Sun revealed a plan Tuesday to reorganize its Java middleware and Web-services software into packages to better compete against middleware vendors such as IBM and BEA Systems Inc. Sun will break up its stack of Java middleware and Web services into five suites priced at $50 per year, per employee.

The new suites being sold at the lower $50-per-year, per-employee price include:

• Sun Java Application Platform Suite, a set of integrated tools for designing, developing, and deploying applications under a services-oriented architecture. It includes Sun Java System Application Server Enterprise Edition and the Sun Java Studio Enterprise development tools. The tools range from Java starter tools, gained through Sun's acquisition of NetBeans, up through an integrated development environment, acquired when it bought Forté Software. Also included in the suite are Sun Java System Web Server, Sun Java System Portal Server, Portal SRA, and Portal Mobile Access.
• Sun Communications Suite, a set of tools and services
  clustered around Sun Java System Messaging Server, an implementation of the Java Community Process standard Java Messaging Services. It includes the development tools in the application platform suite, Sun Java System Calendar Server, and Sun Java System Instant Messaging, which are synchronized with Microsoft Outlook thanks to the recent
  Sun/Microsoft collaboration.
• Sun Availability Suite, a set of software aimed at improving service levels and management of critical applications. It includes Sun's development tools, Sun Cluster load-balancing and failover software for multiple Sun servers, and Sun Agents software for monitoring application performance.
• Sun Identity Management Suite, designed to identify users and assign privilege levels to many users of a system. It includes Sun Java System Identity Manager, Sun Java System Access Manager, Sun Java System Directory Server Enterprise Edition, and development tools.
• Sun Java Web Infrastructure Suite, a set of software
  optimized for Advanced Micro Devices Inc.'s Opteron processor to provide security to existing Web applications. It includes Sun Java System Web Server, Sun Java System Directory Server Enterprise Edition, Sun Java System Access Manager, Sun Java System Web Proxy Server, Sun Java System Application Server Standard Edition, plus development
  tools.

At the same time it unveiled its new software strategy, Sun revealed plans to provide grid-computing services for $1 per CPU-hour and data storage for $1 per gigabyte, per month. To support the initiative, Sun is opening data centers in Texas, Virginia, New Jersey, Canada, and Scotland.

Read more: Sun Launches Java Middleware Suites

I have watched the middleware grew and noticed some troubling trends. The great thing about these products was that you could build architectures using many different products. They were like toy blocks. I could get this guys web server, someone else’s application server, a database here, etc. You could put them all together and build your masterpiece.

Now these markets are dominated by company’s that sell all of the blocks. They want you to buy all of THEIR blocks. If that is the direction, what’s the point of open standards?

This is a smart move for Sun. IBM, BEA, Oracle, they all do this. However, is it good for us?

They want you to buy all of THEIR blocks. If that is the direction, what’s the point of open standards?This is a smart move for Sun. IBM, BEA, Oracle, they all do this. However, is it good for us?

The importance of Open Standards is a great as ever - if you care about having a choice you must insist that your vendor of choice support Open Standards. It should be the customer's choice as to whether they take advantage of a well integrated stack OR choose to plug and play their chosen stack components.

But insisting your vendor support Open Standards isn't enough - the customer needs to take some responsibilty to not get locked in (unless they desire it) - everytime a proprietary API is used the customer is potentially locking *themselves* in.

Rich Sharples
Sun Microsystems Inc.

I think that you have missed my point. I am not against open standards. One advantage to them is that you don't get locked in. However, these large companies are slowly moving use toward single vendor solutions.

Once you never heard of a company that had only one vendor's middleware. There was Oracle database, BEA app server, etc, etc. Now, we're moving toward single vendor solutions. Not because of lock in, but because of bundling. This is great for Sun, BEA, Oracle, IBM. I don't think that it's great for the industry. They'll tell you great.

From what I have seen, you have one gem and you bundle your junk in with it. People want the gem, so the buy it and use the junk because it's already paid for.

If memory serves me right, "Middleware was history", less than a year ago according to Sun (http://www.sun.com/br/0304_ezine/hc_bcbs.html), and now Sun is launching "middleware" suites.

OK, maybe we should give them the benefit of doubt and say that what they really meant was "Middleware was 'shaping' history".

"A slip of the tongue is no fault of the mind." - Anon.

If you read the article (and not just the headline grabbing title) :

http://www.sun.com/aboutsun/media/features/jdj_schwartz.html

IMHO all he's asserting is that businesses would be better investing their time in their core-business and not in integrating 'middleware' - that doesn't seem unreasonable.

All stack vendors are trying to ensure that there is benefit in the whole stack. I think it is perfectly reasonable to 'lock' people into your stack because it is more than the sum of the parts - ie. it all works really well together.

Rich Sharples
Sun Microsystems Inc.

I've read the article - thank you very much.

There is an inevitable symbiotic relationship between "running your business" and "middleware" and one cannot treat them orthogonally. To do so would amount to being in denial of reality. For example, this is apparent from the increase in consolidation of enterprises arising from mergers and acquisitions. Rarely (perhaps never) will the constituent companies involved in these transactions will have identical software stacks. Without middleware, integration of these diaparate stacks would be a nightmare.

Without middleware, integration of these diaparate stacks would be a nightmare.

No, without standards (and vendor's commitment to them) and standards based middleware - integration of these stacks will continue to be high risk and expensive - often uneconomically so. IMO, and I'm not implying that this is your position, investing in often proprietary or 'de-facto standard' middleware to solve the existing 'plumbing' problems will never be a winning strategy.

I've seen this obvious mistake too many times - solve yesterday's propretary-lock-in mistakes of the past with another.

Rich Sharples
Sun Microsystems

If you read the article (and not just the headline grabbing title) :http://www.sun.com/aboutsun/media/features/jdj_schwartz.html

IMHO, the best response to that drivel was published by Steve Vinoski of IONA.
http://dsonline.computer.org/0409/d/w5towp.htm

For crying out loud...don't they get it yet?

The feeling I'm having is like watching the Death Star creeping around Endor to get within range of the Forest Moon. The only difference is that the Falcon (Sun) never makes it to the power generator, and instead slams into the side of the shaft and explodes. Eventually the Death Star (Microsoft) ambles its way (over the course of 4 years) into range and blasts into stardust all hope for the Ewoks (enterprise Java) and what's left of the Rebellion (open source Java).

LOSING THE ENTERPRISE

Much to my disappointment, Sun is already losing Java in the Enterprise to Microsoft .NET. Does anyone really disagree here? Maybe it's localized, but where I live, enterprise Microsoft projects outnumber enterprise Java projects by at least 2 to 1.

I'm really not quite sure what we can do about this now.

LOSING SIMPLICITY

Now what Sun may or may not realize is that they are losing ground on both ends.

Scripty languages like Perl, Python, PHP and Ruby will slowly erode Java's educational, small business and simple solution market. The only reason .NET doesn't have a chance in this space is COST. Java is as free as any solution out there. It's just not widely regarded as being open or simple --WHICH IT VERY WELL CAN BE.

LOSING JAVA

Did anyone notice the simple fact that most of the "Sun Java System" products have nothing to do with Java! They are only resulting in complexity and ambiguity being tied to the Java name. What is Java anyway? Is it a directory server? A messaging server? An instant messenger? A Portal? An "Availability" Server? An Identity Server? An Application Server? A System? Is Java == Sun?

Last time I checked, Java was a language. It was a free, relatively "open", simple language. It was one that could have easily dominated the internet, mobile, enterprise and small business markets alike. And for a brief moment in history, it did.

Sun has had 10 years to make this work. It's not working.

Sun is not Java, and Java is not Sun.

Let's take it...

Clinton

Java is an intelctual property of Sun which
all licensees and users must RESPECT.

Thanks for pointing out the root of the problem more clearly than I did.

Cheers,
Clinton

Java is still doing well, but not SUN products.
SUN always package their "poor quality" (at least in public impression) Java products into a new bundle.
They should not confuse customers in this way when none of their products has no brand.

No Doubt SUN is not doing well.
But everybody recognise that Sun created java and Java is feeding IBM,BEA,Oracle and so many other vendors.

SUN is still recognised as very good technical company and very poor marketing company atleast for Java Products.It is the only company which is holding Microsoft by neck and for all real time critical enterprise projects Java is far ahead of MS.net

Much to my disappointment, Sun is already losing Java in the Enterprise to Microsoft .NET. Does anyone really disagree here? Maybe it's localized, but where I live, enterprise Microsoft projects outnumber enterprise Java projects by at least 2 to 1.

Clinton, I have the opposite sensation. At least in Europe, I hardly see any big .NET enterprise project, there is simply no demand for it. Students study Java and Linux at university CS courses, it is difficult to hire .NET specialists. Sometimes happens that a customer wants to migrate its web apps from ASP to ASP.NET, but many of them are running Java on their brand new Windows 2003 servers for their enterprise projects. Of course I am biased because I work for a Java-centric company, but I still have to hear a company running an entire EAI project on .NET

Sun is not Java, and Java is not Sun. I agree, in fact I think IBM is investing on Java at least ten times the money than Sun. Sun was clearly loosing the application server market, in the enterprise it is perceived more as an hardware vendor. I think this move is for Sun the only logical one, they still have the brand to compete in the software market and to recover a lot of terrain in the enterprise.

cheers,
Maurizio

Much to my disappointment, Sun is already losing Java in the Enterprise to Microsoft .NET. Does anyone really disagree here?

Well yes, I do. I feel qualified to comment on this since the bulk of my developer career was been spent working with MS technologies and I got pretty good first at the Windows API and later at VB/ASP. I've also learnt a certain amount of .NET but I took the decision to switch my company to Java/J2EE as our platform of choice about 5 years ago when I realised that Microsoft were about to make my entire codebase redundant for the 3rd time in 10 years, and I haven't looked back since. Although he reached a slightly different conclusion to me ex-Excel VBA lead Joel made a not dissimilar point in this rather good article:

<ahref="http://www.joelonsoftware.com/articles/APIWar.html">Joel on Software</a>

It's a long quote but worth sharing I think

"…And if you're developing a Windows GUI app today using Microsoft's "official" latest-and-greatest Windows programming environment, WinForms, you're going to have to start over again in two years to support Longhorn and Avalon. Which explains why WinForms is completely stillborn. Hope you haven't invested too much in it. Jon Udell found a slide from Microsoft labelled "How Do I Pick Between Windows Forms and Avalon?" and asks, "Why do I have to pick between Windows Forms and Avalon?" A good question, and one to which he finds no great answer.
So you've got the Windows API, you've got VB, and now you've got .NET, in several language flavors, and don't get too attached to any of that, because we're making Avalon, you see, which will only run on the newest Microsoft operating system, which nobody will have for a loooong time. And personally I still haven't had time to learn .NET very deeply, and we haven't ported Fog Creek's two applications from classic ASP and Visual Basic 6.0 to .NET because there's no return on investment for us. None. It's just Fire and Motion as far as I'm concerned: Microsoft would love for me to stop adding new features to our bug tracking software and content management software and instead waste a few months porting it to another programming environment, something which will not benefit a single customer and therefore will not gain us one additional sale, and therefore which is a complete waste of several months, which is great for Microsoft, because they have content management software and bug tracking software, too, so they'd like nothing better than for me to waste time spinning cycles catching up with the flavor du jour, and then waste another year or two doing an Avalon version, too, while they add features to their own competitive software. Riiiight."

Although I understand the reasons for it I think that .NET was actually a very big mistake for Redmond. They acknowledged that security was a major issue what 5 years ago now and still haven't fixed any of the fundamental security flaws in Windows (many of which were introduced 94-95 to improve the performance of NT4. The performance gains were superseded by the processor speed gains within 2 years) The worst of these decisions was probably moving the GDI to kernel space. This means that your OS security is entirely dependent on a third party screen driver (so if you are an evil genius looking to take over the world you could do a lot worse then getting a job as a coder at nVidia). It also means that the OS stability is dependent on a 3rd party graphics driver. And if the GDI crashes (and traditionally this is the least stable bit of the operating system) you have to reboot the server – whilst even in Linux you can connect to it from another machine and restart the GDI.

It was obvious 6 years ago that enterprise customers were looking for a low cost alternative to Solaris/Aix et al. Gartner and the other analysts were predicting that Microsoft would rule the datacentre by 2005. In order to do this they needed two things: A great development platform and a great, secure, easily scalable OS. They had neither. The odd thing was instead of working on Windows and embracing Java which could have got them to market 2002ish they spent huge amounts of time and resources copying Java and still haven't sorted the OS problems out. So now Linux has quietly established itself as the next OS of choice in the datacentre and Java is almost always the development platform of choice here (Mono – academic crap anyone?). Leaving MS fan boys telling us how great Avalon/Longhorn will be whilst .NET itself looks increasingly lame and increasingly confused (what happened to the hundreds of .NET "My Services" that MS were promising for instance – and remember when Windows 2x was going to Windows .NET?). And most of the movement I see between the Java and MS worlds is from MS towards Java/Unix not the other way around.

I sometimes thing that because MS destroyed Netscape so completely people imagine that they can do this in any market. In actual fact they can't – they've always struggled in the Enterprise (think MSMQ vs MQSeries for example) even when they were first (think COM vs CORBA for instance). And until they grasp the basic and Windows flaws I think this will continue.

I sometimes think that because MS destroyed Netscape so completely people imagine that they can do this in any market. In actual fact they can't..
and most of the movement I see between the Java and MS worlds is from MS towards Java/Unix

Here is some statistic from www.it.jobserve.com,
Available jobs for search terms Java, C#, .NET and J2EE

2003-04-03
Java=1478 C#=282 .Net=635 J2EE=561

2004-01-10
Java=2276 C#=586 .Net=1006 J2EE=1105

2005-01-19
Java=3614 C#=1515 .Net= 2073 J2EE=1635

old Java maintainance jobs is included…

More than 120M copies of the .NET Framework were downloaded and installed.
More than 85% of new consumer PCs is sold with the framework installed.
More than 58% of business PC had the .NET Framework preinstalled or preloaded
http://www.theserverside.net/news/thread.tss?thread_id=31153

COM vs CORBA?
I can not see one single instance of CORBA in Transaction Processing Performance Council (TPC) tests. In Top Ten TPC-C by Price/Performance all ten top results use COM+ even the ones with Linux.

Joel Spolsky.
A demagog who only says whatever the masses want to hear:

Joel Spolsky: "and Sharepoint... Sharepoint! which nobody has!"
Au contraire, mon ami. According to public comments made by Steve Ballmer, SharePoint is on track to be a $400M product for Microsoft, and one of the fastest products ever to get to that point for the company.
which product should you be honing your skills on?
http://www.joemarini.com/articles/sharepoint20041115.php

Security?
Choose Windows 2003 advanced server, Linux is currenlty reporting 35 security breaches per week.

Jan 2004 The truth about the Linux vs. Windows level of security
http://www.zone-h.org/en/winvslinux2
Zone-H- IT Security Information Network

March 30, 2004
Linux vs. Windows: Which Is More Secure?
http://www.eweek.com/article2/0,1759,1557459,00.asp
Forrester Technology research and advice.

As of monday there were 73 critical vulnerabilities listed for Linux vs. 3 for windows 2k/2k3 in ICAT. http://icat.nist.gov/icat.cfm?function=results&startrow=1

"we haven't ported Fog Creek's two applications from classic ASP and Visual Basic 6.0 to .NET because there's no return on investment for us."

If you not offer hasselfree ClickOnce install/deployment to your customers your competitor will. If you want a headstart with XAML buy a third party product like Xamlon.

Regards
Rolf Tollerud

Could somebody please update Rolf the Script code please? It is generating a too repetitive output. The code obviously needs lots of improvements on its FUD-generating algorithms, and major updates to its propaganda links and bad statistics databases. Better yet, append the following line to it: "> /dev/null". :)

Thanks,
Henrique Steckelberg

I am disappointed in you Henrique.

Here am I listing statistic that never before are published and you are saying that I am a script!

Ungrateful *censored*

Best regards
Rolf Tollerud

I am disappointed in you Henrique.Here am I listing statistic that never before are published and you are saying that I am a script!Ungrateful *censored*Best regardsRolf Tollerud

The fact they are new don't make them any better... ;)

Hey, it would be somewhat easy to create a web crawler script which could generate tons of spurious statistics about any subject in the world. It would surely help decrease you posting workload, if indeed you are not this very script itself already! :)

Regards,
Henrique Steckelberg

For crying out loud...don't they get it yet?

On the contrary, I believe Sun *is finally* getting it.

Within the Java community, you have an abundance of really smart people who enjoy the capabilities and features that Java offers in whatever form.

Okay, now how about the other 80% of developers out there who call themselves programmers, but spend their days dragging and dropping applications together (ala MS). While, really, there's nothing wrong with that, it's great that Sun is finally going to appeal to that group of developers.

In terms of simplicity, you have two routes: either simplify the API or use a visual tool to guide development. With all of the requirements that Java users expect-- desktop to enterprise web services, yes things can be quite complex and the API will be difficult to wrangle. So I say start pursuing the second option of visual tools that back standard JSRs, the same JSRs that other vendors support.

To use JSF as an example, I've been seeing lots of developers, brand new to Java technology, putting together simple applications ONLY because of products like Sun Creator for drag 'n drop development. It makes previously difficult tasks quite easy to grasp with Java, and will probably result in new founded interest in other Java frameworks such as IBatis ;-)

Great work Sun.