I just completed the new design of the site.   I’m no longer using just a prefab template, the code is based off of default.

Facebook just got an update.   Is it just me, or is it sort of creepy to be able to stalk every thing all your friends do?   I think the feature is great, amazing and fun, but just a little bit creepy.   Some of the features go to far.   I shouldn’t probably see when you comment on other people’s pictures.   I wouldn’t mind limiting it to only when people change status or post notes.

One thing is for sure, facebook kicks myspace’s ass in general usability.   Myspace still has a lot of nice features for teenagers though, like music integration and customizability.

Random Link:   The 1k Project

I’m moving, so no update today. Also, I am working on a new theme for this site, I am sorta sick of using other people’s themes.

I added myself to the blog tracker Technorati, they made me create a post linking to them.

It is now my theory that they are brilliant, because it is likely many people don’t remove that link, so it is basically technorati built into their website a way to google bomb.

I didn’t want to delete the post entirely, so I’m just editing it to link to a site that links to my blog: Technorati Kaddar  

Brilliant.

It’s interesting: I learned SDL game development a while back.   While learning it, I developed my own model for a graphics memory manager and draw-images queue.

Now I’m working on   a Java2d game engine with a lot of similar features.   I’m finding that more often than not, the features are best implemented in nearly the exact same way.   I’ve done a few changes to my graphics manager and the model of graphic storage for better image ordering and management.
With graphics, in my SDL graphics manager I stored graphics in the graphics manager, and when an entity needed to draw a graphic, it would pass in the x and y coordinates of that graphic, and a reference to that graphic, and ask the graphicsManager to draw it.   This works fine for drawing sprites, but what if I wanted to draw a more complex visual but use the same source image?   For example, animations, color shifted images, blurring, a combination of two images through a filter?
Now, each entity contains a “VisualRepresentation” class.   This class contains the reference to the immutable loaded image , and when I ask the VisualRepresentation to draw, it will allow the graphics manager access to this instance’s properties.     The VisualRepresentation contains only state information relevant to rendering and not information about the entity itself.

Portal looks so goddamned cool. It reminds me of a weapon in the killer quake(1) mod, one that let you make portals to walk through walls.

I just bought a new 300 gig hard drive.   Because of this, I reformatted my hard drive and put the latest ubuntu on it.   I had ubuntu on an earlier setup but to be quite honest, I experienced a lot of strange bugs / memory leaks.   That was before   the release of dapper.

I managed to get dual screens working too!   This is great news for me, as now I can use linux as my programming os.   I’m going to try   it out and see how it goes.

Java game development has been going ok, but honestly, the documentation for java game programming is really bad. I found out the book I was using is outdated and that I should be using the “BufferedStrategy”, but the docuementation for it is pretty sparse. Is the interest just not there compared to flash actionscript / emcascript programming?

A lot of the information I’m getting is off the sun java.net site. I don’t understand why the model of learning things on the internet is as it is. API docuementation is neccessary but not useable for a top-down approach. Tutorials and articles are outdated and too specific to problems. Forums which provide answers to problems over and over, but then do not organize them in a single source location for future understanding. A wiki done right is a good solution, but the problem is wikis just aren’t added too enough sometimes.   I wonder if wikibooks’ problem is that it’s too structured so it prevents teaching anything outside the realm of a book.   I propose we need an intermediary. A documentation style that is top down and bottom up. I’d call it a doctorial, but really it should just be called proper documentation. When people write documentation, the should simultaniously write how to use their API from both the bottom up, and the top down.

I like the whole information hiding in web2.0. To improve the bottom up experience, I could see a documentation system where API’s would have a hidden paragraph that is made visible by hitting a [TUTORIAL +] button, similar to digg’s buried comment show system.

For the top down, an ordered tree of these tutorials (ordered by generalness) with each one having a link to API’s of any referenced classes. Problems would have tags, so you could also find similar problems. This would be the ideal “Doctorial”.

I recently purchased a Wacom Intuos 3 6×8 tablet. It was probably a bit overkill for my art-skill level. I am loving it though. It was worth every penny, compared to the benefits I’d get from say, a new video card. Here is one of the drawings I’ve made with it. I’m getting better at drawing through owning it. It does take some getting used to, however. I like using it with pictionary games like isketch especially.
One thing I’m hoping to use it for is game design, I’d like to draw some of the art for a game I create using it. I have to get further along in a game first, but it would certainly be useful if I did. For some reason, I don’t really like flash compared to java for game development, and I’m looking at what I need to know to program a java2d game.

There are a lot of java books on game design. Killer game programming in java isn’t that bad, but it maybe focuses too much on java3d than I’d like or need. I should probably look for something a little more focused on the latest java2d. I wish wikibooks was a little more professional and complete. I think the incentive to add information to wikibooks is less than for wikipedia, which explains why it has a tendency to suck. Knowledge seems like something that’d be good to keep free. I was trying to learn ruby on rails the other day, and I feel like a lot of that community basically forces you to buy the pragmatic programming books. I guess I understand, it is a pretty good book, but the thing is, there are all these people spending so much time working on open sources libraries for interfaces in ruby, you’d think someone would have the time to write an open source book.

Here is a screenshot of my SDL game written in Dev-c++. I am going to work on getting it complete enough to release this summer. It involves blocks falling and pressing buttons.

I haven’t worked on the site much, but summer’s coming, so things may change. Also, I learned perl recently, so there’s another thing.

It looks like I might be getting on the whole web2.0 bandwagon with loopplay. I was going to do the games as java based, so that I could create side scrollers and single player games too, but let’s be honest, java was not made to make side scrollers (flash was). If I ever get into that, I’d probably end up with flash.
So I am thinking of sticking with board games/card games at the initial point. Assuming I can figure it out, it should be kinda cool, because it will use no slow applets to load data.