Monday, April 30, 2007

ASCAP, Performing Rights, and Downloads

The legislative reports to the 1976 Act indicate that “[t]he[ ] exclusive rights, which comprise the so-called ‘bundle of rights’ that is a copyright, are cumulative and may overlap in some cases. Each of the five enumerated rights may be subdivided indefinitely and, as will be discussed in connection with section 201, each subdivision of an exclusive right may be owned and enforced separately.” The reports also note that multiple rights may be violated in the course of a single activity, “as where a publisher reproduces, adapts, and sells copies of a person's copyrighted work as part of a publishing venture.”

The question of infringement of multiple rights arises most intensely in debates over whether a digital transmission can implicate more than one right, such as the reproduction and performance rights. In the debates over what became the Digital Performance Right in Sound Recordings Act of 1995, the performing rights societies (ASCAP and BMI) insisted on inclusion of numerous “savings clauses,” stating that the grant of a limited digital performance right for sound recordings in Section 106(6) did not affect the Section 106(4) performance right. Music publishers also insisted on extending the “mechanical” compulsory license in Section 115 to include digital phonorecord deliveries.

In my personal view, a “digital phonorecord delivery” under Section 115 is only a distribution and not a performance, while streaming is only a performance and not a distribution. The distinction between a performance and a distribution should be drawn by reference to the type of transmission involved: if the transmission delivers a copy of a work to a consumer in non-real time, it is a distribution. If streaming or other real-time conduct is involved, it is a performance.

This issue was faced head on by Judge Conner of the SDNY, who has long ridden herd over the ASCAP consent decree, as the ASCAP rate court. On April 25th, he issued an opinion on the question of whether the downloading of a digital musical file constituted a public performance of the song. He held it did not. Of particular interest to the Cablevision case (and buffering) is the following language:

[W]e are not persuaded by ASCPA's argument that downloaded music files are indistinguishable from streamed performances because, after a certain amount of digital data has been transmitted to the client computer, the purchaser can begin listening to the transmitted portion of the music file. However, the mere fact that a customer's online purchase is conveyed to him in a piecemeal manner, each segment of which is capable of playback as soon as the transmission is completed, does not change the fact that the transaction is a data transmission rather than a musical performance broadcast.
Opinion at 10-11 (emphasis in original).
(HT to Josh Wattles.)

7 comments:

Anonymous said...

The district court's ruling may be downloaded here from the DiMA website.

While I understand everyone's satisfaction with the outcome, I'm less satisfied with the court's (lack of) reasoning. So, if I simply hold the data in a buffer for some period of time, it no longer constitutes a "performance"? How long? 30 seconds? 5 minutes?

And I continue to maintain that a download is NOT a distribution, as distributions require the passing of a material object from hand to hand (for the full reasoning here, see Prof. Anthony Reese's excellent article). Treating a download as a distribution potentially creates chaos in a variety of contexts involving both compulsory and negotiated licenses -- see, e.g., what XM does with its Inno (digital transmission to a device that makes a reproduction), what cable companies do when they distribute programming to head end facilities (same).

What you have in the case of a download is a reproduction, not a distribution.

Joseph Gratz said...

The opinion is available here, thanks to Public Knowledge. My thoughts are here.

Christopher Fulmer said...

I would attack this problem differently -- is the version in the buffer a 'copy'? Remember, the distribution right applies only to "copies" or "phonorecords." If the recipient does not get a copy or a phonorecord, there can be no distribution.

Sec. 101 requires that both copies and phonorecords be "fixed," or that they must be "sufficiently permanent or stable to permit [them] to be perceived, reproduced, or otherwise communicated for a period of more than transitory duration."

Without going into a heavy analysis, a buffer copy of where the work incrementally plays and is forgotten seems transitory, but a copy on disk does not. (You cannot take the memory/disk distinction too far. Modern memory-management schemes store copies of memory on disk.)

Anonymous said...

I sure hope "pubic performance" was a typo, or I have a lot to learn about copyright law...

William Patry said...

Urijah, I am against open displays of emotion. How about a "punic" performance, though, among warring parties?

Anonymous said...

I know this is a little off topic, but I wanted to share the following with you.

The Information Technology and Innovation Foundation (ITIF) just released a report on Internet Radio and Copyright Royalties at an event on Capitol Hill this morning. In the report, we describe problems with the current copyright royalty system for Internet Radio, and what steps Congress should take to reform this system.

I know you have posted about this issue in the past, and I wanted to make sure you were aware of our paper.

The report is available on our website at - http://www.itif.org/files/InternetRadio.pdf

Anonymous said...

Here's an interesting website from George Washington University Law School looking at these issues

http://www.newcopyrightera.org