Defining Tetris

[BB+]

I got this link from someone interested in the ICGA cases: Defining Tetris: How courts judge gaming clones

[User923005]

I do not see how it applies to chess.
Any copyright to the game of chess has long since expired.
Copyright can be granted to new games like Gothic chess.
I have no interest in those new games.

Personally, I think that granting copyright on the Tetris board dimension is bizarre.
Since look and feel cannot be granted copyright, I feel like the judge was a nitwit in every single exclusion granted to Tetris.
But I am not a legal expert, so maybe it is me who is the nitwit.

[Chris Whittington]

I got this link from someone interested in the ICGA cases: Defining Tetris: How courts judge gaming clones

I take the key message as:

the raw "thing", reduces to an "idea" which is then unprotected. implementations of the idea are protected. For example, the idea of dropping squares in Tetris according to a set of rules is the "raw" thing and may be copied. The N by N dimensions of the Tetris space is however an implementation and is protected. Various special tweaks to the game would be protected.

Crossing from the Tetris analogy back to Rybka-Fruit, we could, for example consider the mobility code in this light. The raw idea of adding up moves to get a mobility score would be unprotected as an idea, whereas the application of specific weights would be an implementation and therefore protected. Various tweaks to the raw idea would also begin to be protectable, for example weighting forward squares more than backward, or not including enemy pawn attacked squares.

Now, specifically, for Fruit, he adds moves, weights them specifically and uses a start weight offset. Ie basic raw idea plus weights.

For Rybka, he adds moves, weights them differently and does not use a start weight offset. ie basic raw idea plus different weights. according the your Tetris analogy, this is fine. Of course, also the codes and program flows are completely different.

Yet according to your COMPEVAL doc Rybka gets a 80% copy penalty (times 4 for 4 piece types), almost 10% of the total score you gave Rybka to "prove" overlap.

To my mind, this anti-Rybka anomoly in your work arises because of a failure to FILTER unprotectable material before your comparison stage.

[syzygy]

I got this link from someone interested in the ICGA cases: Defining Tetris: How courts judge gaming clones

At least under EU law, the copyright on the original Tetris computer program does not extend to other programs implementing the Tetris concept. If other implementations of Tetris infringe any copyright, it is the copyright on the "video game" as a copyrightable work.

The "work" implemented by a chess engine would be the game of chess (chess board, chess pieces), which clearly is free of copyright if only because the game is too old to be copyrighted.

In my view, an evaluation function (or chess engine) abstracted from its implementation as a computer program is not a work. A work needs to be something perceptible. Playing against or otherwise using a chess engine, you don't perceive the details of an evaluation function or search, but only the game of chess.

Software programs are perceptible because you can view the source code or object code, even if it is only a series of 0s and 1s.