By , February 06, 2014.

Previously, I began looking at the legal questions involved in American Broadcasting Companies v. Aereo, currently in front of the Supreme Court. The issue is whether Aereo, by providing a service that allows paying subscribers to watch broadcast television online, is publicly performing the copyrighted programming. The Second Circuit said it is not, based on its 2008 decision in Cablevision, which held that a transmission from a unique copy of a work is merely a private performance. Aereo, relying on the Cablevision decision, designed its service to purportedly assign a unique antenna to each individual subscriber.

The Copyright Act states that to perform a work publicly means, in relevant part, “to transmit or otherwise communicate a performance … of the work … to the public, by means of any device or process, whether the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.”

The seemingly straightforward language belies some of the complexities that result, as evidenced by the Supreme Court’s agreement to hear Aereo. The approach I began in my previous post was to break the definition up into its separate components to understand it better: we have an actor (otherwise unnamed in the Act, we can refer to this as a “performer”), the action (“to transmit or otherwise communicate”), the object (“a performance of the work”), and a prepositional phrase (“to the public”).

To the public redux

I first took a look at what “to the public” means. Having had the opportunity to consider some of the feedback I’ve received on the article, I do want to make a slight modification to my definition. 1I also flubbed when I referred to the three situations as being “mutually exclusive”. What I meant instead is that each situation excludes the previous. For example, a “place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered” could include a place open to the public, but that would make the previous clause redundant. Similarly, the last clause, transmission to the public, could include transmission to the public in a public place, or transmission to a place where a substantial number of persons is gathered, but that would make the previous two clauses redundant.

I think it still follows that each situation is analyzed by exclusive means. We can presume (from the case law) that a “place open to the public” turns on the nature of the place. The “place where a substantial number of persons…” situation doesn’t turn on the nature of the place: if such a place were public, then the first situation covers it, while being a private place is insufficient to tell us if there are a substantial number of persons there. The same holds true for the last situation. The nature of the place is irrelevant (otherwise the first clause would be redundant), the size and relationship of the audience to each other is irrelevant (otherwise the second clause would be redundant). So what is there left to consider? The only thing I can come up with is the relationship between the performer and the audience. There could be others I am not aware of, but the relationship between performer and audience seems to provide a workable rule consistent with the Copyright Act’s structure and purpose.

Originally, I wrote, “Any transmission from one person to another person who is not a family member or a close social acquaintance is a public performance.” The mention of family members and social acquaintances invites a level of specificity that I did not intend and distracts from the original purpose, which is to focus on the relationship between performer and audience. The legislative history of the Copyright Act says that under the second clause of the definition of “publicly”, performances during “routine meetings of business and governmental personnel” would be exempt “because they do not represent the gathering of a ‘substantial number of persons.'” I think the definition of “to the public” should similarly exempt such performances. It seems it would be more accurate to step back one level of abstraction; rather than referring to family members and social acquaintances, we should refer to “public” and “private” relationships. “Public” relationships are those that tend to be described as commercial, arms-length, or impersonal, and are strongly unidirectional, while “private” relationships tend to be described as familial, social, or collegial, and are much more reciprocal in nature.

What is a performance?

With that out of the way, we can turn to the next question in the analysis: what is a performance?

As with “to the public”, “performance” is not directly defined in the Copyright Act. But the Act does define the act of performing:

To “perform” a work means to recite, render, play, dance, or act it, either directly or by means of any device or process or, in the case of a motion picture or other audiovisual work, to show its images in any sequence or to make the sounds accompanying it audible. 217 U.S.C. § 101.

A performance right for dramatic compositions first appeared in US copyright law in 1856. 3Act of August 18, 1856, 34th Cong., 1st Sess., 11 Stat. 138. Nondramatic musical works gained a performance right in 1897, 4Act of January 6, 1897, 44th Cong., 2d Sess., 29 Stat. 694. and the 1909 Copyright Act, the last general revision before the current Copyright Act, provided performance rights for dramatic works, nondramatic literary works, and musical compositions. However, the current Copyright Act is the first to actually define “perform.”

Though the current Copyright Act did not become law until 1978, the language of the definition for performance is substantially the same as the definition included in the 1965 version of the bill. 5H.R. 4347 and S. 1006, 89th Cong. Following the introduction of that bill, the US Copyright Office released a report explaining the bill in detail. The Supplementary Report explained:

Under clause (1) of section 106(b), to ”perform” a work means ”to recite, render, play, dance, or act it.” This includes, for example, the reading aloud of a literary work, the singing or playing of music, the dancing of a choreographic work, and the acting out of a dramatic work or pantomime. A work may be performed ”either directly or by means of any device or process,” and these devices or processes would encompass sound or visual reproduction equipment of all kinds, amplifying systems, radio and television transmitting and receiving apparatus, electronic retrieval devices, and a host of other techniques, undoubtedly including some not invented yet. In the case of a motion picture, performance would mean ”to show its images or to make the sounds accompanying it audible.” It would be clear under this language that the purely aural performance of a motion picture sound track would constitute a performance of the motion picture; but, if the sounds on the soundtrack are reproduced on an authorized phonorecord, performance of the phonorecord would not be a performance of the motion picture.

The definition is relatively clear, and there are very few cases that have have been confronted with issues relating to its interpretation. One of the most relevant and thorough is US v. ASCAP. 6627 F. 3d 64 (2nd Circuit 2010). There, the court was tasked with considering whether a download of a music file was a performance. It looked at the “ordinary sense” of the words used in the Copyright Act’s definition of performance and concluded that performance requires “contemporaneous perceptibility.” Said the Second Circuit:

The downloads at issue in this appeal are not musical performances that are contemporaneously perceived by the listener. They are simply transfers of electronic files containing digital copies from an on-line server to a local hard drive. The downloaded songs are not performed in any perceptible manner during the transfers; the user must take some further action to play the songs after they are downloaded. Because the electronic download itself involves no recitation, rendering, or playing of the musical work encoded in the digital transmission, we hold that such a download is not a performance of that work, as defined by § 101.

In other words, “Transmittal without a performance,” said the Second Circuit, “is not a ‘public performance.’”

Transmission without performance

As support for this proposition, the court cited to Columbia Pictures Indus. v. Prof’l Real Estate Investors. 7866 F.2d 278, 282 (9th Cir. 1989). In Columbia, the operators of a hotel resort offered guests the ability to rent movies on videocassette at the front desk, which they could watch on hotel-provided equipment in their own rooms. Plaintiffs sued, relying on the Third Circuit’s line of cases holding video store operators liable for public performance for operating private viewing booths on their premises. 8Columbia Pictures Industries v. Redd Horne, 749 F.2d 154 (3d Cir.1984); Columbia Pictures Industries, Inc. v. Aveco, Inc., 800 F.2d 59 (3d Cir.1986). The Ninth Circuit, however, rejected this argument, holding that hotel rooms, though offered to the general public, become private spaces once they are rented. 9In support of this conclusion, the court relied on Fourth Amendment case law holding that hotel guests have constitutional protection from unreasonable searches and seizures in hotel rooms. Thus, the operation of the equipment is not a performance in a public place.

Plaintiffs, however, also argued that the act of providing videocassettes to hotel guests implicated the Transmit Clause because the hotel was “otherwise communicat[ing]” the films. The Ninth Circuit rejected this argument as well.

A plain reading of the transmit clause indicates that its purpose is to prohibit transmissions and other forms of broadcasting from one place to another without the copyright owner’s permission. The Act provides a definition of “transmit.” “To `transmit’ a performance or display is to communicate it by any device or process whereby images and sounds are received beyond the place from which they are sent.” According to the rule of ejusdem generis, the term “otherwise communicate” should be construed consistently with the term “transmit.”  Consequently, the “otherwise communicate” phrase must relate to a “process whereby images or sounds are received beyond the place from which they are sent.”

This reading is reinforced by the rest of the transmit clause which refers to the use of transmission devices or processes and the reception by the public of the performance. Devices must refer to transmission or communication devices, such as, perhaps, wires, radio towers, communication satellites, and coaxial cable, while reception of the performance by the public describes acts, such as listening to a radio, or watching — network, cable, or closed-circuit — television “beyond the place” of origination.

In sum, when one adds up the various segments of clause (2), one must conclude that under the transmit clause a public performance at least involves sending out some sort of signal via a device or process to be received by the public at a place beyond the place from which it is sent.

Nothing that La Mancha has done has violated this common sense construction of the transmit clause. While La Mancha has indeed provided the videodisc player, television screens, guest rooms, and makes videodiscs available in the lobby, we are not persuaded that any transmission of the kind contemplated by the statute occurs. If any transmission and reception occurs, it does so entirely within the guest room; it is certainly not received beyond the place from which it is sent. We are not persuaded that the term “otherwise communicate” can be read so broadly as to include the videodisc arrangements at La Mancha.

The reasoning of the Ninth Circuit here and the Second Circuit in ASCAP seems sound. 10And indeed, the Copyright Office’s 1965 Supplementary Report adds support.


In general the concept of ”performance” must be distinguished sharply from the reproduction of copies on the one hand and the exhibition of copies on the other. It has been suggested that some of the internal operations of a computer, such as the scanning of a work to determine whether it contains material the user is seeking, is closely analogous to a ”performance.” We cannot agree, and for this reason we deleted from the definition of ”perform” the ambiguous term ”represent” which appeared in the 1964 bill. A computer may well ”perform” a work by running off a motion picture or playing a sound recording as part of its output, but its internal operations do not appear to us to fall within this concept.

To perform a work includes the transmission of a contemporaneously perceptible rendition of the work and is distinguished from the delivery of a work, including electronic delivery via transmission. 11Thus, uploading of a work to a web site, like downloading, is not a performance.

The nature of a performance

A “performance of a work”, then, follows from this definition. If “performing” means rendering a work so that it is contemporaneously perceptible, a “performance of a work” is something that is contemporaneously perceptible. It is intangible—the Copyright Act refers to the tangible objects that embody works as copies. It is conceptual, an act rather than a thing. 12performance“, Merriam-Webster Online, “an activity (such as singing a song or acting in a play) that a person or group does to entertain an audience”. A performance is not the actor on stage, nor the sound waves emanating from a speaker, nor the photons transmitted across fiber optic data lines. 13“Very few people gather around their oscilloscopes to admire the sinusoidal waves of a television broadcast transmission. People are interested in watching the performance of the work.” Fox Television Stations v. BarryDriller, at 5 (C.D. Cali December 27, 2013).

It is also, perhaps, worth pointing out that it is not a performance that is embodied in a copy but a work. 14Congress separately and outside the Copyright Office has made it a criminal offense to fix, without authorization, a live musical performance in a copy, thus underscoring this distinction. See 18 U.S.C. § 2319A. A performance is, by definition, incapable of embodiment; indeed, one does not even need a copy to perform a work, as is the case of a singer singing from memory.

And it is a conceptual unity. Consider two members in an audience. Neither will be perceiving the exact same thing, both because they are in different locations and because they themselves have variations in their eyes and ears that shape their personal perception. But conceptually there is only one performance of the work, and it is exactly the same for each audience member for copyright purposes.

The same holds true if a performance reaches its audience via transmission rather than via sound and light waves through the air. The Central District Court of California explains why in its decision enjoining FilmOnX (then “BarryDriller”):

The definition section sets forth what constitutes a public performance of a copyrighted work, and says that transmitting a performance to the public is a public performance. It does not require a “performance” of a performance. The Second Circuit buttressed its definition with a “cf.” to Buck v. Jewell-La Salle Realty Co., 283 U.S. 191, 196 (1931), which interpreted the 1909 Copyright Act’s provision of an exclusive right to publicly perform a musical composition and held that “the reception of a radio broadcast and its translation into audible sound” is a performance. But Buck, like Cablevision and this case, was concerned with a copyright in the work that was broadcast. The Supreme Court was not concerned about the “performance of the performance” – instead, it held that using a radio to perform the copyrighted song infringed the exclusive right to perform the song (not to perform the performance of the song).

The Transmit Clause explicitly recognizes this conceptual unity, saying a work is performed publicly when it is transmitted even if “the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.” There could be thousands of separate transmissions, but there is still only the performance.

These may seem like inconsequential distinctions, but I think they undermine even more the Second Circuit’s notion that we can conceptually sever performances of the same work made by the same actor but from distinct copies. If the Transmit Clause spoke about the audience capable of receiving the physical transmission, as the Second Circuit says it does, then it would be possible to have multiple private performances since physical transmissions are discrete and separable. But it doesn’t—the Transmit Clause speaks about the audience capable of receiving the “performance of a work”, which remains a conceptual unity no matter how scattered the audience is. 15Likewise, the Transmit Clause does not speak about “the potential audience of a particular ‘work’,” an alternative interpretation that the Second Circuit raises and dismisses.

This interpretation of performance is, in my opinion, far more consonant with the statutory language and the ordinary use of the terms than the Second Circuit’s “unique copies” interpretation. It also means, as we’ll see in the next installment of this series, that most of the heavy lifting (at least for issues arising in the cloud computing context) occurs in the causation inquiry, i.e., “who is the performer”?

References

References
1 I also flubbed when I referred to the three situations as being “mutually exclusive”. What I meant instead is that each situation excludes the previous. For example, a “place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered” could include a place open to the public, but that would make the previous clause redundant. Similarly, the last clause, transmission to the public, could include transmission to the public in a public place, or transmission to a place where a substantial number of persons is gathered, but that would make the previous two clauses redundant.

I think it still follows that each situation is analyzed by exclusive means. We can presume (from the case law) that a “place open to the public” turns on the nature of the place. The “place where a substantial number of persons…” situation doesn’t turn on the nature of the place: if such a place were public, then the first situation covers it, while being a private place is insufficient to tell us if there are a substantial number of persons there. The same holds true for the last situation. The nature of the place is irrelevant (otherwise the first clause would be redundant), the size and relationship of the audience to each other is irrelevant (otherwise the second clause would be redundant). So what is there left to consider? The only thing I can come up with is the relationship between the performer and the audience. There could be others I am not aware of, but the relationship between performer and audience seems to provide a workable rule consistent with the Copyright Act’s structure and purpose.

2 17 U.S.C. § 101.
3 Act of August 18, 1856, 34th Cong., 1st Sess., 11 Stat. 138.
4 Act of January 6, 1897, 44th Cong., 2d Sess., 29 Stat. 694.
5 H.R. 4347 and S. 1006, 89th Cong.
6 627 F. 3d 64 (2nd Circuit 2010).
7 866 F.2d 278, 282 (9th Cir. 1989).
8 Columbia Pictures Industries v. Redd Horne, 749 F.2d 154 (3d Cir.1984); Columbia Pictures Industries, Inc. v. Aveco, Inc., 800 F.2d 59 (3d Cir.1986).
9 In support of this conclusion, the court relied on Fourth Amendment case law holding that hotel guests have constitutional protection from unreasonable searches and seizures in hotel rooms.
10 And indeed, the Copyright Office’s 1965 Supplementary Report adds support.

In general the concept of ”performance” must be distinguished sharply from the reproduction of copies on the one hand and the exhibition of copies on the other. It has been suggested that some of the internal operations of a computer, such as the scanning of a work to determine whether it contains material the user is seeking, is closely analogous to a ”performance.” We cannot agree, and for this reason we deleted from the definition of ”perform” the ambiguous term ”represent” which appeared in the 1964 bill. A computer may well ”perform” a work by running off a motion picture or playing a sound recording as part of its output, but its internal operations do not appear to us to fall within this concept.

11 Thus, uploading of a work to a web site, like downloading, is not a performance.
12 performance“, Merriam-Webster Online, “an activity (such as singing a song or acting in a play) that a person or group does to entertain an audience”.
13 “Very few people gather around their oscilloscopes to admire the sinusoidal waves of a television broadcast transmission. People are interested in watching the performance of the work.” Fox Television Stations v. BarryDriller, at 5 (C.D. Cali December 27, 2013).
14 Congress separately and outside the Copyright Office has made it a criminal offense to fix, without authorization, a live musical performance in a copy, thus underscoring this distinction. See 18 U.S.C. § 2319A.
15 Likewise, the Transmit Clause does not speak about “the potential audience of a particular ‘work’,” an alternative interpretation that the Second Circuit raises and dismisses.

63 Comments

  1. Can’t wait for part three–the whodunit! Personally, I think a service is transmitting a performance any time its equipment literally transmits a performance–even if the customer stores his own file on the service’s equipment and initiates his own playback. If I stream a video from YouTube, YouTube performs it publicly when it transmits the performance to me (under the “to the public” prong of the Transmit Clause). I then perform the work by rendering it on my computer, which is a private performance. I’m curious to see whether you think there’s any crossover, such as saying it’s both YouTube and I publicly performing the work when I initiate playback. Also curious about how you tie in volition.

    I think your analysis here on the “to the public” language is correct. Conceptually, as I parse the Transmit Clause, there’s three senses of public: (1) public place, (2) semipublic place, and (3) “to the public.” A public place is just any place open to the public. Whether there’s anyone at that public place when the transmission occurs is irrelevant. A semipublic place turns on the nature and number of the audience at a given place. And unlike with a public place, I think the actual nature and number of the audience at the time of transmission counts (since Clause 1 says “is gathered” for this prong)–though I don’t think they actually have to receive the transmission (they just have to be there).

    A transmission “to the public,” however, does not turn on the place where the transmission is received, and it includes a transmission sent to even (and especially) a private place. What makes it “to the public” turns only on the relationship between the transmitter and the receiver. And as with the other two prongs, I don’t think that actual receipt of the transmission is relevant. We don’t need to consider whether the receiver is at a certain place, because the “to the public” prong is agnostic as to the place where the transmission is either received or receivable. You just look at the relationship between the transmitter and the receiver, and that tells you whether it’s public or private.

    • Personally, I think a service is transmitting a performance any time its equipment literally transmits a performance–even if the customer stores his own file on the service’s equipment and initiates his own playback.

      […]

      What makes it “to the public” turns only on the relationship between the transmitter and the receiver.

      So when you transmit a playback of a recording on your Cox multi-room DVR from your den to your bedroom, that is a performance by Cox. And given the nature of your relationship with Cox, it is a public performance.

      And more generally, if you buy a slingbox in order to sling content from your home to yourself, then the transmissions that slingbox makes are private transmissions by you. On the other hand, if you rent a slingbox to sling content from your home to yourself, then the transmissions are public transmissions by the party that rented the device to you.

      I doubt that you would find many people, or many courts, that would agree with you on these conclusions.

      • So when you transmit a playback of a recording on your Cox multi-room DVR from your den to your bedroom, that is a performance by Cox. And given the nature of your relationship with Cox, it is a public performance.

        No. I think there is a geospatial element. When I transmit from my den to my bedroom and watch it on my bedroom television, there are two performances. The transmission from the den to the bedroom is a performance, and the rendering on my television in my bedroom is a performance. As to the first performance, the transmitter is me in my private place transmitting to me in my private place. That is not a transmission to a public place or to a semipublic place, so it’s only a public performance if I am transmitting “to the public.” Since I am not “the public” as compared to me, I am not transmitting it “to the public” and it’s not a public performance. Nor is my rendering it in my bedroom a public performance since it’s in a private place, i.e., not a public place or a semipublic place.

        And more generally, if you buy a slingbox in order to sling content from your home to yourself, then the transmissions that slingbox makes are private transmissions by you. On the other hand, if you rent a slingbox to sling content from your home to yourself, then the transmissions are public transmissions by the party that rented the device to you. I doubt that you would find many people, or many courts, that would agree with you on these conclusions.

        I don’t think whether the Slingbox is rented or purchased matters. Can you explain to me why you think it does or doesn’t matter?

        • I don’t think whose equipment it is should matter. But you certainly seemed to say that you do. In fact, you seemed to say that it is the defining attribute:

          Personally, I think a service is transmitting a performance any time its equipment literally transmits a performance–even if the customer stores his own file on the service’s equipment and initiates his own playback.

          How can you reconcile that with the fact that the Cox DVR you use is Cox’s equipment, or that a rented slingbox is equipment belonging to the party who rented it to you, if you believe that transmissions from those pieces of equipment are transmissions by yourself?

          • It’s the difference between ownership and possession. But before you complain that my distinction is arbitrary, please tell me what your distinction is–what’s your analysis? Or is your claim that it’s irreconcilably unclear such that the world is paralyzed until Congress steps in?

          • Well, if by “possession” you mean “having physical custody or control” then yes, I do think it’s arbitrary, but more importantly, I think it’s unworkable as a legal concept. If physical possession is the key criterion, then when Google colocates a CDN server at Cox, Cox becomes the transmitter of any Youtube videos transmitted from that server, because Cox has physical possession of the server.

            Likewise, if I have a server at my place of business serving content, then I am the transmitter. But when I pay a hosting company to host my server (that I own) within their facility, then they become the transmitter, by virtue of taking physical possession of the server.

            Or perhaps strangest of all, if Cox uses their DVR in your physical possession to transmit a program to a third party, you are actually the transmitter.

            If by possession you mean some concept of control other than physical custody, then it starts sounding to me a lot like the volition standard used by the Second Circuit in determining who the author of copies was in Cablevision and Aereo (and similarly used by the Ninth Circuit in Fox v. Dish litigation over the Hopper) . In that case, it becomes like the customer making a copy at a copy shop, or going online to print pictures at the local drugstore. Somebody else owns and has physical custody of the equipment, but the customer rents use of the equipment (even if temporarily, and even if remotely), and supplies the volition for printing/copying something in particular.

            This seems to me a plausible approach, but it produces results opposite that which you are arguing for, of course. It leads directly to the conclusion that if I rent equipment, even remotely, and exercise control over that equipment to transmit a performance, then I am the transmitter, not the party renting the equipment to me.

            And I don’t think the world is paralyzed. I just think that the model Aereo has adopted is compliant with the Copyright Act, and if that’s deemed a problem (and I’m sympathetic to the argument that it is a problem, inasmuch as it creates an uneven playing field), then Congress is the entity empowered to properly fix it.

          • Let me ask you this. When I go to YouTube and initiate playback of some work, is YouTube publicly performing the work that is transmitted to me? Now, let’s say I instead play back a file I uploaded to a file locker service. Is the file locker service publicly performing then? What if I play back a file that someone else uploaded to the file locker service? If the result is different for each, then why is it different?

          • Let me ask you this. When I go to YouTube and initiate playback of some work, is YouTube publicly performing the work that is transmitted to me?

            I think pretty clearly so.

            Now, let’s say I instead play back a file I uploaded to a file locker service. Is the file locker service publicly performing then?

            If we’re talking about a file locker service that is renting you storage for your exclusive use, then I think you are privately performing the content, just as you would be if you had rented a file locker device from the same company, and placed it within your home.

            What if I play back a file that someone else uploaded to the file locker service?

            Not sure if we need more information here. My kneejerk reaction is to say that if the storage is yours exclusively, and you are the only person who can initiate a playback, and only to yourself, then the source of the file isn’t really relevant to the issue of playback. The act of copying that file into your storage locker may well be an infringement, but under normal “file locker” circumstances, I don’t think the playback is.

            If the result is different for each, then why is it different?

            The two file locker examples are different from Youtube inasmuch as Youtube is “renting” (offering) content as their service, whereas the file locker service is renting you equipment as their service, and you, the user, are using that equipment to manage content.

          • If we’re talking about a file locker service that is renting you storage for your exclusive use, then I think you are privately performing the content, just as you would be if you had rented a file locker device from the same company, and placed it within your home. *** The two file locker examples are different from Youtube inasmuch as Youtube is “renting” (offering) content as their service, whereas the file locker service is renting you equipment as their service, and you, the user, are using that equipment to manage content.

            If I understand your argument, it’s that the identity of the transmitter changes when the transmission comes from remote equipment which is leased by the person initiating and receiving the transmission. This is just the volitional conduct test: Even though the service’s equipment, though leased, is in fact transmitting the work, the lease makes it so that the lessor is not proximately causing the transmission. Only the lessee is directly causing the transmission to occur.

            I don’t think that jives with the case law on public performance. Zediva tried to argue that it was just renting DVDs and DVD players, but the court didn’t buy it.

            Zediva’s idea of a “rental” sounds about as shaky as Aereo’s:

            Defendants describe their service as allowing customers to “rent” a particular DVD and DVD player for 14 days. However, Defendants’ customers do not have access to or control over a specific DVD or DVD player. Instead, Defendants stream the content of the DVD to a customer for a maximum period of four hours, provided that the customer does not pause it for more than one hour during that time. After four hours of total “rental” time or an hour-long pause, whichever occurs first, Defendants use the DVD player containing the same DVD to transmit the Copyrighted Work to a different customer. When the first customer makes a request to resume viewing, the transmission may be sent from a different DVD or a different DVD player than the one originally used to transmit the Copyrighted Work in the earlier “rental” period. According to their website, if all of the copies of a particular Copyrighted Work are “rented out” when a customer wants to view it, that customer “can request to be notified, via email, when it becomes available.”

            Warner Bros. Entm’t Inc. v. WTV Sys., Inc., 824 F.Supp.2d 1003, 1007 (C.D. Cal. 2011).

            And the court didn’t buy the argument that it wasn’t Zediva transmitting the performances:

            Although Defendants are clearly transmitting performances of Plaintiffs’ Copyrighted Works, Defendants argue that their service offers “DVD rentals” rather than transmissions of performances, which is similar to the unsuccessful argument made by the defendant in On Command Video Corporation v. Columbia Pictures Industries, 777 F.Supp. 787 (N.D.Cal.1991).5 In On Command, the court held:

            Plaintiff’s argument that On Command’s system involves not “transmissions” but “electronic rentals” similar to patrons’ physical borrowing of videotapes is without merit. On Command transmits movie performances directly under the language of the definition. The system “communicates” the motion picture “images and sounds” by a “device or process”—the equipment and wiring network—from a central console in a hotel to individual guest rooms, where the images and sounds are received “beyond the place from which they are sent.” The fact that hotel guests initiate this transmission by turning on the television and choosing a video is immaterial. Id. at 789–90 (citations omitted).

            As in On Command, Defendants’ Zediva service transmits performances of Plaintiffs’ Copyrighted Works “directly under the language of the statute.” In this case, the Zediva service “communicates” the “images and sounds” of Plaintiffs’ Copyrighted Works through the use of a “device or process”—the equipment, including various servers, and internet—from a central bank of DVD players to individual customer’s computers, where the images and sounds are received “beyond the place from which they are sent.” See, also, Columbia Pictures Industries, Inc. v. Redd Horne, Inc., 749 F.2d 154, 160 (3d Cir.1984) (in a case involving “in-store rentals” of video cassettes for viewing only on the stores own video cassette recorders, the Third Circuit held that the defendants “never disposed of the tapes in its showcasing *1010 operations, nor did the tapes ever leave the store. At all times, [the defendants] maintained physical dominion and control over the tapes. Its employees actually played the cassettes on its machines. The charges or fees received for viewing the cassettes at [the defendants’] facilities are analytically indistinguishable from admission fees paid by patrons to gain admission to any public theater. Plainly, in their showcasing operation, the appellants do not sell, rent, or otherwise dispose of the video cassette.”). As in On Command, the fact that Zediva’s customers initiate the transmission by turning on their computers and choosing which of Plaintiffs’ Copyrighted Works they wish to view is immaterial. On Command, 777 F.Supp. at 790.

            Id. at 1009-10.

            I don’t agree that Aereo is doing anything different than Zediva. The volitional conduct on Aereo’s part is in its having implemented a system that transmits performances to the general public. You agree that YouTube publicly performs works, even though it only transmits user generated content for free. I see more volitional conduct with Aereo, which supplies the content and charges its users to access it, than I do with YouTube. YouTube was able to find safe harbor in 512(c), but Aereo is not even arguing that the DMCA protects them. Why do you think that is?

          • Devlin, it seems like you very much want to resolve this question only by analogy to other cases. The trouble is, analogies have their limits, and I’m sure (in fact, I think it’s pretty clear) that we could find multiple cases on both sides. I don’t see any particular reason to give Zediva, a district court case involving different facts, greater weight than other cases.

            I have to confess, I’m a bit baffled by your great reluctance to examine this issue in pursuit of a coherent underlying theory, that can then be used to decide cases in a coherent fashion. In a case like Zediva — indeed in many cases — are the courts not frequently simply making their own ad hoc judgment calls?

            I would expect an academician, in particular, to demand coherent theories, and not merely a collection of point decisions, from which we try to infer what the outcome should be for another decision. As a scientist, this seems to me like a astrophysicist who is content to predict the positions of heavenly bodies based on extrapolation from prior measurements, rather than trying to identify the laws of gravity that actually governs their motions. In fact, an astrophysicist who argues that we shouldn’t try to identify the law of gravity.

            This seeming approach just doesn’t make sense to me.

            By introducing a case like Zediva, you are actually complicating the analysis, because you are introducing more variables (e.g., Zediva is about DVDs, where the content is rented, in addition to the alleged equipment rental, and in fact, not even a single copy of the content is necessarily used for a single rental).

            I am trying to do the opposite — to simplify the analysis by separating the variables for separate analysis. That’s why I previously proposed (in a separate discussion) that we first consider a hypothetical case where the exact same equipment is used in-home vs. remotely. After we’ve analyzed that, if we haven’t found a coherent basis for distinguishing the two cases, we can walk through the steps transitioning from the exact same equipment as in-home to the actual equipment used for a case like Aereo or Cablevision.

            I’m well aware that it’s hard to see where the line should be drawn. (In fact. that’s been my point — that drawing such a line will be arbitrary.) But you seem to want to hide from that problem. To my view, that’s not a solution. It’s merely denial of the fact that there is a problem. When you say that’s hard, and refuse to try to identify a coherent criteria for evaluating cases, what I see is somebody who is content to have courts continually making ad hoc judgment calls — effectively, using an “I know it when they see it” standard. That is certainly not what the courts should be doing. And so long as we lack some kind of coherent theory, criteria and legal test, that’s all we’re going to have. And that should not be satisfactory to anyone.

            We really do need somebody to provide a coherent basis for the distinctions that are being made. I’ve been pretty clear that I don’t see a statutory basis for what I believe would need to be arbitrary criteria to distinguish between an in-home equipment rental and a system like Aereo’s or Cablevision’s. And as such, I think it’s something that only Congress can properly provide. But first things first. Let’s forget who provides the theory/criteria/legal test, whether it be Congress or the courts. Shouldn’t we have a coherent basis for distinguishing these cases? Don’t we deserve that?

            Even if lower courts are not charged with doing such, is not the Supreme Court charged with understanding and clarifying the law at that level?

            And even if the Supreme Court were not — in fact especially if it were not — should not legal academics be endeavoring to identify such coherent theories, criteria and legal tests, rather than endlessly forecasting the positions of the planets based merely on extrapolation?

          • Of course a coherent theory would be nice, which is why I tried to pigeonhole your theory into the existing doctrine by identifying it as being an iteration of the volitional conduct test. Whether you agree or even realize that your argument is one of volition, I still don’t know. My method for approaching this problem is the one I learned in law school. You start with the statute, look to the congressional record for guidance when the statute is unclear, look at relevant case law to see how the statute has been applied to concrete applications, etc. You consider the facts, the law, and the policy. The extent to which you rely on each is more art than science. It would be great if we had a coherent theory that let us plug in the variables like a computer program, but that’s not how these things are actually decided. I think it’s hard for some people to understand that the law is far more indeterminate than they think it should be. As I frequently tell my wife, when it comes to the law, I don’t think people really want to see how the sausage is made. I was a math teacher before going to law school, and it was honestly a shock to me that the answer is almost never black and white. While there are of course simple cases where the answer is clear, those aren’t the ones we studied and those aren’t the ones people tend to get all excited about—like with Aereo.

            There’s a lot of moving parts. Take the volitional conduct test. That test is not grounded in the statute. It was basically invented by the district court in the Netcom case. The concern there was that holding truly passive conduits directly liable for infringement would shut down the internet. That test has been stretched and stretched by subsequent case law. The worst offender being the Second Circuit in Cablevision, IMO. What used to protect passive conduits is now used to protect active participants, and the theory is that if someone programs a computer to do a task, then the party that programmed the computer to do the task isn’t responsible for the task once it happens. I think that’s nuts. But that’s the law, at least in the Second Circuit and perhaps also the Ninth. But that’s not the law in every circuit, and the scope of the volitional conduct test seems to change based on which underlying exclusive right is at issue. There is no scientific test of how much volition is enough volition to say that a party is directly causing something. You have to reason by analogy to other cases. You say, “Judge X in Case Y said Defendant Z had sufficient volition when he did A, B, and C.” And the other guy will say will try to argue why it’s not analogous.

            But turning back to Zediva, I think the key to the analogy there is that Zediva was simply transmitting performances of works to any member of the public who signed up for the service. I think the same applies to Aereo. Thus, I’m reasoning by analogy—a main staple in the lawyer’s arsenal of legal arguments. The point I was trying to make is that the idea that the equipment is “rented”—an argument you place much weight on—was rejected by the Zediva court. With your “rented” argument, I think you’re simply trying to disaggregate Aereo’s service into a service that allows customers to make their own copies on leased equipment and a service that allows customers to stream back their own copies from leased equipment. On this view, Aereo is more like a private file locker, and since Aereo’s role is passive enough, it can’t be directly liable for infringement under reasoning, as in the Hotfile decision, where the volitional conduct test was applied. In my opinion, the problem with this approach is that it fails to take Aereo for what it really is—a transmission service. I don’t see it as being two distinct parts, one for making copies and one for streaming them. I see it as a unified whole. On my view, Aereo is simply a service that captures and retransmits/transmits broadcasts commercially and on command. It is “any device or process,” construed broadly, as Congress clearly intended.

            I think the analogy that you’re employing, namely, that Aereo is just a file locker service, breaks down. I realize that you want the analysis of a file locker’s liability to be formulaic to some extent, but it’s just not that easy. Take YouTube, for example. YouTube has videos from numerous sources, including its partners, which it manipulates and transmits in numerous ways. I doubt even YouTube knows the extent to which its actions implicate the copyright rights of others. What’s interesting too is that since the advent of the DMCA, we’re getting less case law about whether copyright rights have been infringed and more case law about whether the service has safe harbor. I could be wrong, but I don’t think the courts in the YouTube litigation ever even reached an opinion of whether YouTube was directly performing the works. Instead, the focus was on whether YouTube had safe harbor and the extent of the “by reason of the storage at the direction of a user” language in 512(c).

            My argument for why Aereo is not like a file locker turns on what can be discerned from the case law, and one significant detail is whether the service provides the source materials. This is why I think it’s wrong to disaggregate Aereo’s service into two parts, each that looks like a file locker. File lockers are services, like YouTube or Megaupload, where people upload and then download/perform/display their files. You can’t upload a file to Aereo because it supplies the source materials for you. What’s interesting about Aereo too is that there are all sorts of doctrines off the table. There’s no DMCA, no fair use, and no secondary liability. I don’t know if this is a good or a bad thing, but it’s certainly interesting. This will force the Court, I predict, to answer the question of who is performing the work that the Second Circuit punted on—twice. Once the Second Circuit’s error in concluding that the transmissions weren’t “to the public” is corrected, as I think it should be, then the tougher question of who is doing the performing comes to the fore. Without the DMCA to consider, the issue will then turn on the extent to which Aereo directly causes the performances, i.e., the volitional conduct test. And which version of the volitional conduct test to apply will be a matter of debate. The more like a passive conduit Aereo is thought to be, the more likely it is not directly performing. On the other hand, the more active it is thought to be, the more likely it is directly causing the performance. Your argument that the equipment is “rented” would be a factor in this analysis. But the extent to which it should be a factor is a matter of debate, and it’s not ducking the issue to point out that it’s part of the issue.

            Both sides, of course, can and will make forceful arguments about where the line should be drawn and on which side Aereo should come down. And the fact that both sides can make good arguments should tell you that it’s not as simple as applying some formula. I think Ray Nimmer captures the issue perfectly: “In an environment involving increasingly ‘intelligent’ systems, the question of ‘online who engages directly in the copying, display or performance of a work’ frequently involves having to draw close distinctions, grounded in technology choices, between the actions of two or more persons.” Raymond T. Nimmer, Law of Computer Technology § 15.6 (2012). The Court will have to weigh the volitional conduct on Aereo’s part, and it will have to confront the issue of how to frame Aereo’s system: as an overly-complicated device or process that transmits performances to the public like any other video-on-demand service, as I see it, or as a just-complicated-enough file locker service, as you see it. And how the issue gets framed will be a function of how the Court analogizes Aereo to other systems.

            And I should mention that I disagree that I’m running from the problem. I’m trying to tackle the problem head on, using the same analytical tools that the courts use. Of course I recognize that the line is difficult to discern, but I also realize that which side of the line Aereo falls on must be determined nonetheless since that is the issue presented. Being that Aereo uses a device or process which lets members of the public watch either live (that is, delayed by 6 seconds, IIRC) or delayed broadcasts, I think it’s exactly the type of service the Transmit Clause was meant to proscribe. And, in fact, I think it does proscribe it. You seem to think my analysis is incoherent, but I think it’s perfectly coherent to say the issue is and will be decided by looking at volitional conduct and analogizing Aereo to other cases, with the backdrop of the Transmit Clause’s purpose lying in the background. I think you want to view one fact in isolation as being determinative, for example, by saying that “rented” = no volition on the service provider’s part. I think the Court will instead do a balancing, where the “rented” argument is but a part of what gets balanced. That we can’t say ex ante exactly what this balancing test is and how all the variables are to be weighed, this is a feature, and not a bug, of the law generally. It’s how the sausage gets made.

          • Devlin, I’m glad you said you were a math teacher once upon a time, because it means I can safely use math analogies. 🙂

            The various prior decisions we have that we try to draw from as precedent are, at best, a very noisy sampling of an unidentified function. (At worst, they are simply random.) Giving them the benefit of the doubt, there is some unidentified underlying function that is the legal criteria for making a determination. The noise is the judgment/interpretation/estimation of the courts, of both the facts of the case, and the underlying function. Resolution of the issue requires separating the noise from the underlying function, and then a determination as to whether the underlying function that the various courts have been applying is actually correct (and if not, specification of the correct function).

            When one attempts to reconcile all the different data points, it’s like a polynomial fitting exercise. You can perfectly fit all the points, but what you get is a very high order polynomial that behaves very, very strangely between the points. Even if we throw out some of the farthest outliers, you almost certainly still get a pretty strange polynomial.

            It seems to me that using such a polynomial is not a coherent 😉 way to decide a case—and certainly, it isn’t what should be happening by the time the case gets to the Supreme Court. I would expect that both parties are going to appeal to some precedents, but they will fundamentally be arguing not for a polynomial to fit all of the data points, or even most of them, but for some much lower order function that fits just a couple of the prior points, and for which there is a separate argument that, philosophically, the proposed function is justified and produces sensible results if applied across a range of cases.

            And I would expect that the Supreme Court will do the same—either accepting one of the parties’ proposed functions, or crafting one of their own.

            This is why I think that the attempt to address this issue entirely by analysis of precedent is mistaken. I think that polynomial fitting is what you do at the trial court level. But by the time you get to the Supreme Court, it is no longer the exercise. Cases generally don’t get to the Supreme Court if the various polynomial fitting exercises within the lower courts produced coherent results. And since the Court, itself, isn’t going to handle the case as a polynomial fitting exercise, nor should we be doing such in evaluating what ought to happen.

            Anyway, to your substantive comments, I think you’ve got my argument, and the analysis in this case, wrong in several regards.

            For one, while I have been satisfied by the Second Circuit’s analysis regarding volitional conduct (and you would be correct, btw, in surmising that I lack expertise in related tests), my view does not revolve around any volitional conduct test.

            Rather, my view revolves around the philosophical view that any distinctions that are made need to be principled distinctions. And what we have happening here (and undoubtedly all too often in cases, generally) are distinctions being drawn, both by commentators like yourselves, and frequently by courts, on bases that are completely arbitrary, often seemingly without even realizing it.

            For example, you talk about the volition with respect to programming a system to perform a specific function. The problem with this is that the programming of an RS-DVR system, for example, is indistinguishable from the programming of an in-home DVR system, in this regard. Both are programmed, by the same party, to perform the same exact function, based on the same exact user control. If an operator’s volitional conduct in programming an RS-DVR renders them responsible for its copies/transmissions etc., then the same must be true of an in-home DVR the operator supplies. Given that, if those two cases are going to be distinguished, it has got to be by something other than by a volitional conduct analysis.

            In other words, it’s only by a lack of understanding of the technologies that somebody like Denny Chin can make a (hand-waving, only) argument that set-top DVRs are somehow distinguished from RS-DVRs in terms of the operator’s “ongoing operational involvement.” The distinction has no basis in fact, whatsoever, and is simply a completely arbitrary distinction, manufactured from whole cloth by Chin.

            Furthermore, my view of Aereo doesn’t rely on an analogy to file lockers. Although the Second Circuit’s analysis relies entirely on Aereo’s use of an RS-DVR to conclude that performances are private (to the point where the fact that they use individualized antennas doesn’t even come into play), my own view would still see the user, rather than Aereo, as the party engaged in transmissions, even if there were no storage, precisely because Aereo does have individualized antennas which are rented to user for their exclusive use, while the user is using them.

            I can certainly draw analogies to file locker services quite easily—and I can definitely tell you that I have yet to see a proposed alternative theory whereby Aereo is infringing, that doesn’t either render file locker services infringing also, or require an arbitrary inconsistency—but my view is not dependent on drawing analogies to file locker services.

            (And btw, Youtube is not a file locker service, in any way shape or form. (I’m less familiar with Megaupload, but at first blush it does not appear to me to be a file locker service.) Youtube is, rather, a publishing platform. And as such, it is one of the things that DMCA was specifically crafted to address, where the content that a user uploads is not private, but instead is made available to the public. And Congress specifically deemed that, in such cases, the platform provider should not be liable for copyright infringements generated by user action, except to the degree that they failed to act once the underlying copyright infringements by users were brought to their attention.)

            My view fundamentally comes down to three things. I don’t believe there is a principled way to distinguish (and certainly none based in the statute):

            between a DVR (or antenna, or slingbox) that I own, vs. one that I rent from a third party, vs. one rented from my MVPD
            between a DVR (or antenna, or slingbox) located within my home, vs. at a remote location rented from a third party, vs. at a remote location rented from my MVPD
            between a conventional DVR (or antenna, or slingbox) where the (critical) components are statically allocated for my exclusive use on a semi-permanent basis, vs. an RS-DVR-type (or Aereo-type) implementation, where the (critical) components are dynamically allocated to my exclusive use, when I am actually using them.

            Furthermore, my view is that to distinguish these cases requires that we be able to identify specific criteria that actually distinguishes one or more of these.

            To my view, to say that one case is clearly private and another is clearly public, without identifying specific criteria for evaluating such, is nothing more than “I know it when I see it.” And any “balancing test” is going to be suspect, absent clear criteria for distinguishing the different pieces to be balanced, and a clear reason why a “balancing” is required. (Something vague like “ownership, location and duration of exclusive allocation are all components to be considered in balancing who the actor is” is just another way of saying “I know it when I see it.”)

            On the other hand, my view is that there is a very simple and coherent way to distinguish cases—the distinction I have mentioned multiple times: equipment rental vs. transmission services (or content rental, in parts of our discussions).

            We know that the Copyright Act was drafted to specifically allow users to receive and perform broadcasts for themselves. We know that the Copyright Act does not in any way distinguish between the user owning vs. renting equipment for such purposes.

            Equipment rental is also a pretty clearly-defined thing, and it is not based on duration. Rather, it is based on exclusive use. If I lease a car for three years from the local car dealer, or if I rent a car for a day from Hertz, or if I rent one for 30 minutes from Zipcar, each is an equipment rental because I get exclusive use of the equipment (the car) for some period of time.

            There’s no basis in the statute for equipment used to receive, record and perform broadcast television signals to be treated any differently. (Or for that matter, the equipment to record and perform channels in a cable package, or the equipment for storing and streaming files.)

            The Copyright Act was also drafted specifically to address transmission services at the time, as well as anticipated future developments; but whether a cable system or an information system, at the time, those were conceived only as systems easily distinguishable from equipment rental in that they utilized shared facilities: e.g., a shared antenna for a cable system potentially serving multiple users simultaneously, or shared storage in an information system, potentially serving the same file to multiple users (simultaneously or sequentially).

            So it seems to me that there is a very simple test that can be applied, completely consistent with the Copyright Act and without requiring any invention. Is it an equipment rental, based on the long-established definition of such based on exclusive use? If so, then the renter is the actor, not the party offering the equipment for rent.

            I believe this very simple approach is coherent in the sense that it “correctly” classifies all instances anticipated at the time the Copyright Act was drafted. (i.e., by “correctly” I mean that it would not produce any surprises, from the standpoint of the drafters, for technologies that existed or which they anticipated.) Furthermore, I believe it also produces the “correct” result in almost all cases that now exist or which are now anticipated.

            There will be some cases, such as Aereo, where some (or even all) will conclude it produce an “incorrect” classification. But that is merely the nature of the beast. No matter how hard one tries to future-proof language in a statute, it is to be expected that it will eventually fail to handle all new, unanticipated situations in the “correct” fashion. (That statutory language is perceived as failing to achieve its presumed intended purpose, when applied to a new technology some 40 years later is not a legitimate justification for the courts to intervene.)

            So this is a very simple and straightforward interpretation, requiring no invention of new and complex criteria, which produces coherent results (even if a few of them are deemed undesirable by many).

            As alternatives, we have yet to see anything presented that is even remotely coherent without the introduction of some new and complex criteria (often unidentified) to “correctly” distinguish between “good” and “bad” use-cases.

            (I’m open to considering alternative theories, but I will remain extremely skeptical that a straightforward and coherent alternative, without the introduction of arbitrary criteria, actually exists unless and until we actually see one presented.)

            From my perspective, what we have are parties who, effectively, think that “incorrect” classification of Aereo is so unacceptable that it is appropriate for the courts to step in and create criteria to distinguish it in order to produce the “correct” result.

            (Actually, this is giving them the benefit of a presumption. So far as I can tell, most of the parties arguing against Aereo don’t even get to that level, because they want to rely on a very high level analysis based on the “any device or process” language to declare it “obviously” infringing, without actually evaluating the implications of that approach. That’s what I referred to, in one of my articles, as reaching a conclusion in one step that they could never have reached in two steps.)

            My view is quite simple. Unless and until somebody can produce an interpretation of the statute that is as simple, coherent, and lacking in the introduction of arbitrary criteria as the simple proposal I made above, then I don’t think there are any other contenders for being a reasonable interpretation, based in the statute.

          • Note that there were supposed to be numbers (1, 2, 3) next to the three things I asserted could not be distinguished in a principled way, but I used an order list, which apparently the comment system can’t handle. Just note that each time “between” appears, it is the start of a new item.

          • I appreciate the comments as always, J.S. Let me sketch out how I think the analysis should go, and hopefully this will address some of your points.

            The issue is whether there’s a public performance when customers watch broadcasts streamed from Aereo. The Transmit Clause tells us a public performance occurs when there is (1) a transmission of a performance (2) that is to the public. The Second Circuit assumed the transmitter was Aereo, and it said the transmissions weren’t to the public because of the one-to-one relationship between the source copy and the customer. On this view, since the audience of each transmission is limited to one person, the performance is private.

            I argue that this is the wrong approach. The one-to-one relationship is not dispositive. There can be a one-to-one relationship where the performance is private, and there can be a one-to-one relationship where it is public. Whether or not the performance is to the public turns on the identity of the performer. The Second Circuit just assumed that Aereo was the performer and nevertheless found the performance to be private because of the one-to-one relationship. This has it exactly backwards. Whether the performance is to the public depends on who is directly causing the performance—and this is determined by the volitional conduct test.

            A service which is a truly passive conduit, providing only dumb pipes, is not a direct performer. On the other hand, a service which is an active participant in the performance, doing more than providing dumb pipes, is a direct performer. There is a dichotomy: either the service is a direct performer or it is not a direct performer. When a concrete case such as Aereo arises, it is necessary to decide which side of the line the service falls on. If there is sufficient volition on the service’s part, then it is directly performing. What constitutes sufficient volition, therefore, is the issue to be decided.

            I brought up YouTube as being a file locker, and this is because I’m using the term broadly. If I upload a video to YouTube and set it to private so that only I can stream it back, this to me is no different than a file locker. And, in fact, I think this is a private performance if I do stream it back because YouTube’s volition is insufficient to say that it’s directly performing. The volition on YouTube’s part is insufficient because here I’m streaming my own content to myself. It’s a private performance even though my relationship to YouTube is a public one because YouTube’s volition is insufficient to say that it’s directly performing the work. Since the direct performer is me and I’m transmitting it to myself, and not to the public, then the performance is private.

            I’m not too hung up on we call the service since I think whether it’s a direct performer depends on many factors which have to be taken into account. The case law helps us identify which factors are relevant, and they rest on the service’s own actions in bringing about the performance. I don’t think they’re at all arbitrary. One such factor is whether the service only lets the customer stream it back to himself (or those he shares the private link with) or whether the service makes it public, indexes it, and draws strangers in to view it. To the extent the service does the latter, as YouTube does with publicly-shared videos, its actions go to the volitional side of the scale. Since YouTube plays such an active role in getting strangers to stream videos, it is directly causing those performances to occur. Its conduct is proximately close such that it is a direct performer. But when a customer streams back his own non-publicly-shared video to himself, I think YouTube is not directly causing the performance.

            I think Aereo’s system is sufficiently volitional such that it is directly performing, notwithstanding the one-to-one relationship between copy and customer. It turns on how much volitional conduct Aereo has engaged in. One factor that is especially important here is that Aereo is supplying the source materials. In fact, I don’t think you can find any case law where the service provides the content and yet it is not a direct performer. And this isn’t some arbitrary distinction. The test is volition, and a service that actually provides the content to be streamed has engaged in much volitional conduct. I’m still working out the details of my theory, but I think there’s an argument that a service which provides the content to be streamed is per se a direct performer.

            Note too that Aereo supplies content in a different way than YouTube supplies content. While YouTube supplies content in the sense that it makes the content available, the content on YouTube is supplied by its users. With Aereo, by contrast, the customer doesn’t supply the content. Aereo itself makes the content available to the customer. YouTube lets people upload whatever content they want. Aereo doesn’t let its customers upload any content, and the only content available on Aereo is the content that Aereo itself provides.

            So when the Second Circuit said that Aereo is directly performing but that this performance is private, this makes no sense to me. Given that the relationship between Aereo and its customers is a public one, a determination that Aereo is directly performing is a determination that Aereo is publicly performing. Instead of presuming that Aereo is the direct performer, the court instead should have realized that whether the performance is public or private turns on the identity of the direct performer. And whether Aereo is the direct performer turns on how much volitional conduct Aereo has engaged in. Given that Aereo’s system only transmits content that it provides, it is the direct performer—or, at least, it is jointly transmitting with the customer. Since Aereo is transmitting a performance to the public, Aereo is publicly performing.

          • Devlin, thanks for laying out your thinking. I do still have several comments and questions/points to make.

            First, I don’t recall the Second Circuit holding that Aereo was the performer, and in a quick search of the holding, I wasn’t able to find evidence of such. Is there something I missed? I know that in Cablevision the Second Circuit explicitly did not make a determination as to whom the performer was. Rather, they concluded it was a private performance regardless of whom the performer was, and therefore decided that they didn’t need to determine who the performer was. I thought they had handled the issue in the same fashion in Aereo.

            (Perhaps you meant to say that their analysis presumed that they could have been the performer, instead of actually assuming they were the performer?)

            Second, Aereo (and I) would dispute your assertion that Aereo supplies the content being performed. Rather, Aereo provides an antenna, which the user may use to tune any of the available OTA channels, for the user to receive content, themselves. Aereo does provide program guide information to assist the user in recognizing what content is available, but that is no different than many other services that no one would assert are suuplying television content.

            To reach the conclusion that Aereo supplies the content requires an unstated assumption/conclusion on your part regarding who is the actor with respect to operation of the individual Aereo antennas.

            (Consistent with what the court observed in, I think it was, the MP3tunes case, to say that Aereo supplies the content in this situation is like saying that Google supplies content that is located using a Google search.)

            There would be a stronger argument that Cablevision supplied content in its RS-DVR, than that Aereo supplies content — because as a service provider, Cablevision of course does supply the content. (Not that I accept your suggestion that this point, alone, is dispositive of who the performer is.)

            Finally, all of the factors you have described and ascribed (rightly or wrongly) to Aereo apply to Cox, with respect to the DVR that they rent to you. Are you therefore prepared to conclude that Cox is, in fact, the performer when you use your Cox DVR to perform a recorded program? And further, given the nature of Cox’s relationship as performer to you as the recipient of the performance, the performance is a public performance, even when you are the only possible recipient, and only receive it within your own home?

            If not, then there must be some other criteria that you are applying, but which you have not discussed (in this last posting). *That* would be the criteria that I would be looking for you to explicitly identify, and which I would assert is, once identified, going to turn out to be arbitrary, and without basis in the statute.

            If so, then I would ask the question of why the case where Cox rented you the DVR differs from the case where Rent-a-center rented you the DVR, or where you purchased the DVR, and *that* would be the criteria that I would be looking for you to explicitly identify, and which I would assert is, once identified, going to turn out to be arbitrary, and without basis in the statute.

          • First, I don’t recall the Second Circuit holding that Aereo was the performer, and in a quick search of the holding, I wasn’t able to find evidence of such. Is there something I missed? I know that in Cablevision the Second Circuit explicitly did not make a determination as to whom the performer was. Rather, they concluded it was a private performance regardless of whom the performer was, and therefore decided that they didn’t need to determine who the performer was. I thought they had handled the issue in the same fashion in Aereo.

            You’re right about Cablevision. The Second Circuit explicitly presumed that Cablevision was a direct performer and then proceeded to focus on the audience of each transmission while applying the one source copy theory (which says that when there is a one-to-one relationship between the customer and the source copy, the performance is private).

            The Second Circuit in Aereo also assumed that Aereo was a direct performer. See, e.g., Aereo, 712 F.3d at 680 (“Aereo transmits to its subscribers broadcast television programs over the internet for a monthly subscription fee.”); Id. at 691 (“Both Aereo and Cablevision are making multiple private transmissions of the same work . . . .”). The court then went on to apply the one source copy theory, running through the same analysis as in Cablevision.

            My argument is that both Cablevision and Aereo are mistaken to focus on who is capable of receiving the transmission instead of realizing that the issue is also who is causing the transmission, since whether it’s “to the public” depends on who is transmitting to whom. If the relationship between the transmitter and the receiver is a public one, then it’s a public performance.

            Second, Aereo (and I) would dispute your assertion that Aereo supplies the content being performed. Rather, Aereo provides an antenna, which the user may use to tune any of the available OTA channels, for the user to receive content, themselves. Aereo does provide program guide information to assist the user in recognizing what content is available, but that is no different than many other services that no one would assert are suuplying television content.

            Without Aereo, the customer would not have the content. Aereo supplies the content by supplying the antennae which make the content available in the first place. The customer isn’t uploading a file to Aereo because he doesn’t have the file to begin with. He uses the Aereo service precisely because Aereo supplies him with that content which he otherwise doesn’t have.

            Consistent with what the court observed in, I think it was, the MP3tunes case, to say that Aereo supplies the content in this situation is like saying that Google supplies content that is located using a Google search.

            Google gives you a link and tells you where you can go to get the content. With Aereo, the customer doesn’t have to go anywhere else. Aereo has the very content that the customer is looking for.

            There would be a stronger argument that Cablevision supplied content in its RS-DVR, than that Aereo supplies content — because as a service provider, Cablevision of course does supply the content. (Not that I accept your suggestion that this point, alone, is dispositive of who the performer is.)

            I think both Cablevision and Aereo supply the content. They aren’t file lockers where people upload content they already have. They’re services where people go when they want to get the content in the first place.

            Finally, all of the factors you have described and ascribed (rightly or wrongly) to Aereo apply to Cox, with respect to the DVR that they rent to you. Are you therefore prepared to conclude that Cox is, in fact, the performer when you use your Cox DVR to perform a recorded program? And further, given the nature of Cox’s relationship as performer to you as the recipient of the performance, the performance is a public performance, even when you are the only possible recipient, and only receive it within your own home?

            When Cox publicly performs by transmitting from their facilities to my DVR, Cox has a license to make that transmission because they need one. When I later perform by playing back the copy on my DVR, I don’t need a license since I’m privately performing. When Cox sends me the transmission, it is the direct performer. When I play back the copy later, I am the direct performer.

          • You’re right about Cablevision. The Second Circuit explicitly presumed that Cablevision was a direct performer and then proceeded to focus on the audience of each transmission while applying the one source copy theory (which says that when there is a one-to-one relationship between the customer and the source copy, the performance is private).

            That’s not what I said. I said the Second Circuit did not hold (or presume) Cablevision to be the performer (emphasis added):

            Cablevision contends that (1) the RS-DVR customer, rather than Cablevision, does the transmitting and thus the performing and (2) the transmission is not “to the public” under the transmit clause.

            As to Cablevision’s first argument, we note that our conclusion in Part II that the customer, not Cablevision, “does” the copying does not dictate a parallel conclusion that the customer, and not Cablevision, “performs” the copyrighted work. The definitions that delineate the contours of the reproduction and public performance rights vary in significant ways. For example, the statute defines the verb “perform” and the noun “copies,” but not the verbs “reproduce” or “copy.” Id. We need not address Cablevision’s first argument further because, even if we assume that Cablevision makes the transmission when an RS-DVR playback occurs, we find that the RS-DVR playback, as described here, does not involve the transmission of a performance “to the public.”

            Saying “even if we assume” (for the sake of argument) is not the same as presuming. As I said, they explicitly did not determine who the performer was.

            Without Aereo, the customer would not have the content. Aereo supplies the content by supplying the antennae which make the content available in the first place. The customer isn’t uploading a file to Aereo because he doesn’t have the file to begin with. He uses the Aereo service precisely because Aereo supplies him with that content which he otherwise doesn’t have.

            And without Google, the customer would not have the content they find via a Google search.

            From Capitol Records v. MP3Tunes:

            MP3tunes users alone choose the websites they link to Sideload.com and the songs they sideload and store in their lockers. MP3tunes does not participate in those decisions. At worst, MP3tunes set up a fully automated system where users can choose to download infringing content. 10 Grp., 586 F. Supp. 2d at 1147 (granting safe harbor protection to a website that automatically created content from user submissions of unauthorized copyrighted work). If enabling a party to download infringing material was sufficient to create liability, then even search engines like Google or Yahoo! would be without DMCA protection.

            Google gives you a link and tells you where you can go to get the content. With Aereo, the customer doesn’t have to go anywhere else. Aereo has the very content that the customer is looking for.

            The same would be true of the sideloading feature of MP3Tunes.

            And furthermore, by this reasoning, it would seem that Google actually is supplying the content, where the Google search user happens also to be using Google for internet access — for example, if the user were a Google Fiber customer. After all, Google is then also providing the connection for the user to get the third-party content (just like Aereo).

            This would seem a bizarre result.

            When Cox publicly performs by transmitting from their facilities to my DVR, Cox has a license to make that transmission because they need one. When I later perform by playing back the copy on my DVR, I don’t need a license since I’m privately performing. When Cox sends me the transmission, it is the direct performer. When I play back the copy later, I am the direct performer.

            I’d be satisfied to drop all the topics before this, and focus on just this one point. We apparently can go back and forth on the rest of it endlessly. Assume for the sake of argument that I accept all your theories above.

            I keep asking you the same basic question, and you keep skipping over the answer. Look at your prior post. What criteria identified in that does not apply to the Cox DVR in your home?

            As I see it, you identified the following criteria:

            1. Volitional conduct in creating/designing/programming the system. This applies equally to the Cox DVR as to the Cablevision RS-DVR, or Aereo system. Check.

            2. Supplies the content. Even if we argue as to whether this actually applies to Aereo, it unarguably applies to the Cox DVR. Check.

            Accordingly, based on the criteria you specified in your prior post, Cox is the direct performer when you play back a program on your Cox DVR.

            So given the previously-identified criteria all apply, what makes Cox’s in-home DVR different?

            You can’t just skip over the answer and say it’s you performing. (That’s nothing more than devolving your argument to “I know it when I see it.”)

            Please tell us: Why is it suddenly you, and not Cox, who are doing the performing, in the case of the Cox DVR in your home? What is the additional criteria you are using, but declining to identify, in order to reach that conclusion?

          • I keep asking you the same basic question, and you keep skipping over the answer. Look at your prior post. What criteria identified in that does not apply to the Cox DVR in your home?

            As I see it, you identified the following criteria:

            1. Volitional conduct in creating/designing/programming the system. This applies equally to the Cox DVR as to the Cablevision RS-DVR, or Aereo system. Check.

            2. Supplies the content. Even if we argue as to whether this actually applies to Aereo, it unarguably applies to the Cox DVR. Check.

            Accordingly, based on the criteria you specified in your prior post, Cox is the direct performer when you play back a program on your Cox DVR.

            So given the previously-identified criteria all apply, what makes Cox’s in-home DVR different?

            You can’t just skip over the answer and say it’s you performing. (That’s nothing more than devolving your argument to “I know it when I see it.”)

            Please tell us: Why is it suddenly you, and not Cox, who are doing the performing, in the case of the Cox DVR in your home? What is the additional criteria you are using, but declining to identify, in order to reach that conclusion?

            It’s two separate performances on two separate systems. Say, for example, (1) Cox transmits HBO to my home, and (2) I play back the recording of the HBO transmission from my DVR. Both of these are performances, but who is the direct performer turns on causation.

            For (1), I think we both agree that Cox is directly performing when it transmits HBO to me (and the reality is that Cox has a license for this transmission because it is a public performance). Cox’s volition is clear. It has not only created, designed, and programmed the system, as you noted, but it also is operating the system. I have no control over that system. Cox does it all. And, of course, Cox supplies the content. It obtains the content from HBO which it then sends to me. Moreover, whether I watch the stream live or whether my cable box is unplugged or whether my DVR is recording the stream doesn’t matter. Cox is publicly performing by sending me the transmission regardless of what I do on my end.

            For (2), the volition changes. The system at issue is now the DVR itself, not Cox’s cable transmission system which was at issue in (1). Sure, the DVR was created, designed, and programmed by Cox, but now I’m operating it. I press play and initiate the playback. With (1), no matter what I did, Cox performed by transmitting HBO to me all the same. But with (2), whether or not the performance occurs at all depends on what I do. In fact, Cox can’t initiate this playback and start the performance. Moreover, with (2), Cox is not supplying me the content. Cox already supplied me with the content in (1) when it performed by transmitting the content to me, which it had a license to do. When I playback that content from my DVR later, Cox is not again sending me that same content. I already have the content because Cox already sent it to me.

          • Devlin, (1) corresponds to content going from an antenna to disk, in Aereo’s system (or from a cable feed to disk, in Cablevision’s RS-DVR). (2) corresponds to transmitting a recorded program from a disk located at Aereo (or Cablevision) to your TV.

            Focus on (2). That is the performance in question in this case.

            For Aereo and Cablevision, you assert that Aereo and Cablevision are the performers. For your Cox DVR, you assert that you are the performer.

            But by your criteria, there is absolutely no distinction between these cases. Cox, Cablevision and Aereo (by your assertion, my assumption for the sake of argument), are each the supplier of content to your disk. Cox, Cablevision and Aereo all exert the exact same level of control respecting the creation, design, programming, and yes, operation also, of the systems.

            Cox can disable your DVR at any time — and certainly will if you don’t pay your bill. They can change the programming at any time, to modify both the device’s autonomous behavior, and it’s behavior in response to your remote control commands (and certainly do, from time to time).

            The only difference between the Cox DVR and Aereo or Cablevision’s RS-DVR relate to things that have nothing to do with the criteria you have identified. (Things like, say, you supply the electricity to the equipment, or the equipment is located within your home.) If there’s actually a difference for the purposes of the transmit clause, it must relate to as yet unidentified criteria regarding one or more of these other differences.

            When you say “now I’m operating it” with respect to the DVR, you’re really just again jumping straight to saying “I am the one performing,” using “operating” and a substitute for “performing” (or “transmitting”), without providing an explanation as to what criteria distinguishes the different cases. (And keep in mind, of no relevance to the matter at issue for you to distinguish (2) from (1) for the Cox DVR. What you need to distinguish is (2) for the Cox DVR from (2) for the Aereo and Cablevision RS-DVRs.)

            In fact, the division between Cox’s role and yours, in using the Cox home DVR is identical to the division between Aereo’s or Cablevision’s role and their user’s role, in using their RS-DVR systems. And by the criteria you’ve provided to this point, these two cases are not distinguishable.

            So what is your basis for saying that, in one case, you, as the user, are the performer (or transmitter, or operator, or whatever other term you may use), and in the other cases, the user is not?

          • If you don’t mind, before we jump ahead to Aereo and Cablevision, let’s stick with (1) and (2).

            Do you agree that in (1), Cox is a direct performer?

            Do you agree that in (2), I am the direct performer? If not, why not? And if so, why?

          • Yes, Devlin, I believe that in (1) Cox is the performer, and in (2) you are the performer. But it’s irrelevant to the question at issue. Comparing (1) to (2) is comparing apples and oranges.

            The only performance at issue in this litigation corresponds to (2), and the question at hand is why is (2) in the Cox in-home DVR case different from (2) in the out-of-home DVR cases.

            No criteria you have proposed to this point distinguishes them.

          • Yes, Devlin, I believe that in (1) Cox is the performer, and in (2) you are the performer. But it’s irrelevant to the question at issue. Comparing (1) to (2) is comparing apples and oranges.

            The only performance at issue in this litigation corresponds to (2), and the question at hand is why is (2) in the Cox in-home DVR case different from (2) in the out-of-home DVR cases.

            No criteria you have proposed to this point distinguishes them.

            Thanks for the response. I’m simply trying to get you to acknowledge the answers I have given you to the questions you’ve asked.

            For example, you asked: “First, I don’t recall the Second Circuit holding that Aereo was the performer, and in a quick search of the holding, I wasn’t able to find evidence of such. Is there something I missed?”

            I responded: “The Second Circuit in Aereo also assumed that Aereo was a direct performer. See, e.g., Aereo, 712 F.3d at 680 (“Aereo transmits to its subscribers broadcast television programs over the internet for a monthly subscription fee.”); Id. at 691 (“Both Aereo and Cablevision are making multiple private transmissions of the same work . . . .”).”

            You didn’t acknowledge that I had addressed your question. Do you agree that the Second Circuit in Aereo treated Aereo as the direct performer? I still don’t know, and this is a critical point.

            You then asked me: “Are you therefore prepared to conclude that Cox is, in fact, the performer when you use your Cox DVR to perform a recorded program?”

            I responded: “When Cox publicly performs by transmitting from their facilities to my DVR, Cox has a license to make that transmission because they need one. When I later perform by playing back the copy on my DVR, I don’t need a license since I’m privately performing. When Cox sends me the transmission, it is the direct performer. When I play back the copy later, I am the direct performer.”

            Unsatisfied with my answer, you then asked: “I keep asking you the same basic question, and you keep skipping over the answer. Look at your prior post. What criteria identified in that does not apply to the Cox DVR in your home? *** Please tell us: Why is it suddenly you, and not Cox, who are doing the performing, in the case of the Cox DVR in your home? What is the additional criteria you are using, but declining to identify, in order to reach that conclusion?”

            I then directly answered your question, giving you the criteria that I thought differentiated the two:

            It’s two separate performances on two separate systems. Say, for example, (1) Cox transmits HBO to my home, and (2) I play back the recording of the HBO transmission from my DVR. Both of these are performances, but who is the direct performer turns on causation.

            For (1), I think we both agree that Cox is directly performing when it transmits HBO to me (and the reality is that Cox has a license for this transmission because it is a public performance). Cox’s volition is clear. It has not only created, designed, and programmed the system, as you noted, but it also is operating the system. I have no control over that system. Cox does it all. And, of course, Cox supplies the content. It obtains the content from HBO which it then sends to me. Moreover, whether I watch the stream live or whether my cable box is unplugged or whether my DVR is recording the stream doesn’t matter. Cox is publicly performing by sending me the transmission regardless of what I do on my end.

            For (2), the volition changes. The system at issue is now the DVR itself, not Cox’s cable transmission system which was at issue in (1). Sure, the DVR was created, designed, and programmed by Cox, but now I’m operating it. I press play and initiate the playback. With (1), no matter what I did, Cox performed by transmitting HBO to me all the same. But with (2), whether or not the performance occurs at all depends on what I do. In fact, Cox can’t initiate this playback and start the performance. Moreover, with (2), Cox is not supplying me the content. Cox already supplied me with the content in (1) when it performed by transmitting the content to me, which it had a license to do. When I playback that content from my DVR later, Cox is not again sending me that same content. I already have the content because Cox already sent it to me.

            Thus, I identified the criteria that I thought mattered, namely, (1) that the issue is volition (which is really causation), (2) that the systems at issue are different, (3) that who is operating the system at issue differs which each, and (4) that who supplies the content differs with each.

            You didn’t acknowledge the criteria I had provided as distinguishing Cox transmitting me HBO and me watching back a recorded HBO transmission on my DVR. At first you complained that I wasn’t providing these criteria, and then when I did provide the criteria, you simply started asking me questions about Cablevision and Aereo.

            I then asked you: “Do you agree that in (2), I am the direct performer? If not, why not? And if so, why?”

            You responded: “Yes, Devlin, I believe that in (1) Cox is the performer, and in (2) you are the performer. But it’s irrelevant to the question at issue. Comparing (1) to (2) is comparing apples and oranges.”

            So at first you complain that I “keep skipping over the answer,” and once I give you the answer, you claim that the answer is “irrelevant” and the comparison between the two is “apples and oranges.” And when asked to explain why the identity of the direct performer is different in (1) than in (2), you offer no reason whatsoever.

            So let me ask you again, because I believe the question is highly relevant to Aereo, why is the direct performer in (1) different than in (2)?

            We both agree that (1) falls on the other side of the line than (2), and I think examining the factors that lead us to that conclusion will help us identify useful criteria for determining which side of the line other systems or services fall on. I’ve identified the criteria that I think are relevant. What are yours?

          • Thanks for the response. I’m simply trying to get you to acknowledge the answers I have given you to the questions you’ve asked.

            For example, you asked: “First, I don’t recall the Second Circuit holding that Aereo was the performer, and in a quick search of the holding, I wasn’t able to find evidence of such. Is there something I missed?”

            I responded: “The Second Circuit in Aereo also assumed that Aereo was a direct performer. See, e.g., Aereo, 712 F.3d at 680 (“Aereo transmits to its subscribers broadcast television programs over the internet for a monthly subscription fee.”); Id. at 691 (“Both Aereo and Cablevision are making multiple private transmissions of the same work . . . .”).”

            I apparently overlooked this specific comment. The answer is no, I do not believe it is reasonable to characterize either of those comments as the Second Circuit holding Aereo (or Cablevision) to be the direct performer. The former comment is in the Court background section, and the latter in its discussion section, and regardless, it’s quite clear that neither is intended to reflect a legal determination (nor are any of the similar statements, elsewhere in the holding). This is merely colloquial phrasing for convenience.

            In fact, the second reference makes this very clear, by including Cablevision in the statement. As I noted before, the Second Circuit explicitly declined to make a determination as to whom the transmitter was, for legal purposes, in Cablevision, and it’s just as clear in the Aereo decision that they were saying nothing to revise the position the Court took in Cablevision.

            You didn’t acknowledge that I had addressed your question. Do you agree that the Second Circuit in Aereo treated Aereo as the direct performer? I still don’t know, and this is a critical point.

            Frankly, I don’t understand how this could possibly be critical to answering the very simple question I keep asking, at all. My question has nothing to do with my interpretation of the precedent. However you interpret it is fine for these purposes. At this point, I’m merely asking for you (again) to provide the criteria you are using to distinguish the performance that occurs during playback in the case of an RS-DVR system, from the performance that occurs during playback in the case of an in-home DVR.

            So at first you complain that I “keep skipping over the answer,” and once I give you the answer, you claim that the answer is “irrelevant” and the comparison between the two is “apples and oranges.” And when asked to explain why the identity of the direct performer is different in (1) than in (2), you offer no reason whatsoever.

            Devlin, comparing apples and oranges is irrelevant. It is not a responsive answer.

            You responded to my question asking you to distinguish the performance that occurs during playback in the case of an RS-DVR system (apple #1), from the performance that occurs during playback in the case of an in-home DVR (apple #2) by distinguishing the performance that occurs during an ordinary cable transmission of linear broadcast television (an orange) with the performance that occurs during playback in the case of an in-home DVR (apple #2). This is, precisely, comparing apples and oranges.

            The only performance in the Aereo case is the performance that occurs during playback in the Aereo RS-DVR system. Suddenly going off to analyze a completely different performance scenario is not instructive. Comparing that completely different performance scenario, as if it were in some way analogous to the performance at issue and therefore suitable for comparison, is worse than not instructive.

            So let me ask you again, because I believe the question is highly relevant to Aereo, why is the direct performer in (1) different than in (2)?

            We both agree that (1) falls on the other side of the line than (2), and I think examining the factors that lead us to that conclusion will help us identify useful criteria for determining which side of the line other systems or services fall on. I’ve identified the criteria that I think are relevant. What are yours?

            It is not at all relevant. It is comparing apples and oranges. Furthermore, at this point, I’m not attempting to make an argument as to what the proper answer is. I’m merely trying to get you to explain what criteria, in your view distinguishes apple #1 from apple #2.

            You see, my view is very simple: I think that whatever the answer is for apple #1, the same answer applies for apple #2.

            You clearly believe otherwise. What is your criteria for distinguishing apple #1 from apple #2?

            Explaining to me your basis for distinguishing an apple from an orange is not responsive to the question, in any way, shape or form. Nor is asking me to debate the distinction between an apple and an orange.

          • I apparently overlooked this specific comment. The answer is no, I do not believe it is reasonable to characterize either of those comments as the Second Circuit holding Aereo (or Cablevision) to be the direct performer. The former comment is in the Court background section, and the latter in its discussion section, and regardless, it’s quite clear that neither is intended to reflect a legal determination (nor are any of the similar statements, elsewhere in the holding). This is merely colloquial phrasing for convenience.

            In fact, the second reference makes this very clear, by including Cablevision in the statement. As I noted before, the Second Circuit explicitly declined to make a determination as to whom the transmitter was, for legal purposes, in Cablevision, and it’s just as clear in the Aereo decision that they were saying nothing to revise the position the Court took in Cablevision.

            I think you’re missing the point. I didn’t say that the Second Circuit held that Aereo was the direct performer. My point is that the court in Cablevision and Aereo held that the performances were private even if Cablevision or Aereo was the direct performer. I think that’s nonsensical.

            Frankly, I don’t understand how this could possibly be critical to answering the very simple question I keep asking, at all. My question has nothing to do with my interpretation of the precedent. However you interpret it is fine for these purposes. At this point, I’m merely asking for you (again) to provide the criteria you are using to distinguish the performance that occurs during playback in the case of an RS-DVR system, from the performance that occurs during playback in the case of an in-home DVR.

            It’s critical, because as I mentioned above, I think if the service is the direct performer, then the transmissions are to the public and they’re public performances.

            Devlin, comparing apples and oranges is irrelevant. It is not a responsive answer.

            I was comparing one scenario where we agree the service is the direct performer against one service where we agree that the customer is the direct performer. Looking at the differences between the two can help us understand what factors may be relevant to determining whether Aereo is a direct performer. The point isn’t to compare apples and oranges for no reason. It’s to determine what makes an apple different than an orange so we can apply that to Aereo.

            You responded to my question asking you to distinguish the performance that occurs during playback in the case of an RS-DVR system (apple #1), from the performance that occurs during playback in the case of an in-home DVR (apple #2) by distinguishing the performance that occurs during an ordinary cable transmission of linear broadcast television (an orange) with the performance that occurs during playback in the case of an in-home DVR (apple #2). This is, precisely, comparing apples and oranges.

            You asked: “what makes Cox’s in-home DVR different?” I misunderstood what you wanted me to compare it to. Still, making the comparison I made is useful, for the reasons already given.

            It is not at all relevant. It is comparing apples and oranges. Furthermore, at this point, I’m not attempting to make an argument as to what the proper answer is. I’m merely trying to get you to explain what criteria, in your view distinguishes apple #1 from apple #2.

            You see, my view is very simple: I think that whatever the answer is for apple #1, the same answer applies for apple #2.

            You clearly believe otherwise. What is your criteria for distinguishing apple #1 from apple #2?
            Explaining to me your basis for distinguishing an apple from an orange is not responsive to the question, in any way, shape or form. Nor is asking me to debate the distinction between an apple and an orange.

            I think the differences between the apple and the orange are relevant to the differences between apple #1 and apple #2. That’s why I want to discuss the differences between the apple and the orange. Once we identify what those factors actually are, we can discuss their relevance to Aereo. But you’re trying to say those factors are irrelevant without even identifying what those factors are. If you don’t know what factors you’re talking about, how can you be so sure they’re irrelevant?

          • Devlin, we have gone back and forth now with more than 60 comments, containing some 45.000 words, between this posting and a prior posting of yours. It is well past the point where it is useful to pursue philosophical tangents into the weeds — particularly ones we have already discussed to death.

            I have been asking you this same basic question since nearly the outset of our exchange in the prior comment thread, and from my very first comment in this comment thread.

            I have asserted from the outset that it will require an arbitrary distinction, not grounded in the statute, to distinguish playback in the in-home DVR system from playback in the RS-DVR system. I have repeatedly supported that contention by explaining the absence of any relevant technological or control distinction between the two cases, and most recently, by demonstrating that the criteria you provided to classify RS-DVR playback as an operator performance apply equally to in-home DVR playback. (N.B., not by challenging your criteria, but merely applying it directly to the in-home case.)

            In response, you have repeatedly declined to identify criteria to distinguish the two cases, instead, introducing extraneous new scenarios for consideration. (As you have, throughout this exchange, repeatedly preferred to argue your view based on analogies to other caselaw, rather than by explicitly identifying criteria or a test to be applied.)

            I can appreciate if in the most recent instance, your focus on two dislike scenarios was, at first, a product of misunderstanding the two scenarios I was asking you to distinguish. But by now there can be no misunderstanding. And yet, you now seem to be explicitly refusing to actually address the fundamental question I have posed, from the outset, unless after some 60 comments and 45,000 words exchanged, I agree to yet again pursue a tangential discussion comparing completely different performance cases.

            If you believe you’re still trying to educate me what the criteria should be, forget that. Right now, just tell me what the criteria is. Criteria that actually distinguishes the two apples-to-apples performance scenarios before us.

            If you present that, and I want to criticize it as mistaken, you can then re-engage in attempting to educate me. (I don’t think that’s what’s going to happen when you actually present your criteria to distinguish these two cases. What I think is going to happen is that either I’m going to demonstrate that you are again mistaken in your technological assumptions about how the systems work, such that your criteria doesn’t actually distinguish the two cases, or I’m going to be easily able to point out that your criteria is arbitrary, and without basis in the existing statute.)

            At this point, if you remain unwilling to identify criteria to distinguish these two cases, I have to assume that it is because either you have recognized that your criteria to do such is arbitrary, and you don’t want to open it up to scrutiny here, or you actually have no identified criteria, and have recognized that your position amounts to nothing more than “I know it when I see it.”

            If for you this is still a serious discussion intended to reach a real understanding of the issues, then you’ll provide a responsive answer my very reasonable (and by now, repeated ad nauseum) question. If not, I can only conclude that, at this point, you’ve devolved to merely litigating the case as an advocate for the plaintiffs.

          • At this point, if you remain unwilling to identify criteria to distinguish these two cases, I have to assume that it is because either you have recognized that your criteria to do such is arbitrary, and you don’t want to open it up to scrutiny here, or you actually have no identified criteria, and have recognized that your position amounts to nothing more than “I know it when I see it.”

            I was trying to get you to acknowledge the factors that separate the apple from the orange so I could then explain why I think those factors apply to Aereo. I’ve been giving it a great amount of thought over the past couple of weeks, and as I mentioned before, I’m working on another post where I’ll present my theory on why I think Aereo is a direct infringer. The post is about 9,000 words at the moment, and I’m not done working on it. I hope to have it finished for next week (so I can stop thinking about Aereo!). It’s got a lot of moving parts. I welcome your scrutiny. In fact, I’ve been enjoying the challenge. At this point, I’d say why don’t you wait for the post and then you can tell me why I’m wrong at that time. But if you want to get a head start on your thinking, I’ll spoil the punchline and tell you that it turns on Aereo’s having provided the content (as well as other factors, but the content is the biggie). I know you think the fact that the equipment is “rented” changes things, but I don’t think it does because the “rental” doesn’t change Aereo’s volitional conduct. But let’s continue this discussion after my next post if you don’t mind. Thanks.

          • J.S., this decision that came out today might interest you: http://www.scribd.com/doc/208095495/Aereo-Order-Granting-Preliminary-Injunction-D-Utah

            Judge Kimball found that “Aereo’s device or process transmits Plaintiffs’ copyrighted programs to the public.” He explicitly stated several times that Aereo is the one transmitting the performances (e.g., “Aereo uses ‘any device or process’ to transmit a performance or display of Plaintiff’s copyrighted programs to Aereo’s paid subscribers,” “Aereo is engaging in public performances when it intercepts and retransmits copyrighted programs to paying strangers”). Thoughts?

          • I’m aware of the Utah opinion. It is no more compelling than the FilmOn X opinions.

            http://cimc-greenfield.com/2014/02/19/utah-district-court-deals-aereo-first-loss/

            And while I’ll surely read your upcoming post with interest, a 9000-word treatise on the issues 😉 doesn’t seem likely to answer the very narrow question I have posed — which is about distinguishing the in-home DVR playback performance from the out-of-home DVR playback performance.

            As I’ve said before, for these purposes, at this point, I don’t care whether you (or I) are right or wrong in the classification of the out-of-home DVR playback performance. I care only about what distinguishes the in-home DVR playback, so that it can be different.

            This doesn’t require (and at this point is not aided by) comparison to any other scenarios. It doesn’t require an explanation of the criteria. What it requires is clear criteria which, when applied to both scenarios, actually produces two different results.

            Once we have criteria that does that, then again, my concern at this point isn’t whether I think the criteria is right or wrong, but rather, whether it has any basis in the statute. (In other words, my concern is not determining whether I think it is correct, but merely whether it’s plausible as an interpretation of the statute, rather than what would be a purely judicial fabrication.)

          • Thanks for the link. I’ll definitely read your post. My post will address the difference between a remote DVR and a home DVR, don’t worry. I think the issue turns on causation, and doctrinally, that is the analysis under the volitional conduct test. You won’t find that test in the statute, and it is in fact a judicial fabrication—as are so many things in the law. Nor is there is any magic formula about how you calculate the volitional conduct. Like proximate causation analysis generally, there’s certainly a normative aspect to it that makes it partially arbitrary. As I mentioned before, how these things get decided is both art and science. I think you’re very mistaken though to resist making comparisons to other scenarios. That’s just legal analysis 101. In my post, I hope to demonstrate that when a service provides the content for streaming, that service is always directly causing the performance. That’s not a wishy-washy, arbitrary distinction. It’s a distinction that recognizes that some volitional acts are so great that they are properly considered to be the direct cause. This is why cable and satellite retransmission companies need licenses. It’s why the DMCA doesn’t give safe harbor to services that provide the content. It’s why Judge Kimball easily concluded that Aereo is the party directly causing the transmission. All of the arguments about “rented” equipment and doing what you could do yourself and the length of the cord strike me as completely arbitrary since they gloss over the actual issue which is causation. Whether the equipment is “rented” or whether you could do it yourself or whether the cord is long doesn’t change any of the facts about what Aereo actually does.

      • Devlin I haven’t dismissed all the things you described. But after a month of exchanging 60+ comments comprising some 45K words, I think we’ve beaten those issues to death, and it is time to get down to brass tacks. (And furthermore, there is zero chance that I will successfully extract from a new 9000-word article the correct criteria that you believe is to be applied. It makes much more sense for you to just deliver it, directly, here.)

        I’m also not arguing about whether volitional conduct is a valid, established basis for determining who the actor is, or whether it is grounded in the statute. (If it’s not grounded in the statute, I’ll accept it as established law, in the realm of that which the courts have no choice but to make a determination.)

        What I am saying is that the two cases are indistinguishable as to volitional conduct. Everything you have cited to this point to classify Aereo and Cablevision as supplying volitional conduct that renders them the actor for purposes of the Copyright Act apply equally to the case of your in-home Cox DVR.

        The only judge to attempt to address this issue (actually, to dismiss it with a hand-wave) was Denny Chin, in his original district court holding in Cablevision, and what he cited to distinguish the two was factually erroneous.

        What I am looking for is not a big treatise on (or justifying) a volitional conduct test, or anything else. What I am looking for is merely a statement of the criteria which, when applied to factually accurate characterizations of the two different systems/scenarios, produces opposite results.

        I don’t see any way that you will ever be able to achieve such in criteria around volitional conduct, save for arbitrarily defining a rule along of the lines of “in this case, the actor is deemed to be” specifically crafted to address this situation and create a dichotomy. And that’s precisely the point.

        If I’m wrong, just give me the criteria, and we’ll see that I’m wrong. (I don’t believe you’ll be able to, but I always remain open to the possibility of learning something new, every day, that changes my view. That will not come from a long treatise discussing the issues. At this point, it will only come from seeing clearly, simply specified criteria — the legal test to be applied.)

        • I don’t think I could be any more clear that the criteria I’ve identified as being pivotal is who supplies the content.

          • Then you must believe that playbacks of recordings on your in-home Cox DVR are public performances, since Cox supplies (and controls the supply of) content to that device, identically to an RS-DVR system.

            (As noted before, we can debate whether Aereo actually supplies the content in its system, but there can be no debate about that, when it comes to Cablevision’s RS-DVR, or Cox’s in-home DVR.)

          • Let’s start with the big picture. Home DVR: The cable company supplies content to the public. The cable company has a license for this. Remote DVR: The cable company supplies content to the public. The cable company has no license for this. Supplying content to the public requires a license, so the remote DVR is infringing.

            Can you explain to me how a cable company can supply content to the public yet not be infringing?

          • Sigh.

            Devlin, we finally got back to a straightforward discussion of criteria. You again cite criteria that doesn’t actually achieve the results you are arguing for…and you immediately want to step back to a broader discussion again. (Can you understand my frustration with this discussion?)

            I will answer you’re question, because it is very simple and quick to do so. Then, I expect we will return to the specific criteria — the legal test to be applied — as that is the real issue here. After 45,000 words exchanged, you can’t credibly argue that your position is even plausible if you cannot provide criteria that actually produces the results you argue are correct.

            Your question is short-cutting again — assuming the answer by constructing the question to suppose the conclusion — that “the cable company supplies content to the public.” More importantly, you are conflating the front-end (input to) and the back-end (output from) of the DVR (again).

            The cable company has a license to deliver content to its customers. Under that license, it delivers content to the customers’ DVRs. In both cases, those DVRs happen to be rented from the cable company. In one case, the rented DVRs happen to be located inside the customers’ homes. In the other case, the rented DVRs happen to be located inside the cabe company’s building.

            The question here is not about the supply of content to the DVR (front end). It is about the playback of content from the DVR (back end). If you wish to define your criteria in terms of who supplies content, that’s fine, but it simply doesn’t distinguish these two cases. (And it breaks in a variety of other cases, also.)

          • The cable company has a license to deliver content to its customers. Under that license, it delivers content to the customers’ DVRs.

            I assumed we were talking about a system like Cablevision’s. Cablevision did not have a license to supply content to customers via the remote DVR: “Cablevision has not obtained permission from plaintiffs, the owners of the copyrighted programs, to reproduce and transmit the programs through its proposed RS–DVR.” Twentieth Century Fox Film Corp. v. Cablevision Sys. Corp., 478 F. Supp. 2d 607, 609 (S.D.N.Y. 2007). Cablevision had a license to supply content directly to customers in real time, but it did not have a license to split the stream and send content to its remote DVR to be recorded and then supplied to customers later. It’s precisely because Cablevision did not have a license for this that the whole Cablevision litigation occurred in the first place. Granting a license to supply content to the public in one manner does not imply a license to supply content to the public in all possible manners.

          • Let me break this down a bit more for you. Since we both agree that the content is transferred from Cablevision to the customer with both the remote DVR and the home DVR, I’m focusing on how Cablevision supplies that content to the customer.

            Remote DVR: Cablevision has a license to publicly perform the content by transmitting it to the public in real-time. It does not have a license to split the stream and send the content to the remote DVR. It does not have a license to make a reproduction of that content, and it does not have a license to publicly distribute that content. There are three possibilities for who makes the copy on the remote DVR: Cablevision alone, the customer alone, or both Cablevision and the customer jointly. If Cablevision alone makes the copy on the remote DVR, then that is a violation of the reproduction right by Cablevision since Cablevision is making a reproduction without a license. If the customer makes the copy on the remote DVR, then that is a violation of the distribution right by Cablevision since Cablevision is publicly distributing without a license. And if both Cablevision and the customer jointly make the copy on the remote DVR, Cablevision is violating both the reproduction right (for its own copying of the content) and the distribution right (for distributing the content to the customer who also copies it). No matter who makes the copy, Cablevision is infringing, and it may not lawfully transfer content to the customer with the remote DVR.

            Home DVR: Cablevision has a license to publicly perform the content by transmitting it to the public in real-time. Cablevision transmits the content to the customer in real-time, and this is a public performance no matter what the customer does with the transmission on his end. Since Cablevision has a license to publicly perform in this manner, Cablevision is not an infringer of the public performance right. And since Cablevision is not infringing, it may lawfully transfer content to the customer with the home DVR.

            There are issues I’m leaving off the table, like who streams from the remote DVR, who performs when that stream is rendered, who makes the copy on the home DVR, and who performs when that copy is rendered. But those are not relevant to the point I’m trying to make. The difference between the remote DVR and the home DVR is that with the remote DVR, it is not possible for Cablevision to supply the content to the customer without first doing some unlicensed act in the process. With the home DVR, Cablevision can supply the content to the customer without doing any unlicensed acts in the process.

            See my point?

          • Devlin, again, you are conflating the input to and the output from the DVR. I didn’t say “the cable company has a license to deliver content through DVRs.” I said “the cable company has a license to deliver content to its customers.”

            (I can also guarantee you that neither Cablevision nor any other cable company have carriage agreements that explicitly give them a license to deliver content to customers through in-home DVRs. Cable (and satellite) companies just took the view that they had the right to do such, and then did it. This is not a distinction between in-home and RS DVRs.)

            The point is that the cable company has the right to deliver content to its customers (not an explicit license to use any particular technology for such), and if the customer is using a DVR, delivering the content to the customer’s DVR is delivering it to the customer.

            You have diverged again into all kinds of topics other than the one topic of focus here: the only performance at issue in Aereo, which is the playback performance — that is, the output of the DVR. I’m still awaiting criteria, the legal test to be applied, that actually differentiates the two playback performances.

            This constant reversion to all other topics is making me think that, perhaps, the distinction you are actually drawing is that in one case you believe that you made the copy, and in the other case, you believe the cable company made the copy — and your view of who is performing flows from who you believe made the copy.

            Is that it? If the RS-DVR copy were legally deemed to have been made by the user, would you then agree that playback of the recording is a private performance by the user?

            If so, we can deem the playback performance issue settled…but we then have to back up to hear what your criteria is to distinguish who is making the recording copy, and we will still have the same basic issue. I assert that it is not possible to distinguish the in-home DVR from an RS-DVR, other than by manufacturing some arbitrary distinction, of whole cloth. And so I’m still going to ask you to identify the criteria by which you believe that these two instances can be distinguished.

          • Part of the problem here is that I read your posts, like the one I’m responding to here, several times, and I still don’t know what you want me to address. Let’s do this. What are the two things you want me to compare? Please be precise, and please don’t mention any extraneous information. I will try my best to answer you directly.

          • I guess that makes sense, since we seem to be talking past each other, but honestly, I don’t know why. Prior to my last post, I have asked for just one thing over and over: for you to provide the criteria you use to distinguish the playback performance of an RS-DVR from the playback performance of an in-home DVR.

            You want to remove extraneous stuff? Ok, let’s really remove extraneous stuff (as I’ve attempted to do before).

            Assume the in-home DVR you are comparing is your in-home DVR from Cox. Assume that the RS-DVR you are comparing is the exact same device. Cox simply now offers to also provide you with space to keep the device, and you take them up on it, relocating the device to their facilities.

            What is the criteria that you rely on to say that playback from one is a private performance by you, while playback from the other is a public performance by Cox?

          • Sorry for the delay in getting back to you. It’s been a crazy weekend.

            First, I need to address your notion that the cable system’s license includes the privilege of transmitting the content to its subscribers from the remote DVR. If it’s the statutory license we’re talking about, it does not include these transmissions. See 17 U.S.C. 111. The cable system only gets the statutory license if it is engaged in the “further transmitting of a primary transmission simultaneously with the primary transmission.” If the secondary transmission is nonsimultaneous with the primary transmission, as the transmission from the remote DVR is, then there is no statutory license. A transmission from the remote DVR is unlicensed, and this is a crucial point to understand. The reason the Cablevision litigation happened in the first place is because Cablevision did not have a license for these transmissions. The issue was whether these transmissions were public performances. If they were, then Cablevision was an infringer and it needed a license to be lawful, and if they weren’t, then Cablevision was not an infringer and it needed no license to be lawful.

            Second, I think it helps to frame the issue by understanding that every link in the chain of transmissions, from the primary transmission to any intermediate transmission and to the final transmission which reaches the public, is a public performance. “Congress intended the definitions of ‘public’ and ‘performance’ to encompass each step in the process by which a protected work wends its way to its audience.” David v. Showtime/The Movie Channel, Inc., 697 F.Supp. 752, 759 (S.D.N.Y. 1988). This chain of transmissions ends when the content reaches the public, and every link in that chain is a public performance provided the final transmission is to the public.

            Consider the follow chain of simultaneous transmissions:

            (1) broadcaster makes primary transmission to passive carrier
            (2) passive carrier makes secondary transmission to cable system
            (3) cable system makes secondary transmission to subscriber

            The broadcaster has a license from the copyright holder to publicly perform the content. The passive carrier needs no license because, even though it is publicly performing the content, it is exempted from liability by Section 111. The cable system has a statutory license under Section 111 to publicly perform the content. Note that the cable system publicly performs whether any subscriber actually receives the transmission or not. Nothing the subscriber does with that transmission, should he receive it, has any bearing on whether the cable system publicly performed. Whether the subscriber doesn’t receive the transmission at all, or whether he receives it and thereafter privately or publicly performs it, the cable system has publicly performed just by transmitted the content to the subscriber. The chain of transmissions is complete once the primary transmission has reached the public; what members of the public do with the transmission thereafter is irrelevant.

            With the home DVR, the subscriber receives the content after the licensed secondary transmission by the cable system to the subscriber is complete. Assuming the cable system complied with the requirements for the statutory license or obtained a license directly from the copyright holder, the transfer of the content from the cable system to the subscriber is licensed. But this is not the case with the remote DVR. As mentioned already, the statutory license only covers simultaneous secondary transmissions, and if the cable system makes an unlicensed nonsimultaneous secondary transmission, it is infringement. With the remote DVR, you seem to think, and the Second Circuit agreed, that there is some statutory loophole that permits a cable system to make unlicensed nonsimultaneous secondary transmissions to the public without infringing. The point I was making in the comment above is that this simply cannot be true. At some point, the cable system has to provide the content that it acquired from the passive carrier to the subscriber. There are two ways to provide content to the public: either by a public distribution or by a public performance. And each of those is infringing if not licensed.

            But to answer your question of who is performing when the content is transmitted from the remote DVR to the subscriber, it is the cable system. The reason is simple. The transmission from the remote DVR to the subscriber is still part of that chain of transmissions from the broadcaster to the public. And each link in that chain is a public performance which is infringing if not licensed or exempted by the passive carrier exemption in Section 111. You seem to think the subscriber obtains the content when it is recorded on the remote DVR, but the fact is that the subscriber has not yet received the content until it’s actually transmitted to him. Until the content has actually been supplied to the subscriber in a way that he can actually watch it, he hasn’t received it yet from the cable system. Once the cable system transmits the content to the subscriber, the chain of transmissions is complete. But the fact that the cable system permits the subscriber to request a time-shift of the transmission doesn’t change the fact that the transmission is still part of the chain of transmissions from the primary transmission to the public. And the fact that this nonsimultaneous secondary transmission from the cable system to the subscriber is unlicensed means it’s infringing.

            The home DVR, however, is different. The content that is on the home DVR comes from the licensed simultaneous secondary transmission from the cable system to the subscriber, and this transmission is the final link in the chain of transmissions since this is the transmission that reaches the public. The performance that occurs when the content on home DVR is played back is a private performance under Section 101 because it occurs in a private place and there is no transmission. It is the subscriber, and not the cable system, who causes this private performance. This follows from the doctrine which says that merely supplying someone with a machine is not enough to hold that party directly liable for whatever that person does with the machine. It also follows from the fact that the subscriber has dominion over the home DVR itself—he has a possessory interest in the home DVR which he does not have in the remote DVR.

            If you rent a DVD and play it at home on your DVD player, neither the lessor of the DVD nor the manufacturer of the DVD player is the one who performs, and the same holds true for the home DVR. While the lessor and the manufacturer are “but for” causes of the performance, their contributions to the performance are not proximate enough to say that they directly caused it. The same holds true for the home DVR. But why can’t we say the same holds true for the remote DVR? It’s because the performance that occurs when the transmission is sent from the cable system to the subscriber is still a link in the chain of transmissions from the broadcaster to the subscriber, and every link in that chain is public performance. The performance that occurs on the home DVR occurs after that chain of transmissions is complete, and so the rule that says that every link in the chain of transmissions is a public performance doesn’t apply. The cable system has already publicly performed with its licensed simultaneous secondary transmission to the subscriber.

            So, long story short, while it’s true that with both the remote DVR and the home DVR, the subscriber presses a button to initiate the playback, the difference is with the content being watched. With the remote DVR, that content is still part of the chain of transmissions from the primary transmission to the public, so it’s a public performance on the cable system’s part. With the home DVR, the content has already been delivered to the subscriber by the cable system, and the chain of transmissions from the primary transmission to the public is completed before this performance, which is private, takes place. With the remote DVR, the performance occurs using the content as its being supplied to the subscriber for the first time. With the home DVR, the performance occurs using the content that had already been supplied to the subscriber by the prior secondary transmission by the cable service.

            I hope this helps!

          • Devlin, that’s a really long answer for what was, at your request, a very simple question, with nothing extraneous. It was seeking a simple statement of the criteria. There’s a lot that’s wrong (as in factually erroneous) in your response. I will address such in a separate reply, to try to keep things focused here. Please respond to the two posts separately.

            I see where you do a lot of explaining, but I don’t see where you actually provided criteria in your response. At best, I am left to infer that your criteria is something along the lines of “the party who possesses the content is the transmitter” and you believe that “possession” of the content cannot be transferred from the provider to the user unless/until the content is perceptible by the user. You also seem to feel that physical location within the home is actually significant in determining possession.

            Am I correct? Can you provide a direct statement of the criteria to be applied? (Not another explanation. Just a statement of the criteria, as you see it.)

            Then, I’d like to make one change to the example you are addressing. Now, instead of the DVR being a Cox DVR that you rent, it is a Tivo that you own. Cox allows you to locate your Tivo in their facilities.

            When you play back a recording from that Tivo, is it private performance by you, or a public performance by Cox?

          • Devlin, that’s a really long answer for what was, at your request, a very simple question, with nothing extraneous. It was seeking a simple statement of the criteria. There’s a lot that’s wrong (as in factually erroneous) in your response. I will address such in a separate reply, to try to keep things focused here. Please respond to the two posts separately.

            I see where you do a lot of explaining, but I don’t see where you actually provided criteria in your response. At best, I am left to infer that your criteria is something along the lines of “the party who possesses the content is the transmitter” and you believe that “possession” of the content cannot be transferred from the provider to the user unless/until the content is perceptible by the user. You also seem to feel that physical location within the home is actually significant in determining possession.

            Am I correct? Can you provide a direct statement of the criteria to be applied? (Not another explanation. Just a statement of the criteria, as you see it.)

            The Criteria To Be Applied

            (1) Performance via transmission: Is the source of the transmission a link in the chain of transmissions from the primary transmitter to the public?

            a. If yes, then it’s a public performance and the service is an infringer if unlicensed.

            b. If no, then whether it’s a public performance turns on the volitional conduct of the service such as whether the service indexes, organizes, and makes the content available to the public, and when the service itself provides the content by actively acquiring it and making it available, it is publicly performing.

            (2) Performance without transmission.

            a. If the performance is in a private place, it is a private performance, and if it is in a public place or a semipublic place, it is a public performance.

            b. It is possible that service’s volitional conduct over the playback device is so great such that it is, either solely or jointly, directly causing the private performance, but the performance is nevertheless private so it doesn’t matter.

            Then, I’d like to make one change to the example you are addressing. Now, instead of the DVR being a Cox DVR that you rent, it is a Tivo that you own. Cox allows you to locate your Tivo in their facilities.

            When you play back a recording from that Tivo, is it private performance by you, or a public performance by Cox?

            It’s the same result as with the remote DVR since the transmission from the Tivo to me is part of the chain of transmissions from the primary transmission to the public. It’s a public performance by Cox.

          • Just want to make clear that my other comment (posted to appear as a separate thread) is not a response to the above comment. My reply to this comment will be forthcoming.

          • I’m attempting to keep this thread very focused on the case at hand, so want to focus on the criteria applicable to the DVR scenarios I asked you to compare. There is, in fact, a transmission from the home DVR in the case I asked you to consider, because we’re dealing with a whole-home DVR (as you previously indicated your Cox DVR is) and when playback occurs to another room, there’s a transmission for the purposes of the Copyright Act. That makes (1) the relevant criteria.

            The Criteria To Be Applied

            (1) Performance via transmission: Is the source of the transmission a link in the chain of transmissions from the primary transmitter to the public?

            a. If yes, then it’s a public performance and the service is an infringer if unlicensed.

            b. If no, then whether it’s a public performance turns on the volitional conduct of the service such as whether the service indexes, organizes, and makes the content available to the public, and when the service itself provides the content by actively acquiring it and making it available, it is publicly performing.

            Since in neither scenario is anybody (service provider, or user) making the recordings available to the public, I presume that (b) amounts to simply “If no, it’s a private performance” for these cases.

            So, in essence, I take you to be saying the criteria is:

            If the source of the transmission is a link in the chain of transmissions from the primary transmitter to the public, then it’s a public performance. Otherwise, it’s a private performance.

            I presume that you (still) intend for this to classify the in-home DVR playback transmission as private, but the RS-DVR playback transmission as public.

            But it’s not clear to me why this is. What is the criteria that allows you to conclude that “the chain of transmissions from the primary transmitter to the public” is broken by the in-home DVR, but not by the RS-DVR? Aren’t you just assuming that delivery to the former completes transmission to the public, while delivery to the latter does not? In other words, aren’t you just assuming the conclusion?

            You haven’t said it explicitly yet, but from this follow-up, it seems as if the key criteria for you is actually location, above all else:

            [jsg] Then, I’d like to make one change to the example you are addressing. Now, instead of the DVR being a Cox DVR that you rent, it is a Tivo that you own. Cox allows you to locate your Tivo in their facilities. When you play back a recording from that Tivo, is it private performance by you, or a public performance by Cox?

            [dh] It’s the same result as with the remote DVR since the transmission from the Tivo to me is part of the chain of transmissions from the primary transmission to the public. It’s a public performance by Cox.

            Here you seem to be saying that even when the cable company delivers content to your device, which you own, and where they no longer exercise any control over the device, but rather, merely house it in their facility, that they still haven’t actually delivered it to you (a member of the public).

            This would seem to make location the only criterion. Volition related to the actual transmission no longer plays any role, whatsoever. (The only volition Cox exercises in this case is the volition to let you place your device in their facility.) Basically, it’s absolute liability for the owner of the facility that the device is located within.

        • As mentioned, creating a separate thread from the criteria discussion, so that we can keep that thread focused on actually trying to identify what your criteria is. (i.e., a succinct statement of the test that you would propose SCOTUS should adopt, if you were making oral arguments for plaintiffs, and the Court asked you for your proposed rule.)

          In this thread, I feel compelled to address multiple errors in your last comment.

          Sorry for the delay in getting back to you. It’s been a crazy weekend.

          No need to apologize. I was traveling this weekend myself. In any case, it’s not either of our jobs to exchange comments here, so there’s certainly no obligation to respond immediately.

          First, I need to address your notion that the cable system’s license includes the privilege of transmitting the content to its subscribers from the remote DVR.

          You keep mischaracterizing what I say. I never claimed that the cable operator has a license to transmit content from the DVR. I claimed that the cable operator has a license to transmit content to its customers.

          This is the distinction between the input to and the output from the DVR that I have now reiterated numerous times.

          To my view (and I think, effectively, that of the Second Circuit), delivering the content to a customer’s DVR is delivering the content to the customer, regardless of whether that DVR is located at the customer’s home, or hosted for the customer at the cable operator’s facility. The cable operator delivers content to the DVR. At that point, “possession” (as you seem to call it) of the content switches to the customer, so that output from the DVR is output by the customer (regardless of where their DVR is located).

          If it’s the statutory license we’re talking about, it does not include these transmissions. See 17 U.S.C. 111.

          It’s not a statutory license we’re talking about. None of the content at issue in either Cablevision or Aereo is carried under a compulsory license. In Aereo, the plaintiffs are all broadcasters who have elected “retransmission consent” so that a negotiated license is required for cable operators to carry it. In Cablevision, the plaintiffs were all cable programmers (no over-the-air signal) and/or broadcasters who had elected retransmission consent.

          The statutory license of 17 USC 111 is really irrelevant. The only stations that elect “must carry” status today, in order to fall within that, are stations that have very little consumer interest, and know that they would be dropped by cable operators but for electing “must carry” status. The stations are looking for the broadcast possible access to consumers. They’re not suing to limit that access.

          Accordingly, any analysis based on your interpretation of statutory limitations is irrelevant.

          Furthermore, the point that I made earlier was very simple. You cited an assertion of plaintiffs in Cablevision to argue that there was a substantive legal difference between RS-DVR and in-home DVR:

          Cablevision did not have a license to supply content to customers via the remote DVR: “Cablevision has not obtained permission from plaintiffs, the owners of the copyrighted programs, to reproduce and transmit the programs through its proposed RS–DVR.”

          Later you added:

          Cablevision has a license to publicly perform the content by transmitting it to the public in real-time. It does not have a license to split the stream and send the content to the remote DVR.

          I made the point that this was merely plaintiffs assertion, not a statement of fact. The fact is, carriage agreements don’t contain anywhere near the kind of detail that you seem to presume, regarding what operators may and may not do. And what happens n practice is that operators looking to do anything different from what has been done before have to evaluate whether they believe that such is within their license and/or legal rights. Programmers, too, have to evaluate such. Often they disagree. Occasionally, those disagreements result in litigation.

          RS-DVR is not unique in this regard.

          As I mentioned, carriage agreements do not give explicitly cable operators the right to use in-home DVRs, for example. Cable operators decided that they had the right to do such. I can pretty much guarantee that some programmers probably held that they did not. It didn’t progress to litigation, however, because the programmers didn’t conclude it to be sufficiently likely that they would prevail.

          Note that the way a typical in-home DVR works is precisely analogous to the RS-DVR. In an SA 8300 (the in-home DVR used by Cablevision), for example, the incoming signal is split, with one copy being retransmitted through the DVR outputs to the attached TV, and another copy being recorded to disk.

          Another example is delivery of live television signals via IP, to devices such as phones, tablets, laptops, game consoles, etc. Carriage agreements did not explicitly grant operators the right to do such. Implementing such also involved splitting the cable signal in order to re-encode the streams for IP delivery. (And btw, the encoding used for IP delivery requires temporary storage of much more video, for much longer, than the RS-DVR of Cablevision. If you watch sometime, I’m sure you’ll find that the IP service from your provider runs 30-60 seconds behind live.)

          Just as with DVRs and RS-DVR, cable operators decided that this form of delivery fell within their license. Some programmers disagreed, and Viacom actually sued both Cablevision and Time Warner Cable.

          The fact that programmers argued that Cablevision needed an additional license to split the stream to supply content to the RS-DVR, does not make it something that you can take as a fact. (In fact, the only precedent on point in this regard is the Second Circuit finding the opposite.)

          A transmission from the remote DVR is unlicensed, and this is a crucial point to understand. The reason the Cablevision litigation happened in the first place is because Cablevision did not have a license for these transmissions.

          This is, of course, true. Playback transmissions from the RS-DVR are unlicensed.

          It’s also true that playback transmissions from in-home DVRs are unlicensed.

          The issue was whether these transmissions were public performances. If they were, then Cablevision was an infringer and it needed a license to be lawful, and if they weren’t, then Cablevision was not an infringer and it needed no license to be lawful.

          Actually, this is a simplification. If the transmissions were deemed private performances, then Cablevision certainly doesn’t infringe. But (your personal views aside), Cablevision doesn’t necessarily infringe, even if they are deemed public performances. For one, if they were public performances by the customer Cablevision would not infringe (at least not directly). Furthermore, even if Cablevision were deemed to be publicly performing, Cablevision might still be acting lawfully, if the action fell within fair use. Before you scoff, consider that the Solicitor General’s certiorari brief for Cablevision suggested precisely this possibility:

          Respondents’ failure to preserve any fair-use defense likewise would hinder this Court’s ability to consider the various issues raised by services like the RS-DVR. This Court ruled in Sony that the manufacturer and seller of VCRs could not be held liable for copyright infringement because “time-shifting” by consumers constituted a fair use of copyrighted broadcasts. […] This Court has never addressed, however, whether a commercial actor who is charged with direct infringement may defend on the ground that he performed the copying at the behest of a customer who himself would have a fair-use defense.

           

          Though I still await an explicit statement criteria, rather than long explanations from which I have to try to infer criteria, it seems that, ultimately, you are fixated on location as definitive in determining “possession” and therefore, volition.

          I have to say, I don’t think such can be justified. And so far as I can see, there doesn’t seem to be much, if any, support for the view that location is dispositive. (From what I can tell, the courts that have deemed there to be infringement in these cases have either gotten lost in details of the systems, or ignored the details of the systems (simply categorizing them as “any device or process”) but I don’t think any have cited location as even a factor in their determination as to volition.)

          Again, from the Solicitor General’s brief in Cablevision:

          …the court of appeals reasonably concluded that the subscriber—who would both select the programs to be copied and press the button triggering the actual recording—would “make” the copies that would be stored in the RS-DVR system. Respondents’ RS-DVR service would replicate the basic capabilities and limitations of a VCR or a settop DVR, and it would lack much of the functionality offered by a VOD system. Like a VCR or set-top DVR, the RS-DVR would permit subscribers to view only programs that already have been broadcast and that subscribers could have chosen to view in real time under the terms of their cable packages. The RS-DVR also would permit subscribers to view only programs that they personally and previously had directed the system to copy, and it would not allow them to view portions of a program that had aired prior to the subscriber pressing the “record” button.

          To be sure, respondents’ RS-DVR service would differ from a set-top DVR in that the tangible devices that would perform the copying and playback would be located in respondents’ facilities rather than in a subscriber’s home. That shift from local to network-based recording and playback, however, appears largely irrelevant to the determination of who would “make” the copies. With respect to the photocopying of written material, an individual who both selects the pages to be copied and operates the duplicating machine is naturally said to “make” the copies, whether the photocopier is located in the individual’s home or at a self-service copy shop. There is no evident reason for a different result here.

          Although the Second Circuit demurred in addressing who the performer was, it’s hard to see why the analysis should be any different when it comes to determining who the performer is, upon playback. The user makes the recording, juts like a home DVR. The recording is stored in storage allocated exclusively to the user, and accessible only to the user, just like a home DVR. It is, indeed, the user’s recording, just as much as that on a home DVR is. And like a home DVR, playback occurs only upon the user’s initiation. It’s hard to see where anybody other than user actually supplies any volition in playing back the recording, let alone sufficient volition to be deemed the performer. And it’s certainly hard to see how location has any bearing, whatsoever, on the question of who the performer is.

          • To my view (and I think, effectively, that of the Second Circuit), delivering the content to a customer’s DVR is delivering the content to the customer, regardless of whether that DVR is located at the customer’s home, or hosted for the customer at the cable operator’s facility. The cable operator delivers content to the DVR. At that point, “possession” (as you seem to call it) of the content switches to the customer, so that output from the DVR is output by the customer (regardless of where their DVR is located).

            I don’t think that’s right. The chain of transmissions ends when a member of the public receives the transmission such that he can perform it without there being another transmission in the interim. He has to receive it such that he could render it simultaneously from the final transmission in the chain of transmissions. When a recording is sitting on the remote DVR, the member of the public has not received it yet since he can’t render it without it first being transmitted to him. Until the customer can render the transmission, the chain of transmissions is not complete, and every link in the chain of transmissions is a public performance.

            Accordingly, any analysis based on your interpretation of statutory limitations is irrelevant.

            That’s good to know that Cablevision had negotiated for consent, thank you, but you haven’t shown how that fact changes what I said. Did Cablevision have a license to make nonsimultaneous secondary transmissions? I don’t think they did, hence the litigation. And if they did not, then everything I said still stands and the point that Cablevision is a link of the chain of transmissions from the broadcaster to the public, and all that implies, remains to be addressed by you.

            I made the point that this was merely plaintiffs assertion, not a statement of fact. The fact is, carriage agreements don’t contain anywhere near the kind of detail that you seem to presume, regarding what operators may and may not do. And what happens n practice is that operators looking to do anything different from what has been done before have to evaluate whether they believe that such is within their license and/or legal rights. Programmers, too, have to evaluate such. Often they disagree. Occasionally, those disagreements result in litigation.

            You’re not telling me much. What did Cablevision have a license to do? Did it specify that the retransmissions had to be simultaneous?

            As I mentioned, carriage agreements do not give explicitly cable operators the right to use in-home DVRs, for example. Cable operators decided that they had the right to do such. I can pretty much guarantee that some programmers probably held that they did not. It didn’t progress to litigation, however, because the programmers didn’t conclude it to be sufficiently likely that they would prevail.

            The public performance they are getting a license for is the one that occurs when Cablevision transmits to the subscriber. It’s a public performer whether the subscriber doesn’t receive it, receives it and renders it simultaneously, or receives it and makes a reproduction on his home DVR to be rendered later. Cablevision isn’t licensing for the acts its subscribers may or may not take down the road. It’s getting a license for its act of transmitting to the subscriber in the first place.

            Note that the way a typical in-home DVR works is precisely analogous to the RS-DVR. In an SA 8300 (the in-home DVR used by Cablevision), for example, the incoming signal is split, with one copy being retransmitted through the DVR outputs to the attached TV, and another copy being recorded to disk.

            And that split happens after the cable system has publicly performed by transmitting the content to the subscriber. Once the content has been delivered to the subscriber in a way that he can simultaneously render it without any more transmissions, then the chain of transmissions from the broadcaster to the public is over, and that means that these renderings are private performances and these reproductions are likely fair uses under Sony.

            Just as with DVRs and RS-DVR, cable operators decided that this form of delivery fell within their license. Some programmers disagreed, and Viacom actually sued both Cablevision and Time Warner Cable.

            If Cablevision’s actions were within the scope of its license, then why didn’t the Second Circuit just say that?

            The fact that programmers argued that Cablevision needed an additional license to split the stream to supply content to the RS-DVR, does not make it something that you can take as a fact. (In fact, the only precedent on point in this regard is the Second Circuit finding the opposite.)

            What case was that?

            This is, of course, true. Playback transmissions from the RS-DVR are unlicensed.
            It’s also true that playback transmissions from in-home DVRs are unlicensed.

            Yes, but playback from a home DVR is in a private place, so it’s a public performance. Playback from a remote DVR requires a transmission, so it’s possible that it’s a public performance. And I think it is, for all the reasons given.

            Actually, this is a simplification. If the transmissions were deemed private performances, then Cablevision certainly doesn’t infringe. But (your personal views aside), Cablevision doesn’t necessarily infringe, even if they are deemed public performances. For one, if they were public performances by the customer Cablevision would not infringe (at least not directly).

            Furthermore, even if Cablevision were deemed to be publicly performing, Cablevision might still be acting lawfully, if the action fell within fair use. Before you scoff, consider that the Solicitor General’s certiorari brief for Cablevision suggested precisely this possibility…

            I wish the Second Circuit had addressed fair use in either Cablevision or Aereo. It’s an interesting argument that I think has some merit. But fair use only matters if Cablevision or Aereo is publicly performing in the first place.

            Though I still await an explicit statement criteria, rather than long explanations from which I have to try to infer criteria, it seems that, ultimately, you are fixated on location as definitive in determining “possession” and therefore, volition.

            I have to say, I don’t think such can be justified. And so far as I can see, there doesn’t seem to be much, if any, support for the view that location is dispositive. (From what I can tell, the courts that have deemed there to be infringement in these cases have either gotten lost in details of the systems, or ignored the details of the systems (simply categorizing them as “any device or process”) but I don’t think any have cited location as even a factor in their determination as to volition.)

            I gave you the criteria in a list in the post above.

            Again, from the Solicitor General’s brief in Cablevision:

            …the court of appeals reasonably concluded that the subscriber—who would both select the programs to be copied and press the button triggering the actual recording—would “make” the copies that would be stored in the RS-DVR system. Respondents’ RS-DVR service would replicate the basic capabilities and limitations of a VCR or a settop DVR, and it would lack much of the functionality offered by a VOD system. Like a VCR or set-top DVR, the RS-DVR would permit subscribers to view only programs that already have been broadcast and that subscribers could have chosen to view in real time under the terms of their cable packages. The RS-DVR also would permit subscribers to view only programs that they personally and previously had directed the system to copy, and it would not allow them to view portions of a program that had aired prior to the subscriber pressing the “record” button.

            To be sure, respondents’ RS-DVR service would differ from a set-top DVR in that the tangible devices that would perform the copying and playback would be located in respondents’ facilities rather than in a subscriber’s home. That shift from local to network-based recording and playback, however, appears largely irrelevant to the determination of who would “make” the copies. With respect to the photocopying of written material, an individual who both selects the pages to be copied and operates the duplicating machine is naturally said to “make” the copies, whether the photocopier is located in the individual’s home or at a self-service copy shop. There is no evident reason for a different result here.

            I wrote a paper about why the copy shop analogy sucks. I don’t think I have the strength to go there today, though.

            Although the Second Circuit demurred in addressing who the performer was, it’s hard to see why the analysis should be any different when it comes to determining who the performer is, upon playback. The user makes the recording, juts like a home DVR. The recording is stored in storage allocated exclusively to the user, and accessible only to the user, just like a home DVR. It is, indeed, the user’s recording, just as much as that on a home DVR is. And like a home DVR, playback occurs only upon the user’s initiation. It’s hard to see where anybody other than user actually supplies any volition in playing back the recording, let alone sufficient volition to be deemed the performer. And it’s certainly hard to see how location has any bearing, whatsoever, on the question of who the performer is.

            You’re not addressing all the differences I pointed out, such as how the source of the content being performed is different. But maybe you’ll get to that in your next comment.

          • [jsg] To my view (and I think, effectively, that of the Second Circuit), delivering the content to a customer’s DVR is delivering the content to the customer….

            [dh] I don’t think that’s right.

            I realize this. I stated it because in your prior comments you have simply presumed that content wasn’t delivered to the end-user until it was inside their home, as if that were axiomatic. But it’s not axiomatic, it’s merely a belief on your part. You can’t simply assume it as a given.

            The chain of transmissions ends when a member of the public receives the transmission such that he can perform it without there being another transmission in the interim. He has to receive it such that he could render it simultaneously from the final transmission in the chain of transmissions. When a recording is sitting on the remote DVR, the member of the public has not received it yet since he can’t render it without it first being transmitted to him. Until the customer can render the transmission, the chain of transmissions is not complete, and every link in the chain of transmissions is a public performance.

            First, I will note that this is certainly a construction entirely of your own.

            Secondly, it breaks very quickly. If, for example, we change your Cox headed (i.e., connects directly to a TV) whole-home DVR to a newer architecture headless (i.e., doesn’t connect directly to a TV) media gateway that only renders content by transmission to networked STBs elsewhere in the home, then by this theory, playback becomes an infringing public performance by Cox (just like an RS-DVR).

            That’s good to know that Cablevision had negotiated for consent, thank you, but you haven’t shown how that fact changes what I said.

            Everything that you argued was predicated on your interpretation of what rights are granted under the compulsory license of 17 USC 111. It’s not relevant, so there’s no need for me to invest time arguing as to the correctness of that interpretation.

            Did Cablevision have a license to make nonsimultaneous secondary transmissions? I don’t think they did, hence the litigation.

            Nobody has claimed that they had a license to that. I claimed that they didn’t need one, and the only courts to rule on the issue agreed.

            And if they did not, then everything I said still stands and the point that Cablevision is a link of the chain of transmissions from the broadcaster to the public, and all that implies, remains to be addressed by you.

            The trouble you have is that you refuse to acknowledge the possibility that the view you have constructed, entirely on your own, could be incorrect. I’ve made clear that I believe the content is delivered to the user at the input to the user’s DVR, wherever that DVR is located. I really don’t have to argue beyond that, particularly since I’ve been just trying to get you to clearly identify your criteria for distinguishing the in-home and remote systems (so that I can can demonstrate that they are flawed).

            You’re the one who is taking a rather odd position that warrants further explanation — or at a minimum, to 1) first, actually distinguish the remote case from the in-home case, and then 2) justify the distinction you draw as actually having some foundation in the statute, as opposed to merely being arbitrary criteria that you fabricated.

            Unless you are now asserting that in-home vs. remote isn’t where the line is drawn, but rather headed vs. headless DVRs is where the line is drawn, you haven’t yet achieved (1).

            [jsg] Note that the way a typical in-home DVR works is precisely analogous to the RS-DVR. In an SA 8300 (the in-home DVR used by Cablevision), for example, the incoming signal is split, with one copy being retransmitted through the DVR outputs to the attached TV, and another copy being recorded to disk.

            [dh] And that split happens after the cable system has publicly performed by transmitting the content to the subscriber. Once the content has been delivered to the subscriber in a way that he can simultaneously render it without any more transmissions, then the chain of transmissions from the broadcaster to the public is over, and that means that these renderings are private performances and these reproductions are likely fair uses under Sony.

            Is it really? The signals on the coax attaching to your DVR are encrypted. They can’t be rendered by the subscriber until they are decrypted by the DVR and output from the DVR. The recording of programs in an in-home DVR happens before the user can render/perform the content. Does that make the in-home DVR infringing?

            [jsg] The fact that programmers argued that Cablevision needed an additional license to split the stream to supply content to the RS-DVR, does not make it something that you can take as a fact. (In fact, the only precedent on point in this regard is the Second Circuit finding the opposite.)

            [dh] What case was that?

            Uh, that would be Cablevision. Is it not absolutely evident that the Second Circuit held in Cablevision that Cablevision did not require any additional license to split the stream and supply content to the RS-DVR??

            Yes, but playback from a home DVR is in a private place, so it’s a public performance. Playback from a remote DVR requires a transmission, so it’s possible that it’s a public performance. And I think it is, for all the reasons given.

            I presume you meant the first sentence to say “so it’s a private performance.”

            Playback of the RS-DVR is also “in a private place” — precisely the same home. Playback from a conventional whole-home DVR often does involve a transmission. And playback from a headless in-home DVR always requires a transmission, just like RS-DVR.

            You’re not addressing all the differences I pointed out, such as how the source of the content being performed is different. But maybe you’ll get to that in your next comment.

            I don’t see where you pointed out that the source of the content is different. The source of the content isn’t different, so far as I can see.

            As for any other difference you pointed out, I don’t know what differences you feel you have pointed out that I haven’t addressed. To my view, I have demonstrated that most of what you perceive as differences are not, in fact, differences.

            The only differences I see remaining at this point are location of the recording equipment, and whether a transmission is required for the end-user to view a recording — however on the latter issue, I’ve pointed out that headless in-home DVRs also always require a transmission. So unless you come back telling me that headless DVRs infringe, the requirement of a transmission isn’t an actually difference, either.

          • I realize this. I stated it because in your prior comments you have simply presumed that content wasn’t delivered to the end-user until it was inside their home, as if that were axiomatic. But it’s not axiomatic, it’s merely a belief on your part. You can’t simply assume it as a given.

            And under your theory, the subscriber can obtain the content from the cable system without the cable system supplying the content to the subscriber. How does the subscriber obtain the content? How is the cable system supplying the content to the subscriber, a member of the public, not an unlicensed public distribution or unlicensed public performance?

            First, I will note that this is certainly a construction entirely of your own.

            Secondly, it breaks very quickly. If, for example, we change your Cox headed (i.e., connects directly to a TV) whole-home DVR to a newer architecture headless (i.e., doesn’t connect directly to a TV) media gateway that only renders content by transmission to networked STBs elsewhere in the home, then by this theory, playback becomes an infringing public performance by Cox (just like an RS-DVR).

            It’s not entirely my own. It’s a theory I base on the statutory text, the congressional record, and the case law. Every time I try to develop the theory, you complain that I’m not just giving you the bottom line. But then if I just give you the bottom line, you complain that I haven’t developed it. Regardless, you can keep changing the facts, and I can keep applying my theory to your new facts. With your new headless DVR, the content that is available on the system is content that the subscriber lawfully obtained in the first place. He obtained that content when the cable system made its licensed secondary transmission to the subscriber. And this is the point I’ll keep driving home. There is a difference between a cable system that lets a subscriber watch content he has legally obtained already, whether via licensed public distribution or via licensed public performance, and a cable system that lets a subscriber watch content he has not legally obtained already—content which is nonetheless supplied by the cable system.

            Everything that you argued was predicated on your interpretation of what rights are granted under the compulsory license of 17 USC 111. It’s not relevant, so there’s no need for me to invest time arguing as to the correctness of that interpretation.

            You totally missed the point. I’m not arguing about whether Cablevision had a license. I’m arguing about what activities cable systems—and Cablevision is a cable system as that term is defined in Section 111—can engage in. Section 111 makes clear that when a cable system makes a secondary transmission of a broadcast to the public, that’s a public performance. You seem to think that Cablevision can transmit broadcasts that it receives to members of the public, and yet it’s not infringement. And, again, this turns on your theory that Cablevision can legally transfer content to its subscribers, all members of the public, without it being a public distribution or a public performance.

            The trouble you have is that you refuse to acknowledge the possibility that the view you have constructed, entirely on your own, could be incorrect. I’ve made clear that I believe the content is delivered to the user at the input to the user’s DVR, wherever that DVR is located. I really don’t have to argue beyond that, particularly since I’ve been just trying to get you to clearly identify your criteria for distinguishing the in-home and remote systems (so that I can can demonstrate that they are flawed).

            How is it legal for the cable system to deliver the content to the input on the remote DVR? How is this not a public distribution?

            You’re the one who is taking a rather odd position that warrants further explanation — or at a minimum, to 1) first, actually distinguish the remote case from the in-home case, and then 2) justify the distinction you draw as actually having some foundation in the statute, as opposed to merely being arbitrary criteria that you fabricated.

            How to distinguish the remote DVR from the home DVR is simple, and I’ve explained it before. With the remote DVR, the content is not lawfully being transferred from the cable system to the subscriber. You agree that the cable system is transferring this content, but you have yet to explain how it’s legal. Where did the subscriber legally obtain the content in the first place? With the home DVR, the content has been lawfully transferred from the cable system to the subscriber. The cable system has a license to transmit the content to the subscriber. And, again, now you’re complaining that I didn’t develop my theory, but as soon as I develop it, you’ll complain that I’m not giving you the bottom line. It’s a bit frustrating.

            Is it really? The signals on the coax attaching to your DVR are encrypted. They can’t be rendered by the subscriber until they are decrypted by the DVR and output from the DVR. The recording of programs in an in-home DVR happens before the user can render/perform the content. Does that make the in-home DVR infringing?

            Nope. Again, this is content that the subscriber has lawfully obtained since the transmission from the cable system is licensed.

            Playback of the RS-DVR is also “in a private place” — precisely the same home. Playback from a conventional whole-home DVR often does involve a transmission. And playback from a headless in-home DVR always requires a transmission, just like RS-DVR.

            I feel like you’re just not listening… or understanding. Playback from the remote DVR requires a transmission. If that transmission is to a public place or a semipublic place, it’s a public performance. If the transmission is to a private place, then it’s a public performance if the relationship between the sender and the receiver is a public one. The remote DVR is a public performance because it’s the cable system sending the subscriber, a member of the public, content the subscriber has not lawfully obtained in the first place and content that was actively acquired by the cable system for the purpose of supplying it to the subscriber. The whole home DVR is the subscriber transmitting content which he has already lawfully obtained to himself.

            I don’t see where you pointed out that the source of the content is different. The source of the content isn’t different, so far as I can see.

            Home DVR: The source of the content is legally acquired from the cable system BEFORE the playback occurs.

            Remote DVR: The source of the content is illegally acquired from the cable system WHILE the playback occurs.

            As for any other difference you pointed out, I don’t know what differences you feel you have pointed out that I haven’t addressed. To my view, I have demonstrated that most of what you perceive as differences are not, in fact, differences.

            The only differences I see remaining at this point are location of the recording equipment, and whether a transmission is required for the end-user to view a recording — however on the latter issue, I’ve pointed out that headless in-home DVRs also always require a transmission. So unless you come back telling me that headless DVRs infringe, the requirement of a transmission isn’t an actually difference, either.

            Let’s hear your theory on how the subscriber obtains the content from the cable system, yet the cable system has not publicly distributed nor publicly performed the content.

          • And under your theory, the subscriber can obtain the content from the cable system without the cable system supplying the content to the subscriber.

            Hardly.

            It may be that the subscriber can obtain the content from the cable system without what is deemed, legally, to be a performance by the cable system—assuming that the courts do not interpret the transmission of signals to an RS-DVR to be a performance. (And I would certainly agree that, to this point, the courts have not interpreted such to be a performance.)

            But your leap to suggest that if there is no performance then there is no distribution is totally without basis, in fact or law.

            Content is distributed all the time without performance. When you download a song or video from iTunes or Amazon, you have distribution without a performance.

            It’s not entirely my own. It’s a theory I base on the statutory text, the congressional record, and the case law.

            You may well have drawn it from your interpretation of these things, but it is certainly your own. That’s my way of saying that, not only are you putting forward a view that has not been adopted by any court, but also one that is, to my view, downright wacky.

            Every time I try to develop the theory, you complain that I’m not just giving you the bottom line. But then if I just give you the bottom line, you complain that I haven’t developed it.

            I can appreciate how you might have interpreted it as being such criticism, but this wasn’t me complaining about lack of development. This was me observing, editorially, that what you’re proposing is just plain wacky.

            Regardless, you can keep changing the facts, and I can keep applying my theory to your new facts. With your new headless DVR, the content that is available on the system is content that the subscriber lawfully obtained in the first place.

            I don’t “keep changing the facts.” You propose criteria that you assert will correctly classify situations. Based on the criteria you propose, I construct a simple scenario where the criteria produces the opposite result that you seem to intend, in order to demonstrate that, in fact, the criteria does not work.

            If the criteria you provide only works for one simplified example I give you (in order to get you to clearly state your criteria), then it doesn’t mean much, does it? Presumably, its supposed to be criteria that works, generally, right?

            He obtained that content when the cable system made its licensed secondary transmission to the subscriber.

            If the system only has headless DVRs, then by your criteria, there’s no transmission to the user until after the recording occurs. The situation is identical to the RS-DVR, with location being the only difference. RS-DVR is just a headless DVR that happens to be located remotely.

            Here’s precisely what you said:

            The chain of transmissions ends when a member of the public receives the transmission such that he can perform it without there being another transmission in the interim. He has to receive it such that he could render it simultaneously from the final transmission in the chain of transmissions. When a recording is sitting on the remote DVR, the member of the public has not received it yet since he can’t render it without it first being transmitted to him. Until the customer can render the transmission, the chain of transmissions is not complete, and every link in the chain of transmissions is a public performance.

            This criteria does not depend on location. Transmission of content to any headless DVR, regardless of where it is located, will never qualify as the final transmission, under the above, because the user has not received the content “such that he could render it simultaneously with the final transmission in the chain of transmissions.” Rather, he can only render it after it is retransmitted out of the headless DVR, to another device.

            Now frankly, I think that this proposed criteria matches even a headed DVR, because (as I briefly mentioned later in my response), the cable transmission to the DVR is actually encrypted. So even with a headed DVR, it cannot be rendered simultaneously by the user, but rather, first has to be decrypted by the DVR. But rather than debate this point, it was much simpler to point out that your criteria produces the “wrong” result in the case of a headless DVR.

            And this is the point I’ll keep driving home. There is a difference between a cable system that lets a subscriber watch content he has legally obtained already, whether via licensed public distribution or via licensed public performance, and a cable system that lets a subscriber watch content he has not legally obtained already—content which is nonetheless supplied by the cable system.

            Again, your assertion is based on a deeply flawed assumption that if, as a matter of law, there’s no performance, then there can’t have been a legal transfer. This is simply false.

            You seem to think that Cablevision can transmit broadcasts that it receives to members of the public, and yet it’s not infringement. And, again, this turns on your theory that Cablevision can legally transfer content to its subscribers, all members of the public, without it being a public distribution or a public performance.

            Yes, I think that Cablevision can legally transfer content without a performance (as “performance” has been interpreted by the courts). And nothing in 17 USC 111 (or related sections) says otherwise.

            I didn’t comment on whether it’s a “public distribution” and I’m not aware of there being any legal definition of such a thing as a “public distribution.” I think that the Copyright Act defined “transmission” (or rather, to “transmit”) and “performance” (or rather, to “perform”) but not “distribution.” And I think the Copyright Act extended exclusive rights to copyright owners to control copying, “transmission” and “performance” but not “distribution” (other than distribution in the form of physical copies/phonorecords).

            If the courts interpreted the distribution of content from Cablevision to the RS-DVR to be a “transmission” or “performance,” Cablevision would clearly have a license.

            If the courts interpret that distribution of content to not be a “transmission” or “performance,” then Cablevision doesn’t need a license. (It may, nonetheless, have a license…but it doesn’t need one.)

            You are attempting to interpret the Copyright Act as if it were enumerating rights for cable systems, and anything not listed is prohibited. In fact, what the Copyright Act does is to enumerate rights for copyright holders. And anything not listed is something that copyright holders lack the right to prohibit.

            The 1976 Copyright Act was specifically enacted to add “performance” and “transmission” to copying as exclusive rights of copyright holders. To the degree a cable company does something that, legally, constitutes neither copying nor performance/transmission, no license is required, because it’s not an exclusive right of the copyright holder.

            Perhaps it offends your sensibilities that this non-performance distribution from cable system to RS-DVR does not fall within a copyright holder’s exclusive rights under the Copyright Act…but you don’t get to just declare it an “illegal distribution” because of that.

            Assuming that lawmakers actually agreed with you, that would merely make the situation analogous to what prompted the 1976 Copyright Act, in the first place: decades after the copyright law is enacted, technological developments introduce a new situation that had not been anticipated, and which is not covered. If Congress feels it should be covered, then they have to enact new legislation to achieve such.

            You certainly could argue that the copy made on the RS-DVR were an illegal copy—but that’s a very different argument from the argument that the “distribution” is illegal. (It’s also an argument that doesn’t seem likely to prevail.)

            And neither argument would distinguish the RS-DVR from a system of in-home headless DVRs.

    • If I stream a video from YouTube, YouTube performs it publicly when it transmits the performance to me (under the “to the public” prong of the Transmit Clause). I then perform the work by rendering it on my computer, which is a private performance. I’m curious to see whether you think there’s any crossover, such as saying it’s both YouTube and I publicly performing the work when I initiate playback.

      I just want to clarify this. In the case of YouTube, the performance of the work is a reproduction cloned copy, another file, from the [copied] file uploaded to the YouTube server. It is a copy, a reproduction, because the original file still exists before and after the upload, a undisputed copying event. In the Aereo case, the performance is the interception of the single-fixed broadcast, not a copy, thus the re-transmission, which is not a re-copy, is the argument. The reason this is important is when J.S. Greenfield states that “[C]ox becomes the transmitter of any Youtube videos transmitted from that server, because Cox has physical possession of the server.” The transmission is fixed from a copy of YouTube which physically resides somewhere else. This is the difference. To receive the transmission, Cox, in this example, still has to COPY a file residing on a server, and the transmit the feed.

      I was going to let that pass, until it became part of the debate. Even so, I too believe that the RIAA, MPAA, etc., suing individuals is wrong. It is Google/YouTube that are the copiers, it is Google/YouTube that has to be ganged up on. A copy is not fixed. It is a reproduction. A transmission is intercepted for re-transmission by another device. Thanks.

  2. It is also, perhaps, worth pointing out that it is not a performance that is embodied in a copy but a work.

    If a performance cannot be embodied in a copy, then when a band performs a song, the only copyright interest in a recording of that performance is that of the composer? Or the recording becomes a work, in itself?

    I’m no lawyer, but I presume it must be the latter, since the former both seems absurd, and seems inconsistent with enforcement of the law, in practice. After all, when music piracy has been prosecuted in the past, it has generally been prosecuted by RIAA. If the former were the case, then it seems we should never see RIAA prosecuting piracy, but instead only entities like ASCAP, BMI, or a rightsholder in the underlying composition.

    It seems to me that your assertion is misleading at best, flat-out incorrect at worst.

    According to the Copyright Act:

    A work is “created” when it is fixed in a copy or phonorecord for the first time;

    Furthermore, the term “fixed” is defined as follows (emphasis added):

    A work is “fixed” in a tangible medium of expression when its embodiment in a copy or phonorecord, by or under the authority of the author, is sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than transitory duration. A work consisting of sounds, images, or both, that are being transmitted, is “fixed” for purposes of this title if a fixation of the work is being made simultaneously with its transmission.

    So while perhaps it is technically correct to say that works rather than performances are embodied in copies, it seems a distinction without a difference, since the fixation of the performance is a work, in itself.

    What’s more, the above definition would seem to explicitly state that the recording of a transmission, itself, is a work.

    The argument that when a performance is recorded, there is no work to perform other than the work underyling the performance seems quite clearly incorrect. And more significantly, the implication that you seem to be trying to make, that when a transmission is recorded, there is no work to perform other than the underlying work, seems at odds with the plain language of the statute.

    In fact, the plaintiffs in Cablevision explicitly disagreed with your view, specifically arguing that Cablevision would be performing to RS-DVR users “the same performance…more specifically, the performance of that program by HBO or another programming service.”

    Let’s take a look at what the Second Circuit (and plaintiffs) actually said in Cablevision. But first, for reference, one more definition from the act (the one including the transmit clause):

    To perform or display a work “publicly” means—

    to perform or display it at a place open to the public or at any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered; or
    to transmit or otherwise communicate a performance or display of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance or display receive it in the same place or in separate places and at the same time or at different times.

    Now, from Cablevision:

    In essence, the district court suggested that, in considering whether a transmission is “to the public,” we consider not the potential audience of a particular transmission, but the potential audience of the underlying work (i.e., “the program”) whose content is being transmitted.

    We cannot reconcile the district court’s approach with the language of the transmit clause. That clause speaks of people capable of receiving a particular “transmission” or “performance,” and not of the potential audience of a particular “work.” Indeed, such an approach would render the “to the public” language surplusage. Doubtless the potential audience for every copyrighted audiovisual work is the general public. As a result, any transmission of the content of a copyrighted work would constitute a public performance under the district court’s interpretation. But the transmit clause obviously contemplates the existence of non-public transmissions; if it did not, Congress would have stopped drafting that clause after “performance.”

    On appeal, plaintiffs offer a slight variation of this interpretation. They argue that both in its real-time cablecast and via the RS-DVR playback, Cablevision is in fact transmitting the “same performance” of a given work: the performance of the work that occurs when the programming service supplying Cablevision’s content transmits that content to Cablevision and the service’s other licensees. See Br. of Pls.-Appellees Twentieth Century Fox Film Corp. et al. at 27 (“Fox Br.”) (“The critical factor . . . is that the same performance is transmitted to different subscribers at different times . . . more specifically, the performance of that program by HBO or another programming service.” (third emphasis added)).

    Thus, according to plaintiffs, when Congress says that to perform a work publicly means to transmit . . . a performance . . . to the public, they really meant “transmit . . . the ‘original performance’ . . . to the public.” The implication of this theory is that to determine whether a given transmission of a performance is “to the public,” we would consider not only the potential audience of that transmission, but also the potential audience of any transmission of the same underlying “original” performance.

    Like the district court’s interpretation, this view obviates any possibility of a purely private transmission.

    Your argument of “unity,” and your suggestion that the language of the transmit clause implies this unity, is the same argument that plaintiffs made that all transmissions must be aggregated under the transmit clause — and it fails for precisely the reason the Second Circuit described above: to so interpret the transmit clause would be to eviscerate the possibility of non-public transmissions.

    Attempting to cure this, as Devlin Hartline has, by suggesting that only transmissions from a single party should be aggregated still fails, because it produces the clearly erroneous result that once a party has engaged in any public transmission of a work, all of their transmissions of the same underlying work are also public. Stranger still, all prior non-public transmissions are rendered retroactively public, by virtue of a subsequent public transmission.

    The “different times” language of the transmit clause implies no such absurd thing, of course. That language specifies that separate transmissions can be used to communicate a performance to the public, not that all cases of multiple transmissions to different parties are necessarily to the public.

    In other words, it tells us that multiple transmissions can be aggregated in some cases, not that multiple transmissions must always be aggregated. The critical question in determining whether to aggregate transmissions is whether they constitute the same performance, of course.

    You want for us to believe that all performances of the same underlying work are the same performance. But for the above reasons, I think that interpretation is pretty obviously wrong.

    • Hi J.S., thanks for the comment. It is thoughtful, and I appreciate that. These are not easily discernible issues, and I don’t claim to know the answers. So keep in mind as I respond that this is my (still evolving) opinion.

      If a performance cannot be embodied in a copy, then when a band performs a song, the only copyright interest in a recording of that performance is that of the composer? Or the recording becomes a work, in itself?

      There are two copyrights present in a sound recording. The musical composition (song) copyright and the sound recording (performer) copyright. Here is why I think it is correct to say a work, not a performance, is embodied in a copy.

      Let’s say Gary writes a song. He has a copyright in the musical composition. Sarah performs the song in a studio with Gary’s permission, creating a sound recording. But under copyright law, Sarah did not engage in a performance when she made the recording, she engaged in a reproduction, a mechanical reproduction. And she needs permission from Gary for every copy (or phonorecord, as the Copyright Act refers to it as) she makes (under current copyright law, there is a compulsory license that handles this if no agreement exists otherwise).

      Now, Luke plays the sound recording to the public. He needs permission from Gary to perform the underlying musical composition. Sarah does not need permission from Gary for this performance. That is, it is not Sarah’s performance of Gary’s work that is embodied in the phonorecord, it is Gary’s work itself that is embodied in the phonorecord. No one would say Sarah is responsible anytime a third party plays a recording of her performing Gary’s song.

      To your second point: I think the aggregation inquiry is a red herring. The conceptual unity aspect is secondary to the main question: what is the relationship of performer to audience? Did you perform to the public? Then it’s a public performance. These are individual inquiries that don’t imply retroactive application. A private performance will always be a private performance. A public performance will always be a public performance.

      Put it this way: let’s say Aereo decided to “recycle” its copies. Subscriber A records Show X and watches it. Show X stays on Aereo’s server. Subscriber B wants to watch it, and Aereo serves Subscriber A’s copy of Show X. Is that a public performance? The audience capable of receiving it is two people instead of one. Now you’re stuck with a quantitative inquiry: how many subscribers can you show the performance to before it’s a public performance? It was private when it was one but it’s public when it’s two doesn’t make much sense. What if there were two people at subscriber one’s house watching the show? What if there were eight people watching the show? And if you finally hit whatever number of people needed for a public performance by recycling copies, does that retroactively make previous performances public? You still don’t get a workable definition if you import a non-statutory “unique copies” definition into the Transmit Clause.

      My interpretation says look at whether there’s a performance, and whether it’s to the public. The “to the public” language simply says you don’t disaggregate performances made in separate places or at different times into private performances.

      And it fits with common sense. You have a copy of a DVD. You transmit it to your parents to watch: private performance. You transmit it to your brother to watch: private performance. You make it available so anyone can watch: public performance. The previous two are still private performances and unaffected even though it’s the same copy.

      • I’m going to skip responding on the embodiment issue for now, because I’m not clear it’s essential to the corequestions here, and frankly, I don’t understand this embodiment question well enough, at this point, to appreciate the distinction you’re trying to make.

        To your second point: I think the aggregation inquiry is a red herring. The conceptual unity aspect is secondary to the main question: what is the relationship of performer to audience? Did you perform to the public? Then it’s a public performance. These are individual inquiries that don’t imply retroactive application. A private performance will always be a private performance. A public performance will always be a public performance.

        I can easily accept the argument that aggregation is a red herring. It is, of course, not my argument, but an argument constructed by plaintiffs, to try to get past the fact that each recording in the Cablevision/Aereo systems is made by and accessible to only a single user. (And once we get into talking about these issues in terms of aggregation, you get into the very strange consequences I mentioned.)

        I would certainly agree that under the law, a performance can be public even if there is only a single recipient — even if there are no recipients, in fact! So we certainly ought to be able to determine the public/private status of each performance, without regard to aggregation.

        Put it this way: let’s say Aereo decided to “recycle” its copies. Subscriber A records Show X and watches it. Show X stays on Aereo’s server. Subscriber B wants to watch it, and Aereo serves Subscriber A’s copy of Show X. Is that a public performance? The audience capable of receiving it is two people instead of one. Now you’re stuck with a quantitative inquiry: how many subscribers can you show the performance to before it’s a public performance? It was private when it was one but it’s public when it’s two doesn’t make much sense. What if there were two people at subscriber one’s house watching the show? What if there were eight people watching the show? And if you finally hit whatever number of people needed for a public performance by recycling copies, does that retroactively make previous performances public? You still don’t get a workable definition if you import a non-statutory “unique copies” definition into the Transmit Clause.

        The fundamental error that you have in the above is that the copy is not Aereo’s to recycle. The copy is subscriber A’s copy. If Aereo were to do such recycling (or even to assert the right to do such recycling, without actually doing it), then their claim to be renting equipment (in this case, storage) would clearly be factually incorrect. They would not be operating their system as an equipment rental.

        And this is, in my view, the true essence of this case. Aereo is renting equipment to users, and the users are using that equipment to receive broadcasts, record those broadcasts, and retransmit/perform those recordings. I don’t see any basis in the statute for distinguishing that from Aereo renting the user equipment to place within their home. And I have yet to see anybody produce anything even plausible for how the two cases might be coherently distinguished (let alone for what the statutory basis for such a distinction would be).

        So while the Second Circuit reached their conclusion regarding the status of the performance without considering who the performer is, for me it’s much easier — and if the status is to be determined based on the relationship of the performer to the audience, then the answer in this case is very simple for me, because the performer is the audience/user.

        Whether there are others also viewing with the user is of no moment under the statute, so long as the others are part of a normal circle of family and social acquaintances. Of course, if the user offers to perform their recordings to the public, then such a performance would be a public performance, but it would be a public performance by the user, not by Aereo, because the user is in control of the recording and whether it is made available to the public — just the same as if it were recorded on a device placed within their home, and they made it available to the public.

        I can appreciate that the plaintiffs, and others, find the Aereo system — renting equipment as a service — to be frustratingly similar to a content rental/retransmission service. But unless somebody can produce a basis for distinguishing a remote equipment rental such as Aereo from a co-located (within the user’s home) rental of equipment, that is drawn from the statute, what we have is simply people arguing for what they think should be done to meet the presumed intent of those who drafted the law, rather than actually interpreting the law as written.

        I have yet to see such, not for lack of looking/asking, and I’m skeptical that any such basis exists. That’s why I have repeatedly said that to the degree this situation represents a problem, it is a problem that requires an arbitrary line to be drawn in some fashion, and to draw such is the role of Congress, not the courts.

    • Attempting to cure this, as Devlin Hartline has, by suggesting that only transmissions from a single party should be aggregated still fails, because it produces the clearly erroneous result that once a party has engaged in any public transmission of a work, all of their transmissions of the same underlying work are also public. Stranger still, all prior non-public transmissions are rendered retroactively public, by virtue of a subsequent public transmission.

      I don’t think that once a party has engaged in one public performance of a work then all performances of that work by that party are necessarily public.

      The “different times” language of the transmit clause implies no such absurd thing, of course. That language specifies that separate transmissions can be used to communicate a performance to the public, not that all cases of multiple transmissions to different parties are necessarily to the public.

      I agree.

      In other words, it tells us that multiple transmissions can be aggregated in some cases, not that multiple transmissions must always be aggregated. The critical question in determining whether to aggregate transmissions is whether they constitute the same performance, of course.

      I mentioned that I was working on a follow up post (it’s about 6,000 words at the moment). I’ll go into more detail there. But the gist of my argument reflects what Terry just above said, namely, that the aggregation theory is a red herring. In fact, I think whether there’s one source copy, many source copies, or no source copy at all, the result is the same. There’s two prongs to the analysis: (1) a transmission of a performance, and (2) that transmission is to the public. The source of that transmission is irrelevant. One can transmit a performance even if one has no copy from which he is transmitting, so the number of copies can’t matter. Nor does the number of transmissions matter. YouTube is publicly performing even if it only sends out one stream to one user since it’s a transmission to the public. I think the crux of the aggregation theory is that the “different times” language shows that it’s irrelevant if the transmissions are asynchronous. But the error is to think that only by aggregating the transmissions do they become “to the public” since one transmission can be “to the public” even if there’s no other transmissions to aggregate it with.

      • I don’t think that once a party has engaged in one public performance of a work then all performances of that work by that party are necessarily public.

        I realize you don’t think that. 😉 My point was that it is the (absurd) implication of aggregating transmissions by the transmitting party.

        I mentioned that I was working on a follow up post (it’s about 6,000 words at the moment). I’ll go into more detail there. But the gist of my argument reflects what Terry just above said, namely, that the aggregation theory is a red herring.

        Since you and Terry both feel the aggregation is a red-herring, and I have no particular investment in it (as I noted, aggregation was plaintiff’s argument — I find that all of the proposed aggregation arguments produce absurd results), let’s just agree to focus on other lines of reasoning.

        In fact, I think whether there’s one source copy, many source copies, or no source copy at all, the result is the same. There’s two prongs to the analysis: (1) a transmission of a performance, and (2) that transmission is to the public. The source of that transmission is irrelevant. One can transmit a performance even if one has no copy from which he is transmitting, so the number of copies can’t matter.

        I would suggest that you try examining this from the perspective that the number of copies (and number of transmissions) can inform the determination of whether the system is being operated as an equipment rental system or as a content rental system, and that this distinction could be determinative of both who the transmitter is, and whether the transmissions are public.

        If there aren’t both separate copies and separate transmissions, for each user, then the system is clearly not an equipment rental system.

        If there are, perhaps that’s not absolutely determinative that it is an equipment rental system. However, if the party operating the system asserts that it is such, I would think separate copies (made and deleted at the control of the user) and separate transmissions (also at the control of the user) are at least prima facie evidence that it is actually operating as an equipment rental system, and the burden to prove that it is something else should then shift to those who claim that it isn’t actually an equipment rental system.

      • Devlin,

        YouTube does not send out a transmission until after a COPY, a reproduction is created first. If it were not a copy, then when the transmission ended the original file would too not exist. Clearly, it does exist. Thanks.

        • I’m not sure I follow you. Where is the copy made that occurs before transmission?

          • Okay. This is where law is behind computer science. Everything on the Internet is a file resting on a server somewhere—-that is the way it is, there is no alternative to this thinking. To view a file, a program has to make a copy of that file for it to be seen, even if it is an exact replica. When two people on two separate devices visit CopyHype.com, each person see a copy of the website (which rests on a server (index.html would be the file that no one can read)). Once that copy is created, each mutually exclusive person can then view the contents via the transmission to their separate PCs, phones, etc. In YouTube’s case, to view a video, their player (which is software) is automatically programmed to play (unlike a video on Bloomberg.com where one has to click the play button before the copy is made). YouTube makes the copy, a reproduction, of video.flv from the auto-play software so that [you] can view it. From that, it is transmitted as a fixed signal to your PC, or via a cable service.

            In the Aereo case, they are intercepting a fixed transmission signal and then attempting to profit from it as a re-transmission—-in fact, they try to profit from unlicensed syndication of fixed objects that reside in TV studio libraries (whether on film, video, or as a file stored on a TV station’s server). Unlike YouTube where x number of people can view the “copies” like an on-demand, Aereo’s retransmission (from the original interception) can only be viewed simultaneously from the single transmission. For example, when Fox News ends the transmission of the O’Reilly Factor at 8pm EST, the public using Aereo have to wait until 11pm EST for the rebroadcast on the same transmission from Fox, unless someone privately copies it on a recording device from the intercepted transmission. There does not exist a copy of the O’Reilly Factor that Aereo can produce for others to see until Fox authorizes its own transmission within a fixed amount of time.

            In radio, Orsen Wells did a live broadcast of War of the Worlds. That transmission was licensed by the radio station, transmitted by an FCC License, and those with radio devices could intercept the fixed transmission by decoding the signal. Because it was fixed, there were no other ways for the broadcast to occur again. In YouTube’s case, they authorize the illegal non-fixed copy to be uploaded to their server, from which it resides. To view it, the service has to make another non-fixed copy. Only the SDNY court in Capital Records v. ReDigi (March 2013) was smart enough to figure it out correctly via current copyright law concerning a fixed digital object and a reproduction copy of a digital object before a transmission.

            Thanks!

          • And thank you! I’ll have to reread the ReDigi case to see what you’re referring to.

  3. I think the nature of digital transmission is throwing people for a loop, completely unnecessarily, so let me try and briefly fill in some technical detail, whilst at the same time explaining why I think the Second Circuit got it badly wrong in US v. ASCAP.
    The standard way of making analog data digital is by quantizing it. For our purposes here, it means encording the value of a signal (audio, video, whatever) in binary form. Binary gives us two possible digits: 0 and 1, so the only way to record larger (or smaller) values is to use a greater number of binary digits (bits). By longstanding technological convention, the basic unit of binary data is a byte (eight bits), but these days bytes are usually grouped into longer ‘words’, consisting of several bytes. Thus, a CD-audio word is 16-bits (2 bytes), meaning that each individual signal value (recorded at a rate of 44.1 kHz – 44,100 times a second) is encoded across 16 binary digits.
    (Aside: I’m sorry if it sounds like I’m saying the obvious, but I’ve learned to be cautious in assuming what people know about computing, digital formats and so forth.)
    When it comes to transmitting digital data, there are two ways to do it: in parallel – where all bits in a word are sent at once, or in series – where one bit is sent at a time and you know you’ve got a complete word once you’ve received enough bits. Parallel transmission has several technical problems – the chief being that whilst all the bits in a word might be sent at the same time, we have no guarantee they shall be received at the same time and thus we cannot say with certainty whether a stray bit belongs in the previous or the following word, so the preferred method these days is serial transmission – especially since the big problem with serial, speed, has long ceased to matter much.
    When receiving serial transmissions, there will be an unavoidable delay between receiving the first bit in a word and receiving the last bit. For the duration of this delay, the received bits will need to be stored in a buffer and may only be used for whatever their purpose is when the entire word has been assembled.
    What that means is that any digital transmission will necessarily involve making a copy of the data transmitted (even if only temporarily), but also that there is no way of going directly from receiving to decoding a digital transmission – buffering will always be involved. This has intereseting implications.
    This brings me to the Second Circuit’s assertion that “The downloads at issue in this appeal are not musical performances that are contemporaneously perceived by the listener.” This is a question of circumstance, rather than the intrinsic nature of downloads. By this I mean that when radio broadcasts music, that also isn’t a “musical performance that is contemporaneously perceived by the listener” – unless they happen to have a radio receiver tuned to the station. Basing a legal analysis on what amounts to asking whether a tree falling in the forest makes a sound when there’s no-one to hear it would seem quaint if applied to radio, but somehow it seems to fly with digital music files.
    The thing is, it is neither impossible, nor very hard, to engineer a solution whereby a downloaded music file is buffered and played back as soon as we’ve captured a sufficient amount of data. That’s how YouTube, but also digital radio and television – that undoubtedly perform transmissions – works. The court has made the error of looking at one possible – and much better known – way of receiving digital transmissions (buffer the entirety of the data and then play it back) and using that to rule out all other possible ways of receiving the same digital transmission.
    We can demonstrate the error by applying it to a different, more familiar technology: the VCR. A TV broadcast – which is unquestionably a transmission – can be watched on a TV set; or it can be recorded on a VCR, even if the TV is turned off. In fact, the timers on VCRs were meant just for this purpose – recording television broadcasts whilst you were away or otherwise unable to “perceive them contemporaneously”, so you could watch them later. Of course, if you used your VCR in this fashion, you would need to “take some further action” to view the broadcast, which leads us to an interesting conundrum: if nobody is watching a broadcast because they’ve all set their VCRs to record it, is it still a transmission? If not, how do we tell if anybody was watching?
    The sensible way out is to assume that the actions of the receivers are irrelevant to the question and only the actions of the transmitter matter. If someone decides to broadcast a performance of a work into the ether, that in itself constitutes a transmission, as long as it is possible to receive it. To consider the question in any other fashion is to stray onto a path fraught with philosophical meanders and logical inconsistencies, It also ties the culpability of the transmitter (in cases where the transmission was infringing) to circumstance rather than intent. It’s not unlike absolving a would-be killer on the grounds that he couldn’t hit the broad side of a barn (example used for its absurdity).

    • The thing is, it is neither impossible, nor very hard, to engineer a solution whereby a downloaded music file is buffered and played back as soon as we’ve captured a sufficient amount of data. That’s how YouTube, but also digital radio and television – that undoubtedly perform transmissions – works.

      That downloaded music file is a COPY that is then buffered. The reason it is a COPY is because you receive data as a reproduction from the server from which it resides. If it were a transmission, then the file would not exist after the transmission ceased. YouTube is that way, a copy from a file that resides on its server. After a COPY is made, then a transmission takes places. The COPY is unauthorized. In Aereo’s case, they are intercepting a transmission from a single FIXED source and then provide it as a re-transmission. If Aereo were just selling antennas, they would be fine. But in reality, they are selling a subscription without paying the “free to broadcast” signal FCC license holders. That is not a rental agreement.

      • Next time you get a speeding ticket you should mention to the officer that technically your car was stationary from your inertial frame of reference.

  4. If you hook up a SlingBox or similar device to an antenna, you get pretty much same exact service as Aereo. The whole setup costs ~$200, and probably could be less in the future. Also without paying a dime in license fees to broadcasters.