Hacker News Reader

Comment by greglindahl | Hacker News Reader

Are you a copyright lawyer? I'm thinking not, because cases like Feist v. Rural Telephone make it pretty clear what "facts" mean in the context of things you can copyright.

https://en.wikipedia.org/wiki/Feist_Publications,_Inc.,_v._R....

View on HN

It's definitely not clear to me that playing chess is an uncreative task like compiling phone numbers. Though, perhaps I don't know enough about chess. This seems like an interesting case to me.

I'm not saying that playing chess is an uncreative task - it's very creative. Describing the chessboard position is an uncreative task. Evaluating the relative strength of the players' positions can be creative. But the position of the pieces? No.

Filming a football field is also uncreative, isn't it?

That's a well-settled issue. Are you really claiming that writing down the positions of chess pieces is similar to operating a video camera looking at 3D objects?

On the other hand, you might want to check out Corel v. Bridgeman.

https://en.wikipedia.org/wiki/Bridgeman_Art_Library_v._Corel....

From a position of total legal ignorance, I don't see a meaningful distinction. In both scenarios, you're recording an account of someone else's performance with enough fidelity for someone to know what's happening in the game. In the linked case, the issue seemed to be the the recorded content had no copyright, so I don't get why that's relevant? Very interested in a more detailed answer though :)

You are not getting a very important distinction: no creative effort (not even a kernel of creativity[1]) is embodied in telephone numbers.

There is something though about the creative expression of chess board positions: they are EXTREMELY limited in terms of entropy. In coordinate notation 64*64 = 4096 choices would define either side's move, so given a dictionary of 4096 words, any typical chess game would be under a few hundred 'words': they're very very short creative expressions.

I would suggest that you compare them with musical melodies. they have about as much entropy. the database of existing melodies (or slight variations) is similar to the database of existing chess games.

in some cases melodies were held to be protected by copyright, so that you could not reproduce them in any form. I am not a lawyer, no. I researched many areas of intellectual property extensively. In general a basic melody might well be a "fact" about a song - yet a fact that you cannot use in your own song. (Similar to a patent.)

The judge, sensibly, chose not to extend the same protection to chess games. I agree with him! I just disagree with your reasoning. There's a reason the judge didn't advance it.

[1] https://copyright.uslegal.com/enumerated-categories-of-copyr...

Chess is far more constrained than that. Pawns generally have 0 or 1 move with the maximum possible of 3 moves followed by piece selection which is knight or queen as bishops and rooks have the subset of a queens moves. There are a maximum of 8 pawns. Thus (8 * 3 * 2) = 48

Kings have a maximum of 8 moves. 2 Rooks a maximum of 14 = 28, 2 bishops 12 = 24, and 2 knights 8 = 16, queen = 36. And this is individually on an actual board there is often less then 50 legal choices (ex: 20 for opening) and the average game is 40 move (pairs).

Thus you could encode most chess matches as a tweet.

I appreciate your analysis - however you need to compare it not with the length of a Tweet but rather with the entropy in a melody: whereas there are typically "less than 50 legal choices" and an average game is "40 moves", a melody that has been held to be protected by copyright has fewer than 50 legal choices for each note, and requires fewer than 40 notes (by far) to be protected by copyright.

I'd like to have you come back and compare chess games as actually played, with what has been held to be copyrighted simple melodies. I'd like to see that comparison and think you're good to go to make it.

Melodies are less constrained than that, timing between notes (which includes 0) note pitch and duration are all open. Which is why musical notation is really complex.

Further they are not patents two people can in theory both have copyright on the same melody.

I wanted you to look at actual case law - the minimal amount of a "melody" that barred others from reusing it. Then to compare the amount of information (or constraint) in that melody, with the case at hand.

If you need a hint, see arguments e.g. here - http://gizmodo.com/5962375/is-it-possible-to-run-out-of-new-...

and elsewhere. While you say that timing between notes and note pitch and duration are all open, in fact there are relatively few choices in actual modern practice.

In terms of choices, high level play also has a tiny fraction of the number of choices in practice. However, this has nothing to do with why it's protected or not.

I can also make a clif notes version of Harry Potter describing what happened in detail. That's not what copywrite protects.

>I can also make a clif notes version of Harry Potter describing what happened in detail. That's not what copywrite protects.

everyone can agree that summarizing a game is fine. We can also all agree that Chess games shouldn't be subject to copyright!

So, I'm not disagreeing with you or anyone.

As a very narrow technical issue, I would like to have a comparison between the number of choices (which key to play in; first note, its length; second note, its length; third note, its length; fourth note, its length) that goes into a melody that has been deemded legally protected, and a chess game.

Here is a protected riff:

https://www.youtube.com/watch?v=1nLCa0YG1ZI&t=57s

Now please focus on just the first 8 notes, after picking a "starting note". The number of choices for where the melody goes, in the key that is in, is rather constrained: what has more creativity, 8-9 of those notes, or an entire chess game?

An entire chess game may well have more creative choices in it. After all, full-length chess games are quite rarely repeated! (independently). There are a rather large number of variations.

As further evidence, consider the term "novelty" which is a single choice at a particular move, which has never been played before (is a new variation/line.) The term "a novelty" in chess is quite specific and also is evidence that even when you are still in the opening, Chess people consider move choices to be quite creative.

All this might seem to imply that I am disagreeing wiht you or others: I'M NOT. Chess games shoudl NOT be copyrighted.

But as a very highly technical point of law, they contain more creative choices or entropy than other creative works that are protected.

Please be more careful in your analysis.

First you are ignoring chords and note timing which is an option. Second there are only 1,327 named openings and variants in chess many of them are known to be inferior so you do see a lot of repition in the beginnings and endings of Master level games.

As an order of magnitude each note in a melody is from a low end estimate of ~1,000,000 completely valid options though musical style will limit this.

Thus from an encoding standpoint a chess game is theoretically a much lower entropy. But, a sufficiently constrained melody may be lower but probably not. Consider an easy example: https://m.youtube.com/watch?v=DtvNAQ8KOqI

This is totally splitting hairs and I don't think we're adding to the discussion anymore. if you'll drop me an email (see my profile, since you don't list one) I'll spit back my analysis but it's really really pedantic and actually probably not worth our time. thanks for the exchange here.

The distinction is that the music is being composed as a creative activity and the notes are the essence of the production. The chess game is a game, winning is the game, and the notation is just one of many possible transcriptions of it.

If you composed a poem from chess notation, it would be copyrightable.

If you transcribed a chess game (mechanically) with a series of notes, it would not be.

And besides, if there was creativity (for its own sake) in the process and the moves were copyrightable it would be the players, not the stenographer, that owned it.

agree with everything you stated, ESPECIALLY the last sentence.

In an alternative reality where the players agreed that they were creating a creative work together, and signed the copyright to it over contractually, would your attitude change?

Can you talk a little bit about why a very short riff such as 10-12 notes of this -- https://www.youtube.com/watch?v=1nLCa0YG1ZI&t=57s

should be subject to copyright? (I also may be mistaken factually - perhaps it's not subject to copyright.)

As you can see from that video, there really are very few choices regarding the "next note". Maybe a few more choices than the possible legal moves on a board - but not by much.

>If you transcribed a chess game (mechanically) with a series of notes, it would not be.

I find this very very hard to believe. If I came up with simple rules for transcribing a chess game with notes and then discovered that for a particular chess game this was pleasant, you really don't think I could copyright that tune?

I'd be shocked if that were the case.

> In an alternative reality where the players agreed that they were creating a creative work together, and signed the copyright to it over contractually, would your attitude change?

Yes, that would probably be the same (in general) as any other work-for-hire.

> Can you talk a little bit about why a very short riff such as 10-12 notes of this [...] should be subject to copyright?

I imagine it is subject to copyright, a Haiku would be.

The grey areas here are that such a short sequence lends itself to brute-forcing which isn't creative and probably wouldn't result in a copyright, and that independent creators would each have their own copyright. You can't just generate all possible books (even if not combinatorially impossible) and block authors from writing them.

> I find this very very hard to believe. If I came up with simple rules for transcribing a chess game with notes and then discovered that for a particular chess game this was pleasant, you really don't think I could copyright that tune?

Your program to do this would be copyrightable, but it would only produce a machine-translation (by definition) of the chess game so while the end results (the tune) be copyrightable, it wouldn't be your copyright.

thanks.

Sure, but if choosing your moves is a creative task, then why wouldn't the moves be jointly copyrighted by the two players?

What's the difference between:

    1. e4 e5
    2. Qh5 Nc6
    3. Bc4 Nf6??
    4. Qxf7#

and:

    fmt:    .string "Hello World!\n"
        .balign 4
        .global main
    main:
        stp    x29, x30, [sp, -16]!
        mov    x29, sp
        adrp   x0, fmt
        add    x0, x0, :lo12:fmt
        bl     printf
        ldp    x29, x30, [sp], 16
        ret

if both are the result of a creative process?

Events that happens are facts, creative representation of facts is copywritable but not mechanical translations. Thus you can't get a copywrite on the S&P's closing value, but you can on a story about it. Further, a program has multiple possible representations, you can copywrite the code in one of them, but not gain a separate one for the compiled code.

PS: There are also many ways to generate code that don't create a copywrite.

You're talking past each other. I think everyone agrees with the decision, but it is indeed a gray area. The S&P closing value isn't comparable because it's the product of a stochastic process.

But even if there's a smart legal doctrine defining the difference between a chess game and two musicians writing a score, lets not pretend that such legal doctrines aren't created to arrive at the result that seems right. Not that there's anything wrong with that – look no further than the "smart contracts" fiasco to see the folly of trying to define rules in a "completely objective" framework.

PS: "copyright", not "copywrite".

Thank you.

Reply on news.ycombinator.com