John O'Conner's Blog

John O'Conner is a software architect, consultant, author, and speaker. After many years developing the internationalization features of the Java platform at Sun Microsystems, he is now actually trying to use those features in real world projects.

Character Conversions from Browser to Database

It's relatively easy to enter multilingual text into a rich client, Swing application. Doing so in a browser-based, web application isn't always as easy. If you're interested in how to avoid data loss as characters move from browsers to database, you may want to take a look at this recent Java EE tech tip: Charset Conversions from Browser to Database.

As always, your feedback and suggestions for new articles and material is always welcome.

JavaOne Tokyo '05: Project Looking Glass

Project Looking Glass must have a special appeal in Japan. Although the project gets plenty of attention in the U.S., it seems especially popular in Japan. This particular project is responsible for attracting standing-only crowds to at least two hands on labs, a couple sessions, and finally a late night Birds of a Feather (BOF). Thinking the crowds would be thinner, I attended the 8 PM BOF on Thursday night (Nov 10). No luck. Again, standing room only. I managed to push through to the front, flashing my "Press Pass" for maximum advantage in the crowded room.

With almost rock-star popularity, both Mr. Hideya and Dr. Sakuraba play their Dukeleles as they sing a Project Looking Glass theme song. Even without the song, these two already have great appeal here...and it's probably a good thing that they already have day jobs too.

The BOF featured four guest speakers, each demoing an application or framework for Project Looking Glass:

• a file manager
• a Scheme based scripting platform
• MXW3D Loader (3D Model loader)
• a presentations framework for multimedia

Right at the moment when pictures could speak a thousand words, my camera battera died, and I have no more picutures. I will, however, provide URLs to more information.

The file manager presentation was provided by Mr. Eiji Inoue. The file manager was visually impressive, representing the file system as a set of stacked boxes, in which each box represented a file. The GUI allows users to grab a box and move it around in all three dimensions...well, virtual 3D anyway. Various file layouts were available, including a top to bottom, left to right, right to left, and random, which brought a laugh from the crowd when they saw file objects strewn randomnly across the desktop.

A Scheme interpreter exists that works well within Looking Glass.

Mr. Eitaro Nishijima showed us a 3D model loader. He modeled a 3D Duke that walked around and waved to us from his Looking Glass world.

Probably the most impressive demonstration was for the Ibrik Presentation Toolkit...and of course the presentation software was written using the toolkit itself. I'll never get tired of seeing pages ripped up and off the screen with new ones flying in nicely after them. The graphics were stunning.

Find out more about the Looking Glass Project here:

• interest_ja@lg3d.dev.java.net
• www.javadesktop.org/forums/forum.jspa?forumID=57

Covering JavaOne Tokyo '05 has been a lot of fun. With this Looking Glass info, I think I've exhausted my blogging material. See you next year, and until then take care! お元気で！

JavaOne Tokyo '05: EJB 3.0 Hands on Labs

Hands on labs were quite popular at JavaOne Tokyo. This particular lab taught how to use EJB 3.0. Many of the labs were filled to capacity. Below, the instructor guides his class through a programming example:

One of the most popular labs was for Project Looking Glass. In fact, the Project Looking Glass labs were so popular that they required not one but two separate sessions to handle all the registrations.

JavaOne Tokyo '05: Layout Managers

Joshua Marinicci and Scott Violet explained layout managers in their session Layout Roundup and Futures. Both shared the speaking duty as they described the various layout manager types available for desktop applications. You already know many of the most common ones:

• FlowLayout
• BorderLayout
• GridLayout
• GridbagLayout

Nothing particularly exciting there. However, they did describe a few newer, easier to use, layout managers:

• Packer
• Form
• Group

The impressive part of the session is simply this: using some of these newer layouts can substantially decrease the code you write. In some cases, you're going to write less than half the code you would if you used GridbagLayout for example.

The Packer layout manager is developed as a java desktop project on java.net. If you like GridbagLayout's power, but think it's too difficult to master, you're going to like Packer.

The Forms layout comes from the JGoodies site as freeware.

Finally, the Group layout is available as the "Matisse" layout in NetBeans. It may be the best reason yet to try NetBeans as an IDE.

JavaOne Tokyo '05: Java Puzzlers with Josh Bloch and Neal Gafter

I'll quickly jump to the most interesting part of the Keynote this morning (Nov 9), the special guests. The guests were Joshua Bloch and Neal Gafter. Not only has Joshua contributed great things directly to the Java language and platform, but he has become a respected author as well.

Josh spoke about optical illusions and why we see them. He showed several optical illusion illustrations. He carefully made the leap from optical illusions to programming illusions, and then from programming illusions to programming puzzlers. It was entertaining for sure, even if it was essentially a long advertisement for his latest Java book, Java Puzzlers.

Below, Josh and Neal sign Java Puzzler books for attendees:

JavaOne Tokyo '05: A Note on Accessibility

On a recent trip to the restroom at the Tokyo International Forum, I was reminded about accessibility. Having been trained well (thanks Mom!), I always wash my hands after visiting the restroom for any reason. However, doing so in a public restroom is almost always frustrating because...well, I'm short and when I reach up to grab a paper towel, the water runs down my arms. No luck with air dryers either; in the US they're often just as high on the wall as the towel dispensers. However, I was pleasantly surprised, almost tickled even, to find an accessible hand dryer in Tokyo. It was low enough on the wall that I could just reach down to use it. Here's a front, side, and top shot of my discovery.

Just extend your hands down into the slot, the warm air begins to blow, and you move your hands up and down in the air to completely dry them. Nice. And no wet elbows or armpits, yeck. That's the way to make and to place a hand dryer...accessible and usable.

This experience made me think about the Java platform's accessibility efforts. The accessibility spec has been available since when, 1.3? This API provides mechanisms to make your applications easier to use for people with certain disabilities. The APIs help other assistive technologies like screen or text readers work better with Java applications. Navigational features help people get around your application in a variety of ways. It's a good thing, and probably should get more attention. I don't know if anything new in the accessibility area will reveal itself here at the conference or not, but maybe it should in the future. If you know of anything here at the conference, please let me know.

I suddenly have a new interest in Java's assistive technologies. If you're interested too, these URLs will help you get started:

• Java Accessibility Overview
• Accessibility Enhancements in 1.5
• Accessibility Utilities
• Accessibility and Swing

JavaOne Tokyo '05: Extreme GUI Makeover

After the rather somber response to the keynotes this morning (Nov 8/Tokyo), I was pleasantly surprised to hear the impressed (but still rather quiet) oohs and aahs from the crowd as Hans Muller (left) and Scott Violet (right) wowed us with visual Swing demos. The session, entitled Extreme GUI Makeover, Episode 1: Lookin' Good was amazing. I realized that one's imagination is the only limiting factor in creating eye-popping, stunning graphics on the desktop. Swing, 2D, and the 3D APIs all collaborate to make great things happen for the developer willing to spend a little more time on their GUI. Many times I whispered to myself, "Wow, I didn't know you could do that."

Next to me on both sides, two attendees scribbled notes, hanging on every translated word that Scott and Hans spoke. Several times attendees started to clap when shown a particularly impressive demo of smooth and fast animations. Then, as if their more reserved nature kicked in, most would quickly settle back down, almost embarrassed for having given in to the urge to reward the speakers with applause.

The demos demonstrated avatar animation and creative displays for a chat application. Hans and Scott promised that we'd be able to download them. They mentioned javadesktop.org, so maybe the demos will show up there. If not, you'll have to check out Scott Violet's blog for more information

We know this session was a success because of the standing room only crowd. Rumor has it that eager attendees were so plentiful that they were clogging the subways.

JavaOne Tokyo '05: Technical Keynote outlines strategies for Java ME, SE, and EE

Java SE Roadmaps, John Pampuch

John Pampuch, Director of Java SE Software engineering outlined some of his team's activities. Here are a few of his comments, summarized of course:

• Name simplification. The Java product names have been changing. The new names are simpler. In summary, Sun will remove the "2" from all names. J2SE will be Java SE, J2ME becomes Java ME, and J2EE will become Java EE.
• Community. All new APIs will go through the community process. The community will lead and drive Java features now and in the future.
• There have been more than 100,000,000 downloads of the J2SE 5.0.
• Looking for 5.1 release. Don't bother; it isn't going to happen. The next major release is 6.0, but there will be 5.0 updates
• The timeline for feature releases will be 18-24 months. Intermediate releases will be bug fix updates.

Some of his Mustang themes:

• compatibility, stability, quality
• diagnosibility, monitoring & mgmnt
• xml & web services
• ease of development
• the desktop. The desktop is critical to Java's success. Expect lots of GUI upgrades, lcd font support, a system tray api, and graphics pipeline boosts using OpenGL and DIrectX

One of the things I noticed is that the keynote slides mention that Sun still wants contributions from the community in the form of bug fixes and features. However, I think it's fair to say that features are frozen. That is, new features are not being introduced into Mustang at this time. Although I don't speak for Sun, my conversations with key engineers on the project seems to confirm this. If you have a new feature idea, don't be surprised if your contribution lies dormant until the next release, Dolphin.

Speaking of Dolphin (Java SE 7), Mr. Pampuch said that the following features are under consideration:

• direct xml support
• friends for cross package refs
• method references
• JVM bytecodes for dynamic languages
• beanshell scripting language
• new io apis

Mark Hapner, J2EE development

Some of his comments:

• There's a big move towards Plain Old Java Objects (POJO) in Java EE.
• Annotations will be used extensively, and one of the biggest results is that deployment descriptors disappear.
• Java EE 5 has simplified web services support.
• EJB development is simplified
• There's a new persistence API.

You can expect this new Java EE 5 technology as a beta in Q4 2005. Finally, in Q1 2006 you can expect the final product.

Glassfish is an open source implementation of all Java EE standards. Mr. Hapner urged everyone to get involved. Start at the Glassfish Project Homepage.

JavaOne Tokyo '05: java.net forum for JavaOne Tokyo

Just in time for the opening keynote, a java.net Forum for JavaOne Tokyo 2005 is now available. Everyone is invited to contribute to the forum, whether you're attending in person or not. Go take a look, share a thought, ask a question, post an answer

JavaOne Tokyo '05: Nov 8 TImetable

Jot it down on a piece of paper, in your PalmPilot, or in your PocketPC...anywhere you'd like...but you're going to need some help remembering the JavaOne Tokyo 2005 Timetable for Nov 8.

Kicking it all off is the Technical Keynote w/ Mark Hapner at 10:00. Following that, well you have lots to choose from. Most sessions are organized into 45 minute slots throughout the day, and they fall into the following tracks/categories:

• Interoperability Track
• Maruyama Sensei's Lecture Series
• Sun Network Computing Track
• Technical Sessions
• Hands On Labs

Check your schedules, write it down, check it off...there's no way to attend everything, so a little pre-planning should help to absorb as much as possible.

JavaOne Tokyo '05: NetBeans Day highlights 5.0

Jeff Jackson, Greg Sporar, and others played to a tough crowd, but I think the message about NetBeans got across even if the jokes or short skits didn't. In the end, the message I got was this: NetBeans has come a long way very quickly, and there's no slowing it down. Everyone has their favorite editor, so I can't expect to change many opinions here. However, just to keep yourself up-to-date with what's going on, you owe it to yourself to evaluate NetBeans 5.0.

Here are a few notable features that I noticed in NetBeans 5.0:

• Improved editor hints not only find problems in your code, but can fix them too.
• Code completion gets turbo charged. You've gotten used to code completion by now...it's been around for awhile. But the NB team has gone a step further. You can now type just the uppercased letter of each word in a method or class, invoke code completion, and get a list of completion candidates that match your abbreviated hint. For example, say your creating an IllegalStateExeption...you can type

  Exception e = new ISE<CTRL>+<SPACE>
  

  This is enough information for NetBeans to bring up a candidate list containing this:
  
  
• more refactoring support, something we can always use more of.
• more J2EE servers are supported right out of the box, including JBoss 4 and WebLogic 9
• online collaboration tools; you can chat with others and can even have them look and change your code all online.
• lots of others I'm just too tired to list

Members of the Java Studio Enterprise and Java Studio Creator teams were there too.

Of course, you can't have a Sun sponsored event without a new t-shirt or two. Jeff Jackson and others launched a few newly minted NetBeans shirts into the crowd:

As if the NetBeans guys understood that descriptions of code completion, server integration, and collaboration tools can get pretty dry, they cooled things down with a beer bash/reception afterwards:

JavaOne Tokyo '05: Warm up with NetBeans Software Day

You haven't missed out on any of the actual conference yet. The conference officially starts tomorrow, Nov 8. However, if you want to get a quick start on things, you can stop in on the NetBeans Software Day WorldTour.

When: Today, Nov 7 @ 13:00 (Tokyo Local Time)
Where: Tokyo International Forum Hall B5
Cost: Free...as in free beer
Registration: http://jp.sun.com/company/events/javaone/2005ex/netbeans/

JavaOne Tokyo '05: First Stop, Akihabara

I'm not a geek, I'm not a geek, I'm not a geek. However, in preparation for JavaOne I have to get a USB cable to connect my digital camera to my PC...you want pictures with these blogs, right? So the natural idea that comes to mind is to hit the stores in Akihabara. If I can't find a USB cable or a Secure Digital (SD) card reader there, then I have to assume that USB cables or SD cards no longer exist.

Since I mentioned geek...ever hear of "otaku"? They are Japan's uber-geek...you can read more about them here:

• Otaku Transform Akihabara Into A Nerd Nirvana -- Planet Tokyo
• In Tokyo, a Ghetto of Geeks

Customize Your JList Display

Have you ever tried to display something within a Swing JList only to be disappointed at the displayed text for that object? Swing JList components don't always display the text you'd like for a specific object list...at least not by default. You can control the text that JList displays by creating a custom ListCellRenderer. Read Customize Your JList Display for more information.

JavaOne Tokyo '05: Map of Tokyo International Forum

Trying to find your way around the International Forum? Use this TOKYO INTERNATIONAL FORUM Map to find out just how lost your are.

JavaOne Tokyo '05: Hotel rooms and technology

Several weeks ago, I booked my hotel room using a fancy, online reservation system. Everything looked great. Good area in Shibuya with a room overlooking Shinjuku, just minutes from the Shibuya station. After an unusual panic attack, I decided to actually call (after all these years with email, I can still use the telephone) the hotel to confirm my reservation.

"Hello, this is John. I would like to confirm my reservation for ..." I said nervously.

The response was somehow already in my mind before she spoke, "We're sorry. We don't have your reservation on file."

"Oh no...I made these reservations online weeks ago" I weakly replied.

In the most humble, it's-all-our-fault-but-there's-nothing-we-can-do, sweet voice, she said "I'm sorry."

There I was, phone in hand, jaw on floor, wondering where I would stay in just 24 more hours when I arrive in Japan.

"OK, gotta think quickly," I thought. "Get on the internet...no, that's how I got into this problem...an internet reservation. No, no, just get a few phone numbers."

I quickly found a few numbers of hotels in the same Shibuya area, made a couple calls, and BINGO! I found another room and booked it immediately. And I made sure to get the confirmation number too.

The odd thing in all of this is that the online systems quickly pronounced me hopelessly out of luck...no rooms available in all of Asia. When I finally decided to call the people directly, I found a place almost immediately. I don't get it...online, nothing; on the phone, I make a reservation immediately.

I suppose there are still some things technology (and online reservation sytems) still lack, and sometimes whether I like it or not, I'll have to resort to the old but extremely useful, dependable telephone.

Close call, but I'll actually have a place to stay after all. If you haven't found a place by now, let me give you a quick tip...forget the online reservation systems, get online, find a few hotel numbers, and make some direct, old-fashioned phone calls.

JavaOne Tokyo '05: Keynote Addresses

Beginning Nov 8, you can attend the keynote addresses...apparently for free if you happen to be in the area.

Here's the line-up:

Bidirectional Text Inconsistencies: Bug # 4701238

I received an email from one of my good friends in the Java internationalization group at Sun. He's asking for feedback on how his group should fix a bidirectional text layout bug in Swing. His email explains it quite well, so I quote it below.

Please allow me to solicit your opinions on a bidi problem in Swing
that we have been facing for a long time.  The problem is regarding 
the default text run direction in Swing components.  The gory detail
of the problem itself is described in the bug report 4701238, but 
here is the summary of it:

1. In Swing text components, such as JTextField or JTextArea, the 
default run direction of the text follows the component orientation. 
So, for example, if those text components are laid out Left-To-Right,
the text in those components start from the left edge of the 
components.

2. Other Swing components that deal with text, such as JLabel, draw 
text based on the first character in the text which has strong 
direction property.  So if the first character of the text is one of 
the alphabets, the text starts from the left edge of the component, 
and if the first character is one of Arabic/Hebrew, it starts from 
the right edge.

So if I draw a swing application with RTL text, it looks like this 
(ARABIC is the Arabic text in logical order):


LTR component orientation:
+-+-------------------------------+
+-+-------------------------------+
| JTextArea:                      |
| +---------------+               |
| |CIBARA         |               |
| +---------------+               |
|                                 |
|                                 |
| JLabel:                         |
| +---------------+               |
| |         CIBARA|               |
| +---------------+               |
+---------------------------------+

As you can see, the text direction is inconsistent.  We need to 
resolve this inconsistency, but just don't know which one is more 
acceptable for those who speak bidi languages natively.  Of course, I
would not be surprised if there is no definitive answer, because the 
initial assumption (RTL text in LTR component orientation 
environment) is somewhat inconsistent in the first place.

I would appreciate your feedback. 

Got any ideas, help, suggestions? Let me know!

JavaOne Tokyo '05: NetBeans Software Day

You could win a NetBeans t-shirt! But more importantly, you'll find out why Sun thinks NetBeans is the ベスト (best) IDE and application development platform.

Attend NetBeans Software Day on Nov 7 from 1:00 PM in Hall B of the Tokyo International Forum.

This event is 無料 (free), so you can attend even if you don't have a JavaOne Tokyo '05 conference badge. Although attendance is free, you still have to register. See you there!

NetBeans 5.0 Beta...ouch

I recently discovered Sudoku and decided I'd use NetBeans 5.0 to develop a Java version of the game. I thought it would be a good way for me to play with the latest NetBeans and its new GUI layout tool called Matisse. Turns out Matisse is easy to use, and seems to do the right thing most of the time.

However, I do not like IDEs that generate code that is surrounded with "Do not edit this section. This code is autogenerated." If you work within NetBeans, the editor won't let you directly edit the code anyway. However, despite the "do not edit" warnings, NetBeans does provide a way to insert your custom code into this section. Using the graphical design and "Inspector", you can select any component and view its properties. Some items have properties like this:

Pre-Init Code
Post-Init Code
Pre-Creation Code
Post-Creation Code

You can use these "properties" to add your own code to this "do not edit" section of Netbeans "initComponents" generated code. Fine. However, I have experienced a bug in which after saving and redisplaying my component, this custom code disappeared. Gone. Vanished. I was angry because I had spent more than a few hours writing the code.

I want to use NetBeans, and I want to be a proponent. I'll continue to use NetBeans...but I don't think I'll use the GUI layout tools anymore, at least for a while. I think I'll just hand code my GUIs...what a shame. JBuilder really gets this right. Their GUI layout tool works well, doesn't generate any "do not edit" sections in your code...and it does *not* delete anything you put there either. I like that...sigh...but it's a bit too expensive for me.

Come on NetBeans...how can we get rid of those special "do not edit" sections in the generated code? And if you're going to allow a back-door approach to actually putting code into that section, please, please don't delete it.

JavaOne Tokyo '05: Session schedule

Patience, patience...good things can't be rushed. But we're not so far away from the conference, and I'm ready to create my session schedule. Unfortunately the schedule isn't posted...we don't know who will speak or when. We don't know the topics or keynotes...but I'm sure it's all on its way.

One other little note...the session schedule won't be translated into English. That's what I hear anyway. That's all the more reason to catch some of the news and schedules here. I'll translate as much as I can, but you have to understand that I'm not a pro. I'm a Japanese hobbyist; I play around with the language a little. If we depend too heavily on my translations, someone's bound to get hurt. Nevertheless, I'll do what I can.

Contributing to Mustang: Contributor status granted

Just a little surprised, I got an answer to my faxed contributor agreement within two days. I had heard of unanswered faxes lingering, perhaps unseen or forgotten, for as long as a couple months...but that isn't my experience.

With my new contributor status, the next part is to hunt down a bug. There's a little catch...I don't have a MS Visual Studio environment. Does that prevent me from contributing? No it doesn't. As long as the problem is in documentation or Java code, I can still fix a bug and contribute. In the past, I was a little bummed about this...I can't contribute fixes to the native C/C++ code. However, I realize that there's a lot of Java code left...I doubt I'll need to dabble in the C/C++ code. And as Tom Ball commented in one of my blogs, if I really want to do any low-level Windows work, I have to "drink the Redmond kool-aid, like it or not," which means purchasing the SDKs, the developer subscription pack, etc. I don't like it, but he's right. Programming at the C/C++ level in the base VM isn't for a weekend Windows programmer like myself...perhaps I'd better stake out the higher ground and stay in the Java code.

Hey, wait a minute...while I've been playing around, Mustang work has continued. The build #53 code that I downloaded just a week ago is now old, outdated...history. Now build #55 is available. OK, I promise, after I get these latest bits, we'll go to the next step.

If you could submit a fix right now, what bug would it be for?

Contributing to Mustang: The Process Map

Two days ago I faxed my contributor agreement to gain "contributor" status for the Mustang project. This step is necessary to fully participate in Java SE 6 (Mustang) development. Those Mustang guys have lots of work on their hands, so it's still way too early to expect a response.

Speaking of "expectations," the participation instructions don't mention how long it should take to gain contributor status. I suppose it's a case by case task for each potential contributor. So, I don't really know what to expect for a response. However, most businesses ask for 5-10 business days, so I think that's probably reasonable in this case too...right?

In the meantime, there is a lot of work to do. I have to find a bug, fix it, and prep it for submission. The complete process isn't as simple as some might want, but it doesn't seem too bad either. I found a process map that might interest you. I'm just at the beginning.

Contributing to Mustang: Waiting on a fax reply

If you're new to this particular blog thread, let me review. First, Sun has invited everyone to participate in Mustang development. Take a look at the java.sun.com site...invitations everywhere to participate. Second, I'm experimenting with the invitation in public via this blog. I'm following the steps to become a contributor to the Mustang project. Third, hopefully I'll be able to submit a bug fix that gets included in the Mustang product...that's the goal!

So...where were we in this process? I'm at the "fax in your signed contributor agreement" step. I sent off the fax today at 2:45 PM PST. At some point I should get a response (by email?) that says I'm now able to promote myself to "contributor" status on the java.net site.

In the meantime, I have some excellent news. Dr. Cay Horstmann, author of Core Java Vol 1 and Vol 2, has agreed to help me in my Mustang experiment. In a response to my previous blog, Dr. Horstman mentioned that he had several bug fixes that he'd like to contribute. However, he apparently hasn't had a lot of success getting Sun's attention with a previous bug fix. I realized that I didn't need to search for any new bugs to fix while he had already discovered and fixed several. When I asked if he'd like to participate in my very visible, public experiment, he graciously agreed. Thanks, Dr. Horstmann, I hope we're successful!

Contributing to Mustang: Signing the agreement

I downloaded the source and binaries over the weekend, took a look around, and have decided that yes I do actually want to continue this experiment. I want to make a contribution to the product. I want to fix a bug. However, downloading the code is just the beginning of the process. Like any good adventure game, contributing to Mustang requires that I find and engage the gatekeeper, perform an act to show my loyalty, and then, only then, do I get the keys to continue.

Next, then, is the contributor agreement. In order to continue, I need to sign and fax this back to Sun. OK, that's tonight's task.

See you again tomorrow...and hang in there. We'll get through this.

Contributing to Mustang: The Source and Binary Download

To make any contribution to the working Java product, you need the source and binary distributions. Turns out you can get that without any sort of group or java.net membership at all. Just a quick review of the license...ahem...a careful study of the license, and I'm just a click away from all that Mustang code.

I feel like I should apologize for some reason...I'm using XP Pro for this experiment. So, I download the Windows 83 MB jar containing the debugable binaries. Why this choice? I might want to make a change in the C/C++ code itself...probably won't, but I don't know where this experiment is taking me yet. Good. All done. The fact that I'm 4 blocks from my DSL provider's "headend" must account for why this takes less than 8 minutes.

The source is in a separate location, and I'm already slightly confused. It appears that I need 3 of the 5 possible downloads from this location:

• JDK 6.0 Source under the JRL license
• JDK Binaries for Source Build 6.0
• Mozilla Binaries for Source Build 6.0 (Windows)

Just a quick question...why not a choice between just 2 possible downloads? This isn't really difficult, but a choice between Windows and Unix would be easier. Stick the JDK 6.0 source and JDK Binaries and the Mozilla Binaries in one bundle for Windows downloads, and create just one other download for the Unix people. Or maybe it could be arranged like this: 1 generic source and binary bundle and a choice between a Windows or Unix Mozilla binary bundle... Never mind now, the download's done.

Now I have downloaded 4 jar files:

• jdk-6_0-ea-bin-b53-windows-i586-debug-22_sep_2005.jar
• jdk-6_0-ea-bin-b53-jrl-22_sep_2005.jar
• jdk-6_0-ea-mozilla_headers-b53-windows-i586-22_sep_2005.jar
• jdk-6_0-ea-src-b53-jrl-22_sep_2005.jar

Depending on when you download, you should have the same number of jars but with slightly different names depending on the date of the build.

Whew...that's it for now. I'll report back again later after I unjar this stuff and take a look at what I have.

Contributing to Mustang: Got java.net?

I feel like I may have started the experiment unfairly by locking some of you out of the experience. My steps so far to contributing to Mustang worked for me, but they may not work for you. Why? Because you may not be a java.net registered member.

I decided to logout of java.net and follow my own instructions. There were immediate road blocks when I tried to apply for the jdk.researcher status. You have to have java.net membership apparently...silly me, I forgot to tell you that step. Funny thing...I never even suspected that one might have to be a registered java.net member. Hmm...are there other websites that will provide access to Mustang, or does java.net have the exclusive gatekeeper status? If I have an account over at javalobby or other pro Java site, can I participate. Hmmm..I dunno. I'll check it out though.

Contributing to Mustang: An Experiment Begins

Sun wants the community to contribute to Mustang (Java SE 6) development. The call to arms is all over the java.sun.com website with click through icons that send readers to the Mustang community site.

I'm a developer, and I want to contribute. Starting today, I'm going to blog my way through the process and share the experience with you. There is a process, and I'm going to follow it starting now, today.

OK...pop on over to community Mustang site...let's see, I want to fix a bug, what do I need to do? There it is, I need to visit the collaboration page. Got it...now what?

1. Download the Mustang source and binaries.
2. Request "jdk.researcher" status.
3. ....more steps that I can't do yet..

So far, so good. I haven't downloaded the source and binaries yet. I'll do that tonight after work...and maybe my download times won't be too bad. I have to bring over a whopping how many MB? Hmm...can't quite tell yet. Looks like I might need the 83+ MB debug binaries and probably just as much in source...but more on that later. Until then, what can I do? I'll request jdk.researcher status.

Read the JRL license...good, no problem. Click on over and request "jdk.researcher" role. Click the correct radio button...done. It takes a day to get approved, so I'll wait it out until tomorrow. Until then, I've got plenty of source and binaries to download

See you tomorrow with an update.

JavaOne Tokyo '05: Getting There

Considering that the event is actually in Tokyo, Japan, it's no surprise that most information about the event is also in Japanese. Hmmm...that may be a little problem for many gaijin (外人) or foreigners. Trying to be helpful, I'm sharing what I know about how to get to Tokyo International Forum (東京国際フォーラム).

If you're already in the country, I'll assume you can already get there...this blog isn't for you. However, if you're like me, I'll arrive in Japan via the Narita Airport. You have a few choices from here, but instead of providing you bad directions, I'll let someone else do it. You can find decent instructions on how to get from Narita to Tokyo using any of these on-line sites:

• The Japan Guide
• About.com's Japan for Visitors

Once you get to Tokyo, you're just a hop, skip, and a jump away. However, those hops, skips, and jumps will probably involve a train or subway. Hint: just ask someone. My experiences with Japanese people have all been fantastic. Speak slowly...not loudly...and someone will surely help you find your way to JR Tokyo station or Yuraku cho station. If you're at the JR Tokyo station, you're only about 5 minutes away by foot. An underground concourse will take you directly to the forum from the Tokyo station. If you're at JR Yuraku Cho station, you're only 1 minute away by foot.

Got tips for getting there? Share them here.

JavaOne Tokyo 2005

When you get a chance to go to JavaOne, you don't pass it. And when you get a chance to go to JavaOne Tokyo, well, you really make an effort to make the most of it!

On assignment in Tokyo from Nov 8 through Nov 10 2005, I'll attend the JavaOne conference, cover as many sessions, attend as many BOFs, and discover as many new, cool Java toys as possible. Just how many things can a single person cover? We'll find out, and I'll blog it all right here!

OpenOffice Blog Extension

On a recent trip to Blogger, I noticed a new tool: a blog toolbar for Word. How convenient! You can review, edit, and even post new blogs to Blogger from Word...nice. There's only one problem...what if I can't use Word? What if I don't want to use Word? What if I want to use OpenOffice instead? What if I want to blog to java.net instead?

I've been hunting the internet for an OpenOffice plugin that would allow me to blog to java.net or some other MovableType or Blogger compatible blog host. Ideally it would use the OpenOffice UNO APIs with Java and JAX-RPC to communicate...great idea right! But I'm finding nothing, nada.

That's where you come in...the community. Do you know where something like this exists? If not, maybe it should start up as a java.net project right here...just an idea.

December 2005

Search this blog:

Archives

December 2005
November 2005
October 2005
September 2005
August 2005
July 2005
March 2005
December 2004
November 2004
July 2004
April 2004

Recent Entries

Character Conversions from Browser to Database

JavaOne Tokyo '05: Project Looking Glass

JavaOne Tokyo '05: EJB 3.0 Hands on Labs

This work is licensed under a Creative Commons License.