Monday, May 30, 2011

A Taxonomy of Intellectual Property

In the interest of adding some more substantial content to this blog, I thought I would present a tutorial series. I have already made reference to issues in intellectual property law, but some of my readers might not be familiar with the basics (let along the intricacies) of intellectual property in the US. To that end, this post will be the first in a series of tutorials on intellectual property. Let's begin with a taxonomy of intellectual property.

US intellectual property law recognizes four types of intellectual property.

Copyright concerns creative works fixed in a tangible medium of expression, including musical works, novels, poems, paintings, sculptures, etc. Since it concerns creative media, copyright is likely the form of intellectual property that people have encountered most. Copyright originated as a way to censor seditious or heretical publishers. Once enshrined in British law, from which the American statutes take their original form, copyright allowed publishers to enforce statutory monopolies on books. As of the 1998 Copyright Term Extension Act (CTEA, or the Sonny Bono Act), copyright lasts 70 years after the death of the author for works of individual authorship, and 120 years from the date of publication for works of corporate authorship. Considering the history of copyright these terms are fairly excessive. The earliest copyright term in the United States was a mere 7 years from publication. For much of the 19th century, copyright lasted for 14 years, again counted from the date of publication. Copyright began approaching such long terms only in the 20th century. No matter the term, copyright gives the copyright holder the sole right to sell the work, duplicate it (hence, copyright), and prepare derivative works (ie translations, adaptations to different media, etc). Fair use provides a loophole for certain kinds of use (for instance, reproducing an excerpt of a novel for a book review). For those curious about other such exceptions, fair use will have to be handled in its own tutorial.

In many ways, patent law is to inventions and discoveries what copyright is to creative works. Patent provides innovators, inventors, and researchers, with a temporary monopoly in their creations. Unlike copyright, patent terms are fairly short and have remained mostly unchanged throughout the 20th century. Patents last for 20 years from the date of filing. The Patent and Trademark Office clears patent applications, checking the new invention against “prior art” (previous inventions that might have led to the current invention). Innovation really is the word of the day for patents; to grant a patent, the applicant must show that the invention is both novel and useful in some fashion, not merely a new version of an existing invention. The patent application must disclose prior art, and in some jurisdictions, Europe for instance, the patent holder must actually utilize the invention, licensing it for production or industrial use. The US allows for “patent trolls” to sit on patents, holding them without licensing them until a possible infringement appears, at which time the troll files suit for patent infringement to make money. In practical terms, patent gives the rights-holder the exclusive right to license the invention, in exchange for disclosure. As such, while one might think that patents protect from reverse engineering, in practice the disclosure requirement essentially negates any need for reverse engineering. Nevertheless, the patent holder does have the exclusive right to license the invention for production or use, so patents are a way of securing a living by way of technological innovation.

Somewhat different from the other forms, trademark does not concern creative or technological innovation. Instead, the best way to think of trademark is as a kind of fraud protection. A firm can trademark its logo, name, or mascot, allowing the firm the exclusive use of the mark for the purpose of representing its business. As such, trademark lasts as long as the firm uses the brand for its business. A firm can lose a trademark if the brand becomes disassociated with the firm. The most famous example of losing trademark is likely the case of Bayer and their trademark on “aspirin.” At the time, patents were not granted on chemicals, only new processes for extraction or synthesis, so Bayer took out a trademark on “aspirin.” Nevertheless, aspirin became a household word for the extract of willowbark, so Bayer lost is exclusive claim on the word. Trademark law has evolved along with trends in marketing and advertising, requiring new rules for “non-representational” logos, such as the brown shade used by UPS and the blue/orange combination used by FedEx. Perhaps the most disturbing aspect of trademark is the ability of the mark's owner to control public display of a trademark. As such, a firm can invoke trademark law to suppress a work of art critical of the company if that work features the trademark in some fashion.

Trade Secrets
The final category is perhaps the most unusual category. Trade secrets can be any important business strategy held as confidential within the firm. There is no application for trade secret protection. Instead, the firm must take measures to keep the information confidential, such as requiring employees privy to the secret to sign Non-Disclosure Agreements or Non-Competition Agreements. Trade secrecy is not a very robust form of intellectual property. The primary protection offered is protection against corporate espionage. Once a trade secret is released to the public, there is no recourse. In cases of espionage, the victim firm can sue for an injunction preventing the offender from making use of the trade secret. If the secret is released by other means, such as an irresponsible employee, the secret is simply out. The most famous example of a well-kept trade secret is the formula for Coca-Cola. The formula has been kept secret for over a century, but should it ever find its way into a newspaper, the secrecy would come to an end.

Friday, May 20, 2011

Digital Distribution

Continuing on this week's topic of first sale and digital distribution, I thought I would discuss emerging distribution strategies for digital media. The outline below comes from my observations on new media technologies, some of which can be found in an earlier entry here. As far as I can tell, digital distribution strategies can be divided into three categories according to salient features.

  • Access-Based distribution (“cloud” based services)
    • the customer subscribes to a service
    • the subscription entitles the customer to access content stored on the provider's servers
    • content is remotely stored, though some items may be remotely cached for offline use
    • when the subscription is terminated, the customer loses access to all content
    • the content provider can exercise a great deal of control over what content is offered; the selection of content may vary over time, meaning that the customer is only guaranteed access to the cloud, not any particular item in the cloud
    • typified by
      • Hulu+
      • Netflix streaming service
      • Amazon's Instant Video service
      • Pandora
      • Napster
  • Direct Distribution
    • the customer purchases a digital file from the content provider
    • once downloaded, files are stored locally
    • the customer retains the files regardless of any relationship with the provider
      • DRM technology may or may not restrict the customer's access rights to the files
    • typified by
      • iTunes
      • Magnatune
      • Napster
  • Device-Based distribution
    • content is purchased, delivered, and consumed via a device sold by the content provider
    • the customer downloads content directly onto the device; DRM technology ensures that the files are useless without the device or software from the content provider
    • files are purchased individually, so the customer can continue to use any purchased files regardless of continued relationship with the content provider
    • content cannot be backed up or transferred to another device; once the device is abandoned, access to the content is lost
    • content provider has access to the devices, so content downloaded is not always secure from remote management by the provider (see Kindle and removal of Animal Farm)
    • typified by
      • Amazon's Kindle and other ereaders
      • PS3/Xbox 360/Wii marketplace games

These strategies are not mutually exclusive. As noted above, Napter now provides a subscription service for access to its cloud alongside offering individual songs for purchase. Netflix continues to offer rental of DVDs in addition to its online streaming service. Now, to see what implications these strategies have for first sale, we must keep in mind that first sale seems to serve two key functions. First sale makes media more available, by allowing the secondary market where copies can be bought and sold even after the original distributor has ceased production. In addition, first sale makes content more affordable by facilitating the existence of lending libraries and, once again, the secondary market. The extent to which these goals can be realized on the digital marketplace depends heavily on the distribution strategy employed.

Access-Based Distribution
There is no way to talk about first sale in the traditional sense here because there is no formal sale. The customer has merely paid for access. It is as if the customer had a subscription to a non-lending library. The subscription gives the customer the right to enter the library, peruse the collection, and read or consult any of the available books, but subscribers may have no control over the content of the library's collection. The library might remove a book from the collection before a customer gets the chance to read it, or it may be very slow to acquire new works. The lack of control makes the subscription less valuable than purchasing works individually, but subscribers may get access to more content for their money. Nevertheless, the provider's control means that content will not be available should the provider go out of business or end the subscription service, so there is no chance of archiving content.

Direct Distribution
Here, the case may be improved somewhat. In such cases, the customer has downloaded files, so access to the content is not contingent on any relationship with the content provider. In the absence of DRM, the customer can access content using any appropriate device or software, make backup copies, and convert the file into another format should changes in technology make such conversion necessary. Nevertheless, content cannot be re-distributed lawfully under the current intellectual property system. The typical means of transferring files involve simply making a copy (from the hard drive to a removable medium, from one hard drive to another across a network), so such transfers are presumed infringement under current law. Of course, content providers might also include DRM technology with such files, potentially hampering all such uses. Free culture advocates often express frustration at both DRM and the state of current intellectual property law because direct distribution of digital files should improve availability of content, both upon release and in archives. While digital files require some storage medium, transferring between media is trivial, allowing many copies to be made. Computer networks also speed distribution compared to distribution by physical medium. Even so, DRM technologies and current intellectual property law prevent making content more available and affordable. The potential of the technology is going unrealized because the law has lagged behind technology, and content owners have been very effective in lobbying legislators to protect their current business model.

Device-Based Distribution
Here, we have a mixed bag of the other two strategies. On the one hand, files are transferred to the customer, so there is no reliance on a continued relationship with the content provider to ensure access to content. On the other hand, DRM is standard for this distribution strategy, making the files more difficult for the customer to transfer, convert, or even use (since a particular device is required). Archiving is impossible since the files are tied to the device; once that device becomes obsolete, the content will be inaccessible. The possibility of transfer will depend on the content provider's DRM; the nook, manufactured by Barnes and Noble, allows limited lending of some books, but no permanent transfer, even on a “send and delete” basis. Furthermore, some content providers may choose to retain access to the customer's device storage. Amazon recently used such access to remove copies of George Orwell's Animal Farm (a public domain work) from all Kindle devices at the request of a publisher (as seen in the New York Times, among other places). As such, content stored on such devices cannot be considered entirely secure from outside interference. All told, device-based distribution is the worst for preserving any first sale benefits, unless content providers offer content at a much lower price in exchange for inherent limitations. Nevertheless, archiving of content distributed in this way is at best unreliable, at worst utterly impossible, and in any event reliant on the provider's continued support of the device.

From personal observation, Access-Based and Device-Based distribution are the strategies preferred by content providers, likely due to amenability to control. If those strategies come to dominate the market, first sale itself will become obsolete. In that event, the availability of content, and the distribution of that availability will be entirely depended on the content providers. There will be no method to archive content, resell it on a secondary market, or alter it for one's own purposes.

Wednesday, May 18, 2011

First Sale and Digital Content

The previous entry ended on the note that to discuss file sharing, there must be some discussion of first sale in the age of digital content. At base, first sale is a straightforward concept. A copyright-holder has the exclusive right to make and sell copies of the work, but that right to sell is exhausted after the first sale. Once a consumer has purchased a copy, she has the right to resell the copy, give it away, lend it out, or even destroy it. First sale typically makes content both more available and more affordable. Content is made more affordable by the creation of secondary markets for used copies. Libraries also rely on first sale rights to lend copies to the public (or to members, students, etc), lowering the cost of access by distributing the cost of copies over a population (of tax-payers, tuition-payers, etc). First sale increases availability of content by allowing the owners of copies to continue to circulate (through sale or gift) their copies even after the work is no longer in production (publication, distribution, etc) by the copyright-holder. Works are therefore less likely to become “lost” when they are no longer available on the primary market.

The first thing to notice here is that first sale is predicated on the requirement of a medium, some form by which content is both delivered and consumed. The copyright-holder has the right to make and vend copies. Those copies serve as a means to transfer content between the publisher and the consumer (books, records, DVD's, etc) and as the means by which the consumer consumes the content (reads the novel, listens to the music, watches the film). At the point of purchase (or other form of alienation, such as winning the book in a raffle), the copy becomes the property of the consumer, and she may dispose of it as she wishes. In so doing, no copy is made, so there is no infringement, merely a transfer of chattel property. As such, it is coherent to talk about the sale of the copy as something other than the sale of the content.

The digital age changes the picture somewhat. When content is distributed digitally, no medium is required for transfer of content from distributor to consumer. Instead, the consumer simply downloads a copy from the server to her own computer (or other device). The consumption of the content then takes place on an appropriate device or with the use of appropriate software. Of course, there is still a medium, the hard disk drive or other storage device (flash drive, mp3 player, etc), but no medium changes hands. Furthermore, the role of medium in media consumption is trivialized by the ease of transfer. If I have an mp3 on my hard drive, I can play it using any one of several programs, or I can transfer it to my mp3 player for later listening.

In such cases, it is not immediately clear how first sale might apply. All of the standard ways of transferring digital content involve making a copy of the file. In an instance of digital distribution, there is no medium to resell or transfer to another. I could make a copy, but media producers appear to be certain that such copies constitute infringement, especially when disseminated via peer-to-peer networks. As such, it would seem that first sale might simply pass away with the ascent of digital content. Unfortunately, in such cases, there is chance that the benefits of first sale will be lost.

Consider the relatively recent case of Amazon deleting copies of George Orwell's Animal Farm from Kindle devices due to a request from the publisher. Without notice, “owners” of a public domain text found that text missing from their libraries. Likewise, paid subscribers to the Hulu Plus service may queue an episode or movie for later viewing, only to find that the content provider has removed the episode from the service. In these cases, the content providers have made clear that access to digital content is something that they want to control, and very tightly, but tight control is more likely to lead to loss of works. If a content provider removes all episodes of a series from Hulu, refuses to release it on DVD, refuses to sell rebroadcast rights, and refuses to sell or otherwise release any digital copies, the series will quickly be forgotten. Now, one might also say that such activities are antithetical to the content provider's interest. After all, if the series cannot be seen in any form, it will not become popular, so there will be no demand and no way of making any further profit from the work. Nevertheless, there is nothing to stop a content provider from releasing the series, allowing access via Hulu or perhaps Netflix for a time, then removing the content entirely once the profits begin to decline. In such cases, there may be demand for the series, but the content provider prefers to focus its efforts in developing other media, and does not see letting any particular series fall by the wayside as especially tragic.

The challenge, then, is to fine some way of preserving the benefits of first sale, finding some way to distribute access to content on a more egalitarian basis (as libraries as secondary markets do) and allowing individuals or organizations to archive content to make it available once it is no longer offered on the primary market.

Monday, May 16, 2011

Types of File Sharing

I have neglected this blog due to a wave of work. In an attempt to turn that neglect around, I offer up some somewhat sketchy notes for my dissertation. I have been thinking through some arguments about file sharing and first sale. To get a better grip on the issues, I have a short taxonomy of file sharing cases. Other scholars have done this before; this is just my attempt to get some clear on the fine points of the file sharing debate.

Instances of file sharing can be divided into four categories:

  • Public Domain content (PD)
  • Protected works still in circulation (PC)
  • Protected works out of circulation (P~C)
  • Unreleased works never in commercial circulation (NC)

For each category, there are different issues at stake. PD cases are the least problematic. A work in the public domain can be freely copied, so there are no legal interests at stake. PC cases are perhaps the hardest cases for justifying file sharing. The work is still under legal protection and is still available for purchase on the primary market. In other words, the work is still readily available, and it might be the case that the proliferation of free copies will undermine the market for commercial copies, depriving creators of profits and therefore undermining the incentives provided by copyright. To argue that PC file sharing can be justified, one must address those concerns.

P~C and NC cases are even more complex. For P~C cases, file sharing might be justified on an Archiving Argument. One might argue that even though the work is still under copyright, if it is no longer available for purchase, the work might be lost. If the work has some cultural significance, or even just personal significance to a sizeable audience (and these two may be one and the same, really), the work's falling into obscurity may be seen as tragic. Devotees of the work may then counteract the loss of the work by making digital copies available and encouraging others to do the same. In addition to archival purposes, some of the fans who have downloaded the work might have been willing to pay for it if it were available commercially (ie if a copy could be located on the secondary market, or if the original distributor released the work again), but since it is not, file sharing is their only means to access the work. These argument all appeal to some sense of creative works bringing a benefit to the audience, and the Incentives Argument is rooted in securing that benefit. In such cases, one might argue that copyright has failed because the work is both protected, the monopoly still legally guaranteed and enforceable, but unavailable to the audience. If one has ever felt any animosity directed at the Disney Vault, such arguments might have already come to mind.

NC cases are perhaps the most difficult to address because they are not uniform. Some NC works are recordings of concerts made by attendees, some are works the creator never decided to release, and some might be early versions or drafts of works released later. One might argue that the creator's decision to not release a work must be accepted by the audience; releasing such works might undermine the creator's artistic integrity or distort the creator's sense of his/her own work. On the other hand, there are some Archiving Arguments here as well. One might argue that preserving and sharing unreleased works is in the interest of documentation for future study. A close study of NC works might give the audience a more rich appreciation for the creator, providing some valuable insight into the creative process, perhaps. Once again, there seem to be two sets of concerns competing: the creator's control over the release of works, and benefits sought by the audience but potentially blocked by copyright and the creator's unwillingness to release the work.

P~C and NC cases seem to test the Incentives Argument. To address such cases, providing justification for those types of file sharing or to show that such justification is impossible, the limits of the Incentives Argument must be better defined. While PD cases are entirely unproblematic, PC cases might not be justified if free copies do indeed undermine the market for commercial copies. Now, some scholars have argued that there is no such effect. If it is the case that free distribution of digital copies either does not affect the commercial market, or have a positive effect on the commercial market, market-based arguments simply fail. At that point, it seems that the only remaining argument against PC file sharing is that first sale justifications cannot apply, since a new copy is created at every transfer.

It seems that discussing file sharing requires some discussion of first sale. More to follow, hopefully soon.