This is a good article. Kudos for being exactly what the title suggests, and not the sarcastic opposite.
I had some big problems with certain parts, however:
InnocentSam wrote:#3 - Good at only one thing? Do not form a team.
Isn't the point of a team to have each person fill in one or more of the parts of a game, depending on each's expertise? One guy does the scripting, likely using a placeholder box and placeholder platform-and-bunch-of-walls to test his script, another replaces the walking (or flying, perhaps) box with a model (original or not), someone builds the world and designs obstacles or architecture or puzzles or whatever, and someone playtests any beta(though usually on larger projects)
If you know how to complete all the parts of the game, you can just make it yourself without a team if you'd like.
InnocentSam wrote:If you suck at modelling, or any part of development, find your way around it! Find a copyright-free model of what you want, or use a model from another game (it's a free mod for a free game, no one's going to care that you lifted the model from another game).
What if you are making an original concept? You can only go as far as making a stickman-humanoid hybrid, or perhaps have basic architexture that looks like an apartment or a maze.
InnocentSam wrote:#6 - Random testers suck
[...] And only let them test it when you feel the mod is finished.
So, no beta testing? If something is finished, it means further adjustments and updates are unnecessary, and its creator may leave it be. If you haven't had others test the game, the monster chasing you (for example) is most likely too fast, but it's just you able to react instantly after it appears, too strong but you know how to dodge it every time, or perhaps the player can easily get lost in the place but you perfectly know your way around.
Otherwise, good job on the guide.