David Van Couvering 's Blog

May 2007 Archives

Google Gears - The Giant Speaks

By now most of you have heard about Google Gears. Yes, I heard about it too. I actually tried installing it into Firefox on my Mac, I wanted to try Google Reader offline. But it just is not working for me. First of all Google Reader keeps telling me I have lost the connection when I haven't (or probably did for a microsecond). And I can't tell it to shut up or increase its timeout.

Secondly, the "download" icon is not showing up, although it seems to be for other folks. So I can't even use it, and it's annoying, so I uninstalled it. I have posted to the Google Reader forum, we'll see if I get an answer

But this is just growing pains. The concept is great, and it looks like they've done a good job. SQLite is an excellent little engine. SQLite is also the basis of Mozilla's storage engine. You can also find SQLite embedded in Solaris as the database used by the Service Management Framework.

I've met the author, Dr. Richard Hipp, and this kind of success couldn't happen to a nicer guy. I just checked the web site, and the license has changed to be a more official form of public domain, but when I was first looking at this as part of evaluating databases for use within Sun, it used to be a single paragraph that said something like this software is free for your use and is given to you fully with a blessing

As I heard it second-hand, this license text set our lawyers in a tizzy :) but finally it was decided it was OK to use.

All that said, OK, yes, I'm a little bummed. It's great to see this kind of architecture making it out there, and it looks like Google did a good job (for instance I think it's right to leave synchronization to the application). But, yes, I had hoped JavaDB would have been in the mix.

But I understand why Google may not have chose it. Java is not famous for being fast and easy in the browser, even though it is very portable and has a strong set of APIs. It looks like we are working to change that, but too late for Google Gears.

I think Mark Mclaren had a good point too, that this was shoved onto browsers in a non-standard way, and only Google could do this, because they could. But I also recognize that sometimes you have to do this - get the implementation out there first, standardize it second. It's encouraging that it's in open source under a BSD license.

I was wondering how this will impact the very interesting work Brad Neuberg has done with the Dojo Offline Toolkit. I watched his demo and it was quite impressive how simple it all was, and how he has tried to tackle the Synchronization Problem. I like his vision of keep things very simple for the user and don't confuse them with too many choices.

But then I saw that Brad's presenting at Google Developer Day about the integration between Google Gears and Dojo Offline Toolkit. So there you have it. Good to see that they've been talking.

Ethan Nicholas Walks the Walk

I remember noticing when Ethan Nicholas said he was coming over to Sun from Yahoo! after whining about the issues with Java on the desktop and proposing some solutions. When Sun hired him, I was hopefully holding my breath because (a) I didn't know Ethan that well and was hoping he was up to the challenge and (b) I didn't know if Sun would let him do his job (it's not unique to Sun, all big companies can get in the way of the innovators sometimes).

Well, both Ethan and Sun appear to be delivering. Ethan's latest blog is not just fluff about a Consumer JRE but real details. I look forward to seeing this stuff in action. Good luck Ethan and the entire team! And congratulations to Sun for giving him the support he is needing to make this happen, and happen quickly.

Update: I re-read Ethan's original blog. I had thought he was hired to create and lead the team, and now I see that's not so clear. I still think Ethan's doing great work, but I want to make it clear I absolutely know there is a whole team behind this and not just Ethan, and hats off to all of you.

Meet the new life, just like the old life

Michael Arrington talks about the dark side of Second Life. Some of what he talks about is sad and ugly (such as issues around virtual rape and pedophilia). But the antics of the SLLA (Second Life Liberation Army) made me laugh, although I am sure they are damaging and annoying, and probably cost people time and money.

Proving that Second Life is indeed a reincarnation of the Wild West, the Australian Broadcasting Commission (ABC) was literally blown off the face of Second Life Tuesday by “miscreants” according to The Sydney Morning Herald. ABC Island featured a number of venues where visitors could listen and dance to music, attend talks and watch videos of selected ABC programs before it was reduced to something more like Hiroshima circa 1945.

Sadly, this type of attack is not new. The self styled Second Life Liberation Army have previously attacked Reebok and American Apparel stores in Second Life and stormed the stage at the January 2007 meeting of the World Economic Forum in Second Life.

The image of a liberation army prowling around a virtual world just made me giggle. It reminded me of the Life of Brian, with the Judean People's Front and the People's Front of Judea. People are people, and the dynamics are the same wherever you go, even into your Second Life.

Great news for Java DB on Internet Clients : Consumer JRE

Great blog by Chet Haase on the new Consumer JRE. I'm very excited about this because the issues they're grappling with and addressing are pretty much exactly the issues that can make Java DB on a consumer desktop a pretty hard sell.

People like Zimbra are still using Java DB for this because the solution is so compelling, but solving these problems are going to make the barrier to entry a lot lower.

I also like this because it feels like the Java team is really "waking up" and are going after areas that have for a while felt like they were relegated into a dusty corner of the platform.

Great to see this, and good luck to Chet and the Consumer JRE team!

Connecting to an embedded Derby database from another process

A common question we get on the derby-user list is how to talk to a Derby database running in embedded mode from a separate client, such as a report tool or visual query tool.

You can do this by starting a network server in the same VM as the embedded database.

The code is pretty simple

import org.apache.derby.drda.NetworkServerControl;
import java.net.InetAddress;
NetworkServerControl server = new NetworkServerControl
	(InetAddress.getByName("localhost"),1527);
server.start(null);

Then you just connect to it using the standard network client URL, e.g.

"jdbc:derby://localhost:1527/sample"

Glassfish in less than a second

I know I mentioned this in my overview of the JavaOne keynote, but if you blinked your eyes, you may have missed it. It took just a second, but I keep remembering my wonder and amazement at seeing it, so I thought I should give it its own blog entry.

I've both built and used app servers for many years. I am so used to line after line of log output as the beast starts up, loading all of its various components, 99% of which I will never use. A good three to five minutes later (if I'm lucky), it's up and running.

The Glassfish team has done some impressive work to modularize the behemoth called an appilcation server. Before my very eyes they showed a startup time of v3 of less than a second. Less than a second!.

According to the person doing the demo, this magic occurs because only a basic kernel that knows how to serve up static pages is loaded initially. Then, when you deploy an application, Glassfish figures out what other modules you need and automatically loads and starts them. Each time, the deployment time was no more than a second. This works for JRuby, for Phobos (JavaScript on the server) and for regular Java EE applications.

Later I had lunch with an old buddy who now works for a competitor to Glassfish, and he told me that when he saw this he sent an email to his coworkers saying "I warned you! We're f***ed". I have to say I think he has a point there...

Relevant Relatives

My little sister, Elizabeth Van Couvering, is in the graduate program at the London School of Economics (lucky cad, she was born in England and has dual citizenship). She is doing her work on the social impact of search engines, and just recently published an article entitled "Is Relevance Relevant? Market, Science, and War: Discourses of Search Engine Quality."

From the abstract:

The evidence presented here suggests that resources in search engine development are overwhelmingly allocated on the basis of market factors or scientific/technological concerns. Fairness and representativeness, core elements of the journalists' definition of quality media content, are not key determiners of search engine quality in the minds of search engine producers. Rather, alternative standards of quality, such as customer satisfaction and relevance, mean that tactics to silence or promote certain websites or site owners (such as blacklisting, whitelisting, and index "cleaning") are seen as unproblematic

I find it not more than a little ironic that my big brother, Antony Van Couvering, has been in the Internet business for a while, and recently has a venture entitled Names@Work which, among other things, helps people promote their business through search engine optimization.

In a complete non-sequitur, my older sister Anne Van Couvering is an excellent massage therapist living in Long Island, and is about to go to school in Seattle to become a naturopath.

Disclosure: I am related to these people, I grew up with them, and I think all three of them are the bees knees.

That Little Issue of Durability

I think there is a lot more to choosing a database than performance. Yes, performance is important, and all other things being equal, it's a very good way to help you choose a database vendor.

What's problematic is that published performance numbers can often be misleading.

As an example, one place where Derby often gets compared is with other embedded databases, often showing where it's a big cumbersome oaf next to these screaming meemies. Take for example this comparison between H2 and some other databases out there. Derby in particular looks like a dog. Sigh...

But wait! Brian McAllister brings things back down to earth when he makes H2 and MySQL actually write to disk on transaction commit.

-- 1024 bytes, 1000 times --
    derby:		2192 millis
    bdb:		1849 millis
    h2: 		2221 millis
    h2b:		2351 millis
-- 2048 bytes, 1000 times --
    derby:		2199 millis
    bdb:		2129 millis
    h2: 		2578 millis
    h2b:		2414 millis

Hm, now Derby isn't looking so bad...

Now, I'm more than ready to say Derby has room for performance improvements, and I do know that the Derby team is working on this. But it's good to keep the story straight. You may not care about durability, in which case not writing to disk is a good choice to improve performance. But it's good to know what you're getting in to.

By the way, one of the folks I talked to at the Java DB pod during Java One wrote his own in-memory database, and is interested in adding support for an in-memory storage engine for Derby. In which case we could see some kick-ass numbers.

There is also the poor-man's way of running Derby in memory: set derby.system.home to /tmp or use the URL "jdbc:derby:/tmp/mydb".

Running a SQL script from Java in Java DB

Most databases provide a command-line tool to run SQL that you've put in a file. But sometimes you want to do this straight from Java.

A recent thread on the derby-user list included an email from Dan Debrunner about a Java API that exists in Java DB that lets you do this. Something to keep in your back pocket in case you need it.

Say "No" to Schwag

I just learned the term "schwag" at this JavaOne. I am so behind the times. I don't even know if I'm spelling it right. But as I understand it, it's all the "stuff" vendors foist on you at these conferences -- T-shirts, pens, USB hubs, laser pointers, memory sticks, badges, little flashing logos, you name it.

I mentioned in an earlier blog the lengths I went to (including giving away my soul) just to get some water-bottle schwag. On an internal list at Sun, somebody said it didn't make a lot of sense to hand out big green plastic bottles that end up in landfills as a motivational force for bikers. That hit home; I had to admit it felt kind of odd. On the same thread, somebody said that perhaps we should eliminate schwag altogether.

Well, now that's very interesting. Sun and other companies talk a lot about reducing carbon footprint and the overall ecological footprint a conference has. Well, let's start with the schwag.

Here's my idea. Let's put those Java Cards to use that we get at the conference. When we register, rather than picking up our goodies (the nifty backpack and all the ads and CDs), we scan our card over a "no thanks" reader, and we get a certain number of "eco-points." Each time we say "no thanks" to a vendor for some schwag, we get another "eco-point." Eco-points for riding to the conference. Eco-points for not grabbing one of those massive program guides.

At the end of the conference, we cash in our eco-points for some real $$$.

The only problem is, vendors love handing out schwag as it increases awareness for their product. I think JavaOne could just say "too bad, that's how this conference runs." But we could also reward vendors for reducing waste by giving them kickback for how many eco-points they hand out rather than schwag...

Anyway, that's the high-level idea. Lots of details to work out. But it's time for us to move away from the huge amounts of waste generated by schwag to another, more eco-friendly model.

By the way, thanks, IBM for the pen (they did a drive-by schwag drop at the Apache Derby booth in the .orgZone). I can "take notes" with it at my 1-1 meetings with my manager, just to keep him honest :) :)

Technology as a positive force

You may have read my blog about my concerns around technology. At the same time, I also believe that if you are conscious and committed and vigilant about how you use it, technology can be a huge enabler for helping make a difference in people's lives. That was actually one of the key reasons I chose computers as a career path when I first started in the mid-eighties.

I think I'm not alone in this belief, nor in the desire to make a difference and not just working for The Man.

One person who is on this wavelength is James Governor of Redmonk. A few weeks ago he blogged about Shai Agassi leaving SAP to work full-time on providing an infrastructure for electric cars. In it he says

Shai may have just tired of the shackles of SAP’s success. The future of the planet is surely a more important Dip than the success of SAP. Like many in the emerging green economics Shai is setting out to save the world or get rich(er) trying.

This blog got me thinking about how more and more people at fairly high levels in the tech industry are seeing they have a duty and obligation to try and turn things around, while not sacrificing the more mundane goals of keeping bread on the table.

Then last night at the Apache Derby BOF, James was telling some of us about his plans to kick off a social networking site dedicated to helping the environment, and today he announced it!.

This is great stuff, and I think we'll see more of this as we in the tech industry become increasingly aware of the gravity of the situation we're in. I know I'm thinking about this, and I hope to find some way to use my skills to make a difference while still fulfilling my (very joyful) duties as the provider for my family. So, stay tuned, and maybe some day you'll see me make an announcement like James' :)

Java DB Booth is hopping!

The first year we had a booth for Apache Derby/Java DB at JavaOne, it was kind of sad. Nobody had heard of us, or even understood what we were about. We spent most of our time chatting amongst ourselves and with other Sun employees.

The second year was a little busier, but mostly it was "so, what is Java DB anyway?"

This year couldn't be more different. I thought I'd drop by today just to see what was up, and I was there talking to people and answering questions non-stop for about three hours! It was like that at lunch time yesterday and today too. A number of people with questions just walked away because I wasn't able to get to them on time - I told our pod organizer that next year we need at least three people at the pod during busy times, preferably four. We just couldn't keep up.

There were still questions about "what is Java DB" but there were also tons of folks using Java DB in production. One guy I talked to had heard about Java DB last year at Java One, and I had talked with him at our BOF, and he had said "I think I really want to check this out." I didn't think much of it until he came up this year, and he's fully in production with Glassfish and Java DB, and he just loves it, couldn't stop raving about it. He said he's trying to convince others on his team to move off of HSQL. They are having all sorts of problems because every time someone shuts their laptop and it goes into standby, the database gets corrupted. He said they put JavaDB through a very intense serious of crash tests, killing it in the midst of processing, and it just never ever corrupted the database.

Another guy says he bumped into it when they were trying to build a system for a card house where the dealers needed to record what they were doing, and they couldn't guarantee a connection to the back-end server. So they cache data in Derby in their on-the-table systems (real old, cheapo boxes), and it works like a charm, he says he couldn't be happier.

A number of mobile providers also came by wanting to learn more about putting Java DB on CDC.

There were also quite a few folks who wanted to get involved as contributors to Apache Derby. I met one guy who had built his own in-memory database and now he wants to work on contributing an in-memory storage engine for Derby. Another guy is very interested in really fast bulk loads, and when I said we really didn't have much in that area, he said this was very important to his company, so he'll make sure he gets involved and contributes this functionality into Derby. Wow.

So, I was quite happily surprised. The derby-user list is actually pretty darn quiet, and this can make you think that nobody is using Derby. But actually I'm beginning to believe what this really means is that lots of people are using Derby, but they just don't have a lot of issues. I know that's true for us in NetBeans - it Just Works.

JavaOne Session: What's New in JPA 2.0

I went to the Linda Lemichiel's session on what they're planning JPA 2.0. I thought I'd summarize what I heard. Note - at the end of the blog I provide an email address you can use to provide feedback for what you want to see in JPA. This is really important. JPA is a crucial technology for Java, and we are all responsible for making sure it is delivering what we need to build successful database applications using JPA.

Continue Reading...

Leave your bike, lose your soul

I took my bike to JavaOne yesterday. I didn't really need to, it's a short walk from BART, but I was late, and I wanted to get the free water bottle offered for anyone who brought their bike to the valet service being offered by the San Francisco Bike Coalition.

That all went well, and I got my bottle.

But I may have gotten it at a very great price. I had to sign a little agreement on the valet ticket. As always, you have to be careful what you sign:

We will give your bike to the holder of this ticket. Do not lose it.

Because we cannot verify the condition of your bike nor inventory individual items, the SFBC valet bike parking service is not responsible for scratches or attached items. Also, we keep your soul.

Now what? When I die, will I find myself in Bike Rider Hell, where I am forced to ride on tiny shoulders with speeding drivers whizzing by and cursing at me and dogs chasing me and nipping at my heels and my chain slipping off every ten minutes? All for a stupid water bottle...

(P.S. if anyone knows how make Mac PhotoBooth show letters the right way instead of mirror image, please let me know...)

JavaOne Party Conversations on XML and JRuby

Well, I did go to the Derby UnBOF last night, good to see some of you there. It was quite hopping, and I got to see a lot of old friends and meet some new folks. I met some folks from an insurance company who told me that what they want out of a database tool and framework is the ability to map directly from XML into a database and then pull elements out of the XML tree directly from the database. We got into a discussion over the semantic dissonance between XML, which is hierarchical, and the relational model, which definitely is not. I proposed that perhaps the right approach is to use the law of transitivity (A=B and B=C means A=C) and make use of existing existing tools and APIs to go from XML to objects (JAXB) and then from objects to databases (JPA). The problem with this approach, of course, is performance -- you're doing two transformations instead of one, with the associated data copies. What do you think. Is tooling for direct mapping from XML to databases a high priority?

Also a great conversation with Charlie Nutter and Thomas Enebo from the JRuby team. I mentioned to them that although I like the instant feedback you get with dynamic languages like JavaScript (I haven't worked with Ruby yet), I don't like the number of code/run iterations you have to go through to debug basic issues around syntax and correctness, whereas with Java, because it is so strongly typed, if it compiles you have a very good chance it will run correctly.

Thomas argued that well, that may be true, but with Java you have to write so much more code to get the job done, and Charlie argued that getting it to compile and run is just the beginning of your development cycle -- something I have to think about further. I'm not sure Java vs. Ruby affects the effort to build an app beyond coding.

That said, they also said they are working hard on reducing the number of iterations for JRuby by giving as many hints as possible in the IDE. That's encouraging. People I talk to really like working in Ruby, so perhaps it's not as bad there as I have experienced in JavaScript. I'd love to hear your experience with this.

Convergence and Java FX

David Berlind writes a very interesting blog with his take on convergence onto a rich Internet experience on mobile devices. Check out his video interviews of Jonathan and Rich.

One comment that David made in his blog was very interesting to me:

Of the three companies all vying for a converged market, Schwartz believes that the current global footprint of Java across devices and operating systems along with Java's open source nature makes it the leading candidate over Adobe's Apollo and Microsoft's Silverlight. ... That could be the case. On the other hand, with Adobe, Microsoft, and Sun coming from their relative areas of strength, the one equalizing factor may end up being Moore's Law. If for example, Sun can now fit Java SE into handsets in a way that it could not before, whose to say that hanset manufacturers won't soon be able to fit all three runtimes in their devices (something that desktops already do with grace). In other words, for developers, it may not be either/or.

Well, to me, that does sound very possible. And that takes me to a conversation I had at the Apache Derby unBOF yesterday evening, I chatted with some folks from Nitobi. These guys are delivering rich components in AJAX, particularly components that handle large data sets.

I asked them what they thought of Java FX, and they asked the question I suspected would be asked a lot: "how is this different from Flash?" Not being a scripting/Flash developer, and being very new to Java FX, I didn't have an answer. They were concerned that as developers they would now have to support multiple solutions with no standard between them. And David Berlind seems to be saying, this could very well be the case. The hardware platform may not be the forcing function everyone is hoping from just because it's going to powerful enough to support all the popular runtimes.

But then they proposed a solution that was very interesting to me. Flash, Java FX and Silverlight all employ a declarative model for defining/configuring rich components. What they would do is define their components in XML, and then use XSLT to output the appropriate code for Java FX, Flash or Silverlight depending on what's available on the client. That sounded pretty compelling to me.

What do you think? I'm just brainstorming; I don't do this for a living.

JavaOne Tuesday Afternoon General Session - Java Technology

Well, there was a lot of interesting stuff at this session.

One thing great to see was the ability to run Rails apps on Glassfish, and being able to build and deploy this straight from NetBeans. Great demo from Tor Norbye and Charlie Nutter showing how to take advantage of Java in a Ruby app, code completion, instant rename of variables, and so on. I do have to learn this Ruby stuff soon. So much to learn, so little time.

Danny Coward talked about the future of Java SE 7. The two areas that were very interesting to me is the work on modularization support and the focus on enabling Java on the Internet client.

The planned work for Java SE 7 includes "superpackages" and "super jars". Super packages will allow you to give much finer grain control of what a package exposes and to whom. Super jars is an upgrade of jar files to allow for really using jars for application and module packaging. This includes versioning, defining dependencies, and to make sure the module system is interoperable with existing frameworks such as OSGi.

Something I have been wanting to see for a while appears to be planned for Java SE 7, and that's to spend some focused time for Java on the Internet client. This includes defining a "Java kernel" which is the base VM required to run Java, and then the rest is downloaded "in parallel" (I'm not sure what that means, but I think it sounds good). As someone who has worked on applets and seen the cost of downloads, this is very exciting.

There was also a demo of Glassfish v3, which has gone fully modular. What this means is there is a tiny kernel that is sufficient to serve static HTML, which starts up in 463ms (wow!). Then, when you deploy an app, be it Ruby, Phobos, or a standard Java app, Glassfish figures out what you need, and then loads the appropriate code. Very nice!

Bob then surprised me by dropping my name about the blogging I have done around running Java DB in an Internet desktop, and said that with Glassfish v3 you can now have an app server small enough to deploy to the desktop as well, which truly enables a fully offline application. I need to look into this more, and I'll get back to you with my take on this.

NASA World Wind - Wow, this was amazing!. An open source, 100% Java geospatial package. This is Google Earth that you can embed and "mash up" within your Java application using standard Swing programming. I love this because it is opening up the geospatial platform and making it available to everyone, without it being owned by a single company like Google or Microsoft. Definitely you need to check this out!

There was also a great demo of Project IRIS. A very nice demo showing the power of Java on the desktop. The 3-d browsing was beautiful, you can go full screen, and you can drag-and-drop photos directly from the desktop.

What was very interesting was that it was a lot of work to build this with HTML and CSS. Chris Oliver, the inventor of Java FX Script, was able to rewrite this in Java FX Script in 3 days. Wow.

There was an in-depth demo of how Java FX Script worked, but I couldn't see the code where I was sitting and it was just a bit too much to take in in this format. I need to sit down and read this stuff, and you should too.

Someone sitting next to me asked what I think will be a very common question - how does this compare to Flash? Having never used Flash, I didn't have an answer to this, but I'm sure there will be lots of blogs and articles trying to answer this specific question.

Java One Tuesday Keynote - Announcements, Announcements, Announcements

Rich Green had a number of very interesting announcements today. You'll hear about them from others, but I wanted to share the ones that interested me most and give you my take if I feel like spouting my own opinions

First of all, Rich announced that Sun has completed the process of open sourcing Java, a promise they made a year ago and which has now been finalized. Also of interest was a vague-ish announcement about the opening of the TCK. I couldn't fully understand the implications, but I'm sure we'll learn more in the days to come.

Another very interesting announcement was of a new "bundle" of tools, technologies, and programs called JavaFX. The intention of JavaFX is to make Java available to "humanity" (e.g. consumers) on a much wider scale. Around this were two big announcements.

First, Rich and James Gosling announced a new scripting language called JavaFX Script that is intended to make it very easy for creative content producers to build rich internet content on the Java Platform. This seemed to me to be a very strong response to the other RIA platforms out there such as Adobe Flex and Apollo. I'm interested to see more of what this looks like, but what I saw demonstrated looked very interesting indeed.

The other big announcement was Java FX Mobile, which takes a full Java SE platform and puts it on the mobile platform. Exactly what this means is a bit vague to me, but what caught my attention was that it is the full Java SE platform and that architecturally it provides an operating platform that runs on top of the device hardware. As a non-mobile developer, this surprised me, but I guess traditionally the hardware and the operating environment are very hard-wired together. So Java FX Mobile is like a Java OS for many different devices. It will be produced in an open fashion and available for anyone to take and deploy to their device.

Jonathan then got up with a gentleman from the UN who is in charge of bringing about positive social change through youth sports. Their vision is to use Java and Java FX Mobile and make it available on mobile devices that are affordable to everyone, and then this platform can be used to enable social change. As an example, Scott McNealy stood up and talked about Curriki, something he has been behind for a while now to bring the full K-12 educational curriculum to anybody anywhere, absolutely free. A very interesting vision - using technology to enable social change, rather than just a new game. I hope it's more than just marketing spin on a greedy goal to get everyone on the Internet and buying stuff, but a true vision where everybody wins.

More later, off to the session with Francois and the Zimbra folks talking about using Java DB to take Zimbra offline. I have what I call a "leftovers badge", where I can get it only if the true paying attendees have all gotten in and there is still space. So wish me luck!

NetBeans 6.0 Preview Release Available

The NetBeans 6.0 Preview Release is available and you can download it here.

I have been playing with NB 6 for a while, and there are tons of new features I have never touched. One of them is support for Beans Binding, which I am very interested in but haven't had time to look at. Josh Marinacci talks about it here, and the comments on his blog are impressively positive and excited about this technology.

What do I like about NB 6? I have to say I really like the editor. It's faster with its completion, and it has all sorts of new hints. I'm still learning the ins and outs, but I am very impressed.

I also like how integrated the Visual Web functionality is. It's no longer an add-on pack - it's part of NetBeans proper. You just create a Web project, and you choose a "Visual Web JSF" framework, and all the functionality of the Visual Web Pack is there for you. No separate download, no separate installation of a "pack."

Another area I am excited about but haven't had time to investigate much is the UML support. I always like defining my architecture in terms of UML, just as a way to help me understand it, and now it's in there with the rest of NetBeans - I don't need a separate tool.

So, check it out, and if you're in town next week, come to NetBeans Day on Monday (it's totally free), and go to the session on NetBeans 6 to get a full overview of what's coming in the next release.

Apache Derby UnBOF and Blogger's Bash at JavaOne

Just a few quick notes about some fun events going on next week at JavaOne.

On Tuesday night from 7pm - 11pm at Jillian's in the Sony Metreon there will be an Apache Derby UnBOF, and you're all welcome! Last year we did this and it was a rousing success, with as many database geeks as you could shove into the restaurant. Some analysts snuck in too, including James Governor from RedMonk and David Berlind.

On Wednesday night at 6:00 at the Thirsty Bear (just down Howard from Moscone) is a big Blogger's Bash, and I hope to be there too.

So if you're in town, drop by to one of these and come and say hello -- I've talked with many of you through comments and all I know is your obfuscating java.net login - it would be nice to meet you in person and get to know your real name.

Taking Zimbra Offline using Apache Derby: David Berlind

When Francois Orsini demoed how to use Derby for local storage in a browser, and after some prodding from Simon Phipps, this in turn got picked up by David Berlind of ZDNet. He was very excited and was predicting Great Things for the web because of this.

Fast-forward about a year. I was in the room at the Web 2.0 Forum when Zimbra announced and demonstrated offline support. I was impressed and interested, and wondered how they did it. I was quite surprised and pleased to later find out that they used Derby to accomplish this, and that they based their implementation on Francois' demo.[1]

In this article, David Berlind follows up by talking with the Zimbra folks and giving his take on what this means for Zimbra and Internet applications in general. He is definitely quite excited about this:

By equipping it's Web-client with offline capabilities, Zimbra has put itself within reach of a Holy Grail that many said couldn't be found. Other vendors with similar interests and with the sorts of resources that dwarf those of Zimbra — companies like Google who will want the off-line problem solved for their own Web-apps as well — can't be far behind.

[1] Francois and Dan Karp of Zimbra are going to be giving a talk about this at Java One if you're interested in finding out more.

This work is licensed under a Creative Commons License.