joegratz.net

The Copyright Office today released a lengthy report on the orphan works issue. I expect I’ll have more to say about the document after I’ve fully digested it, but the key section is the Copyright Office’s legislative proposal. It’s a bit like skipping to the end of a mystery novel to see how it all turns out, but this represents a major step forward in the debate and merits commentary. The Copyright Office recommends the addition of a new section 514 to the Copyright Act, which would read as follows:

SECTION 514: LIMITATIONS ON REMEDIES: ORPHAN WORKS

(a) Notwithstanding sections 502 through 505, where the infringer:

(1) prior to the commencement of the infringement, performed a good faith, reasonably diligent search to locate the owner of the infringed copyright and the infringer did not locate that owner, and

(2) throughout the course of the infringement, provided attribution to the author and copyright owner of the work, if possible and as appropriate under the circumstances,

the remedies for the infringement shall be limited as set forth in subsection (b).

(b) LIMITATIONS ON REMEDIES

(1) MONETARY RELIEF

(A) no award for monetary damages (including actual damages, statutory damages, costs or attorney’s fees) shall be made other than an order requiring the infringer to pay reasonable compensation for the use of the infringed work; provided, however, that where the infringement is performed without any purpose of direct or indirect commercial advantage, such as through the sale of copies or phonorecords of the infringed work, and the infringer ceases the infringement expeditiously after receiving notice of the claim for infringement, no award of monetary relief shall be made.

(2) INJUNCTIVE RELIEF

(A) in the case where the infringer has prepared or commenced preparation of a derivative work that recasts, transforms or adapts the infringed work with a significant amount of the infringer’s expression, any injunctive or equitable relief granted by the court shall not restrain the infringer’s continued preparation and use of the derivative work, provided that the infringer makes payment of reasonable compensation to the copyright owner for such preparation and ongoing use and provides attribution to the author and copyright owner in a manner determined by the court as reasonable under the circumstances; and

(B) in all other cases, the court may impose injunctive relief to prevent or restrain the infringement in its entirety, but the relief shall to the extent practicable account for any harm that the relief would cause the infringer due to the infringer’s reliance on this section in making the infringing use.

(c) Nothing in this section shall affect rights, limitations or defenses to copyright infringement, including fair use, under this title.

(d) This section shall not apply to any infringement occurring after the date that is ten years from date of enactment of this Act.

After an initial look at the proposed statute, it appears to strike a desirable balance between lessening the risk to users of orphan works and providing appropriate compensation to resurfacing copyright holders. If this statute is enacted, the most interesting part will be the first few years, in which courts will need to:

• set the boundaries of a “reasonably diligent search”
• tell us when attribution is “appropriate under the circumstances”
• figure out how to arrive at a rate of “reasonable compensation for use of the infringed work”
• define when a derivative work includes “a significant amount of the infringer’s expression”
• decide what sort of continuing use of an infringing derivative of an orphan work is “reasonable under the circumstances”
• craft injunctions that “to the extent practicable account for any harm that the relief would cause the infringer.”

I suspect that the judicial interpretation of the proposed statute would play a significant role in the debate, ten years after its enactment, about whether to make it permanent or allow it to expire.

The proposed statute bucks the trend in copyright law to micromanage individual cases through statutory drafting. (See, e.g., 17 U.S.C. § 114.) It allows the courts to consider the most important equitable issues on a case-by-case basis — which is what courts do best. The lack of bright-line rules is likely to frustrate prospective users of orphan works in the short term, but could prove critical to the long-term utility of the statute, since, if enacted, it will be adaptable to new circumstances and new technologies.

The Free Software Foundation has released the first discussion draft of Version 3 of the General Public License.

Notable among the new elements in this draft are:

• An aggressive stance against software patents. Software patent holders who distribute GPLv3′d programs would be granting a license to all downstream users to practice their patents. And software patent holders who sue over the use or modification of GPLv3′d software automatically lose the right to use that software.
• An aggressive stance against DRM. The draft license explicitly states that no GPLv3′d work ever “constitutes part of an effective technological protection measure,” and thus can never the the subject of a section 1201 circumvention claim.

Also notable is the impenetrability of certain sections. I’m glad this is the first draft, since there are a few sentences that I just can’t parse. The “propagation” / “distribution” dichotomy is giving me trouble, and I think section 7 will be a lot clearer once I figure out why it’s there in the first place.

Overall, however, this promises to be an excellent input into the lengthy, distributed “constitutional convention” of the free software movement the FSF plans.

CBC Distribution & Marketing, a fantasy baseball company, is suing in the United States District Court for the Eastern District of Missouri for a declaration that its use of baseball statistics does not require a license from Major League Baseball. Coverage is here and here; I couldn’t find the case on PACER in a quick search.

CBC had been paying the MLB Players’ Association 9% of revenue in order to use players’ stats until a new agreement between the clubs and the players gave MLB itself the right to license player stats. MLB has declined to issue CBC a license.

CBC says they never needed a license in the first place. They argue that sports statistics are uncopyrightable facts, at least once the game is over. MLB says that permission is required in order to “commercially exploit the identities and statistical profiles” of players.

It is an axiom of copyright law that facts are not copyrightable, but original compilations of facts are. The real question is whether the compilation of facts created by Major League Baseball, to the extent the selection and arrangement of that compilation is copied, is copyrightable. The runs, hits, and errors themselves are not the “works” of an “author”; if they were, the game would be fixed. As it were.

Even if the right to prevent misappropriation of “hot news” established in INS v. AP still exists — a dubious proposition, in my opinion — stats are primarily useful to fantasy baseball companies after the game is over.

It seems likely that MLB will assert a raft of non-copyright theories, since their copyright case would be so weak. They’ll almost certainly say that the names of the teams and the players are their trademarks — but their use in the fantasy baseball context is almost certainly fair use, since the names are being used solely to describe things that actual people actually did, not to appropriate goodwill. MLB will probably also claim that licenses are required because of state-law rights of publicity — but no state, to my knowledge, has allowed a person to control the use of his or her name when it’s used to describe things he or she actually did.

I agree with the lawyer quoted in the LA Times story who said this seems likely to settle quickly, since a decision on the merits is unlikely to favor MLB.

UPDATE: Thanks to Adam, I’ve found the district court record. (It’s docket number 4:05-cv-00252-MLM.) It’s a bit more complicated than it seemed, since the Players’ Association has intervened as a defendant and is asserting counterclaims against CBC for breach of contract, and is asserting defenses against declaratory judgment based on terms of the previous license. The Players’ Association has moved for summary judgment.

Sure, I live in San Francisco now, but I grew up in the great state of Wisconsin — and today, Governor Jim Doyle made me proud of my Wisconsin roots. He signed into law Assembly Bill 627 [Ed: See update below.], mandating a voter-verified paper audit trail for all Wisconsin voting machines and requiring that municipalities disclose the source code of the software that runs the machines.

The law is somewhat inartfully phrased, since “[t]he coding for the software” is not a technologically unambiguous synonym for source code, but the remainder of the provision — specifying that it’s referring to that which “may be used to independently verify the accuracy and reliability of the operating and tallying procedures to be employed at any election” — probably gets rid of any ambiguity.

UPDATE, 6 January 2006: My sources pointed to the wrong bill; it was substantially amended before passage. The enacted bill is here. Happily, it takes care of the “coding of the software” ambiguity, referring specifically to “vote−counting source code, table structures, modules, program narratives and other human−readable computer instructions used to count votes with an electronic voting system” — which is much, much better. Sadly, the requirement of truly open source code was removed, replaced with a code escrow requirement. In the event of a recount, each side gets to deisgnate some geeks to sift through the source, as long as the geeks “exercise the highest degree of reasonable care to maintain the confidentially of all proprietary information.” It’s not as good as requiring real open-source voting machines, but it’s probably enough to prevent serious chicanery, as long as there’s some mechanism to make sure the code submitted to the Board of Elections is really the software that’s running the machines on election day. Having the voting machine companies promise really hard doesn’t cut it, when — in security terms — they’re the attacker in this law’s threat model.