No Sun Is An Island

As if driven by the cycles of the moon, it seems like the Java community gets a monthly visit from that special topic that divides and angers us more than any other: the question of whether Sun should "open source" Java? Most recently a top open source advocate named Eric Raymond (author of " The Cathedral And The Bazaar ") published an open letter to Sun urging them to "Let Java Go." I have huge respect for the achievements of the open source community, but I tend to stay away from the pissing contests that these discussions about open sourcing Java often devolve into.

I think the issue is something of a bugaboo, anyway. The source code for Java is readily available to anyone who accepts the Sun Community Source License (SCSL.) You can fix problems and submit patches to your heart's content. You can freely use the source code to better understand where problems in your own code are occurring, and you can also look to the Java source for useful examples and implementation patterns which you can emulate in your own code. Most of the technical benefits of source code availability are present to developers under the SCSL, and they are a very significant set of benefits. Furthermore, the Java Community Process (JCP) does a fine job of driving Java technology innovation in a balanced way that meets the needs and serves the interests of many vested participants. The problem definitely isn't that the source code to Java is unavailable or that the community has no voice in the platform's ongoing evolution.

What you cannot get from the SCSL and JCP is any reason to trust Sun regarding the disposition of the Java brand, their ultimate lever of control. I hope I am not the only one who cringes every time I hear Java described as a product of Sun Microsystems. Java has long since transcended it's origins at Sun to become a globally preferred platform, an industry and a community powered by the tireless work of countless individuals and organizations. Java today is not much more a product of Sun Microsystems than the telephone is a product of AT&T. But the ownership and control of the Java[tm] BRAND is 100% firmly and resolutely in the hands of McNealy and crew, and that's why they are so utterly alone and without allies in the platform-level marketing of Java.

Sun's over-insistence on total control of the Java brand has created a situation where nobody else is willing to help them with the general platform marketing of Java. Although this technology is useful and attractive to many strategic industry leaders, very few of them would be so naïve as to place any trust in the goodwill and beneficence of Sun Microsystems, Inc. Why should Intel, Hewlett-Packard, Sony, Apple, Oracle, BEA or IBM expect Sun to do anything but look out for itself? Why should they invest valuable resources in the consumer marketing of Java platform advantages when they reasonably fear that Sun could turn a cold shoulder on them at any time?

The simple answer is: they won't! They don't trust Sun, and they won't engage cooperatively as long as they feel that their resources would be invested for the benefit of a Sun-owned and Sun-controlled product brand name. I don't blame them. As long as they feel the risk that Sun could perform an about-face on key positions and start to charge usurious licensing fees (or even revoke licenses!) they will not invest in helping to increase the consumer appeal of the platform and drive its adoption. I have personally spoken with the CEO's and heads of marketing for dozens of Java industry players, and virtually none of them is willing to shoulder any of the load for marketing the benefits of the Java platform as long as they feel disenfranchised and powerless where strategic marketing decisions are involved. If Java is merely a product of Sun Microsystems, and the benefits of owning that product accrue exclusively to Sun, then why should any other company or organization devote any resources to helping with the consumer marketing of Java - marketing which Sun is apparently unable or unwilling to do?

So there's no place where all the many interested parties invest and work together to ensure that our Java industry is strategically well-positioned against competitors. That's the real heart of the problem. Isolated Sun lacks the consumer marketing skills and budget resources to promote Java at the platform level the way the vastly more successful monopolist in Redmond promotes Windows. The core messages of Java, the ones that attracted most of you years ago, are no longer being marketed to the general public by anyone at all! Nobody is telling average people about the benefits of WORA and platform independence, freedom of choice in vendors, community-driven innovation, or Java's ease of learning and use. Without these core messages forming a foundation for consumer acceptance, we really don't have to worry about whether people will buy their Java products from Vendor A or Vendor B - they won't be buying Java products at all.

Most consumers (you know, those people who will gladly pay $8.95 for funky little dolls that say "Intel Inside") do not have a clue what Java is or how it benefits them? Their lack of awareness is hurting you because average consumers are much more involved in the economic big picture than we typically give them credit for. It's making it harder for you to get a good Java job. It's making it harder to sell your Java-based products. It's increasing the risk that the Java job you now have will be gone in two years. Today, more than ever, we need to organize ourselves as an industry and engage in marketing efforts to cooperatively protect, sustain and grow the market share of the Java platform.

Eric Raymond has focused on the wrong issue. Making Java "open source" might placate a vocal contingent in the software development world, but it would not significantly raise consumer awareness and acceptance of our platform. I simply don't care all that much whether Java is "open sourced" because I don't consider that to be the central problem. I do, however, very much want to see the Java platform more competently and competitively marketed so that this great industry and community can grow and provide all of us with a foundation for thriving prosperity.

The only way that will happen is for Sun to yield some control over the Java brand in the same way it has yielded a great deal of control over the platform api evolution. Sun needs to learn from its own success with JCP, and it needs to learn quickly. They simply cannot continue to go it alone on platform marketing. If McNealy can be shrewd enough to trade a modicum of control now in order to drive the formation of a Java industry marketing coalition, then the tide would rise for everyone in the Java world, and his shareholders would profit greatly. If Sun continues to doggedly insist on total control, then it will never be able to enlist the allies needed for success in such a daunting marketing challenge. Sun may retain 100% control, but everyone knows 100% of a marginalized market can be a lot less than a lion's share of a mainstream, successful industry.

There's powerful magic in the Java platform, but we need more powerful resources than Sun alone can muster if this industry is ever to reach its full potential. I hope you'll join me in urging Sun to provide meaningful incentives to rally this industry into action and to create a cooperative industry alliance for Java platform marketing. Just as the dairy industry jointly funds the "Got Milk?" campaign, all of us in the Java industry need to work collectively to promote key messages that lay a foundation for consumer acceptance of Java and for our long term economic success.

Please click here to join me in a discussion of this subject at the Javalobby forums. I look forward to hearing your thoughts.

Until next time,
Rick Ross
rick@javalobby.org
AIM or Yahoo Messenger: RickRossJL

Entering the Embedded Java World
If you've been following my column the past few weeks then you know that I've become interested in embedded Java. Last week our design team decided to finally bite the bullet and go with a micro-controller from Dallas Semiconductors running the TINI Runtime Environment – Java! We decided to go with their boards after much deliberation and finally decided they were the right company based on a few key things. The first one being that they provided a moderately powerful development board with Ethernet at a reasonable cost. The second key thing was that they had great customer service and their engineers replied to three or four of my emailed questions in one day. Now if they get their act together and get our stuff shipped out within the next day or so, they might just make my “companies with great support” list. The final key item that pushed us over the edge towards the TINI product was the fact that it ran Java. While in my heart I would rank this a bit higher, the team as a whole decided that this shouldn't be a major factor. In the end, it turned out that having Java is just going to make all our lives easier and hopefully will allow us to have very similar network communication code on both the embedded side and the server side.

I want to thank everyone who wrote in after my previous columns providing insight into the state of embedded Java and providing links to help me in my research for a Java microcontroller. Without everyone's help, I don't think I would have been able to zero in on something, although its always fun to learn new things! I've been longing to work with embedded java for quite some time now, and hopefully within the next week or so I'll finally get the chance. If everything goes well, perhaps I'll finally build that hardware companion to iRelate when I finish my design project.

The Votes Are In…
Last week I put a question to all our readers and I was overwhelmed by all the responses I received – by far more than for any other question I've put to you guys. I'm glad to hear so many new voices that we don't normally see in the forums answers the questions we put in the newsletters. So I'm sure you're wondering what everyone else thought about whether Sun should change the Java naming convention. Well the answer, my friends, is that most of us are torn on the issue. There was a pretty even split among you as to whether you wanted to get rid of the Sun marketing moniker (Java 2) and leave the engineering number (1.5), whether you wanted to go with something completed different (Java eXperience…the neXt generation of user technology), or whether you just wanted everything to stay the same. I have to say that I tend to agree with those people who wanted to drop the Java 2 moniker.

What is that you say? Sun's marketing people are the best on the planet? We all remember the new logo that Java received last year at JavaOne that cost some insane fortune I imagine. I have yet to see this new logo circulating on Java hardware. The pages that I purchased my new embedded Java board from did not have this new logo on it and we've found that most places do not have the new logo (Apple just upgraded the logo with 1.4.2). Now, here I could be wrong, but have you see the new logo having any effect? Was it worth the money that could have been spent in better ways? At JavaOne we want BIG announcements, not fluff over a new logo. This year give the developers who are paying a LOT of money something to really cheer about, but I'll save what I really want to see at JavaOne for another column though.

Always Check Your Facts
As any writer knows (or should know), its best to check your facts before you print something and I should be no exception. Last week I blamed Sun for the UI known as CDE when in fact they were not to blame. Thank you to everyone who wrote in to correct me, next time I'll be sure to double-check what I'm writing about!

‘Till Next Time,
Matthew Schmidt
matt@javalobby.org
Yahoo: mattschmidtjl

Thanks to Michael Emmel for pointing out that one of the most prominent open source advocates has just published an "open letter" to Sun, urging the company to "Let Java Go." Eric S. Raymond, author of the seminal work "The Cathedral and the Bazaar" and president of the Open Source Initiative, published the open letter in apparent response to Sun CEO McNealy's recent comments about how much Sun loves open source.

ESR has credibility in the subject, and his open letter makes some points about how the choice between "control" and "acceptance" has impacted Sun in the past. The open letter is not long (approx. 700 words), so I recommend you take the time to read and consider it.

What do you think? Has the time come for Sun to take ESR's advice?

I'm pleased to announce that the NetBeans 3.6 beta release is now available for download and testing. The NetBeans.org team would like to thank the Javalobby community for all the input they've provided to us in discussions here over the past few months. This release contains many items suggested by JavaLobby members. In particular we recieved some feedback at JL that people wanted better Mac support. We've really tuned up the new Windowing system under the Aqua look and feel, so Mac users please take a look.
Download: http://www.netbeans.org/downloads/ide/next.html

Some of the new features in NetBeans 3.6 beta include: A new windowing system, providing native look and feel Enhanced navigation and workflow 2-tier J2EE 1.4 support (servlet 2.4 and JSP 2.0) Improved JSP debugging with JSR-45 Support Tomcat 5 Integration Code editor enhancements such as Smart Brackets and Code Folding Integrated testing support through JUnit Improved online help with JavaHelp 2.0 integration and more ...

Some of you may remember me as the host of the cajo project at java.net. Well now I have a problem, and maybe some gracious souls at the Javalobby can help me out. I am in active discussion with the Free Software Foundation to transfer my copyright of the cajo project over to them. (Further discussion of this is also welcome in the replies)

However, in order for the FSF to accept title to any Java project, it must be able to run on a GNU/Linux/Java platform. The current front runners are gcj and Kaffe. I am not yet proficient enough at Linux to do this myself, so I would like to ask if one of you would 'carry the torch' and verify this, on either or even both platforms. It would be a great gift to the entire open source community. Continued...

Every now and then I tend to poke around in SUN's Java source code. Often I find quite a lot of defensive programming. The question is, does it hinder performance and/or download size to any notable degree? Should we ask for a debug/release (or normal/optimized) version of the API? What I mean is tests like these:

    private Entry entry(int index) {
        if (index < 0 || index >= size)
            throw new IndexOutOfBoundsException("Index: ex+ ", Size: " + size);
      
      ...
    }

I've been experimenting with a new look and feel for an application I'm working on. When I was looking at the new Synth look and feel, I was surprised that they don't allow you to incorporate the Java2D more easily, at least not without some (non trivial) custom painters.

I am working on what I think is the trend in recent look and feel using vectorial instead of bitmap, like SVG, PDF, flash or Avalon. You can check out this applet for proof of concept.

One of the nice features of Java2D look and feel, besides the flash-like, smooth look it provides, is its scalable property. Try to resize the palette, which uses a grid layout, to see what I mean. It's something I wasn't able to do with my all bitmap based look and feel (oyoaha lookandfeel)

You can find more info about avalon here: for a java perspective and here for some screenshots.My primary goal was to develop this vectorial look and feel mostly for my own application. But now that I've seen the specs for Synth, I'm not sure if I should port these features to Synth or continue to develop my own look and feel. Are other people interested in having an all vectorial look and feel?

I have released the version 1.3 of the JGoodies Swing Suite.

:: Introduction ::
The JGoodies Swing Suite provides components and solutions
that complement Swing to solve common user interface tasks.
It advocates a UI production

Version 0.9.0 of HelpSetMaker has just been released. It features many enhancements over the formerly announced version 0.6.2 from October of 2003. Among them:

- Output not only as JavaHelp HelpSet and HTML file hierarchy, but also as one big HTML

http://www.javasight.com ,Java's new portal, is online in a beta version.
The world of Java increases the need to be constantly updated. Javasight makes this possible by assembling news and articles from major

This is a quick note to inform you of the availability of the first performance-tuned public beta of Excelsior JET for Linux on the x86 platform. (In case you have not heard of this product, it is a Java VM supercharged with ahead-of-time

YourKit Java Profiler 2.0.1, the maintainance release, is available. The release contains several bug fixes.
With help of YourKit Java Profiler you can easily solve wide range of CPU and memory related performance problems.
Please visit http://www.yourkit.com to learn more about the profiler, get a free evaluation license key and download full-functional version of the profiler.

OpenReports is a complete web based reporting solution that provides the ability for users to view dynamically created PDF, HTML or XLS reports in a browser. It is developed in Java and uses JasperReports as the reporting engine.

Leading Global Internet Company Pairs with TopCoder
to Determine Collegiate World Champions of Coding

Glastonbury, Conn., February 16, 2004 – TopCoder, Inc., the leader in online programming competition, skills assessment and competitive

SearchBlox is a J2EE Search Component that delivers out-of-the-box search functionality for quick integration with your websites, applications, intranets and portals. SearchBlox uses the Lucene Search API and incorporates integrated HTTP and File System

I'd like to announce the availability of Faces Console 1.0 Beta 3. This new version supports JSF 1.0 beta and contains a few bug fixes and enhancements.

The Faces Console is a FREE standalone Java Swing application for managing JavaServer

Alma is a software with the following functionalities :

* Reads several sources (languages, models, ...),
* Helps to design for object-oriented modeling (definition of classes, relations, patterns, ...)
* Modifies the structure

EditiX is a Swing XML editor designed to help web authors and application programmers take advantage of the latest XML and XML-related technologies such as XSLT / FO and XSD Schema. EditiX provides users with an extensive range of XML functionality

IntelliJ IDEA is a comprehensive development environment for Java programming.
It provides a robust combination of enhanced development tools, including: refactoring, J2EE support, Ant, JUnit, CVS integration. Packaged with an intelligent editor,

New and noteworthy in M7:
http://download.eclipse.org/downloads/drops/S-3.0M7-200402122000/eclipse-news-M7.html

jIRCii is a cross platform java Internet Relay Chat (IRC) Client. The client is fully scriptable using a perl-like language called sleep (to be released as a separate project soon). Supports CTCP/DCC protocols, connects to SSL servers, and supports a

After some delay we are pleased to anounce the release of
the latest version of jConfig.
Managing configurations is getting even easier with this release. Here is a short list of new features:

- added the new ExtensibleCategory (can extend

WebSphere developers can now offer their clients an enterprise solution, without the need to spend US$100,000 on a dedicated WebSphere environment.

GroupWare offers a shared WebSphere Hosting service for your client's demanding J2EE applications.

Hi Javalobby,

JamP is a Webstart enabled Mp3 Player. It can change the skin on the fly via a skin browser. The newest versions boasts a stabler and improved gui with smoother desktop integration. It works on many platforms, Solaris

JamP

Datamata SQL Portal is a J2EE server side product. It provides easy to use, secure SQL access to all project databases for all project members from a single web location.

Datamata SQL Portal allows managing data over the web. It is free for light

The goal of Simple is to bring the power of simplicity to the world of server side Java. The primary focus of the project is to provide a truly embeddable Java based HTTP engine capable of handling extreme stress. Simple has consistently out performed

Recently, motivated by feedback from our users (in-house and out), a lot of incremental changes were integrated to Bossa. We have enough changes now to warrant a release.

Most notable is the ability do query and set the case state, the