EE/Java 2 combo before it was complete, and promise backwards compatibility back to Java 1.0, so the platform is being deprived of necessary upgrades in order to maintain compatibility with previous versions.

so, at this point, as sun's market valuation decreases, their relevance does too.
they are beginning to take desperate measures, and this may include seeding the market with inferior, but ridiculously cheap technology in order to fend off competition.
this is in leiu of high quality technology which requires more time and money than Sun has.

they were backed into a corner, and they failed to find a way out.
it's a shame, because they once had the time, money, marketshare to do anything they wanted.
they just kept playing the same ridiculous hand.

BTW - recall how Java was created.
Jim Gosling was going to quit Sun because he felt they lost their ability to build software, so Scott M. spun off JavaSoft.

Again, Sun lost it a long time ago.

Get your facts together

[Anonymous Coward]

So now Sun is taking on .NET and they're teaming up with Oracle for it ? What a load of nonsense. According to Sun themselves the whole partnership is almost entirely based on Oracle choosing Solaris 10 as their preferred platform. You can read more about that here.

IMO some "reporters" only read what they want to read. Sun already has Java and it has got quite a big foothold to last. Solaris 10 is also kicking some serious ass. Why on earth would they want to directly confront a company like MS when they can easily expand their own market and slowly strengthen their position ? IMVHO the big competitor for Sun is Linux at this time. Something clearly displayed when looking at Novell which almost immediatly started "OpenSuSE" after the release of OpenSolaris. Coincedence? I wonder...

This step has IMO nothing to do with .NET, and if you take the effort to skim the Sun news articles I'm sure you'd conclude the same. What about this: Linux with either MySQL or Postgres vs. Solaris 10 with Oracle, or MySQL/Postgres if you so prefer. And all based on almost the same price / options.

Utopia? Then why is Oracle also jumping on the "opening up some products" bandwagon ?

No, I don't think MS has much to worry, Sun is targeting another audience here.

Re:Get your facts together

[Anonymous Coward]

You really believe that Oracle will drop Linux?
LOL.

Re:Get your facts together

[Anonymous Coward]

Why so narrow black/white minded ?

Yes, I believe that they will if there's more money to be made. Just like I think they'll drop Solaris if there's more money to be made on Linux.

Re:Predictions

[Anonymous Coward]

Ugh, why is this modded flamebait. MS FUD is a fact.

This is how Sun should promote Java

[Anonymous Coward]

GNU Classpath, has completely implemented over 98% of JDK1.4 and is building steam. Most packages are now 100% complete, and momentum is building.

GNU Classpath, Essential Libraries for Java, is a GNU project to create free core class libraries for use with virtual machines and compilers for the java programming language.

More discussion is here.

Re:This is how Sun should promote Java

[Anonymous Coward]

GNU/Java still sucks donkey dick at this time. And if you're right about having more than 95% done then I think its going to be a major dissapointment.

Don't believe me? Then try to follow the Sun Java tutorial and try to compile one of their easier Java examples (no, "HelloWorld" doesn't count). You'll notice that in many cases it compiles cleanly, but when you try to execute it it'll barf with a lot of exception errors.

Being a Java newbie myself and running Debian this had me baffled for an hour, also because I made some manual modifications here and there. After trying to compile an exact example and seeing it crap out on me too I finally installed the Sun JDK for Linux and used that. Guess what? No problems what so ever.

So if you claim they got it done for over 95% now (I really wouldn't know the exact numbers) then I think the whole project totally stinks.

C# 3.0 is your God like the US-dollar!

[Anonymous Coward]

C#-3.0 is the SQL-killer embedded into C#-3.0.

I'm processing terabytes of Platinum Information Technologies with C#-3.0!!!

I've a RAID5 of 20 hardisks!!!, 500 GB each!!!, total 9 TB!!!

Re:C# 3.0 is your God like the US-dollar!

[slashk]

i kinda kicks ass, actually
i can't believe they actually added lambda functions to the language!
for those of you who know lisp/scheme/xml, you might recognize some of the syntax

and someone even hooked up WMI as a provider.
someone else hooked up WinForms as a provider as well :-0

and it sure beats EJB 3.0 Entity Beans

the alpha's a bit buggy though.

it's supposed to RTM with Orcas sometime late this year.

Oh darn...

[TrappedByMyself]

Java-centric...competitively priced and based on robust frameworks

I thought this was going to be an article about Spring, Hibernate, & PostgreSQL

Dot Java Programmers Wanted!

[Anonymous Coward]

Requirements
Qualifications:

5+ years Java, J2EE
4+ years Microsoft Dot Net
4+ years Sun Dot Java
3+ Internet Explorer Programming

** Attention to detail
** Likes to work on mulitple projects simultaneously
** Excellent communication skills (written, verbal and other)
** Must be able to work 50+ hours per week
** Up to 90% travel
** No benefits!

Re:Dot Java Programmers Wanted!

[Anonymous Coward]

Forgot some variation of 'team player'

Or has that one gone out of vogue?

Although the 'written, verbal, and other' is truely excellent. I'll probably add this to my 'fuck with the headhunter/hr drone messages'

Funny in many ways

[Anonymous Coward]

I was laughing since .NET still seems to be merely a J2EE clone (and Java a C# clone) so I thought all along MSFT was copying/parodying Oracle and Sun.

So does this make this new project a parody of a parody of J2EE while LAMP platforms contine to do the real work in the world?

Re:Funny in many ways

[Bert64]

Actually C# is more of a java clone, since java has been around a lot longer.

Why another alternative for .NET

[guruevi]

Why another alternative for .NET? We already have LOGO, Pascal and Basic.

Of course if you want to become more advanced than .NET, you could start using Cobol or Fortran

If you want to outsmart .NET developers, learn PHP, Python or Perl

If you want to outsmart AND build faster/better/less buggy/stable software with datacenter/airfare/military specs you should get to start on C, C++ or Java.

Re:Why another alternative for .NET

[renrutal]

"If you want to outsmart AND build faster/better/less buggy/stable software with datacenter/airfare/military specs you should get to start on C, C++ or Java."

Not wanting to sound like Captain Obvious, but some people really need to be reminded of this:

It's the coder who makes a faster/better/less buggy/stable software with datacenter/airfare/military specs, not the language.

A bad software comes from unskilled coders(and/or from projects with bad management), not from languages other than your favorite one.

Re:Why another alternative for .NET

[Anonymous Coward]

Exactly how is Java a better alternative than .NET to create stable applications??

I can't think of anything Java has over .NET/c# apart from being cross platform. Eventually when mono ships .NET will be more open than java anyway.

Re:Why another alternative for .NET

[aXis100]

I think C++'s lack of memory management and bad programming has lead to a shitload of buggy software. .Net manages that for you, so bad programming is less damaging.

Re:Why another alternative for .NET

[jma05]

Nah! He is just trying to tell everybody all the languages he has heard of. Too bad he does not understand the difference between a language and framework.

Re:Why another alternative for .NET

[jbplou]

You can't serously believe programming in C/C++ for the web would be less bugy than .NET. The protection from buffer overflows alone makes .Net less bugy than C/C++ for most web development.

Who needs Sun ONE, when you can have Sun ZERO!

[abelikoff]

Here we go again, two firms absolutely unfit to compete against Microsoft in their products' price/performance, one basically in its death throes, the other one continually losing the market to the SQL Server... What a great duo! I'm sure they will come up with a real .NET killer. At least this time (they sadly have been failing this task on several noble attempts).

Seriously, I doubt world needs yet another infrastructure (even assuming this one will be Java-based). The only thing that I see happening by this effort is further discrediting of the Java technology, which after all those years still cannot provide an enterprise infrastructure compelling to developers (the hodge-podge of Struts, Spring, JSF, etc. only makes things worse).

Last but not the least, let's go back to the basic question and ask ourselves "cui prodest?" I see what Sun has in it, yet I fail to see what benefits Oracle is supposed to realize for it. Oracle is not a player in this field. Speculation and rumors may fly high but it will be interesting to observe the real motives of Ellison in this case.

Mono is Already a Serious Alternative

[illectro]

Mono is a more complete environment than any open source version of Java has ever managed - Miguel et al have done great things - now even commercial products like imeem are using it for their data center and Unix ports.

Re:Mono is Already a Serious Alternative

[Anonymous Coward]

In case you haven't noticed, Sun's Java implementation is open source. It's just not "free software."

Re:Mono is Already a Serious Alternative

[Anonymous Coward]

Sun's J2SE implementation is 'Shared Source' like Microsoft's .NET stuff in Rotor: useless, as it does not give you the freedom to learn from the code, use, modify and redistribute it freely.

cheers,
dalibor topic

Re:Mono is Already a Serious Alternative

[Anonymous Coward]

*sarcasm*

Yes, Mono has duch a great WinForms implementation and is so legally safe for the next 10+ years that everybody should choose to build some enterprise applications on it. And don't forget to use mysql instead of postgres (I heard teaming up with SCO and "selling" InnoDB to Oracle will bring them millions)!

*/sarcasm*

Microsoft Alternative?

[Austin Milbarge]

Man, it seems everyone in the world is trying to make alternatives to Microsoft's wares. First off, the above article does not explain how Java is going to somehow provide an alternative to .NET. Java has been around for a good 10 years and I don't how the involvment of Oracle is going to magically revive the language. .NET has been around for less than half that, and already is taking over Java. Why?

Very few have managed to do is what Microsoft has done for years. Make software and software development easy to use! Ease of use for the computer user and just as importantly, ease of use for the developer. As a developer who has worked with C/C++ C#, VB and Java for years, I've found all but the Java language to have had a well designed and programmer friendly development environment. Java has tons of IDEs out there and they all pretty much suck.

Programmers are people and people are inherently lazy. Lazy in the sence that people don't want to go out of the way if they don't have too. In programming terms, a good IDE can make all the difference. Microsoft doesn't always create the best technology but they make it friendly and thats why they sell. I've been waiting years for alternatives to Word, Access, Visual Studio, and Windows and I haven't seen it yet. In my opinion, if Sun and Oracle are going to try and make a .NET, alternative they should be focusing on an easy to use and powerful development environment. The Java language is powerful enough. Now it's time for a useful IDE.

Re:Microsoft Alternative?

[simonsayz]

Haha you make me laugh.. vs.net is like being in the stone age. Take a look at intellij idea .. or even eclipse even though i prefer the first. http://www.martinfowler.com/bliki/PostIntelliJ.htm l Even Jbuilder6, years ago was a better IDE than vs.net 2005 is today... i hate every time i have to work with it.. it doesnt have half the features that most Java IDEs do.. and the ones it does have is poorly implemented compared to most Java IDEs.

Re:Microsoft Alternative?

[Anonymous Coward]

I don't know. Picking a language based on an IDE is probably one of the worst things you can do. There's a reason CS programs don't allow IDE's to begin with. I've seen too many 'developers' that can't even initialize a damn array without an IDE. When your software breaks, and you have no idea where to look because your IDE has loaded every single library and bloated all of your code, it's time to go back to the real way to debug: print out or echo statements in vi. At least then you might understand how your code works rather than depending on an IDE.

Re:Microsoft Alternative?

[Austin Milbarge]

> it's time to go back to the real way to debug: print out or echo statements in vi

ummm.. that was a great way to debug in 1987 but in 2006 it's better to let the machine help you then waste precious time with print statements. Also vi is not considered an editor most people want to spend time learning. It's powerful but it's not practical anymore. Take a look at visual slick edit.

Re:Microsoft Alternative?

[Austin Milbarge]

> Even Jbuilder6, years ago was a better IDE than vs.net 2005 is today.

JBuilder 6??? Are you talking about that "has been" company called Borland. Borland's technology was great for my 16bit DOS C/Pascal and assembly projects. Thats was about it. Lets be honest here. Visual Studio has been the #1 development environment for over 10 years. In the early 90's Microsoft created the most successful programming language ever. Visual Basic, from which all IDEs are now measured against. VB wasn't such a great language. In fact it was kinda weak. But it was easy to use, and thats spells power in my book. Visual Studio today is still easy to use, powerful built-in debugger, powerful built in help (MSDN) and fully expandable via add-ins. Besides, programming is hard enough without needing to have to learn how to program your editor. (ie. emacs LISP)

I firmly believe if VS was available for Linux tomorrow, you'd see a lot more software (and better software) written for Linux.

Re:Microsoft Alternative?

[ZenShadow]

I firmly believe if VS was available for Linux tomorrow, you'd see a lot more software (and better software) written for Linux.

Nail on the head, baby.

--S

Re:Microsoft Alternative?

[ClosedSource]

"There's a reason CS programs don't allow IDE's to begin with."

Is it because many CS professors learned to program before IDE's were available and haven't done any real-world work since then?

Re:Microsoft Alternative?

[Anonymous Coward]

Nine out of ten devs I've worked with at Microsoft use gvim or emacs, even though installing Visual Studio costs them $0. The editor is crippled, the debugger is a joke compared to WinDbg (which the public can also download), the integration works poorly with every source control system worth using (SourceSafe databases routinely shit themselves), and MSDN Library works perfectly well standalone.

Re:Microsoft Alternative?

[simonsayz]

>Visual Studio has been the #1 development environment for over 10 years. In the early 90's Microsoft created the most successful programming language ever. Visual Basic, from which all IDEs are now measured against. It's about time you visual studio people take a look at what is happening outside vs.net, by Java developer standards its primitive and haven't moved much forward. "Visual Studio today is still easy to use, powerful built-in debugger, powerful built in help (MSDN) and fully expandable via add-ins" Tools from the post intellij era have that and much more, done much better. Even simple things like curly bracket matching is done better in most tools i know. It's a shame that so many developers are satisfied with vs.net, it's ofcourse because they don't know any better, there is ofcourse the ones that came from the Java world, and thats why there is room for things like Resharper.. but thats just a patch and it should be integreated... it's a shame because if people is satisfied then things will not move very fast.. and i personaly wish i had a decent IDE for C# as i do for Java. >I firmly believe if VS was available for Linux tomorrow, you'd see a lot more software (and better software) written for Linux. Yes and with a post intellij era vs.net we would see even more!

Burn Out

[Ftizzle]

In astronomy class they said the sun wouldn't burn out for a few billion years. It looks like teacher was wrong. First of all .NET was the Java alternative. Sun has been inept at managing Java. .NET should never have been a serious alternative to Java. With .NET Microsoft looked at Java's shortcommings; and did a great job of answering those issues. All Sun needed to do was answer back in kind in the next release of Java. Yet here we are years later; and its only just now catching up. Why should we expect this press release to amount to anything from Sun? They've been promising to turn it around in the next version now for 10 years; but they keep getting farther behind. 10 years ago I told my Sun sun rep their boxes were too expensive to buy and upgrade compared to Linux. It's ten years later and I could have the same talk with my rep today. The only thing left in my racks with the sun logo are some monster 64 cpu machines running (big suprise) a massive Oracle database.

What, like Office 12 XML?

[WebMink]

Rather than teaming with Larry Elliscum, a better move for Sun would be to open Java up to the ECMA/ISO for standardization.

Why exactly would that help? Right now the Java standards are open to input from a wide range of voices, from individual developers through open source communities like Apache to corporations like Oracle and IBM. No voice has overall control, no-one can force through self-serving capabilities and everyone gets to use the specifications royalty free. All of them know their contributions can be implemented as open source yet that the market in which they operate can't be monopolised by any single company.

Sun started ECMA standardisation and then realised half-way through the process that it was going to produce the worst of all worlds; a rubber-stamp for the work Sun had done, with no input from any communities and a freezing of the specs by the ECMA dinosaur, combined with a loss of the ability to enforce the Java trademark and an inevitable embrace-and-extend by companies like Microsoft and IBM. Sun should have worked this out before starting with ECMA but fortunately realised in time and pulled out of the process. The result was the creation of the JCP and the most open, competitive software market the computer industry has yet seen.

Microsoft fully understands the PR value of ECMA and is cynically using it to rubber stamp it's Office 12 XML format to undermine the openness of OpenDocument. That action has done us the good service of showing us just how intellectually bankrupt ECMA actually is. What the Java platform needs is not the destruction ECMA would bring, but rather the further evolution of the JCP, which is working better than pretty much any standards body before it and is only hampered by the public perception of Sun control.

Re:What, like Office 12 XML?

[Anonymous Coward]

I'd disagree. As someone who's working on Kaffe and GNU Classpath, and after looking deeply into it decided not to join the JCP I'd say the JCP is hampered by many things:

a) it's not transparent

Unless one is on the EG of a JSR, or in bed with one of the companies on the JSR, there is no way to know what's going on on the majority of JSRs. I was recently looking for information about JSR277, which is one of the really badly needed features and highly touted as the way into the future last year. There is absolutely nothing anywhere on the JCP web site, or anywhere else that would let me know if that JSR is dead, alive, or just an inside joke by some bored dude who applied for a JSR, save from a half-secret site with an empty mailing listy archive. Unfortunately, most JSRs are run that way: closed doors, smoke filled rooms, NDAs, saussage making.

b) the results are not very useful

The JCP produces a spec, a RI and a test suite. In general, none of that is available to the public under sane licenses, contrary to, for example, W3C's output. I can use W3C's DOM implementation freely in my products, and run the compliance test suites freely, and even implement the spec freely. With the JCP, the test suites are not available in general, the results may not be published, the specs mandate a full implementation or imply death by software patents, the RIs are proprietary, etc, etc.

c) the quality of the specs is bad

A lot of the specs coming out of the JCP for the core JSRs are surprisingly sub par. JavaMail is a running joke, forcing projects like OpenExchange to use Sun-specific classes to get anything done. Swing specs have been miserable for years. In Java 1.5, it is still not documented what the subset of HTML 4.0 is that Swing actually implements. It's impossible to use parts of J2SE portably, because their specifications are of such embarassingly low quality and have been so for years.

d) lack of a useful patent covenant

Just as Microsoft does with Office XML 12, the JCP also plays lovely games with software patents: unless you are fully compatible with some NDA-covered test suites and rules, you don't get any of the benefits of the patent covenents. Contrast that to the ODF patent covenant, which allows you to implement the specs without fear of submarine patents.

e) bad specification strategy

The JCP tries to codevelop the spec, the RI, and the test suite, leading to a close coupling of the three. A better process would be to ratify a spec only if a second, independant implementations exists, and passes the test suites.

f) self-selecting club

With a few noteable exceptions, the JCP suffers from being a place where solely the people in violent agreement with the house rules get together. Rule number one of the JCP club is: you don't talk about the JCP club. The NDA-infested way JCP operates in practice makes it very unattractive for a lot of 'freedom-loving 'individual developers, and organisations like the FSF, Debian, Gentoo, and companies.

g) it's stuck in a rut

The last set of bugfixes to the JCP itself has been applied in March 2004, resulting in JCP 2.6. There is currently nothing that I am aware of underway to fix the above mentioned problems. It'll be two full years in March.

All that being said, the JCP has potential, in theory, to become a useful place for the wider Java developer community if the problems above were fixed, say in a new JCP 3.0. I'm not keeping my hopes high, though.

cheers,
dalibor topic

Re:What, like Office 12 XML?

[penguin-collective]

No voice has overall control, no-one can force through self-serving capabilities

You're dreaming; the entire JCP is there to "force through self-serving capabilities". Worse, because Sun's mandatory compatibility requirements, if you don't implement all their shit, you can't implement any of it. And the result is that there are no third party Java implementations--only Sun's implementations and its derivative.

and everyone gets to use the specifications royalty free.

"Royalty free" in that you don't have to pay money for it; you do have to give away your firstborn son, in effect, if you want to access Sun's Java specifications.

Microsoft fully understands the PR value of ECMA and is cynically using it to rubber stamp it's Office 12 XML format to undermine the openness of OpenDocument. That action has done us the good service of showing us just how intellectually bankrupt ECMA actually is.

ECMA requires disclosure of intellectual property claims, licensing requirements, and publication of the standard. Indeed, that doesn't make a standard an "open standard", but Sun has been afraid even to agree to those limited terms. The sad fact is that Sun refuses to comply even with the very limited degree of openness ECMA requires: Sun doesn't want to disclose their intellectual property claims on Java, or their licensing requirements, and they don't want the Java standard to be published without their restrictive license.

So, I'd be happy if Sun at least lived up to the limited level of Java standardization, but in reality, Sun Java is more proprietary than ECMA C# and Microsoft Office XML formats.

Re:What, like Office 12 XML?

[WebMink]

I'd disagree. As someone who's working on Kaffe and GNU Classpath, and after looking deeply into it decided not to join the JCP I'd say the JCP is hampered by many things:

We've discussed these things many times before, Dalibor, and as you know we largely agree. I'm not saying the JCP is ideal, just that I believe the path through the JCP is much more promising than the path through ECMA, which does not even permit membership by individuals and which, as Redmonk point out, exists to offer "a path which will minimise risk of changes to input specs ".

While there is (as usual) much to agree with in the points you're making, they do reflect a single view of the world, one in which there are no large carnivores, and some of the fixes you suggest would possibly remove freedom from the market at the same time as they appear to give it to the individual FOSS developer. Where we do agree for sure is there's a need for further evolution of the JCP governance, and personally I am more hopeful than you that it will happen.

SAP, not .NET

[jt2190]

Read the source (article), Luke!

According to the article linked to by arsdigita, this is not about .NET at all, but about SAP. It looks to me like Oracle is actively porting its middleware to Java in order to claim that they are easier to develop for and less proprietary than SAP's counterparts. Sun and Oracle will promote each other's non-competing products as a part of this deal.

Java frameworks: Which are good for web apps?

[Jesus_666]

Concerning that... I have a few questions and maybe someone here might want to help me by bashing/hyping something.

I'm a CS student (still doing the foundation courses) and one of my courses is a one-year software project. We have to design and implement a replacement for an online bibliography. As the CS department is somewhat Java-centric we have to do it with JSP (or pure servlets, if we dislike JSP for some reason). That by itself is not much of a problem, although Java might be a bit heavy for a site getting about twenty unique hits per day... What bugs me is that we're forced to use MySQL; alternative databases like Postgre are not allowed for some reason. If you want to tell me how exactly this is going to make my life worse, feel free to do so (this is the bashing part).
For example, does MySQL support a transaction log? I thought something like this might be useful in case Bad Things(TM) happen to the database (support for it wouldd also look nice on our feature list).

If you feel like hyping something, I'd appreciate it if you could enlighten me on frameworks which might be useful in the development of a web application. My main concern is reducing development time as that's the resource we have the least of.

Re:Java frameworks: Which are good for web apps?

[adolfojp]

You most likely know this already but here it goes.

Java's web presentation framework is JSF, Java Server Faces. It runs on top of standard JSP pages. It is quite similar to WebForms on ASP.NET and they even give you a "rather good" free IDE with a visual form designer. http://developers.sun.com/prodtech/javatools/jscre ator/index.jsp

To facilitate your database access try to use Object Relational Mapping. Ruby on rails guys get Active Record by default. You can try one of these: http://java-source.net/open-source/persistence

About MySQL. The best way to argue against using MySQL is by explaining to your "superiors" how the GPL works. A lot of them probably think that MySQL's open source licence does not require them to leave the code open. Even the JDBC connectors are GPL so no, you cannot cheat. Be caregull with MySQL null dates and always enforce your constraints at the database level and you should do fine.

Good luck and happy coding
Adolfo

Re:Java frameworks: Which are good for web apps?

[TrappedByMyself]

Learn Spring. The rest of the pieces will fall into place.

As for MySQL, I don't bother with it because of the license issues. Basically, pay for a license or GPL your code.

Re:Java frameworks: Which are good for web apps?

[Jesus_666]

Thanks, I'll take a look at it. Now that I see it mentioned on /. I remember that I've read an article about it in a computer mag a few months ago. *goes looking fot the mag*

Re:Java frameworks: Which are good for web apps?

[Jesus_666]

Actually, no, I didn't know anything about JSF or Object Relational Mapping. This is either going to be covered in the next couple weeks (we only have lectures throughout the first semester of the project), which is unlikely, or we're supposed to pick up the technical knowledge ourselves. Good thing there's Slashdot. ;)

By the way, the GPL is no point against MySQL, as we are supposed to opensource our app anyway. However, the fact that even the JDBC connectors are GPL'd will make picking a license much easier. This is good to know (and I should probably share this with the project newsgroup).

Thanks for your reply, I'll look into JSF and Object Relational Mapping.

Re:Java frameworks: Which are good for web apps?

[misleb]

MySQL does have transaction logging if you make sure to use the InnoDB table ISAM when you create the tables.

As for reducing development time, you may want to look into using Ruby on Rails. I know the Java fanboys here will throw a fit, but for the size of project you are talking about, Rails sounds like a perfect fit. You should be able to have something useful up in a matter of days if not hours. Just look at the of the demos and follow a couple tutorials and see what I mean. Java is NOT the way to go for reduced development time. Especially if you are a novice.

-matthew

Re:Java frameworks: Which are good for web apps?

[Jesus_666]

Unfortunately Java is a non-negotiatable requirement. But InnoDB is a good idea, thanks.

Re:Java frameworks: Which are good for web apps?

[Decaff]

Rails sounds like a perfect fit. You should be able to have something useful up in a matter of days if not hours.

Yes, but there are major issues. Having your data model determined in the database is one - it can potentially make portability between databases a problem (Rails is starting to deal with this, but it's support for portability is not good).

Almost all other object-relational mapping tools in other languages do things the right way - by at least allowing the definition of database tables from an object model, and providing a mapping layer than can isolate your code from changes in the schema. Java has good mature, fast ORMs - JDO, Hibernate etc. Of course, Rails abandons decades of experience in this matter and assumes it is doing things correctly.

Secondly, don't fall into the 'ruby is fast enough' trap. I have done that in the past with Cold Fusion and PHP, and have regretted it.

Java is NOT the way to go for reduced development time.

Anyone with experience of a good java JDO tool with reverse mapping and schema generation facilities (like Kodo) knows this statement is nonsense. You get reduced development time by knowing a language and its tools well. This is the case for Ruby on Rails. It is the case with Java, JDO and JSF or whatever.

Re:Java frameworks: Which are good for web apps?

[Anonymous Coward]

MySQL munges data rather than giving errors (truncating strings, ignoring trailing whitespace, range-limiting numbers). After every update, do a query to make sure what it stored matches what you specified. And given the developers' general attitudes, you might want to lock all your tables during each transaction rather than assume they managed to implement ACID correctly.

still dont like java

[synx]

After spending part of 2005 writing an app in Java, I must say, I don't really like it. I started out very excited about eclipse, and hibernate, and I wrote a huge pile of code, and got many things working quickly. But in the end, I didn't feel like I had much control. The ability to tune and analyze performance was not really there. Furthermore, the whole community of Java-heads seem to be "performance, not a problem". My problem domain involves large data sets (~ 1 GB in ram or more), so I really couldn't go with this general outlook. It's not that Java performance isn't a problem, most people doing Java tend to do heavily SQL/IO bound apps, so most of the advice I got or read was along the lines of "create an index" and so on. Getting information on what the VM is doing in Java is a little difficult, even though in theory its all there, and Sun should be providing really cool tools to see what your app is doing.

Of course I wasn't prepared to buy the very expensive performance tools, but if I had, maybe I'd feel better.

I'm moving onto C++/Python and erlang. See ya later Java.

Re:still dont like java

[Anonymous Coward]

OH, cmon, there are gazillions of profilers for JVM, including those Sun gives for free (jconsole). As far as commercial ones go, I can't recall any of them costing more than 50$. There are also a ton of "tune this, tune that" advices all over the web, including some of the best written technical articles about VM tuning available on Sun sites (which of 5 GCs to use, how to fine tune for ones needs etc). I'm sure that it will come as a surprise to you, but you are not a first person on a planet that had a need to fine tune a Java app.

Multiple Vendors is the PROBLEM

[ellisDtrails]

IT organizations don't choose .NET for any language advantage, or price. They choose it because it is from ONE VENDOR. The IDE, the server, the support, all from one vendor who has done a TREMENDOUS job of supporting developers despite all the FUD around their business tactics.
Oracle/Sun: its still two vendors, and it is STILL CONFUSING

It would have been better if they backed Mono.

[Ian_FBNS]

Seriously, I don't see this as anything more than Ellison's continued attempts to have a shot at Bill Gates - this has zero to do with technology, it's one man's grudge match gone awry. As for an alternative to .Net? How backwards can you get? Why not support Mono, and help build the best of all worlds - a fully cross platform implementation of .Net with incredibly rich development tools (VS2005) running across all the major platforms? I.

Re:It would have been better if they backed Mono.

[Anonymous Coward]

That's because you're stupid. Their announcement has nothing to do with .NET. All the .NET comparisons in the article were the musings of some random Ars Technica contributor.

Re:Predictions

[masklinn]

Sun can die for all I care, and allying with Oracle will probably help them down that path.

Not in my experience....

[Anonymous Coward]

I've been doing Java development for a couple of years now (and software development for 20+), and this hasn't been my experience.

We've got a couple of java data-collection and presentation programs that run basically forever. Ok, it took a while to get the last memory leak out, but now that that's done, they just run.

Once you track down any memory leaks, java pretty much does what you tell it (well, except for my repeated run-ins with custom class loaders).

-- ac at home

Re:Predictions

[ClosedSource]

I'd say Sun won the battle, but lost the war. Taking MS to task for making a Windows-optimized version of Java resulted in a big payday for Sun, but killed Java's chances on the Windows desktop.

Re:Mod down article.

[Anonymous Coward]

Well, he's kind of right, and maybe half insightful.
The article is biased for .NET vs. Java holywars, as
illustrated by the posts.

are they going to open the vm?

[Anonymous Coward]

if they want to combat the possibilities, .net provides, they'll need more than another oracle + j2sdk bundle. one way would be to *really* open the java vm and finance ports of other languages. java is a great language, but it is not general purpose. even microsoft has learned that pushing a framework and more than one language is the way to go. now seeing free ports of php for mono and others. do web-frontends in ruby on rails or php, embedded cronjobs (with access to certain direct interfaces) in perl or whatever while writing the main app in java - sounds good.

doing all this in java, where the same condition leads to different errors, any developer costs about the same money, sysadmins can't really do their automatization stuff for themselves, comes pretty late.

when the buzz comes down to reality, it's the vm, the developer pool using it and the enterprise support you can provide. in at least one point, sun+oracle won't catch up. the vm is very java centric atm - but still it's possible.

It's the developers! (Re:As a sysadmin... )

[MisterP]

I'm in the same boat. I look after everything under the sun. Everything from shitty little 2 server ASP websites to 20 server clusters with TB's of backend disk.

I have java servlets used by over 2000 people 24x7. When was the last time I had to restart the JVM? Dec 2002. I also have 8 java (jsp) web applications used by 200,000 ISP customers 24x7. JVM uptimes range from 2 years to several months. On the flipside, i have applications that need to be restarted every week.

The difference? The developers.

Re:It's the developers! (Re:As a sysadmin... )

[slashk]

i've seen java SEGV many a time, actually.
on revenue generating production applications - huge systems.

In once case it was crashing in some part of the JVM related to weak references, although the app wasn't using them.
In another case it was in a WebLogic JNI library.

An alternative .NET??

[Anonymous Coward]

How about starting with a language with properties, indexes and events.

Mainframe technologies move to the PC.

[CyricZ]

Come off it, look at the core .Net technologies (before they were re-branded): COM (implementing the multiple interfaces per object idea without multiple inheritance) predates Java, ODBC predates JDBC etc. etc.

And those familiar with 1970s mainframe or minicomputer technology would know of such concepts by different names.

Look at almost any VMS on VAX installation. You'll see interoperability between languages (BLISS, C, COBOL, FORTRAN, PL/1, plus many others) that worked quite well. Frederik Data Products offered a Smalltalk system in the mid 1980s that embodied many of the OO related traits of COM.

Not only that, but products like the DEC MDP database system offered the core concepts and benefits of ODBC far before ODBC did!

Not surprisingly, much leading edge technology was developed in the mainframe/minicomputer world. It was only later that it found its way onto lower-end servers, workstations and desktops.

Re:Mainframe technologies move to the PC.

[Anonymous Coward]

Unless DEC or VMS are releasing frameworks this year, being advertised as 'alternatives to .NET', you're post is irrelevant and yet more trolling.

No one is claiming Microsoft got their technologies ex nihilo, rather that many of them pre-date Java (even if re-branded and/or re-developed since).

alternative? FIX JAVA

[idlake]

OK, so here's what Sun needs to do in order to provide an "alternative" to .NET:

• add missing features to Java: multidimensional arrays, true generics, portable unsafe code
  
• remove bloat from Java, like JNI and many of the APIs
  
• remove compatibility requirements
  
• make Java an ISO and ECMA standard
  
• stop being so goddam arrogant about WORA and cross-platform compatibility: as a Linux developer, I want high quality Linux GUIs and Java GUIs suck on Linux
  
• open source the JRE and JVM under the LGPL, BSD, or Apache license (this wouldn't be necessary if Sun hadn't alienated open source developers, but they have)
  
• dedicate the patents Sun holds on Java to the public domain
  
• fire the people responsible for the current mess, McNealy and Schwartz, and stop lying about Java and open source
  

But, frankly, even if they do all that, it's too little too late. Sun is doomed anyway--they have proven conclusively that their management is incapable of adapting to changing environments. As for Java, Java has missed its opportunity: Java will be the new Pascal and the new Cobol rolled into one, which isn't bad, but it will never become the language of choice for more than those two niches.

Re:alternative? FIX JAVA

[Anonymous Coward]

Pascal still lives on - see Chrome, FreePascal, GNU Pascal, Delphi, Lazarus and more

Re:alternative? FIX JAVA

[penguin-collective]

Yes, and Java will "live on" just like that.

mono?

[monkeyboy87]

why not just pour some real money into mono and then beat them at their own game ? and all the while PHP continues to drive something like 40% of the globes websites why is microsoft even a consideration here ?

Mkay.

[AFCArchvile]

...the new platform is competitively priced...

We'll see. Consider that it's coming from the software and hardware company legendary for their high prices.

Java in Enterprise

[davcrock]

Java may be great to program in, but it cannot compete with .NET in a large environment. I spend the same amount of time screwing with fifteen different and conflicting version of the JRE for one program distributed to three users than five .NET programs distributed to one hundred users. Larger java webstart applications are the downright devil when they're cached into a user's windows profile and it's only recently that Sun instituted a buggy implementated system cache.

The dearth of JAVA ...

[willtsmith]

Java never turned out the way people said it would. It was supposed to be a cross platform web centric UI technology. As it turned out it did that job terribly. Now the entire purpose for Java seems to have been eliminated. Namely, there are really just two platforms left that have a real future, Windows and Linux.

The advantage that Oracle will gain by moving to Java interfaces is slower interfaces.

Both of these companies have similar problems and they are just teaming up to keep Microsoft from biting their heels off.

Re:Predictions

[twiddlingbits]

Really? I have it on my XP desktop right now, and on my Cell phone too. It's behind many many web sites, a hell of a lot more than .NET. Java makes it so easy to port from one platform to another that running Java on Windows is trivial. There are 1000's of Java apps that run on Windows. I think you need to re-examine the Sun-MS battle. MS was trying to pirate the Java langauge and make it run on NOTHING BUT Windows by adding proprieraty extensions as part of the "Standard". They are now trying the same trick in the EU with MS-Office.

Flamebait? /. is fading..

[msimm]

I can see where your coming from. We use Oracle on our database already. Thats got a pretty hefty price tag, especially as you scale up. We are weening ourselves OFF of sun hardware after about 6 years of it. Of course the punchline is the new Opteron systems look great.

But at the end of the day keeping our tech budget low means making compromises, sometimes the right ones, sometimes the wrong ones. The less closed-source software we are locked into the better. After all, the point is being profitable and we've been tied to one or two (or more) products already that didn't really do much good for us in reality (Coldfusion, MS Sql) compared to open alternatives.

Anyhow, long and short of it is I agree. Where their appropriate and both companies have produced (and continue to) excellent (although expensive) products.

Right now I just don't see the need for yet another development platform, much less a closed one. Not in my budget any time in the near future anyway.

Javas biggest coup

[willtsmith]

Javas biggest implementation will be application/menu/programming layers in the forthcoming Blue Ray DVD standard.

To .Nets features, you have an incredible amount of run time type and linking information that can do things with network enabled applications that were simply not possible before. Forget client/server, think about "clouds" of information logic that can be passed over networks and applications that are incredibly customizable on the fly.

you're dreaming

[penguin-collective]

Something clearly displayed when looking at Novell which almost immediatly started "OpenSuSE" after the release of OpenSolaris. Coincedence? I wonder...

The reason for OpenSuSE was Fedora Core, and the reason for Fedora Core was competition Debian and other free Linux distributions.

oh my

[penguin-collective]

When they're talking about a ".NET killer", they're actually talking about a suicide pact.

Oh for the love of FSM -- Scott & Larry, GROW

[CFD339]

Grow up boys, you're not Bill Gates. You're not going to be Bill Gates. That's not such a bad thing. People don't like Microsoft anyway.

QUIT running your companies like vengefull morons bent on doing everything with a goal of hurting Microsoft. Its getting predictable, boring, and less and less profitable.

Larry, you're not selling enough of your application suite to matter to anyone. Sorry to break that to you. Market share wise, you're still a database company and lately you're getting your clock cleaned on that front by MYSQL, MS SQL, and DB2 of all things. Why? Because you're so focused on your applications that nobody wants, that you make it impossible to do business with Oracle. You bought out Peoplesoft -- and trashed it. The folks I know in Munich can't stand what you've done to their work environment and projects. The problem, Larry, isn't Microsoft.

Scott.. WTF? Can't fine enough cool technology to give away this year? Your model is broken. Its been broken for years. The people who want non-microsoft servers are using Linux, not Solaris for new work. Its proving itself capable. Your hardware is too expensive for its performance -- its modeled after an old school "big blue" style sale with its service revenue coat-tails. How 70's and 80's. You've turned one of the most innovative and power companies in the world, into this decade's version of Silicon Graphics. Remember their last ditch effort? They came out with a high end graphics workstation with a non-standard embedded video rig that performed better but wasn't upgradeable. They were trying to compete with PC based workstations. Ooops. Now you're trying to compete with PC based servers without competing against IBM or Hitachi (et. al.). Ooops again.

Does anyone else but me remember the "Netscape-Sun-Aol Alliance" that was going to bring us J2EE based replacements for Windows servers and change all of us into J2EE freaks? Did ANYONE think an alliance like that could possible be stable? What are iPlanet sales like now?

The "Everyone but Bill" treehouse club is for 10 year olds. Its not for people running major corporations with a fiduciary responsibility to shareholders. Thank the FSM that I don't have any stock with you guys. You're like the poster children for why egomaniacs shouldn't be running public companies.

I want function destructors

[willtsmith]

Personally, i'd like to be able to specify a code block at the end of the function that will execute after a return statement. That way you only have to write cleanup code once.

Re:I want function destructors

[zjbs14]

You mean like:
  try { ...
          return;
          }
  finally { ... } ?

Re:I want function destructors

[willtsmith]

More like

if ( ... )
        { // do some stuff // cleanup local variables/resources
              return()
        {
else if (....)
        { // do some other stuff // cleanup local variables/resources
        }
else
        { // do some stuff // cleanup local variables/resources
              throw ( my error );
        }

I guess I would just like to be able to write

func()
{ // declare my local variables // allocate/open resource (like files/DBs, sockets, threads) // do my logic with multiple control paths and perhaps some try/catch statements

      cleanup
      { // clean up all my local variables and free resources
      }
}

If an anonymous delegate with closure can access the encapsulating scope, I think that would do the job though.

String func( someObject obj, anotherObject obj2 )
{
      int a;
      myComplexObject B;

      delegate A () Cleanup = {
                                                          B.freeResources(); // lots more code
                                                      };

      B.OpenSomeExpensiveResource();

      if ( obj.x == 5 )
      {
              Cleanup();
              return (obj2.Whatever);
      }
      else if (obj.bling == "sillystring" )
      {
              Cleanup();
              return (obj2.DerivedValue)
      }
      else
      {
              Cleanup();
              throw ( new SillyException( "oogaooga" ) );
      }

}

Re:I want function destructors

[ZenShadow]

try/finally is exactly what you want. The finally block will be executed regardless of whether or not an exception is thrown. To use your example:


String func( someObject obj, anotherObject obj2 ) {
      int a;
      myComplexObject B;

      try {
            B.OpenSomeExpensiveResource();

            if ( obj.x == 5 )
                            return (obj2.Whatever);
            else if (obj.bling == "sillystring" )
                            return (obj2.DerivedValue)
            else
                            throw ( new SillyException( "oogaooga" ) );
      } finally {
            B.Cleanup();
      }
}


More or less, anyway. The only wart is that (AFAIK) you can't access variables scoped to the try block from within the finally block. That would be a kickass option.

Maybe they should just add a finally clause to methods.


void someFunction() {
        byte[] memory = AllocLotsOfUnmanagedMemory();

// Do stuff...
} finally {
        FreeUnmanagedMemory(memory);
}


The scoping issues ARE kinda unnatural though.

--S (too lazy to format any better than that!)

Re:I want function destructors

[willtsmith]

Well yes, the entire purpose is so I can cleanup local resources ONCE.

Stop dissing Sun and Java !

[Anonymous Coward]

Since when did Microsoft become a bigger friend of Linux / Open source community than
Sun. Stop biting the hand that feeds you fools ! It was Sun that gave more code to
open source community, if it was not for them, you would have a pretty lame Networking
in Linux, along with many other things. Next time you pick on Sun, Why dont you show
your true colors by identifying who you work for, IBM, HP or yet another M$ sympathizer !

Your Stallman and Eric where just a bunch of weenies when Sun was duking it out
with MS for Unix !

Fools ! Slashdot and the Linux community has turned into a bunch of Bill Gates
fans !

This from an avid Java user and former Sun employee. The only fools who pick on
Sun work for IBM or HP or Microsoft. Hey, just incase you have not realised
Java is now the #1 language for Open source development.

Re:Stop dissing Sun and Java !

[ultrabot]

The only fools who pick on Sun work for IBM or HP or Microsoft.

I don't think so, many probably remember all the theatrics and attacks Sun has made against Linux and Open Source community. Many probably also understand that they need to do that to make money, but it still makes Sun an unreliable partner. IBM business model is just a better match for Open Source community, IBM is not desperate to sell something Open Source community provides for free.

Hey, just incase you have not realised
Java is now the #1 language for Open source development.

Yeah, right. Typical Linux distro won't be able to ship Sun's Java VM, and the others are not as good. The only Java program I use is Eclipse, the rest seem to revolve around the web server space.

This is a joke for Sun/Oracle

[Anonymous Coward]

It's obvious that these guys are running out of oxygen. That's all we need, another technology from Sun/Oracle. I thought that's what they have now with Java 1.5 and Oracle 10g.

Prediction: This will be a disaster for Sun/Oracle. Nobody want's their shit anymore, at least no-bleeding edge developers. All the hot language action has moved on to Python, and more specifically Ruby on Rails. And you've got the 2.0 release of .NET now, so the Microsoft crowd is busy moving to that.

This is Sun/Oracle trying to reinvent themselves, however, very few people are listening anymore. Wakeup Sun/Oracle, this is 2006, not 1996 anymore.

Re:Oh for the love of FSM -- Scott & Larry, GR

[Anonymous Coward]

Grow up, Andrew. You're not a software industry analyst. You're not going to be a software industry analyst. That's not a bad thing. People don't like software industry analysts anyway.

The partnership between Oracle and Sun has nothing to do with Microsoft. Sun is gaining by Oracle making Solaris their favored platform again (back from linux). Oracle is gaining by Sun endorsing their J2EE products even though some of Oracle's J2EE middleware competes directly with Sun's products. (To Sun, the fact that Oracle is using Java makes Oracle a better company than SAP.) The only way .NET got into this thread is through the (totally off) musings of the linked Ars Technica contributer.

Java is on its way out

[mythz]

Here are some features of c# 3

        Implicitly typed local variables, which permit the type of local variables to be inferred from the expressions used to initialize them.
        Extension methods, which make it possible to extend existing types and constructed types with additional methods.
        Lambda expressions, an evolution of anonymous methods that provides improved type inference and conversions to both delegate types and expression trees.
        Object initializers, which ease construction and initialization of objects.
        Anonymous types, which are tuple types automatically inferred and created from object initializers.
        Implicitly typed arrays, a form of array creation and initialization that infers the element type of the array from an array initializer.
        Query expressions, which provide a language integrated syntax for queries that is similar to relational and hierarchical query languages such as SQL and XQuery.
        Expression trees, which permit lambda expressions to be represented as data (expression trees) instead of as code (delegates).

Meanwhile sun is just getting around adding features like being able to turn off echo on the Console - While continue to ignore their own naming conventions. Java will still lack properties and indexers before c# 3.0 features are in full effect!

ugh

[bmajik]

i want to call your message a horrible troll, but it doesn't even make sense. trolls are usually a particuarly obnoxious, but plausible, point of view, presented as-fact with no supporting evidence.

nothing about what you're written is plausible.

The most ridiculous thing you wrote -- that really convinced me to respond -- was the bit about non-buggy code for reliable systems/aircraft being written in C/C++.

With this gem, you immediately disqualify yourself from any possible further serious consideration. Managed platforms like java and .net were invented precisely because of the difficulty of C/C++ environments. I'm not saying java/.net are showing up in aircraft (yet?), but C/C++ is NOT known for its non bugginess and stellar reliability :)

In any case, the one avionics system i have a dim knowledge of is the F-15 program, and my understanding is that it primarily used Ada... which you dont mention at all.

My last little lecture here, to try and set you straight:

people that take religous or ideological positions about technology ("C is the best language ever", "Fortan is better than C#") are hurting themselves and the industry. Any given technolgy has good and bad points about it, for a particular targeted use, implemented by a particular team/engineer, and deployed/supported for a specified time.

With programming languages and development environments, there is no hammer that makes every problem a nail.

Competitively priced?

[ben_1432]

Does that mean they'll pay people to use it? .NET Framework is free.

Demystifying class loading problems

[javaDragon]

Apparently you were caught in a class loading problem you didn't understand. You need to have a look at the article series Demystifying class loading problems, on the IBM DevelopersWorks site.

Too bad... just work with Mono

[Auri]

The only reason I don't like Java is the only language choice in Java. At least in .NET I can choose what I want to code in. Why not run with Mono? Like Sun having a monopoly on one programming language is any different than Microsoft? Bitter, party of one. Best, -Auri

Re:Too bad... just work with Mono

[Anonymous Coward]

Tcl, Lisp, Prolog, Eiffel, Sather, Smalltalk, COBOL, Ada, ECMAScript, and Forth aren't enough? Here's the full list of systems which support languages other than Java on the JVM; it's always been much longer than the list for CLR.

Bill Gates is lauging already

[Anonymous Coward]

Whatever they produce, it is doomed to fail. .NET will be a huge and collossal success, not because being initially better or by some marginal feature superior, but because it is included in Windows (in XP it's one click in Windows Update away, in Vista it will be included) that is still the best (performance+usability+simplicity+"not wasting time on configuration"+"geek help in case of trouble"+familiarity per price [of entire system w/hardware]) available operating system for general (corporate and Joe Sixpack) use and thus has and will continue to have the largest market share in foreseable future.

The main reason for .NET progressing faster than Java is that .NET brings revenue to Microsoft (it is a part of operating system that you have to pay for). Java doesn't generate money for Sun directly, so they are not as motivated as Microsoft to improving it for its full potential. They are more focused on improving its usability in server scenarios, but they dont't make a serious effort for it being a good general desktop applications platform (no, Swing is not a serious effort, it's just there so that they have something).

The desktop applications will be increasingly written in .NET. The server applications will follow. Why learning, installing, keeping up to date on, supporting, etc a new framework, if we already know .NET? While .NET may be behind now, it will catch up with Java and practically kill it. Even if it is slightly worse - Java would have to be dramatically better, because it doesn't benefit from being there by default.

The same problem will plague the new product:

• If it's free, it won't have the desktop part of the stack that would be usable to majority of potential customers (just like Linux isn't usable for tipical Joe or my grandma), because they will not be willing to commit significant resources to it (as for open sourcing it: OSS developers will be too busy playing with Linux kernel and "leaving its stabilization to distribution vendors" to bother with difficult testing and bug fixing work on this new thing).
• If it costs money then why paying for another thing that's similar to one already paid for (=included in a Windows license with 90% of world's desktops)? Don't mention Linux on desktop here, Linux on desktop is only good for certain specialized situations. Also there will not be development expertise: .NET will have plenty, because the basic development environment is included with the cost of new computer.

Note: I'm not bashing Linux - I have Linux on all my x86 computers and it's my primary OS at work. I'm just saying that it is a specialized platform for certain usage scenarios (I'm using it because I'm a geek).

Larry is completely defined by this

[b7j0c]

your post is dead on. too bad for larry he has let this issue completely define him. he's missed a lot of markets because he was obsessively chasing microsoft. in any case it seems every two or three years larry makes some claim to the meme du jour - corba, application service providers (asp), java, linux, soa, etc and tries to redefine the industry based on this so-called vision. but oracle has never been able to move the industry in this way no matter how hard they try, which is usually not very hard. larry has a critical dilemma - his core product is being rapidly commoditized and he has not been able to transform his company into something with greater future value. oracle isn't going away - those database installations will be paying his jet fuel bills until he dies, but oracle's days as a growth story are long gone.

Re:Predictions

[ZenShadow]

The battle never ended. You just got used to the sound of the cannons.

Microsoft was not trying to make Java proprietary; Microsoft was trying to make it integrate properly into Windows. If Sun had worked with them instead of against them, Java would probably be a very popular platform for desktop applications. Instead, it's gotten itself pidgeonholed on the server side.

Meanwhile, as a direct result of Sun's lack of cooperation, Microsoft decided to build their own wheel. .NET is a *direct* response to Java. And Microsoft took the time to do it right, which has a sinigifcant possibility of killing off new Java development in the long run. .NET is light years ahead of Java in terms of ease of development -- which is what matters to people who are responsible for making money.

Technologies that work well on the desktop do tend to find their way onto the server. Just look at Windows for proof of that.

--S

How about delivering today's products first?

[koreth]

Yeah, .NET alternative, whatever. Wake me up when I can actually run the current Oracle database release on the current hardware from Sun! We just bought a bunch of Sun's new dual-core AMD boxes and we have to install Linux on them to run Oracle 10g in 64-bit mode -- Solaris AMD64 support is supposed to come "in the first half of 2006." Kind of sad considering the way Sun and Oracle sing each other's praises all the time. Scott McNealy even spoke at an Oracle conference over a year ago about how Solaris 10 was the Oracle platform of choice! Um, yeah, right, if it actually worked it might be.

Re:How about delivering today's products first?

[Anonymous Coward]

I was a bit sceptical about the "platform of choice" thing when I had an Oracle installation on a Solaris box the other week... The only Sun compiler Oracle supports for Pro*C compilation is version 8, not 9, 10 or 11 (which is free now)... I guess they must come out every year or so, so if they're 3 years behind on their platform of choice, I hate to think how long certification for a non-choice platform would take!

Re:How about delivering today's products first?

[jbplou]

To be fair, Solaris on Sparc runs 64bit and that is considered the platform of choice. Sun makes boxes with over 100 processors using sparc.

slashbot morons

[kaffiene]

...out in force again, I see.

First of all, the spin in the original article is pure FUD - Oracle are claiming that Java gives them an advantage over SAP. The author's relating this to .NET is pure spite and FUD.

BIIIG suprise that the slashbot crowd fall for it hook line and sinker. Get a clue you morons! Java is now the leading language for Open Source software yet you keep bleating about Sun as if they're the anti christ. They're a company that have open sourced a tonne of their IP, they give their tools away for free they also allow anyone to build a Java stack and yet you malcontent wankers still bleat about Sun being evil.

Fuck the lot of you halfwit morons.

Re:Predictions

[ClosedSource]

"I have it on my XP desktop right now"

And perhaps you're one of the very few who actually run java applications on your desktop, most people don't.

"and my Cell phone too"

Which has nothing to do with the Windows desktop.

"It's behind many many web sites"

Which also has nothing to do with the Windows desktop.

"Java makes it so easy to port from one platform to another that running Java on Windows is trivial. There are 1000's of Java apps that run on Windows"

The question is not whether one can create Java apps that run on Windows but rather whether developers use Java when their product is specifically required to run on Windows. In most cases, platform-independence is not required. Windows customers would much rather have a responsive product that runs only on Windows than a slower product that could run on a platform they have no intention to use.

"MS was trying to pirate the Java langauge and make it run on NOTHING BUT Windows by adding proprieraty extensions as part of the "Standard"."

My speculation is that once Standard Java was established and was popular on many platforms including Windows, Sun planned to sell proprietary hardware to accelerate Java's performance to native speeds. MS's Windows-specific implementation undermined that plan by making Java faster on Windows without special hardware, so Sun decided to fight them.

Hello.

[Anonymous Coward]

Ehm...

$ cat > ~/bin/my-app
#!/bin/bash
exec java MyApp.class "$@"
^D
$ export PATH=$PATH:$HOME/bin
$ my-app

Now that wasn't so hard was it? And you call yourself an administrator. For shame!

Re:Hello.

[Anonymous Coward]

You're executing a Bash script from the command line. The point was, you cannot execute Java programs from the command line. And that is a limitation.

What crap

[mdm42]

From TFA: "Rather than emphasizing open standards -- an area where both companies fall significantly short of competitors..."

What a load of crap. Whilst Oracle may (or may not) fall foul of this accusation, Sun have probably the most long-standing and solid commitment to open standards of any tech company. Almost everything they've done has been based on open standards through their entire history.

Mod this article "crap".

"Technology Stack"

[zanderredux]

So... I see people ranting about how EJB are a failure and about Java being better than .NET and how Mono is not really .NET, only a small subset of it.

But what are exactly the components of the said "tech stack" that Java does not have? I mean, is there anything below the Sun (the celestial body, not the company) that Java does not have?!

Still....

[Debiant]

My point was that how people work.

Maybe VB has changed a lot, but it can be easier to management to send experienced VB programmaer to learn VB . NET than to hire altogether new one. It can be also more motivating to go forward with envoriment and programming syntax you're atleast familiar with.

Another important point is the spread of programming languages and easiness to get support in forms of sensible editors and ready enviroment. Programmers in fast few years or a decade, propably have learned C++ or JAVA. And lot of companies still hava lot of VB applications and code. Dot NET is also supported by MS so much that editors and are enviroment quite ready.

I've done enough with Java to know I like I, but I also say it's many times resource hug and not-so-elegant from a developores point of view. On the web front I'd wager my pets on Ruby, PHP or even .NET ASP gaining much speed, not JSP or Java application servers. On the desktop, I've not yet seen swarm of formidable Java graphical applications. Anyone who has done things in both fields, is not hard pressed to figure out why it would be so.

I'm not saying that everybody should do .NET, but I'm saying that Java has real life weakness where .NET is much better. I also think that denying that would not help anybody else than Microsoft.

So propably Ellison & Co are going to right direction. Another question of course is that will anything tangible come out of it?

And let's not go 'yes this language can do anything' arguments either, we bit experienced programmers propably know that almost anything can be done with most programming languages. It's eally besides the point, because real limitation is time and how much it is sensible to debug something to get it work.

There are also lot of programmers out there who don't control well the enviroment they work in. Those do or are just really dilligent, may also developed utter disintrest of debugging contiously the enviroment(I have). So lot of people do have a tendency to gravitate where everything is set ready for them, reason or another. So having support for something, doesn't tell is the support good enough so people will really want to use it.

Re:Still....

[Decaff]

On the web front I'd wager my pets on Ruby, PHP or even .NET ASP gaining much speed, not JSP or Java application servers.

Well the evidence is to the contrary. J2EE takeup continues to grow.

On the desktop, I've not yet seen swarm of formidable Java graphical applications. Anyone who has done things in both fields, is not hard pressed to figure out why it would be so.

You aren't likely to see a swarm of Java graphical applications, just like you aren't likely to see a swarm of WinForms applications. Virtually all client-side applications are for internal company use. Mind you, there are some Java graphical applications that are successful and getting quite well known (like Moneydance).

Re:Still....

[Decaff]

I'm not saying that everybody should do .NET, but I'm saying that Java has real life weakness where .NET is much better. I also think that denying that would not help anybody else than Microsoft.

But also wrong would be to deny that .NET has fundamental weaknesses where Java is much better. One is performance - in all high-performance tests JVMs generally do better than the CLR. The other, of course, (and this is a real show-stopper for many of us) is it's lack of cross-platform portability). Many of us have been through the single-platform single-vendor route before and will never risk it again.

they all are in the same range!

[sogod]

Look all above mentioned technologies are in the same range! I see Ruby and Ruby on Rails as the REAL ALTERNATIVE to both Java and .NET. It does not mean we stop using strong typing class of languages, but when you need to do the best you can in shortest period of time you have, you cannot overlook RoR.

Java way more open than ECMA or Microsoft

[Anonymous Coward]

Do you have an ax to grind or are you just clueless? Java is not "proprietary to Sun" as you're trying to spin it - each spec is owned by all it's contributors. Now, the JCP is not FSF-Free in its licensing, but you're smoking something if you think ECMA, or C#, or anything Microsoft does, is more open than Java.

Re:Java way more open than ECMA or Microsoft

[Anonymous Coward]

Do you have an ax to grind or are you just clueless? Java is not "proprietary to Sun" as you're trying to spin it

No, but you are clearly clueless: Sun controls Java. They control it through ownership of key specifications, patents, trademarks, conformance requirements, and the only J2SE and J2EE implementation in existence.

The fact that they weren't willing to undergo ECMA standardization even given ECMA's already low standards speaks for itself.

Now, the JCP is not FSF-Free in its licensing

Sun's Java implementation is not FSF-Free. The JCP is far worse and claims something rarely claimed before: control of the language and its specs by a private body.

but you're smoking something if you think ECMA, or C#, or anything Microsoft does, is more open than Java.

C# is definitely more open than Java. One can argue about whether it's open enough, but that doesn't change the fact that Java is another SCO waiting to happen, only that Sun's claims on Java and all public Java work as derivative works will have a lot more legal teeth than anything SCO ever had.

Re:Predictions

[twiddlingbits]

You MS FUD guys are SO Funny. Grow up and join the REAL IT world where portability IS important. Your application may need to support Windows, Unix, handhelds and other devices. You can't do that with .NET. Oh, and all the Java tools and software from Sun are FREE and soon a lot of it will be open source. MS would never do that. I can see you don't understand programming at all. The speed of the application is more commonly a factor of how badly or well the programmer wrote the code than the systems Architecture (J2EE or .NET). If he/she wrote the code to take advantage of certain features of the .NET app server and/or operating system (which makes it non-portable and hard to maintain) it will run faster than code that does not such as J2EE. SUN still owns the Java standard, but has NOT introduced a "Java accelerator" nor has anyone else. Java doesn't NEED it so I don't forsee anyone doing it. I'm tired of arguing with those who are just re-hasing MS FUD. This thread is dead, I have better ones to waste my time on.

Re:Predictions

[Anonymous Coward]

Microsoft added proprietary methods to java.lang.* classes in a blatant attempt to trick developers into writing unportable apps. They also dropped the standard Java support for RPC and native methods, which does nothing other than hinder interoperability with other VMs.

"competitively priced"

[Guspaz]

So, then, it's free, like Microsoft's .NET is?

Microsoft doesn't charge for the barebones C# compiler (csc.exe), last I checked (I think it comes with the runtimes and one of the service packs), nor did they charge anybody who wanted to download the .NET runtimes.

Also, Microsoft's Visual Studio 2005 express edition is currently free.

Re:"competitively priced"

[Random Guru 42]

The Visual Basic.NET compiler (vbc.exe) also rides along for free as part of the framework runtime.

For those who care.

Re:Predictions

[GISGEOLOGYGEEK]

You seem to have missed the boat (again).

Sun and Oracle are simply working together with their own FUD Response to the release of .NET2 and the new version of SQL Server.

Get with the game.

Perhaps if you weren't being such a spelling Nazi all the time, you would have noticed that.

Re:Predictions

[ClosedSource]

"Microsoft added proprietary methods to java.lang.* classes in a blatant attempt to trick developers into writing unportable apps"

Were there really any developers who believed they were writing a portable application using J++ only to find when they released their product, that some of their customers couldn't run it in on a Unix machine?

Generally speaking there were primarily two kinds of developers at the time J++ was introduced: Those targeting Windows exclusively and writing their applications using tools running on Windows and those who cared about cross-platform apps and developed them on Unix.

The Unix folks had no interest developing applications using MS tools and the Windows folks had no interest in developing applications that ran on Unix. There were exceptions of course, but this was generally true.

Re:Predictions

[ClosedSource]

"SUN still owns the Java standard, but has NOT introduced a "Java accelerator" nor has anyone else."

From Byte.com in November 1996:

"Sun is also dev eloping a more expensive (approximately $100) chip called ultraJava, which will be for desktop systems. Sun officials won't say whether or not ultraJava chips will use a picoJava core. However, these chips could include multimedia capabilities such as JPEG decompression and the graphics-processing optimizations now found in Sun's UltraSPARC RISC processors.

BYTE couldn't obtain actual silicon samples of Sun's Java chips at press time, so we don't know how well picoJava succeeds at boosting Java performance. According to Sun, these chips will run Java programs about 12 times faster than the same code executed by Sun's current Java interpreter."

Fast Swing development -- can vs does

[Latent Heat]

You argue that developers can pound out a Swing interface in a few seconds -- yes they can, but in practice does this work?

An old VB/Delphi/whatever Windows developer downloads NetBeans, fires up the GUI designer and asks, "this Java Swing stuff, how hard can this be?" So they click on a button widget, drag it to a form, and fire up the application. They get this big honkin button which takes up the entire form. "No problemo, I will simply drag this button over into a corner on the Form Designer." Fire up the program, and you still have a big honkin button taking up the entire form without any apparent way of changing its location.

You see, the VB/Delphi/whatever Windows school of GUI development is pixel-based and folks are used to using the Form Designer to place a widget on the form and then tweak locations and placement on a pixel level to satisfy their client/bosses/customers. Yeah, yeah, when you resize the form, now what? But for a lot of people, they do pixel-level layout using the form designer and kind of assume no one ever resizes the form. Also, a switch between large font/small font on Windows can jazz up a lot of Delphi widgets, but we will assume no one ever makes that switch either.

The Java Swing people know about layout managers, and layout managers solve all of the above problems -- font size changes, form resizes. But the layout manager world is a bizarro world to the form designer pixel-placement world. It is kind of like the gulf between WYSIWYG and TeX/LaTeX -- Word and its WYSIWYG cousins are like the pixel-placement world while the TeX world is related to layout managers. But just as TeX word processing is largely generic text editor plus document preview -- does anyone have a workable TeX/LaTeX WYSIWYG -- do layout manager GUI more properly belongs in the vi/emacs/pico and javac world -- does it even make sense to have a graphical form designer in the layout-manager world.

So to VB people, the .NET Form Designer is quite comfortable and familiar and Windows Forms is pixel-placement friendly while Swing/Netbeans is in the bizarro world of layout managers. I am not saying layout managers are bad -- I do a lot of word processing using LaTeX -- it is just that people who are not indoctrinated in them get frustrated real quick on Java Swing and return to what they are used to.

Re:Predictions

[ZenShadow]

Your application may need to support Windows, Unix, handhelds and other devices. You can't do that with .NET.

Uh, I do that on a daily basis with .NET. Where have you been? There's Mono for Unix, and there's this thing called the .NET Compact Framework for mobile devices...

Oh, and all the Java tools and software from Sun are FREE and soon a lot of it will be open source. MS would never do that.

Which is why they released the spec and let the Mono guys build their *own* open source version.

I'm tired of arguing with those who are just re-hasing MS FUD. This thread is dead, I have better ones to waste my time on.

I would politely suggest that, instead of wasting time being a zealot, you go get an actual job as a programmer. The real world tends to cure zealotry in all but the most hardcore cases.

--S

90%?

[ttfkam]

I would have to admit, far less than 90% of the Java code I've seen out there -- and I've been using it since v1.02 -- has catch (Exception e) {}.

I would even venture to say that the majority of code I've seen that I have written and has been written be others didn't do that.

Re:90%?

[slashk]

well, then you must work with good programmers
most of the junior programmers i've seen do so, and there are a lot of junior java programmers

what else can you do if you are implementing an interface that doesn't declare the exception you have to handle?
catch, wrap and rethrow as an unchecked exception.

and if you find this acceptable, other posters think that this practice is a sin.

in other words - there is little or no consensus on how to handle this.

hence, EJB 3.0 uses unchecked exceptions - whether people believe in it or not

Re:90%?

[slashk]

well, maybe not 90%, but i see it far too often

other than that, i see a lot of try/catch/wrap/rethrow with an unchecked exception

why does EJB 3.0 move to unchecked exceptions?
because checked exceptions cause too many issues for programmers.
really - doing this in EJB 3.0 is a major shift for them.
i'm sure it was a huge debate.

Re:90%?

[Anonymous Coward]

You catch the specific exceptions and wrap them in whichever of the interface's checked exception types best fits the concept "your request is valid but this implementation {has a bug|is misconfigured|failed unexpectedly}, try again {immediately|later|never}". If the interface doesn't provide for that, you fire its designer using unnecessary roughness.

.NET designers doesn't bother to think about what can go wrong because they settle for a platform on which anything can go wrong at any time. I haven't followed the EJB clusterfuck for a long time, but they're probably just competing for mindshare with the lusers who actually think that's normal. Quality is not the way to win in this industry.

Re:90%?

[ttfkam]

Let's be sure we're talking about the same thing.

catch (Exception e) {}

catch (IOException ioe) {
    throw new ExtendsFromRuntimeException(ioe);
}

catch (IOException ioe) { // do cleanup and log the error
}

These are three VERY different things. The first is a blind catch that gobbles any error without regard to applicability. The second is a declaration that we cannot intelligently handle the error where we are and are looking for a fast track to the bottom of the stack (such as for servlet code where an error occurs and we want to send the 50x error code and message without an intermediary. The third is the standard exception we all know and love/hate. The last two are completely valid depending upon the situation. If you can handle the error locally, handle it locally. If you can't, send it on the stack unwinding fasttrack.

My argument was that the last two were simply two options in your toolbox. The first one is just laziness and never acceptable.

My apologies if I was arguing with what you said if in fact we agreed. I read "unchecked exception" as "ignored exception." Unchecked exceptions are fine... in moderation and where they are appropriate. Just like checked exceptions.

Re:90%?

[slashk]

ha ha, well put! quality is NOT the way to win in this industry, this is certain.

Unsafe code is a feature?? and a desirable one?

[wheresjbob]

You're kidding about adding "portable unsafe code", right? That's what keeps the virus in business and why Windows is such a petri dish.

Why does the license matter for an alternative to .NET? If people who used .NET cared about open source, they would not be using .NET, so why should an alternative be open source. Sounds like an unsubstantiated religious rant to me.

I love how people get all "McCarthy" at Sun and Java over the license, while they continue to give M$ a hall pass on the same topic.

This is where the truth cannot be hidden within the BS because actions speak louder than words. The continued support and adoption of .NET means nothing more than "quality, security, and freedom do not matter".

• If quality and stability mattered, people would not use Windows or .NET.
• If security and viruses mattered, people would not use .NET.
• If Openness, Freedom, and Cost mattered, people would not be using Windows and .NET.

I do not and never will use Windows or .NET