If everybody could make games, then we would need a much more thorough system to identify which are worth playing. Presently there's very little between word of mouth and major, general-purpose review sites and review aggregators. A more populist development process means that there will be a much greater proportion of terrible, unfinished, or otherwise undesirable games. That's not a bad thing in and of itself, since there will still be a greater absolute quantity of excellent games. However - and this is the problem faced by all distributed media (which is increasingly what the Web is best at) - there needs to be a more robust way to bring games to the people who would appreciate them. Mere popularity is not good enough, since it can't account for taste, but something like the Digg model would need to appear.
