If you answered (3):

Your views are kinda shared by: Roger Ebert; and Hideo Kojima.

There are two major difficulties when trying to make art games: firstly, one has to question whether or not they strictly have to be "fun"; Secondly – and more pertinently – one has to consider whether or not the interactivity of these games hinders their value as an art form.

Film critic Roger Ebert acquired a certain notoriety within the game development community a few years ago with his claim that "games could never be considered art" on the same level that film and literature were attributed the label.

His most oft-quoted comment on the matter goes as follows: "Anything can be art. Even a can of Campbell's soup . . . [but] games could not be high art, as I understand it."

Outraged gamers everywhere started flooding him with messages, trying to persuade him otherwise, but he stood by his words on the grounds that art is understood to be controlled by the author, whereas the level of interactivity inherent in most games forbids a suitably controlled environment – suddenly, the artistic message is controlled and subsequently diluted by the player.

Since this controversial statement, however, Ebert has generally fallen silent on the gaming issue and approaching him for questions on the matter seems all but impossible. Yet his point still plagues developers and sparks numerous discussions on its own. Even Rohrer admits that the old chap has an interesting argument.

You never know what a player is going to do, so how can you craft a meaningful experience with any degree of certainty? About 50% of the people who played Passage never realized that they could walk DOWN and just pressed RIGHT the entire time, missing about 75% of the game and about 90% of the point . . . and there's no way I can prevent that."

Even Hideo Kojima, the great hiding-in-boxes, red-bandana-wearing mind behind freakin' Metal Gear Solid, has made controversial statements in the "games as art" debate. He related the entire debate to cars, of all things – though he got his point across clearly enough: "art" could most closely relate to concept cars and similar elite ideals, whereas "games" were more similar to mass-produced models which needed to work for everyone. For Kojima, the core idea of games needing to pander to their audience was the restriction when it came to classifying them as art – a point which is essentially very similar to Ebert's, and very difficult to empirically refute.

Admittedly, this argument sounds like it applies mostly to mainstream games – after all, indie devs don't necessarily have to appeal to anyone, right? Maybe. But if art games aren't sufficiently fun, and nobody likes them, are they still holding to the meaning of what it is to be a game? Is it possible that the moment a game becomes art, it stops being a game?

Take for example this super-duper intense discussion on a Global Game Jam entry known as "4 minutes and 33 seconds of uniqueness" – an art game which has... well, have a look for yourself, you lazy bugger. This game provides a convincing case for art, but not everybody agrees that it maintains its ties to gamedom in the first place. Unfortunately, Petri Purho, (he also created Crayon Physics) couldn't be reached for comment on his offering, but there's an interesting interview with him in which he indicates his belief in the fact that the project satisfies the "game" criteria.