1. be fair to the people here at newgrounds: if this is your 4th collab and you didn't finish even one then don't do another one
2. to do a game tutorial collab is a good idea (at least in principle) but you should take care of its quality. Don't let people just code something somehow
- give some advice on how to code, i.e.: no object scripts; use a simple framework (i.e. code blocks: import, vars, functions, init)
- use classes for repetetive tasks (or: don't use classes at all)
- insclude some theory your work is based on (i.e. steering behavior: what is it? Why is it useful? What kind of behaviors are possible? Then go on to the specific steering behavior you want to talk about)
3. to do a game tutorial collab on a specific type of game is a better idea since you can focus on one subject and handle it in depth instead of getting lost in 1001 subjects, i.e.:
- tile based games:
- what is it?
- some infos on the history and relevance of tile based games (non-Flash and Flash) to get some background knowledge
- kind of tiles (rect, hex, iso, ...)
- maps (organising data, ...)
- some basic enemy ai
- sprite sheets, blitting
- automatically generate levels
- level editor (how to use an existing one like mappy, how to code your own)
- at least one complete game (including how to write a simple game design doc)
- further sources (from the famous tonypa to gamedev.com etc.)
4. or you could deal with subjects that are relevant to many games without focusing on one sepcific game type (like random, physics, trigo, ai, wayfinding, spawning and so on).
Yes, I know, there are tuts on all these subjects out there in the web (and even here on newgrounds). But this is true of almost every game that will be included in this collab. In that case either give up or do a collab that is better than many tuts you may find with google.