The notion of Tetris having no winning condition seems to have progressed from annoying to presumptuous, so let me briefly explain.
If properly arranged blocks did not disappear from the Tetris screen, then perhaps Tetris would have no winning condition. But they do disappear.
The winning condition in Tetris is to properly arrange falling blocks and cause them to disappear, thereby enabling a new game of Tetris.
These successive games of Tetris can be made easier or harder to win by manipulating a number of factors — most commonly the speed at which the Tetris blocks fall. (And here, I should note, making a game so difficult to win that it is very unlikely to be won does not, of itself, preclude that game from having a winning condition.)
In analogy, a chess match of infinite chess games is not a game without a winning condition. It is an arbitrary extension of the game of chess, wherein each game of chess has precisely the same winning condition as the ones before and after it.
In the case of Tetris, we can clearly see that this sort of extension is arbitrary by the common interposition of “levels” to restore the (most enjoyable) integrity of the game form.
And further, just to be clear in reference to McGonigal’s claims in Reality is Broken, Bernard Suits (in The Grasshopper) directly addresses games that are not arbitrarily extended in his discussion of “open games.”
I would define an open game generically as a system of reciprocally enabling moves whose purpose is the continued operation of the system. p. 124
With this definition, Suits does not admit games within which game players have no “state of affairs they are trying to achieve” (i. e., winning conditions), only that some take an “unnecessarily narrow view of what constitutes a state of affairs” (p. 124).
Regardless of this definition, however, in the specific case of Tetris, it seems clear that the objective of the Tetris game player is not to clear blocks in consort and cooperation with the continued operation of the system, but rather to clear blocks in opposition to and in competition with the continued operation of the system. If so, then Tetris is quite a conventional game and has quite a conventional winning condition. And, even if not, even if Tetris need be construed as an open game, then it would still have a winning condition: the continued operation of the system.
I see no intermediate position available to those who claim Tetris has no winning condition.