My Recent Article may be of interest: Why your games are unfinished and what to do about it

There is a very definite stage in a game developers life where they hit a "well" - they make tons of games and find themselves unmotivated and end up not finishing it. The article hopes to expedite the process

The reason why there is that "well" is because most new game developers try to think of things that are not important. This is my 2 cents to your problem:

What do you want to get out of making games? Do you want to be a programmer? Do you want to be an artist? or Do you want to make/design games?

I'd always say start with something simple and small.Your #2 question is a popular one among newbies, the "it's too simplistic for a 'real' game" excuse. Throw that sort of egotism out. Some people will call you out for it, but these too will be game developing wannabees who use the 'best' tool they could find - but end up not releasing any games. You might have started with a simple tool, but at least you have a finished game in the end.

Also, don't believe anyone that says GameMaker is a 'simple' tool, it has grown (an example of a game from GM: http://godswillbewatching.clay.io/) Note that I personally do NOT use GameMaker, and I am not trying to popularize it in anyway.

Finally, and on your last question, if you can create one game in one day, then you're well on your way as a game developer!

Start small and work your way up.

1. Build yourself a couple text based games. Start with something like Hangman and Tic Tac Toe. If those games seem beneath you, you're wrong. If those games ARE beneath you, then it will take you very little time at all to develop them, and it will still be a useful experience. That experience should encompass the full gamut of basic, required, game functionality. At the bare minimum a Start screen, game menu, robust and flexible Console and File IO, saved games, Computer AI, and high score boards. There's a lot to be learned here, and if you haven't done all of that with minimal debugging required, moving on to more complex games will likely be quite frustrating.

2. Once you're comfortable with console based games you can step up to simple 2D graphics games. Best to start with single screen style games ala something like the old Atari Tank game. This will introduce you to working with 2D graphics and collision detection; and also provide a good introduction to some real Computer AI. Navigation and Combat AI specifically. One of the distinctions of this style of game with regards to AI is that it's fine to give the Computer complete knowledge of the environment because the player also has complete knowledge of the environment. If you're feeling ambitious, go ahead and try your hand at network multiplayer as well. Scrollers can be tackled here as well, but it's probably better to leave these as your second or third 2D game.

Once you've done all that, the answers to pretty much all of your questions should be completely obvious to you.

Thanks for all of the good advice.

I'm a competent programmer, so making the game myself isn't an issue. I ended up going with Flash as a platform, using FlashDevelop. I'm just making a basic shooter for now, so far it works with random enemies and one type of ship, but it's been good practice and I'm slowly adding in more and more features as I go.

I think a shoot'em up game is a good starting point, because it doesn't take very long to have a working game, but then you can just keep adding in more enemies, more player ships, more weapons, upgrades, levels, and so on. It's been great to be able to have a basically 1:1 ratio of me doing stuff to the game improving, unlike some other game types I've tried where you need to know exactly what you're doing from the beginning, plan things out, and it takes a long time before you have something playable and fun.

I'm not much of an artist, but I can do some basic pixel art, so that's defined the style of the game. I don't want to hire out an artist (or request a freebie) on my first game, so I've just kept it simple. Here's a screenshot of the game so far:

At the moment, I'm planning to put it up on Kongregate once it's done, since that seems to match the quality and size of the game fairly well.