"I am willing to agree that a video game could also be a serious work of art. It would become so by avoiding most of the things that make it a game, such as scoring, pointing and shooting, winning and losing, shallow characterizations, and action that is valued above motivation and ethical considerations. Oddly enough, when video games evolve far enough in that direction, they will not only be an art form, they will be the cinema."
The way I read it, he's saying an artist is a decision-maker. This makes sense to me. To whatever extent you leave decisions for the audience to make, you are not an artist. In other words, a video game is not a work of art; it is a medium.
Of course, there are plenty of examples of audience participation in music and theater; think Drood or Shear Madness. Such examples are usually perceived as gimmicks. Surely no one would seriously suggest Drood was comparable with, I don't know, the Ring Cycle. And it doesn't usually turn the audience into great artists, either.
(Hmm... does having to think about a movie constitute participation? Like when a movie is "involving" and "draws you in"? Answer: no.)
So could a particular "performance" of a game be a great work of Art? I think maybe so, but probably not yet. For one thing, art forms need to evolve in parallel with their media. I've noticed this problem with newly invented musical instruments: you can't tell how good the instrument is, because it won't have any great players for at least a century or so. Players suggest improvements to the instrument, they find new techniques, they teach them to younger players, who improve the instrument a bit more, and eventually you get a person and an instrument who together can play the Shostakovich Cello Concerto (and before that, of course, the concerto couldn't have been written, either). Totally new instruments tend to be relatively easy to play (since otherwise there would be no one to play them), which means they don't allow the player as many degrees of freedom as a traditional instrument, which means they won't allow as much expression in the hands of an eventual master.
Most current games don't allow the player very many degrees of freedom for the same reason, and also for another reason: they're games. Games, practically by definition, keep people involved by being at the right (increasing) level of difficulty and by providing instant feedback (cf. Flow), usually by evaluating how well they're doing according to a set of objective rules. Computer programs are good at objective rules; not so much at artistic criticism.
The media that allow Great Art allow a lot of freedom (although constraints can help stimulate the imagination) and can't reasonably be judged by any objective rules that we know of. So maybe the games I'm hoping for, the ones that become Great Media, won't really be games by most definitions. They will, in fact, be more like specialized editing applications, like Photoshop or PowerPoint. Some games are already well on their way. The Sims and its kin allow a great deal of creative expression and don't pass much judgment on the player. Maybe they're toys, not games. Maybe Ebert is right, at least, that games that are art will not be games.
Some games manage to be both design programs and true games by allowing the user to design an automatic "contestant", and then setting one contestant against another according to some rules. Mind Rover is a great example; Core War is an older and better known one. It may be hard to imagine a Mind Rover robot or a Core War program impressing anyone as an artistic achievement (though many people say computer programming should be considered an art), but the physical robots used in 6.270 and other robot competitions definitely could, and future virtual robot battles could allow more of the kind of design freedom that is necessary for artistic expression.
Of course, even in straightforward FPS games, there are opportunities for expression. Some games include a "level editor", which is basically a CAD application for designing virtual worlds. This is really a form of sculpture (one in which, interestingly, works are meant to be seen from the inside), although levels are usually judged in terms of "playability" as well as "eye candy". Some people enjoy designing levels more than they enjoy playing them.
At the very least, this reminds us that even if games can't (yet) be art, they can contain it. In fact, practically every other art form has been used inside of computer games. Hence the word "Multimedia".
Art can contain other art; it can contain itself; the boundaries between container and contained are not always clear; categories are not distinct. A game can be mostly art and a little bit medium, or the other way around. It's all just semantics, which is probably what you've been thinking since you started reading this.
What we really want to know is: is it (a game, a movie, whatever) worth my time? And the answer is: you won't know until you've tried it. This is one of the fundamental problems of AI, and hence beyond the scope of this, er, article. But the above-referenced research on Flow suggests that, since computer games are more consciously designed to induce flow than anything else in history, they might potentially be able to make us happy.