Weird Looking: actionscript still sucks

Stick with FlashDevelop.  I get completely lost in Flash Studio, but I rather like FlashDevelop.  Then again, I’m a better at programming than drawing.

Yeah, I am pretty much the same way.  I just seem to have trouble positioning and sizing a video with it.

When I resize the movie to 320x240, the video is displayed at 256x192.  It maintains the aspect ratio, but it’s always at a size smaller than the movie.  I’d like it to take up the whole thing.

I’ve tried setting the width and height in the swfmill XML, I’ve tried many combinations of _width and _height for the video and the movieclip that it’s embedded in from actionscript…  It’s just a bit frustrating.

Pirate? Well, that’s hardly the image we want for Long John Silvers!

I love the duplicitous nature of people who make money writing software pirating software.

Hey, I wrote some software pirating software once!

I used an event hook to run code in the target’s address space.  The code rewrote its import tables to replace various functions used to uniquely identify my computer to the remote server.  Returning random id’s allowed me to be in a perpetual free trial.  I don’t know if this is a normal method, but surprisingly it worked.

That reminds me of a long story.

I was once trying to write a program to randomize system sounds.  Since there’s no system event for this in Windows, I decided to try to hook the PlaySound function in all user processes and change the registry setting for that sound immediately after playback.

It worked well enough, but it turns out that a great deal of system sounds are not generated by user processes, so all it ended up randomizing was the Empty Recycle Bin and New Mail sounds.  Sounds played by message boxes and by clicking on modal windows are apparently generated by a kernel process.

So I figured out how to hook the kernel process, but it involved changing a registry key to cause my library to load at system startup.  I did this, rebooted my machine, and almost immediately hit a blue-screen-of-death STOP error.

Apparently I did not realize that my debugging code, which displayed a message box to indicate that my event hook was firing, was causing the kernel to crash.  This was because the kernel does not have a desktop in order to perform any kind of user interaction.

I was unable to boot my machine, even in safe mode.  I had to put in my Windows disc, run the recovery console, delete my library, and reboot.

At that point I decided that I was not comfortable with my code running in the kernel, and what I was trying to accomplish was not worth creating a potentially unstable system.

I think if I ever try to write the program again, I’ll use an idle-priority thread to randomly change the registry settings for system sounds.  Since frequently-read registry settings tend to stay in memory and not on disk, it should perform pretty well.