Wednesday, February 06, 2008

Mechanical Licenses

Last week, William Patry wrote on his participation in the royalty rate setting hearings going on in DC. Today, articles from Digital Music News and Billboard discuss the same issue.

The National Music Publishers Association ("NMPA") continues to argue that, not only should users pay a performance royalty for the stream, they should also pay a mechanical license fee for the reproduction. Most people would call this "double dipping."

Admittedly, these questions arise because a liberal amount of duct tape has been applied to an antiquated copyright system that hardly made sense before the new technologies emerged. Rather than applying more bogus fixes, we should instead take a step back and look at the fundamentals of the desired result. Namely, how can we ensure writers and publishers are fairly compensated for uses of their works without unduly burdening the user?

In order to do this, we have to look at how the music is provided, how the provider is being compensated, and how the user consumes the music. There are a couple of clear cases - pure downloads (DRM'ed or unencrypted) are virtually identical to CDs and records. A copy is made that can be performed at will by the user, and the user is paying for that particular copy.

Non-interactive streams are also identical to their predecessor in that when the music is performed is dictated by the provider, not the user. Whether the provider earns income from advertisers or subscription is, for all practical purposes, immaterial because that income is not directly attributable to any single performance or single song. The whole of their income can be fairly attributed to the whole of the music performed. Therefore, it is impractical to levy a license fee strictly on a per-use basis.

The NMPA has been selective in how it would like to see certain uses be treated based on what would create the most wealth for its members, even if its position is contrary to logic or the common good. For instance, in 2006, it sought to have the Register of Copyrights find that ringtones were not digital phonograph deliveries ("DPD") but were in fact performances and, therefore, did not qualify for the statutory mechanical license. It went further to argue that these performances were unlicensed derivative works.

The Register issued her findings on October 16, 2006, and after a thoughtful analysis, concluded that ringtones were DPDs and qualified for a mechanical license under Section 115. The Register observed the definition of a DPD, namely, "each individual delivery of a phonorecord by digital transmission of a sound recording which results in a specifically identifiable reproduction by or for any transmission recipient of a phonorecord of that sound recording."

A ringtone, like a download, CD, or record, results in the same thing: a specific copy of the composition on a sound recording that is controlled by the user. The NMPA tried to argue against this by pointing out that ringtones are not played at the will of the user but instead are only performed when the phone rings. It further argued that because the performance could be perceived by others in public, it constituted a public performance.

A couple months after the Register's opinion was published, I asked the head of the NMPA whether this meant that publishers would stop insisting on a performance license for ringtones, since the Register made it clear that ringtones were DPDs. My presumption was that mechanical licenses for DPDs compensate the "mechanical" performance of music, while a "public performance" license compensates for "live" performances. In other words, the two are mutually exclusive - like sedans and pick-up trucks are both automobiles but no sedan can be a pick-up (El Caminos notwithstanding).

The NMPA took advantage of the Register's opinion and muddied the waters even further. The response to my question was that the NMPA did not agree that ringtones were DPDs but so long as that was the Register's opinion, they would insist on a mechanical license. The NMPA also did not agree with me that public performances and DPDs are mutually exclusive and pointed out that the Register did not say that ringtones weren't public performances, only that they were DPDs. Therefore, the NMPA's position was (and as far as I know, still is) that a ringtone requires both licenses.

Such far-fetched arguments are precisely why the debate has gotten to where it is today, and these same types of arguments are why the copyrights system was already flawed.

Now the NMPA is using its ringtone-based logic to maintain that interactive streams require two licenses and two fees. On the other side, digital radio services attempted to have the Register issue a ruling as to whether on-demand streams require a mechanical license at all. The royalty board refused to submit the question to the Register, finding that it was an issue of fact and not law.

The semantics are getting in the way of reaching a real solution. We should stop trying to put square pegs into round holes and look at what is really going on. Unlike ringtones, which are clearly and undeniably a DPD and DPD only, on-demand streams are a hybrid because users are getting to perform the music when they want, but yet never take title to a tangible medium that embodies the work. However, their use is limited like that of radio because they must be connected to the service in order to play the music, thus limiting their ability to perform the music. They must also continue paying a subscription fee (presuming there is one) and once they stop paying, they no longer have access to the music.

As much as I dislike the ASCAP and BMI royalty models, perhaps it is best to suggest a third, wholly distinct class of uses akin to their licenses. After all, our "mechanical" license owes its origin to the player piano, and it's name is indicative of what it permitted - the mechanical performance of music. A third class of use could shed the shortcomings of both performance licenses and mechanical licenses. It could be designed specifically for what it is intended to authorize.

This third class would include any use that allows users to select their performance but stops short of providing them with a permanent copy. The licenses for these uses could be set on a scale, the cost for which depends on these factors:

(1) the degree of access, i.e., unlimited performances versus a single performance per user, and/or the availability to the user (must user be connected to the service?);

(2) the duration of access, i.e., virtually indefinite versus a short time; and

(3) the payment method, i.e., free to the user and/or ad-supported, one-time fee, or a monthly subscription.

Considering each one of these factors will likely account for all the possible delivery methods and will allow enough flexibility so that the license system may grow. They are all relative to the same two extremes and where they fall in that continuum will determine what rate is appropriate.

No matter what happens, I hope that all parties involved stop thinking solely with their wallets and start focusing only on what is right and fair.