April 2007 Archives

This story will without a doubt get a lot of coverage today, and rightfully so. This is a big deal. The premier RIA platform is now the premier open-source RIA platform.

There are so many reasons why releasing Flex as open source software is a good decision. There has been good coverage already, and the story isn't even that old yet. The bottom line is that, in the end, everyone wins.

By releasing Flex as open source, there is no more barrier for adoption. There is no vendor lock in. There is no reason not to give Flex a try and find out just how great it is for building web applications.

One thing that I haven't seen discussed yet is the public facing bugbase. This is, perhaps, one of the greatest side effects of the announcement.

Currently, there is no public facing bug base. When you submit a bug, you have no idea what it's status is... it just gets lost in the black hole that is the wish form. Through the public facing bug database, you can submit a problem and finally be able to monitor its status. When a fix is submitted for it, you can grab that fix right away from a nightly build or through SVN instead of having to wait months for the next official release. Not only that, but you don't even have to wait for Adobe to write the code. Anyone can submit modifications, enhancements, fixes. You can even fix your own bugs and contribute the goodness back for all to enjoy.

We all benefit from each other's work. We all have a chance to make the platform we rely on daily even better.

This announcement will definitely take awhile to sink in. I know the full scope of it hasn't hit me yet... but I know it's big.

Developers! Developers! Developers! DEVELOPERS!

Congrats Adobe! I'm sure this wasn't an easy decision, but it was definitely the right one. We all win because of it.

Relevant links:

• Open Source Flex
• Press Release
• FAQ
• Open Source Flex mailing list
• Tamarin Project

I'm pleased to announce that I've got the as3cards Google Code Project up and running. The as3cards project provides a skinnable ActionScript 3.0 card engine for creating playing card games, with an example implementation of Klondike Solitaire.

You can view the example implementation of Klondike here.

This is essentially a continuation of my over-a-year-and-a-half old ActionScript 3.0 Klondike game, that was showcased on labs back in the early days when we still had Flash Player 8.5.

Originally I had included this in IFBIN. When IFBIN disbanded I wasn't sure what I wanted to do with the code. I spent a lot of time writing it and wanted to try to get something out of all the time that I put into it, so I didn't want to just blindly release it as open source.

Time passed, and I become involved in a lot of other work as my consulting business really picked up. The Klondike code was left to gather dust on my hard drive. After mostly forgetting about it, I've been contacted by a few people recently that put it in the front of my mind again. I figured that since it's been so long and I haven't done anything with the code yet, I might as well just put it out there as open source and let others run with it.

So, I went through the (very old) code base and updated it to work with the latest build of Flash Player. I've committed the source code into the as3cards Google Code Repository, and released it under the MIT license.

Also, I've built a simple Flex 2 shell that loads the ActionScript-only game and plunks it on the stage. You can use as3cards in both ActionScript-only (plain ol' Flash) projects, or you can inject the card engine into a Flex application.

This project was somewhat hard to part with because of my time invested in it, and the potential market of selling it to people that want Flash poker sites... but in the end, I think this is the right decision and I hope that you find it useful. If you'd like to help contribute (with maybe a FreeCell implementation, or a new skin, or anything), let me know. Donations (and/or ad clicks) are welcome, of course, but not expected.

Enjoy!

This morning I just checked in a FlowBox container for FlexLib. FlowBox implements a flow layout algorithm to fluidly lay out children, similar to an HBox but able to span multiple rows.

View the Example Application here.

Download the latest FlexLib distribution from here.

There's nothing really fancy or special about this component. The FloxBox itself pretty much does nothing other than instruct the layout manager to use a FlowLayout instead of a BoxLayout. The FlowLayout code isn't that complicated, but it gets the job done.

Some improvements would be to:

• Write a better measure function. Because the FlowBox can adapt it's children for any width/height, I didn't do anything special with measure. I'm not sure what the preferred measurement should be (1 row for max width min height, 1 column for min width and max height, or somewhere in the middle?), so I just defaulted to the Box measurement.
• The layout algorithm doesn't respect the includeInLayout property. This is a relatively easy change, maybe the next release will see it implemented.
• Children with a percent width and height are probably not laid out like you think they would be. I'm not doing any checking here, so if you have to use percent layouts, FlowBox is probably not the right container for you.
• The verticalAlign style applies on a row-by-row basis, as you can see from the example. When you set the align the "middle", the children in each row are placed in the middle of that row vertically. The entire contents of the container are not centered vertically in all of the available space.

I hope someone finds this useful. I've seen some people asking for a layout container like this, so I thought I'd whip it together this weekend and keep FlexLib growing.

Remember, if you'd like to contribute to FlexLib, see the HowToContribute page.

Enjoy!