Hope it goes better for you than it has for me. Every time I've tried to use Rails, the experience has been a weird mix of wonder and revulsion. Some parts are super cool, some parts are unbelievably horrible.

Digdogger in Oracle of Seasons is like the worst boss ever or rather I hate that fucking ball.

http://www.youtube.com/watch?v=1fUb1JQwL8M

I actually really liked that fight. The guy in the video is clearly having a lot of trouble controlling the spike ball, though...don't think it gave me anywhere near that much trouble. Seemed like a cool mechanic, but I guess it's not for everyone.

i read an average of 3 books a day when i was 15

Holy shit. I read probably an average of 3 books a year. Wish I had that kind of time and focus.

In one of my games, I showed an animated screen with all of the important stuff on it the first time a player tries to start a new game (after that, the screen is accessible from the main menu). I was happy with it, and it was pretty well received. The text updates with any changes the user makes to their key bindings, and the character animates exactly the way things look in the game. Probably wouldn't work for a game with a lot you need to know to play it, but for this simple one it did the trick beautifully:

There is potential there to be revealed. There are some games that will only be possible with the kind of cloud server system that Microsoft is capable of providing.

Any examples come to mind? All this cloud gaming stuff I've been hearing about has always seemed woefully impractical and mostly useless to me, but maybe you're seeing an upside I'm not.

I assume you're talking about the black seams at the edges of your textures? That's an annoying issue not specific to Windows 8 or LWJGL, with a variety of possible solutions...the easiest may be to simply pad your texture atlas a bit by making sure each individual thing in it has at least a pixel of empty space all the way around it. You might also be able to fix it by insetting your texture coordinates slightly, but that can cause other artifacts too. To fully understand what causes it and how to fix it, you'll have to get familiar with exactly how texels are sampled and what kind of rounding affects it how (which I was once, but I've forgotten the exact details).

A crash usually comes with some information about what went wrong (and if it doesn't, better diagnostic tools are needed). What does your stack trace look like?

I've tried programming but I think it may be a little too hard for me. I've also tried designing in photoshop or 3D animations software but I don't think im creative enough.

Are these things interesting to you? Do you have a desire to create stuff with them, but it's just out of reach due to not being familiar enough with your tools? My position is that any person can learn any skill, creative or not; it just takes time and dedication. If you're not dedicated or don't have enough time to spend, you won't be able to get anywhere. As long as you're passionate about something, if you stick with it, you'll eventually master it.

I don't think 15 is too early to be thinking about this stuff. It might help not to think about it in terms of career, though - what do you really enjoy doing, or really wish you were able to do? If you have some sort of distant future goal, it can potentially help you figure out what you need to do today to get there. If you're not sure what you want to do yet, maybe it's not time to choose - try out a variety of new things without worrying about whether you want to dedicate your life to them, and when one really grabs your interest, maybe that'll become your passion.

When I was 15, I had some vague ideas about what I wanted to do with my life, but nothing definite. I was just thinking about how to earn money. I was lucky enough that there was a software company in my town that wasn't too stringent at the time on who they hired, so I was able to get a low-level audio editing job with them. It got me into the work environment, and became a stepping stone that allowed me to switch to a programming position a few years later. I was self-taught and didn't know what I was doing in a lot of areas, but they guided and educated me, and by the time I left the company (9 years or so later), I was regarded as one of the best programmers they had. So, keep your eyes open for opportunity - even if it doesn't seem great at first, it can eventually lead you to a really good place in the future.

In line with my functional design style, I try to avoid sound effect pollution. Usually there's only SFX for things that need the player's attention. Jumping doesn't need a loud "BOING" to inform the player that they've successfully pressed the jump button, and steps shouldn't sound like high heels on a marble floor. I understand some games use SFX to create atmosphere, but music is my way of doing that.

Interesting perspective. My take has always been that when you're talking about sound effects, more is better. I try to attach a sound effect to pretty much every possible event in my game that could be used to trigger one, and each one I add makes the game feel more alive. Sound effect pollution can certainly be an issue, but I try to address it by using quiet or subtle sounds, keeping frequencies from overlapping too much, and other things rather than choosing not to attach a sound at all to a particular event. Sounds related to important events (usually) need to be the most noticeable, but unimportant events contributing quietly to the soundtrack can add quite a lot of texture to the game.

Maybe having the game engine already done makes developing a sequel easier? If Fez 2 even uses the same 2D/3D mechanic, that is.

Based on the video, I get the impression it's 1D.

New Super Mario 3D Land Bros. Wii U 64 looks pretty cool. I'll probably buy one for that. Nothing else on it has really caught my eye yet, though...

My AS3 is pretty rusty, but as I recall, there's an optional argument you can pass to addEventListener to make it use a weak reference. That way, you don't have to remove the listener explicitly to avoid leaking. Might help a bit.

Rendering with a narrower FOV when zoomed in will probably give you the effect you want.

More specifically, I would like to have a basic idea of how much such data is usually reasonable to hold in memory at one time... could I get away with doing this for a few hundred entries? A few thousand?

Hundreds of thousands, most likely. Even using the most inefficient way of storing it, this kind of data will be a tiny drop in the bucket compared to any audio, bitmap, or video data.

Your approach sounds pretty reasonable. If it were me, I'd make it more explicit by creating an ItemModel class, which knows how to parse XML to fill out its ivars. By doing this explicitly, you can catch errors in your data at startup when all of the models are initialized, rather than having to wait until you access a particular resource to find out it's missing one of its required attributes. Depends on how much you want to trust your data, though - if you're OK with assuming it will always be correct, you can get away without the extra safeguard. I like to make a habit of assuming any data that's not physically part of my executable can and will be wrong, which I've found makes my life easier in the long run.

Well I'm grumpy because I'm working in an organisation that uses SVN. We've joked about getting a "I broke the repository" hat that would be passed along to the latest person SVN decides to mess with

Weird, svn has always worked extremely well for me. I've been using it both professionally and on my own projects for several years and never had the issues you guys are describing. It sounds like the main problem is the vulnerability of having just one central place where the repository lives? It seems to me like consistent backups would solve this just as well as using a distributed version control system. I'm not at all convinced of the benefits of git/etc. over svn.