Outside of moral considerations[1], a game can't be good or bad. Only a "game experience" can be good or bad. I would be among the first to say that different games are likely to be better or worse at providing good game experiences for a particular audience at a particular time.
I wouldn't enjoy a roll-and-move game about Powerpuff Girls with no player decisions on almost any occasion. Noone would enjoy playing it 24 hours a day for the rest of their life. But for SOME people (young children, PPG fans, drunk people) on SOME occasions (have never played the game before, slumber party, drinking party) a roll-and-move PPG game will provide a good - even great - game experience. Maybe there exists other games that - for the same people and on the same occasions - would produce as good an experience, or an even better one. But that is beside the point.
I understand people who don't like a game or encourage people to play different games. But some of the comments - from supposedly respected sources - about Flappy Bird are just stupid. Kotaku actually called the game a "terrible game". A game that was being played happily - frustratingly - by millions of people. Commenters on Kotaku's post exhort people to stop playing the game and journalists to stop writing about the game.
Think about that for a minute. People should stop enjoying a wildly popular game because the game Isn't Enjoyable? The designer should be shunned for creating a wildly popular Bad Game? Other people imply that the game's popularity is a result of deceit on the designer's part, or that he ripped off graphics and mechanics from other games (as if there were millions of different ways to draw a pipe using 8-bit graphics or there were not already millions of other scrolling platformers with similar visual design and mechanics).
Holy camole, Batman. The "wrong kinds" of games shouldn't be tolerated? Games can't be fun unless they take more than the "right amount" of time to make by a big development studio? And they have to provide the right levels, systems, graphics, sounds, and whatever else doodads are in your "design by the numbers" game design manual?
There is no "right kind" of game, because there is no right game. It doesn't matter if the game experience was fun only because of a stupid meme (for example, a game was seen being played by Brad Pitt in a recent movie). It doesn't matter if the game is only fun for two plays and then you're done. It doesn't matter if the game is only fun for certain types of people, or just a few people, or only one person. And it doesn't matter if the game is only popular at a particular moment in time. Want to play it? Fine. Don't? Fine. If you don't like the game, get on with your life, and let the people who like the game get on with theirs.
Even market or monetary success or failure is not proof of a game being good or bad. A game can succeed by accident or because it was published at just the right moment, was played by just the right people, or offers one new challenge that other games don't (as is obviously the case for Flappy Bird). Instead of hating on "undeserved success", try reflecting on the fact that a) life isn't always fair, and b) you could be making something instead of complaining about that.
Also see: Radiator, Polygon, Guardian,
[1] Such as a "game" about rape or murder.