It sounds like you are wanting to have your startup be built around a single game. If you are wanting to do that, you probably shouldn't use a casual game. Games with persistent worlds seem to be popular. Also, text based mmorpgs can attract a fair amount of people.

http://Travian.com

http://Kingdomofloathing.com

If you want a casual game, why have only one? There is one huge problem with casual games. You will have to code the game in such a way that it will be hard for the average person to find the memory address for the game variables.

Cheat Engine is an easy to use memory scanner that allows people to easily cheat at games. It will index all addresses for a program, for example firefox, and then let you search for them. So I can start out by indexing all addresses while playing a game, then increase a variable. Then I can search that index for all memory addresses whose values have increased. So basically I can edit almost any value in the game as it's easily observable So if you want to make the game hard to cheat, it's going to be somewhat difficult. What you need to do is to keep track track of the variables and check to see if they have suddenly changed.Maybe map x to y using a non linear transformation, and if there is ever a time where x does not map to y, you know the player is cheating. Or you could just accept input from the player and do the calculations concerning the score on your server.

http://cheatengine.org/aboutce.php Cheat Engine in use:http://www.youtube.com/watch?v=zd4kf6ID1Hc&feature=relat...

multi player 'snake'

Because ranking is very important in my startup idea. In real time games like "snake", players' performance depends too much on their network connections.

The idea game (javascript code) would be able to be downloaded via players' browser, and when the game is over the browser send final scores to the server, of course I will do the something to ensure they can't trick the server by hacking on the client side. Javascript tetris can work like this, but as I mentioned the it's just a little too ubiquitous.

> real time games like "snake", players' performance depends too much on their network connections.

That sounds like a challenge to me ! Solve that one and you've got it made, I think some kind of pipelining of game state changes should be possible. Remember, the only requirement for a real time multi player networked game is that it should make sense to whoever is looking at this screen, if you put two screens side-by-side then you are allowed to have differences. As long as it all works out again by game-over.