I have several ideas for some games I think would be really fun, so I'd like to get started with game development and learn how to make games the right way. I have some experience with programming, but most of my experience lies within web development which is what I've been doing for the past 5 years. I've dabbled with game development in the past with C++ and the SDL library (before I started webdev) a little, but nothing serious -- just a pong and tetris game, and a (very) simple 2d shooter.

But now I'd like to take this more seriously, and actually get started on real game development. The problem is I'm not sure what to do right now. Should I re-learn C++ (I haven't used it for several years) and a graphics library like DirectX or OpenGL, or should I use another language like Python? Also now that HTML5 technology is supported on most up-to-date browsers, would developing games with a library like LimeJS be practical?

Any ideas?

Want to just make a game then use an already built engine like Unity, which is easy for beginners or UDK which is a bit more tricky, to some. These are 3D engines. Torque 2D for 2D games. Here are a list of engines if you want some in specific http://en.wikipedia.org/wiki/List_of_game_engines all have there pros and cons and type of game they are suited for, but can also work for different types of games. Such as, UDK is a fps engine, but it is capable of making a board game or RTS tile based game.

if you want to understand basics of how games are made then you want the low level stuff like OpenGL or DirectX are the popular ones. I personally was interested in making a game so I went straight into an engine, Unity since I was new with an engine environment. After a couple of weeks I know a lot about it and can crank out games pretty quick provided I had some assets.

Here is a simple game that was made for global game jam this past weekend. http://globalgamejam.org/2013/super-morbid I know it isn't AAA quality, but it was fun making and hey I actually made a game.

Design the game completely before you actually start making it. That way it's harder for the project to actually fall apart, because even if team members don't contribute and/or leave, you'll still have the end goal of the game written down. That means that, yes, you'll be doing more writing in documents and having meetings with people before you actually start drawing anything or coding anything, but less time will be wasted overall once you get to the actual creation stage.