UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

What role for copyright in podcasting? : a study of crowdfunding and advertising models in an emerging… Rei-Anderson, Cody 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_may_rei-anderson_cody.pdf [ 965.4kB ]
Metadata
JSON: 24-1.0378227.json
JSON-LD: 24-1.0378227-ld.json
RDF/XML (Pretty): 24-1.0378227-rdf.xml
RDF/JSON: 24-1.0378227-rdf.json
Turtle: 24-1.0378227-turtle.txt
N-Triples: 24-1.0378227-rdf-ntriples.txt
Original Record: 24-1.0378227-source.json
Full Text
24-1.0378227-fulltext.txt
Citation
24-1.0378227.ris

Full Text

WHAT ROLE FOR COPYRIGHT IN PODCASTING?: A STUDY OF CROWDFUNDING AND ADVERTISING MODELS IN AN EMERGING MEDIUM by Cody Rei-Anderson B.A.&Sc., McGill University, 2012 J.D., University of British Columbia, 2017 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF MASTER OF LAWS in The Faculty of Graduate and Postdoctoral Studies   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2019 © Cody Rei-Anderson, 2019   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the thesis entitled:  What Role for Copyright in Podcasting?: A Study of Crowdfunding and Advertising Models in  an Emerging Medium             Submitted by  Cody Rei-Anderson    in partial fulfillment of the requirements for the degree of   Master of Laws    in  Law       Examining Committee: Graham J. Reynolds, Law           Supervisor Jon Festinger, Law            Supervisory Committee Member   iii  Abstract Podcasts are a relatively new form of media which have grown immensely in profile and popularity in recent years. From the standpoint of copyright scholarship, however, one characteristic of podcasting stands out: despite being protected by copyright, podcast episodes are largely released for free. On one influential theoretical account, copyright’s purpose is to provide creators and publishers with a financial incentive to produce and distribute works. In particular, the exclusive right to reproduction allows the copyright owner to sell copies of a work without being undercut by competitors who could sell the work at a lower cost. The free distribution of podcasts may present a challenge to this “copyright-as-incentive” model and further raises the question of how podcast creators make a financial return in the absence of selling copies. With this in mind, this thesis will consider the following two-part research question: How do podcast creators make a financial return on their work without excluding non-paying users, and what role if any does copyright play in realizing these financial returns? This thesis will establish a theoretical framework based on Nicolas Suzor’s concept of abundance models. Next, a legal analysis of Canadian copyright law will show that podcasts are in fact protected by copyright. With the relevant copyright law and theory established, this thesis will use a study based in content analysis methods to generate data on how podcasts make money. This study will focus on non-exclusionary alternatives to funding creative work such as advertising and crowdfunding, as well as evidence of the use of copyright.  The data gathered for this study confirm that all of the surveyed series had free episodes available. Despite existing copyright law protecting podcasts, the study showed little evidence of iv  the use of copyright by podcast creators. This study does not indicate why creators are choosing to make their work available for free. But if abundance models based around crowdfunding and advertising are more effective for podcasts than selling copies, this complicates the utilitarian theory proposition that “copyright incentivizes creativity” and suggests that copyright is insufficient as the only or primary creative policy lever.      v  Lay Summary Podcasts are a relatively new form of media which have recently grown in profile and popularity. However, one aspect of podcasting stands out: despite being protected by copyright, podcast episodes are largely released for free. This presents a potential challenge to one influential justification of copyright: that copyright gives creators and publishers a financial incentive to make and distribute works by allowing them to exclude non-paying users. Free distribution of podcasts also raises the question of how podcast creators make money without selling their work. With this in mind, this thesis will consider the following two-part research question: How do podcast creators make a financial return on their work using free distribution, and what role if any does copyright play in realizing these returns? This thesis will answer this question by determining the extent to which copyright protects podcasts, and by studying how podcasts make money through crowdfunding and advertising.    vi  Preface This thesis is the original, unpublished, independent work of the author, Cody Rei-Anderson.   vii  Table of Contents Abstract ......................................................................................................................................... iii Lay Summary ................................................................................................................................ v Preface ........................................................................................................................................... vi Table of Contents ........................................................................................................................ vii List of Tables ................................................................................................................................. x List of Figures ............................................................................................................................... xi Acknowledgements ..................................................................................................................... xii Dedication ................................................................................................................................... xiii 1. Introduction ........................................................................................................................... 1 1.1. Podcasting Terminology and Definitions ....................................................................... 3 1.2. Review of Relevant Scholarly Literature on Podcasting ................................................ 5 2. Copyright Theory ............................................................................................................... 10 2.1. Introduction ................................................................................................................... 10 2.2. The Utilitarian Account of Copyright ........................................................................... 12 2.3. Nicolas Suzor on Abundance and Scarcity Models ...................................................... 15 2.4. Integrating Theoretical Approaches: Incentives and Abundance ................................. 18 2.4.1. Copyright as the Appropriate Economic Incentive ................................................. 19 2.4.2. How is this Project Responding to Utilitarian Theory? .......................................... 21 2.4.3. Applying the Suzor Abundance Framework........................................................... 22 2.5. Conclusion .................................................................................................................... 22 3. Copyright and Podcasting .................................................................................................. 24 3.1. Introduction ................................................................................................................... 24 3.2. Subsistence of Copyright in Podcasts as “Works” under Canadian Copyright Law .... 25 3.2.1. Definition of a “Work”: What Kind of a Work is a Podcast? ................................. 26 3.2.1.1. Unlikely Works Categories for Podcasts: Artistic and Musical........................ 27 3.2.1.2. Video Podcast: Dramatic Work ........................................................................ 28 3.2.1.3. Scripted Podcast: Literary Work ....................................................................... 28 3.2.1.4. Podcast Drama: Dramatic Work ....................................................................... 30 3.2.2. Originality ............................................................................................................... 31 3.2.3. Fixation ................................................................................................................... 32 3.2.3.1. Fixation of Unscripted Podcasts........................................................................ 32 3.2.3.2. Fixation of Live-streamed Podcasts .................................................................. 33 viii  3.3. Creators’ Rights in Podcasts ......................................................................................... 34 3.3.1. Economic Rights ..................................................................................................... 35 3.3.2. Moral Rights ........................................................................................................... 36 3.4. Users’ Rights in Podcasts .............................................................................................. 37 3.5. Neighbouring Rights ..................................................................................................... 38 3.5.1. Performers’ Rights .................................................................................................. 39 3.5.1.1. Podcast Guests as Performers ........................................................................... 40 3.5.1.2. When do Performers’ Rights Arise? ................................................................. 40 3.5.1.3. What do Performers’ Rights Protect? ............................................................... 41 3.5.2. Sound Recording Rights ......................................................................................... 41 3.6. Conclusion .................................................................................................................... 43 4. Methodology ........................................................................................................................ 44 4.1. Introduction ................................................................................................................... 44 4.2. Content Analysis ........................................................................................................... 46 4.2.1. Methodological Concerns in Working with Digital Media .................................... 47 4.3. What is Being Studied? ................................................................................................. 49 4.3.1. Defining the Population: Podcast Networks and Crowdfunded Podcasts .............. 49 4.3.1.1. Defining “Podcast Network” ............................................................................ 51 4.3.2. Defining the Message Set ....................................................................................... 53 4.3.3. Why Not Study a Sample of All Podcasts? ............................................................ 55 4.4. Study Implementation ................................................................................................... 57 4.4.1. Search Process and Selection Method .................................................................... 57 4.4.2. Development of Codes ............................................................................................ 58 4.4.3. Coding Process........................................................................................................ 59 4.4.4. Testing Reliability ................................................................................................... 60 5. Results .................................................................................................................................. 62 5.1. Introduction ................................................................................................................... 62 5.2. Crowdfunded (Patreon) Podcasts Group ...................................................................... 63 5.2.1. References to Patreon or Other Donation Pages ..................................................... 65 5.2.2. References to Additional Content Available to Donors ......................................... 66 5.2.3. Discussion of Non-Podcasting (Secondary) Activities ........................................... 67 5.2.4. Advertisements ....................................................................................................... 68 5.2.5. Intellectual Property ................................................................................................ 70 5.3. Podcast Networks Group .............................................................................................. 70 ix  5.3.1. References to Donation Pages ................................................................................ 74 5.3.2. Discussion of Non-Podcasting (Secondary) Activities ........................................... 74 5.3.3. Advertising .............................................................................................................. 75 5.3.4. Podcast Network Promotion and Cross-Promotion ................................................ 76 5.3.5. Intellectual Property ................................................................................................ 78 5.4. Podcast Series Copyright Notices and Licenses ........................................................... 78 6. Discussion ............................................................................................................................ 80 6.1. Introduction ................................................................................................................... 80 6.2. Issues ............................................................................................................................. 81 6.2.1. Issues with Detecting the Presence of Dynamic Ads.............................................. 81 6.2.2. No Discussion of Patreon Payment Issues .............................................................. 83 6.2.3. Series in the Podcast Network Set Inactive during Relevant Timeframe ............... 83 6.2.4. Overlap between Data Sets ..................................................................................... 85 6.2.5. Inclusion of Podcast Series associated with Other Media ...................................... 86 6.3. Reliability ...................................................................................................................... 88 6.3.1 Measures of Reliability ........................................................................................... 88 6.3.2. Calculating Reliability for This Study .................................................................... 89 6.3.3. Calculating Reliability for Study Variables ............................................................ 91 6.4. Analysis of Findings on Fundraising in Terms of Abundance and Scarcity Models ... 93 6.4.1. Crowdfunding ......................................................................................................... 93 6.4.2. Advertising .............................................................................................................. 96 6.4.3. Secondary Activities ............................................................................................... 96 7. Conclusion ........................................................................................................................... 98 Bibliography .............................................................................................................................. 102 Appendices ................................................................................................................................. 107 Appendix A: Example Code Sheet ......................................................................................... 107 Appendix B: Episode IDs, Series, Sources, Dates, and Titles ................................................ 108    x  List of Tables Table 1. Sampled Patreon Podcast Series with Exclusions .......................................................... 64 Table 2. References to Patreon or Other Donation Pages ............................................................. 65 Table 3. References to Additional Content for Supporters ........................................................... 67 Table 4. Discussion of Non-Podcasting Activities ....................................................................... 68 Table 5. Advertisements for Products and Services not Affiliated with a Podcast (Patreon Set) 69 Table 6. List of Podcast Providers Derived from iTunes Front Page ........................................... 70 Table 7. Sampled Podcast Network Series ................................................................................... 73 Table 8. References to Donation or Membership Pages in Podcast Network Series ................... 74 Table 9. Discussion of Non-Podcasting Activities in Podcast Network Series ............................ 75 Table 10. Advertising in Podcast Network Series ........................................................................ 76 Table 11. Podcast Network Promotion and Cross-Promotion ...................................................... 77 Table 12. Copyright Notices and Licenses on iTunes Store Pages............................................... 78 Table 13. Recent Updates of Inactive Podcast Network Series (as of January 14, 2019) ............ 84 Table 14. Percent Agreement and Cohen’s kappa for Audio Variables ....................................... 91    xi  List of Figures Figure 1. Formula for Percent Agreement .................................................................................... 90 Figure 2. Percent Agreement: Coded Audio ................................................................................. 90 Figure 3. Percent Agreement: Coded Metadata ............................................................................ 90 Figure 4. Formula for Cohen’s kappa ........................................................................................... 91   xii  Acknowledgements The author is deeply grateful for the invaluable feedback and inexhaustible patience of his supervisor, Graham Reynolds and for financial and administrative support from the Peter A. Allard School of Law.   xiii  Dedication  Dedicated to the memory of my grandfather, Dr. Terence W. Anderson (1927–2018) “What you get out of [life] depends on what you put into it.”     1  1. Introduction Podcasts are a relatively new form of media which have grown immensely in profile and popularity since the term “podcast” was coined in 2004.1 From the standpoint of copyright scholarship, however, one characteristic of podcasting stands out: despite being protected by copyright as discussed in Chapter 3, podcast episodes are, for the most part, released for free.2 In one influential theoretical account of copyright, rights in creative works are provided so that creators and publishers are incentivized to produce works.3 In particular, the exclusive right to reproduction allows creators and publishers to sell copies of a work without being undercut by competitors who could copy the work and sell it at a lower cost.4 The free distribution of podcasts may present a challenge to this “copyright-as-incentive” model. Free distribution further raises the question of how podcast creators make a financial return off of podcasting in the absence of selling copies. With this in mind, this thesis will consider the following two-part research question: How do podcast creators make a financial return on their work using free distribution (i.e. not excluding non-paying users), and what role if any does copyright play in realizing these financial returns?5 Each chapter in this thesis will answer an aspect of this research question. The second chapter of                                                  1 See Ben Hammersley, “Audible Revolution”, The Guardian (12 February 2004), online: <www.theguardian.com/media/2004/feb/12/broadcasting.digitalmedia>. The runaway success of Serial, an audio podcast first released in 2014, is credited by some for the medium’s recent surge in popularity: Miranda Katz, “Podcast Listeners Really Are the Holy Grail Advertisers Hoped They’d Be” Wired (29 January 2018, online: <https://www.wired.com/story/apple-podcast-analytics-first-month/>; Richard Berry, “A Golden Age of Podcasting? Evaluating Serial in the Context of Podcast Histories” (2015) 22:2 J Radio & Audio Media 170 at 171 (describing reactions to Serial’s success that saw it as “perhaps a sign that we had reached the tipping point for a medium in maturity”). 2 See Chapter 3, below. 3 See Steven Shavell, Foundations of Economic Analysis of Law (Cambridge, Mass: Belknap Press, 2004) at 137–74. 4 Ibid. 5 The assumption that podcast creators do in fact use free distribution will be substantiated through the study described in Chapter 4. 2  this thesis, “Copyright Theory”, will address the relationship between copyright and the exclusion of non-paying users as it is understood through utilitarian theory. This will introduce copyright’s theorized function as an economic incentive through which creators and publishers can make a financial return from producing and disseminating creative work. The theory chapter will go on to introduce Nicolas Suzor’s concept of abundance models. These models describe non-exclusionary alternatives to funding creative work such as advertising and crowdfunding. The third chapter, “Copyright & Podcasting”, will describe the legal reality of copyright in podcasts with respect to Canadian copyright law. This will serve to answer one aspect of the “copyright’s role” part of the research question, as well as providing necessary legal context for a discussion of possible legal reforms later in the thesis.  The fourth chapter in this thesis, “Methodology”, will describe how this thesis will use content analysis methods on podcast audio and metadata to generate data relevant to the research question. The methodology will focus on two groups of podcasts which will serve as populations for the study: podcasts crowdfunded through the Patreon platform and podcasts affiliated with podcast networks featured on the iTunes front page. The fifth chapter, “Results”, will present the findings of this investigation with respect to two surveyed podcast groups. The data gathered will include prevalence of advertising, use of “bonus content” such as additional podcast episodes for crowdfunded podcasts, and discussion of intellectual property issues and inclusion of licenses. The sixth chapter, “Discussion”, will elaborate on these results and discuss their theoretical and legal implications (drawing on all of the above chapters). The seventh and final chapter will conclude this discussion and indicate where future research in the context of podcasting may be valuable to further explore the role of copyright in this area. 3  However, before addressing the research question it is important to define and characterize podcasting. The first section below will provide a short list of relevant definitions. This will include some discussion of relevant technical aspects of the medium. The second and final section in this chapter will review scholarly literature on podcasting relevant to this thesis. Relevant research literature relating to copyright theory and content analysis methodology will be reviewed in Chapters 2 and 4, respectively. 1.1. Podcasting Terminology and Definitions This section will provide a list of terms and definitions that are relevant to the discussion of podcasting in the remainder of this thesis.  podcast: In content analysis studies, defining a medium is itself a difficult task.6 Defining “podcast” is no exception.7 Podcasts can consist of audio or audio-video,8 can be delivered live or downloaded or streamed on-demand,9 and can take many different formats including lectures, conversations, and improvisational comedy.10 Podcast series are often distributed using a unique RSS feed (see below) which is updated with new episodes as they are released.11 Users may                                                  6 See Kimberley A. Neuendorf, The Content Analysis Guidebook, 2nd ed (Thousand Oaks, CA: Sage Publications, 2017) at 111–13. 7 See Andrew J. Bottomley, “Podcasting: A Decade in the Life of a ‘New’ Audio Medium” (2015) 22:2 J Radio & Audio Media 164 at 166 [Bottomley, “Podcasting: A Decade in the Life”]. 8 Research conducted for this study found examples of podcasts released in audio and audio-video formats. See Chapter 5: Results, below. 9 Some podcasts are recorded live at physical venues and some podcasts are recorded and streamed live over the internet. In the sample set for this study, the series All About Android and Bizarre States appear to regularly stream live episode recordings, while several podcast episodes from various series consisted of recorded live events: episode IDs DOLL34, WWC41, WWC47, BIZR75, and WWC95. See Appendix B, below. 10 See Bottomley, “Podcasting: A Decade in the Life”, supra note 7 at 166–67; Kris M. Markman, “Everything Old is New Again: Podcasting as Radio’s Revival” (2015) 22:2 J Radio & Audio Media 240 at 241–42. 11 The research conducted for this project found RSS feeds for every podcast series in the sample. iTunes/Apple Podcasts, a major podcast platform, requires creators provide an RSS feed in order to list a podcast: Apple, “About Submitting a Podcast”, online: <help.apple.com/itc/podcasts_connect/#/itc4f0f5ac7d>. 4  access a podcast series directly through its RSS feed, through a series’ website, or through a podcast listening app. episode: The “episode” is the basic unit of the podcast series, consisting of a single (audio or video) file distributed using an RSS feed.  series: This term denotes a set of episodes released under one show title and (usually) one RSS feed. Podcast series have no effective upper or lower limits to the number of episodes they can include, with some having reached over 1,000 episodes. Many are open-ended, while others have defined end-points and may be organized into seasons. season: Some podcast series release a number of episodes as a “season” distinguished by a common theme or narrative.12 Podcasters may refer to these as “seasons” or “series”.13 A podcaster may choose to do this because starting a new series under a new title usually entails  starting a new RSS “feed”, which existing subscribers will have to newly subscribe to (potentially resulting in a loss of subscribers).14 subscriber: A listener who uses a podcast listening app or RSS feed reader to be notified of new episodes in a podcast series, and (with many podcast apps) have them downloaded automatically to a device.                                                  12 See e.g., Mike Duncan, Revolutions, online: <www.revolutionspodcast.com> (seasons of 20-100 episodes on different political revolutions); Karina Longworth, You Must Remember This, online: <www.youmustrememberthispodcast.com> (seasons of 10-30 episodes on different themes and stories in 20th century Hollywood history).  13 For the sake of clarity, this project will exclusively refer to these as “seasons”. 14 However, some podcasters do start new series for various reasons, usually where there is a significant departure from the first show’s theme, format, or narrative—as an example, before starting the Revolutions podcast, Duncan produced a series on the history of Rome: Mike Duncan, The History of Rome, online: <www.thehistoryofrome.typepad.com/>. 5  RSS: An open standard for syndicating content over the internet. The standard is “open” in the sense that it can be used by anyone: there are many “aggregator” applications available that can import any RSS feed which adheres to the standard. This can be contrasted with closed platforms such as Twitter or Facebook, which are under the control of one entity and allow third-party applications through limited APIs (application programming interfaces) if at all. podcast listening app or podcatcher: An aggregator that specifically works with podcasting RSS feeds. These often provide the option to download podcasts automatically as they are updated. iTunes and Apple Podcasts are very popular podcast listening apps; other apps include Stitcher and Spotify.  1.2. Review of Relevant Scholarly Literature on Podcasting This section will provide a review of relevant scholarship on podcasting to provide context for further discussion of the medium later in this thesis. First, I will briefly introduce the development of podcasting. This will lead into a discussion of early legal scholarship on podcasting which focused on the use of copyrighted music in podcasts, and which may have foreseen a different future for the medium than the talk format which has come to dominate it. Next, I will review more recent scholarship which addresses law-related aspects in podcasting, particularly around the “true crime” genre. Finally, I will briefly address one aspect of podcasting which multiple scholars reference: the sense of “intimacy” that can be created by the medium. 6  In podcasting’s early history, the medium’s name itself was contentious, with Apple attempting to assert its iPod trademark against podcasting app creators.15 At that time, prior to the rise of smartphones, the iPod and similar portable media players were one of the main ways for listeners to listen to podcasts, which gave the medium its name.16 But podcasting outlasted the iPod: the years after the iPhone’s release saw the decline of portable media players in favour of smartphones, leading to the eventual discontinuation of most of Apple’s iPod line.17 Smartphones, with data connections and “podcatching” apps, greatly reduced the friction for users to subscribe and listen to podcasts.18 It was in this context that podcasting saw its recent boom in popularity, often connected to the 2015 release and runaway success of Serial.19 This historical context is valuable to lay the groundwork for a review of podcasting literature. It emphasizes that podcasting took some time to become what it is today as a medium, even if the predominant format (digital audio released serially) has stayed more or less constant.20 Some early legal scholarship on podcasting focused on legal considerations around the inclusion of copyrighted music in podcasts and the applicability of US statutory licensing schemes for recorded music to podcasting.21 Similarly, existing legal guides for podcast creators devote                                                  15 See Jonathan Sterne, Jeremy Morris, Michael Brendan Baker & Ariana Moscote Freire, “The Politics of Podcasting” (2008) 13 Fibreculture J, online: <http://thirteen.fibreculturejournal.org/fcj-087-the-politics-of-podcasting/>.  16 Ibid. 17 David Pierce, “Goodbye iPod, and Thanks for All the Tunes”, Wired (27 July 2017), online: <https://www.wired.com/story/goodbye-ipod-and-thanks-for-all-the-tunes/>. 18 Richard Berry, “A Golden Age of Podcasting? Evaluating Serial in the Context of Podcast Histories” (2015) 22:2 J Radio & Audio Media 170 at 173. 19 See ibid; Bottomley, “Podcasting: A Decade in the Life”, supra note 7 at 165. 20 See Markman, supra note 10. 21 See Edward L. Carter & Scott Lunt, “Podcasting and Copyright: The Impact of Regulation on New Communication Technologies” (2006) Santa Clara Computer & High Tech LJ 187; Jared Barrett, “Podcasting Pop Songs?: Licensing Concerns with Podcasts that Contain Mainstream Music” (2006) 3 Shidler JL Com & Tech 3; Michael N. Lang, “The Regulation of Shrink-Wrapped Radio: Implications of Copyright on Podcasting” (2006) 14 CommLaw Conspectus 463.  7  significant attention to issues around the inclusion of copyrighted material in podcasts.22 In fact, podcasting has remained a primarily talk-based medium, which may be in part attributable to the legal and financial difficulties associated with using copyrighted music.23 In this focus on talk, podcasting diverged from radio, a medium to which many scholars trace podcasting’s roots.24 Some more recent scholarship on podcasting has focused on podcasts’ content and impact in law-related areas. Brandon Golob, in a 2017 law review article, considers how “new media” might impact legal consciousness around wrongful convictions.25 In particular, his article focuses on the podcast Serial and the Netflix television series Making a Murderer. Golob expresses concern that these series might lead audiences to mistaken conclusions about the criminal justice system, including making out the cases as merely “isolated instances of injustice”.26 Taking a broader scope, a 2018 media studies article from Kelli S. Boling and Kevin Hull explores the motivations and demographics of the audiences of “true crime” podcasts generally.27 Their findings indicated a preponderance of female listeners in the true crime audience (73%) despite other cited research finding that a majority of podcast listeners are male (56%), as well as a majority of male users on the site on which they circulated their survey (Reddit, 67%).28 The Boling and Hull study found that “entertainment, convenience, and boredom” were the most                                                  22 See e.g., Colette Vogele, Esq & Mia Garlick, “Podcasting Legal Guide: Rules for the Revolution”, online: <wiki.creativecommons.org/wiki/Podcasting_Legal_Guide>; Kathleen Simmons & Andy Kaplan-Myrth, “Podcasting Legal Guide for Canada: Northern Rules for the Revolution”, online: <cippic.ca/sites/default/files/Podcasting-LegalGuide-Canada.pdf>. 23 See Christopher Cwynar, “More than a ‘VCR for Radio’: The CBC, the Radio 3 Podcast, and the Uses of an Emerging Medium” (2015) 22:2 J Radio & Audio Media 190 (describing CBC’s Radio 3 podcast as a rare, music-based exception). 24 See Berry, supra note 18 at 171; Sterne et al., supra note 15.  25 “Un-Making a Murderer: New Media’s Impact on (Potential) Wrongful Conviction Cases” (2017) 54 Cal W L Rev 137. 26 Ibid at 149. 27 “Undisclosed Information—Serial is My Favorite Murder: Examining Motivations in the True Crime Podcast Audience” (2018) 25:1 J Radio & Audio Media 92. 28 Ibid at 99, 103.  8  prominent motivators for true crime podcast listeners.29 Outside of this popular genre, other podcasts deal with law as well. An episode of a podcast series concerned with the US Supreme Court, More Perfect, provided the basis for Nancy S. Marder’s discussion of peremptory challenges.30 For this article, the podcast episode in question contributed to an understanding of peremptory challenges by engaging persons involved in the Batson v Kentucky case and including input from practitioners and academics on whether peremptory challenges should be abolished.31 One theme that comes out of some podcasting scholarship is the intimacy and immersion of podcasts.32 Podcasts are often experienced by a listener alone, through headphones.33 Podcasting is also characterized as a space in which marginalized or less well-established voices can be heard, literally.34 These themes come through in Andrew J. Salvini’s discussion of history podcasts as folk or “DIY” histories, Vincent M. Meserko’s discussion of fan identification with the Mental Illness Happy Hour series, and Sarah Florini’s article on series by Black podcasters which “reproduce a sense of being in Black social spaces” outside of the white gaze.35 The latter piece makes a particularly interesting point that the fact that spoken audio is not as easily searched as text effectively makes podcasts more “resistant to easy intrusion” by outsiders (that is, those outside of a podcast’s listening community) than some social media.36 The intimacy of                                                  29 Ibid at 92. 30 “Batson v. Kentucky: Reflections Inspired by a Podcast” (2016) 105 Ky LJ 621. 31 Ibid at 624. 32 See Vincent M. Meserko, “Going Mental: Podcasting, Authenticity, and Artist-Fan Identification on Paul Gilmartin’s Mental Illness Happy Hour” (2014) 58:3 J Broadcast & Electronic Media 456; Andrew J. Bottomley, “Podcasting, Welcome to Night Vale, and the Revival of Radio Drama” (2015) 22:2 J Radio & Audio Media 179 at 186 [Bottomley, “Revival of Radio Drama”]; Sarah Florini, “The Podcast ‘Chitlin’ Circuit’: Black Podcasters, Alternative Media, and Audio Enclaves” (2015) 22:2 J Radio & Audio Media 209 at 210. 33 See Florini, supra note 32 at 215. 34 See Markman, supra note 10 at 243; Florini, supra note 32. 35 Florini, supra note 32 at 210. 36 Ibid at 214.  9  podcasting and its potential to elevate marginalized voices may in turn be factors in the development of strong communities or fandoms around some podcasts.37 This project will look at crowdfunding as a means for podcast creators to fund their work. Since crowdfunding is generally voluntary, the feelings of intimacy and community around podcasts may underpin the success of these models. Although the methodology chosen for this study does not include the perspectives of listeners, future research might follow up on this aspect of podcasting.                                                    37 See Markman, supra note 10 at 243. 10  2. Copyright Theory 2.1. Introduction One major theoretical account of copyright holds that copyright is necessary to incentivize creative production.38 In the utilitarian account of copyright, an important function of copyright is to allow creators to sell their works without competition from unauthorized copiers. As Chapter 3 will establish, there is copyright in podcasts.39 However, many podcasts are released for free. Relatively few creators sell individual episodes, and many instead rely on crowdfunding or advertising to support their work.40 Through a legal analysis of copyright in podcasts and a data gathering study, this thesis will argue that Nicolas Suzor’s concept of “abundance models” provide a better basis to understand copyright’s role in podcasting than the utilitarian account of copyright.41  This chapter will first describe utilitarian copyright theory. Utilitarian copyright theory holds that copyright plays a necessary role in incentivizing creative production. It does so by providing copyright owners with a bundle of exclusive rights in works. One of the most important of these rights is the exclusive right to reproduction. The exclusive right to copy works allows copyright owners to sell individual copies without competition from unauthorized copiers. The sale of individual copies of works therefore relies on the exclusive right to copy. While not every successful medium has relied on selling copies, this model has remained important for traditional                                                  38 Shavell, supra note 3 at 137–74. 39 See chapter 3, below. 40 The study results support this, but note that there are podcast series outside the sample which do sell individual episodes. Further, some podcasts in the sample offer access to archives and bonus podcast episodes only to paying users. See section 6.4.1, below. 41 See Nicolas Suzor, “Access, Progress, and Fairness: Rethinking Exclusivity in Copyright” (2013) 15 Vand J Ent & Tech L 297.  11  media such as music, movies, and television in the digital age.42 In contrast, though the sale of individual podcast episodes appears to be fairly rare, podcasting is booming.43 The utilitarian theory account of copyright does not immediately provide an explanation for this conundrum. Instead, I will suggest in this chapter that podcasts should be understood as operating on Nicolas Suzor’s concept of “abundance models”.44 These models use the effortless and nearly-free copying enabled by the internet and digital technologies to achieve wide distribution without relying on copyright to exclude non-paying audiences. This raises the question of how podcast creators make a financial return on their work, which will be further addressed through discussion of the content analysis study in this project. I will argue that the crowdfunding and advertising used by podcast creators can be understood within the framework of “abundance models” theorized by Nicolas Suzor.45 Under these models, non-paying users are not excluded from enjoying works and creators or publishers rely on means of raising revenue which do not require exclusion of non-paying users. Similarly, many podcast creators raise funding through crowdfunding or advertising.46 That podcast creators do not choose to rely on copyright to sell copies of their work may show that this method of generating income is less preferable to podcast creators than advertising or crowdfunding.47 The scope of the study in this thesis will leave some questions unanswered, as it does not have access to creators’ motivations. However, Suzor’s abundance model framework will help in drawing some                                                  42 For example, broadcast radio and broadcast television are distributed for free as radio signals. 43 By “sale” I mean charging for access to individual episodes of a podcast. 44 See Suzor, supra note 41 at 314–15. 45 Ibid. 46 See Katz, supra note 1; Graphtreon, “Patreon Podcasts”, online: <graphtreon.com/patreon-creators/podcasts> (showing a list of podcast creators using the crowdfunding platform Patreon). 47 A subsequent chapter will explore possible reasons for this, informed by the results from a content analysis study. See Chapter 6: Discussion, below.  12  tentative conclusions from the study results about the advantages and disadvantages of distributing works for free.48  Section 2.2 of this chapter will provide a sketch of the utilitarian account of copyright and its influence on contemporary copyright law. Section 2.3 will introduce Nicolas Suzor’s concept of “abundance” models, which take advantage of the near-free copying and cheap distribution allowed by digital technologies. I will argue that abundance models describe how many podcasts are distributed. In section 2.4 I will elaborate my hypothesis that podcasts should be understood as operating on Suzor’s concept of “abundance models”. I will also address how this hypothesis responds to the incentives-based understanding of copyright put forward by utilitarian theory.  2.2. The Utilitarian Account of Copyright The utilitarian justification of copyright holds that the purpose of copyright is to provide an economic incentive to prospective authors and publishers of creative works. In this account, copyright incentivizes production of creative works by providing exclusive rights in creative works. The goal is a system which is “welfare maximizing”: that is, one in which the greatest amount of aggregate welfare is produced, considering both the welfare consumers enjoy from using creative works and the welfare creators enjoy from the profits of selling their works.49 The rights provided under copyright include, critically, reproduction of the work. (Modern copyright also protects other uses of a work, such as adaptation or translation of the work.) The exclusive reproduction right provides an incentive to produce original copyrighted works because the rights owner can sell the work at a price above the marginal cost of production.50 In the absence                                                  48 See Suzor, supra note 41 at 309. 49 Shavell, supra note 3 at 2.  50 In addition, the exclusive rights provided under copyright can be licensed or assigned to provide revenue to the creator.  13  of an exclusive reproduction right, a competing party could sell the same work at a lower price because they would not have to recoup the costs of producing the original work.51  However, utilitarian copyright theory also recognizes that strong copyright protection has costs. High transaction costs make certain kinds of uses more difficult to contract for.52 Transaction costs can therefore be used to justify limits on copyright term length, and the need for user rights such as fair dealing.53 Some potential users of copyright works are excluded from works because of their inability or unwillingness to pay, resulting in “deadweight loss”.54 This loss can be greater where copyright encompasses more uses of the work.55 The extensive scholarship on these (and other) issues demonstrates that there is ample room within the utilitarian justification to argue for and against stronger copyright protection. Utilitarian theory is important to elaborate in this thesis because the idea that copyright law’s purpose is to act as an incentive for creative production is central to both the theory and this thesis. Further, this idea and utilitarian theory generally have been very influential in law, legal scholarship, and policymaking. Shyamkrishna Balganesh, a US legal scholar, in a 2009 article called the “economic theory of creator incentives” the principal justification for contemporary copyright law and asserted that “most agree [that] copyright law’s primary purpose lies in providing individuals with an incentive to generate creative expression”.56 The Hargreaves                                                  51 Shavell, supra note 3 at 138. 52 Some examples of these types of uses include parody and criticism. 53 Shavell, supra note 3 at 146–47. 54 Ibid.  55 Ian Hargreaves, Digital Opportunity: A Review of Intellectual Property and Growth (May 2011), online: <https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/32563/ipreview-finalreport.pdf> at 11–12; Amy Kapczynski, “Intellectual Property’s Leviathan” (2014) L & Contemp Probs 131 at 133; Oren Bracha & Talha Syed, “Beyond Efficiency: Consequence-Sensitive Theories of Copyright” (2014) 29 Berkeley Tech LJ 229 at 240. 56 “Foreseeability and Copyright Incentives” (2009) 122:6 Harv L Rev 1569 at 1571–72.  14  Report, an independent report commissioned by the UK government in 2010 and released in 2011, frames intellectual property rights generally as intended to “fulfil the economic incentive role described in the Statute of Anne, Britain’s first copyright legislation”.57  The “copyright as incentive” idea is similarly prevalent in Canadian law, policy, and scholarship. In a 2005 article, Canadian scholar Daniel Gervais summarized the central question of a trilogy of then-recent Supreme Court of Canada decisions on copyright using language that directly evokes utilitarian theory: “what is the impact of copyright on the level of general welfare?”58 More recently, the 2013 Supreme Court of Canada decision Cinar Corp v Robinson states bluntly that the Copyright Act “seeks to ensure that an author will reap the benefits of his efforts, in order to incentivize the creation of new works.”59 A 2017 letter from federal government ministers framing a statutory review of the Copyright Act does not explicitly mention copyright as an incentive, but does emphasize the economic goal of copyright legislation to “foster a marketplace” in copyrighted content.60 While it may be argued that Canadian copyright law and policy is not beholden to the utilitarian justification to the exclusion of other justifications, that theory’s influence is clear and unmistakable.61                                                  57 Hargreaves, supra note 56 at 16. 58 Daniel J. Gervais, “The Purpose of Copyright Law in Canada” (2005) 2 U Ottawa L & Tech J 315 at 318. 59 Cinar Corporation v Robinson, 2013 SCC 73 at para 23 [Cinar]. This is not the only purpose of the Copyright Act which the Court has cited: see Society of Composers, Authors and Music Publishers v Bell Canada, 2012 SCC 36 at para 21 (stating that “dissemination of works is also one of the [Copyright] Act’s purposes”). 60 Letter from Navdeep Bains and Mélanie Joly to Dan Ruimy, Chair of the Standing Committee on Industry, Science and Technology, online: <http://www.ourcommons.ca/content/Committee/421/INDU/WebDoc/WD9341854/421_INDU_reldoc_PDF/INDU_DeptIndustryDeptCanadianHeritage_CopyrightAct-e.pdf>. 61 For example, the quote from Cinar above alludes to authors’ rights arising from their labour, though the final clause seems to subordinate this to the instrumental purpose of copyright to incentivize creation of new works: see supra note 59. Other copyright theory approaches include author’s rights theories, analysis of copyright grounded in human rights, and critiques of copyright based in distributive justice. While these approaches have distinct characteristics, particularly in their formulation of copyright’s purpose, they are not always at odds. While this  15  2.3. Nicolas Suzor on Abundance and Scarcity Models Nicolas Suzor’s work departs from familiar copyright dichotomies of author’s rights as opposed to user rights and strong copyright as an incentive to produce more creative work as opposed to a wider public domain.62 His analysis focuses on the use of copyright to exclude as forming the basis of a traditional “scarcity” model of providing creators with returns on their work. This model relies on excluding non-paying users—the same approach which is an important aspect of the utilitarian understanding of copyright. He contrasts scarcity model of creative production with “abundance” models which fund projects by means other than selling copies of or licensing access to works. These models are made possible by the effortless copying and cheap distribution which computers and the Internet enable.63  Suzor cites advertising and crowdfunding as two ways to fund the production of creative works under abundance models.64 These models “suggest methods of funding cultural production” that do not rely on the scarcity of copyrighted goods. They take advantage of the near-zero marginal cost of copying digital works to envision broad access to works not constrained by users’ ability to pay.65 Instead of access meaning “access to a smoothly functioning commodity market” for works, abundance models see the potential of digital distribution that “all connected individuals could have immediate access to almost perfect reproductions of the entire wealth of recorded cultural expression”.66 The challenge faced by abundance models is to find ways to fund creative                                                  project will focus on one aspect of utilitarian theory (that copyright is the best way to incentivize creative work), it is worthwhile to introduce as context other approaches to copyright theory which may provide their own insights. 62 Suzor, supra note 41 at 301–10. 63 Ibid at 299–300. 64 Ibid at 298–99. 65 Ibid at 299–300. 66 Ibid at 313, 314.  16  production that are fair to audiences and creators, and which resolve coordination problems inherent in the production of capital- and labour-intensive creative works.67 Copyright directly supports and enables scarcity models, but its relationship with abundance models is less clear. Simply put, the exclusive right to copy works granted by copyright is the legal means through which scarcity models are enforced. The copying right and the scarcity model are conceptually linked: scarcity models require copyright owners to be able to stop unauthorized copiers and thereby create scarcity in the market, and the exclusive right to copy exists so that copyright owners can use scarcity models. This is essentially part of the utilitarian account of copyright: that copyright provides an incentive by allowing copyright owners to exclude non-paying users from works. However, the exclusive right to copy is not of central importance in abundance models in the same way it is under scarcity models. The control that copyright grants creators over their creations may still be valuable to creators even if they are giving their works away for free. Copyright may help to support abundance models insofar as control remains important: for example, if an ad-supported podcast could be freely re-cut and re-released without ads, that would certainly have the potential to undermine that podcast’s business model. However, the methods creators might use to make financial returns from their work under an abundance model—such as crowdfunding or advertising—do not have analogous legal rights in the same way selling copies under a scarcity model does.68 Copyright’s relationship with scarcity models is therefore much stronger than its relationship to abundance models.                                                  67 See ibid at 321; Julie E. Cohen, “Copyright as Property in the Post-Industrial Economy: A Research Agenda” [2011] Wis L Rev 141. 68 This is not to suggest that legal rights are necessarily the most effective way law could support abundance models.   17  For users, the potential benefits of abundance models over scarcity models are apparent. As Suzor notes, “the great social cost of excluding users who cannot afford to pay the monopoly price of copyright expression, which is costless to distribute, is deeply troubling.”69 This concern mirrors some criticisms of copyright grounded in distributive justice that “[t]he advantages of copyright are reaped primarily by those already privileged: affluent consumers, the most successful creators, and major publishing houses and other copyright holders located in industrialized countries”, while the “burdens of copyright protection, in the form of higher prices, fall hardest on the already disadvantaged.”70 The abundance model framework has the potential to address some distributive issues because under abundance models users are not excluded from works based on their ability to pay. However, whether abundance models can truly break out of the incentives-access dichotomy depends on whether creators can be sufficiently incentivized to do creative work.71 Without an adequate economic incentive for creative work, works would be underproduced (in the language of the utilitarian model). This problem can also be understood in distributive terms: if creators are unable to realize an adequate economic return on their creative work, those without the resources to create for free or at a loss would be excluded from opportunities for expression. The success of abundance models under the distributive or utilitarian accounts of copyright therefore depends on whether creators can generate enough of a return on their work without excluding potential consumers from their works. The appropriate return for creators could be based on factors such as consumption of the work, the time and labour invested in the work, or                                                  69 See Suzor, supra note 41 at 316. 70 Lea Shaver, “Copyright and Inequality” (2014) 92 Wash UL Rev 117 at 141. 71 See Bracha & Syed, supra note 55 at 237–40.   18  the value derived by users from the work.72 Suzor contends that reducing the cut of creative profits enjoyed by intermediaries could make abundance models more attractive for creators.73 However, intermediaries also provide benefits for creators, such as marketing and promotion, which intermediaries may be able to perform more effectively than individual creators. In the absence of intermediaries providing services such as advertising, these costs are borne by the creators themselves. Further, in currently existing examples of crowdfunding (one of the abundance models Suzor identifies), there is at least one major intermediary, the technological platform which facilitates the crowdfunding (e.g., Patreon). Along with introducing costs that come with being an intermediary, these platform holders can create other issues that might negatively affect creators, such as disrupting the inflow of financial support. (The recent payment handling issues experienced by Patreon and creators on that platform is an example of this that will be further discussed in a later chapter.)74 However, these criticisms do not fully offset the benefits abundance models have with respect to access. If podcasts are successful operating on abundance models, that suggests that there may be types of works for which abundance models are a more attractive choice than scarcity models from the perspective of creators. 2.4. Integrating Theoretical Approaches: Incentives and Abundance As demonstrated by data collected as part of this thesis, many podcasts use business models other than the sale of copies or licensing of access to copyrighted works.75 In Suzor’s terms, this suggests that podcasts are based on abundance models, and take advantage of near-free digital                                                  72 See Suzor, supra note 41 at 324. 73 Ibid at 331–32. 74 See section 4.3.1, below. 75 See Chapter 5, below. 19  distribution and effortless copying. The following sections will further address this project’s relationship with utilitarian theory and Suzor’s abundance model framework. 2.4.1. Copyright as the Appropriate Economic Incentive In the utilitarian view, it is necessary to provide an economic incentive for creative production in order to achieve a welfare-maximizing outcome. However, copyright is not the only way to accomplish this objective: other measures could provide an incentive for the kinds of information production that intellectual property laws promote, such as public procurement or prizes.76 The question, for the utilitarian, is whether these measures can promote welfare more efficiently than intellectual property rights.  This project is concerned with a specific context—podcasting—in which copyright’s role as an economic incentive is in question. However, the utilitarian conclusion that private intellectual property rights (such as copyright) are the best way to incentivize information production has also been critiqued in more general terms. Amy Kapczynski argues that the utilitarian justification for private intellectual property rights has as its “linchpin” a neoliberal conception of the state.77 The utilitarian argument that incentives are needed for information production to be produced at a welfare-maximizing level does not on its own point to the necessity of private property rights in information as the instrument to provide that incentive.78 For example, public procurement could provide incentives for information production without the need to propertize information.79 In the neoliberal view, however, the state is expected to “generate the conditions of private market ordering” by creating and enforcing intellectual property rights, but is                                                  76 See Shavell, supra note 3 at 162–66. 77 Kapczynski, supra note 55 at 133–35. 78 Ibid. 79 Ibid at 133.   20  “characterized as incapable of effectively engaging more directly in the organization of information production.”80 This incapability is allegedly justified because “the state is unlikely to have sufficient information to determine what goods ought to be created” and is “imagined to be uniquely vulnerable to capture.”81 However, Kapczynski points out that both of these objections seem to apply to the private intellectual property rights regime as well: current intellectual property regimes in the United States are enormously complex and even advocates of private intellectual property rights recognize the need for “a state that plays an active and interventionist role in the information economy”.82 This project is concerned with the same linchpin of utilitarian theory that Kapcyznski identifies, but from a different angle. Rather than addressing the argument that private property rights in information in the form of copyright are the most efficient way of incentivizing (creative) information production, this project looks at a situation in which copyright is present but may not be performing an incentivizing role. This project addresses an aspect of utilitarian theory: copyright’s function as an economic incentive. This function may be recognized in other approaches to copyright theory, in the same terms as it is discussed in utilitarian theory or otherwise. As such it is not necessary for this project to explicitly work within one approach or another. The empirical portion of this project will include codes that speak to the question of whether copyright acts as an economic incentive in the context of podcasting. Further, this project’s approach shares with utilitarian copyright theory a focus on the role of copyright in society as a way to promote creative production. However, this project will rely on a different                                                  80 Kapczynski, supra note 55 at 134. 81 Ibid. 82 Ibid at 142.  21  framework for understanding and explaining the results of the study described in Chapter 4. The abundance and scarcity models proposed by Nicolas Suzor are particularly valuable to understanding how distributing content for free with the aid of digital technologies can work.83 2.4.2. How is this Project Responding to Utilitarian Theory? As discussed above, the utilitarian approach has been very influential as a justification for copyright. The subject of this project is an area in which the logic of the utilitarian approach—that copyright provides a necessary incentive84—does not seem to hold. The study will attempt to shed further light on what makes the business of podcasts work. However, this project is not trying to adapt or disprove utilitarian theory. It may be that a more complex economic model compatible with utilitarian theory could account for podcasts. It is not the aim of this project to formulate such a model.  Utilitarian theory has been introduced in this chapter because understanding this theory is important to understanding modern copyright. Further, podcasts stand out under the lens of utilitarian theory as something that needs explaining, because the business models which creators use in this medium do not seem to rely on copyright exclusivity to sell copies or license access. However, the critiques of utilitarian theory introduced above have been effective at pointing out some of the shortcomings of utilitarian copyright with respect to distributive justice, human rights, and utilitarian theory’s connections with neoliberal ideology. Instead of trying to understand and account for podcasts in utilitarian terms, this project will adopt a different framework.                                                   83 See Suzor, supra note 44 at 298–99. 84 More precisely, that exercising copyright through the licensing or assignment of exclusive rights or selling copies is necessary in order to incentivize the creation of expression. 22  2.4.3. Applying the Suzor Abundance Framework The advantage of abundance models is access, and free podcasts are accessible to anyone with a computer or mobile device and an internet connection. The Suzor abundance framework provides a good fit for podcasts because it specifically contemplates the conditions under which podcasts are created (digital creation and distribution) and raises relevant questions about how free media can make money. The study described in Chapter 4 will gather data on how two groups of podcasts generate income. This data will be presented in Chapter 5, and discussed further in Chapter 6. The discussion will in part focus on whether the data on the podcast series supports the hypothesis that they are working on abundance models, and highlight where scarcity models may still support creative work in this area.  2.5. Conclusion This chapter has considered utilitarian theory and an analytical framework (Suzor’s) that theorizes different models for incentivizing creativity. This project takes a theoretical approach which draws on the framework of abundance and scarcity models put forward by Suzor to provide helpful concepts and language to frame how podcasts (and the business models on which they operate) are different from traditional media. This is important in part because the promise of abundance models is that technology can be leveraged to produce more social good by lowering barriers to access. Podcasts released for free and supported by other means may be a realization of this vision.  The relationship between copyright and podcasting is of interest in part because many podcasters use a business model which does not rely on copyright’s exclusive right to make copies.85 That                                                  85 See section 6.4, below. 23  is, many creators in this medium do not use copyright to exclude non-paying users from their content. Many podcasters release episodes for free, and rely on advertising or crowdfunding for financial support.86 Whatever role copyright may play in podcasting, it is not generally used to create an artificial scarcity in order to generate profits.87 Rather than trying to sustain scarcity models, copyright policy should help abundance models work (or work better).  Advertising and crowdfunding constitute two categories of “abundance model” cited by Suzor. The two groups of podcasts that were examined in this study broadly line up with these categories: series on podcast networks and series crowdfunded through Patreon. The results of the study as presented in Chapter 5 should therefore be helpful in describing how abundance models work in practice, whether they run into problems, and if so what kinds. However, it is important first to establish that copyright does indeed exist in podcasts, which the next chapter will demonstrate.                                                     86 Ibid. 87 Ibid. 24  3. Copyright and Podcasting 3.1. Introduction This chapter will address the following questions: under current Canadian copyright law, what is the scope of copyright protection for podcasts, and what rights does this protection grant to users and creators? Answering these questions will inform the discussion of the study’s results in Chapter 6 as well as serve as a potentially useful guide for creators themselves. The application of copyright to podcasts ties directly into this thesis’s research question, which asks why many podcast creators do not sell their works. It is necessary to establish that copyright exists in podcasts to exclude the possibility of a gap in the law, and confirm that podcast creators could use copyright exclusivity to exclude non-paying users.  Further, an understanding of the legal rights copyright currently grants to podcast creators and users provides necessary context to a discussion of what copyright could or should do in this area. Whether copyright is an important financial incentive for podcasters at present will be partially addressed through the results and discussion of the study in Chapters 5 and 6. However, this thesis also recognizes that copyright’s relevance to podcast creators could change as the business of podcasting changes.88 In the event that a copyright “scarcity model” becomes more important to the business of podcasting, the scope of protection which copyright law offers to podcasts may similarly become more important.                                                  88 It is possible that in the future podcasting becomes more like other media. For example, they may start operating on paid or “freemium” subscription/library models like Spotify. Notably, Spotify recently acquired two podcast companies, Gimlet (one of the networks in this study) and Anchor: Wendy Lee, “Spotify Acquires New York Podcasting Companies Gimlet and Anchor”, Los Angeles Times (February 6, 2019), online: <www.latimes.com/business/hollywood/la-fi-ct-spotify-acquires-gimlet-20190206-story.html>.  25  Drawing on the Copyright Act (the “Act”) and relevant case law, this chapter will consider whether different types of podcast will qualify for protection as a “work”.89 As I will discuss, depending on the format of a podcast, it may be protected as a dramatic work (cinematographic or otherwise) or a literary work. However, Canadian copyright law may offer less protection to improvised audio podcasts, which seem to fall into neither category. This chapter will also discuss the scope of creators’ and users’ rights as they apply to podcasts, as well as the subsistence of sound recording rights and performers’ rights in podcasts.90 3.2. Subsistence of Copyright in Podcasts as “Works” under Canadian Copyright Law Under the Canadian Copyright Act (“the Act”), the requirements for subsistence of copyright in a work are originality, that the creation falls under one of the definitions of a “work”, and that the creator fulfill a residency requirement.91 A fourth, judicially-required condition is fixation.92 If the author is a citizen or subject of, or a person ordinarily resident in a treaty country at the time of the making of the work, then this fulfills the residence requirement.93 This section will explore potential difficulties with the “work” and fixation requirements as applied to specific types of podcasts. It will also address how the originality requirement has been elaborated in Canadian law.                                                   89 Copyright Act, RSC 1985, c C-42; Gould Estate v Stoddart Publishing (1996), 30 OR (3d) 520 (SC), aff’d (1998), 39 OR (3d) 55 (CA). 90 This chapter will focus on the rights granted through the Copyright Act; however, podcast creators’ rights in their work can also be affected by contract. For example, podcast platforms may have terms of service which creators agree to in order to have their show distributed.  91 Copyright Act, supra note 89, s 5. Note that neighbouring rights protection such as sound recording rights and performer’s rights have different requirements and are within the Act’s definition of “copyright”. See section 4, below. 92 See Gregory Hagen, Graham Reynolds, Cameron Hutchison, Teresa Scassa, David Lametti & Margaret Ann Wilkinson, Canadian Intellectual Property Law: Cases and Materials, 2nd ed (Toronto: Emond Montgomery, 2018) at 18; Gould v Stoddart, supra note 89. 93 See Copyright Act, supra note 89, s 5. “Treaty country” is defined as a “Berne Convention country, UCC country, WCT country or WTO Member”: Copyright Act, s 2 “treaty country”.   26  3.2.1. Definition of a “Work”: What Kind of a Work is a Podcast? The rights granted in section 3 of the Copyright Act apply to “works”. The four main categories of work for the purposes of this legislation are literary, dramatic, musical, and artistic.94 The Act also defines subcategories such as cinematographic and choreographic works (both included under dramatic works).95 “Compilations” are also defined in the Act, consisting of “a work resulting from the selection or arrangement of literary, dramatic, musical or artistic works or of parts thereof”.96 Podcasting as a medium is primarily defined by its means of distribution, rather than form (podcasts can be audio or video) or content (some examples of podcast content include scripted non-fiction, semi-scripted interviews, unscripted conversations, and improvisational comedy). Podcasts may be released as audio, as video with a soundtrack, or as separate audio and video releases.97 They may be broadcast live, recorded in a single session, or produced from multiple segments recorded at different times.98 They may be based on a script, conducted as an interview, or wholly improvised.99 As a result, not every podcast will fall into the same category of “work” as defined in the Act. A video podcast, for instance, will automatically attract protection as a cinematographic, and therefore dramatic work, while an audio-only podcast will not.100 The copyright protection for a scripted podcast will include protection for the script as a                                                  94 Ibid, s 3. 95 Ibid, s 2. 96 Ibid, s 2 “compilation”. A podcast which consisted of multiple segments, each of which fall into the definition of a work would qualify as a compilation. Further, subsection 2.1(1) deems compilations of more than one type of work to be “a compilation of the category making up the most substantial part of the compilation”. If for example, a podcast which was based around scripted segments also included some musical breaks, it would still be protected as a literary work under subsection 2.1(1). 97 See note 8, above. 98 See note 9, above. 99 Most of the series in the study set seemed to be conversational and not based on a set script. For examples of scripted podcasts, see Duncan, supra note 12; Longworth, supra note 12. 100 See section 3.2.1.2, below.  27  literary work, while an unscripted podcast may only be protected as a sound recording.101 These distinctions will be explored further below. 3.2.1.1. Unlikely Works Categories for Podcasts: Artistic and Musical Two categories of work defined by the Act are clearly inapplicable to podcasts and can be dispensed with briefly. The definition of “artistic work” focuses on visual arts such as paintings, photographs, sculpture, and architecture.102 The format in which podcasts are distributed (as digital audio or video files) precludes podcasts from falling into this category.103 “Musical works” include “any work of music or musical composition”.104 While podcasts are generally distributed as digital audio or video files and generally emphasize audio, few if any are comprised solely of musical performances.105 This is not because of any technical limitation underlying podcasting. Podcasts are distributed as digital audio or video files which could contain musical performances. Simply as a matter of fact, few are exclusively or primarily                                                  101 However, if the podcast is structured as a lecture, it may also fall within the definition of a literary work. See section 3.2.1, below. 102 Copyright Act, supra note 89, s 2 “artistic work”. 103 However, note that elements such as a logo or series art that accompany a podcast may be protected as artistic works. 104 Copyright Act, supra note 89, s 2 “musical work”. 105 The emphasis on audio is clear from the survey results. Most of the podcasts found were audio-only, and some of those with accompanying video were released in audio-only format as well. In the sample, All About Android and Bizarre States were released in both audio and audio-video formats. In contrast, no podcasts known to the author are only video with no audio.   28  musical recordings.106 Many podcasts do make use of music to begin or end an episode, or as segment breaks. This usage may fall within the definition of a “compilation”.107 3.2.1.2. Video Podcast: Dramatic Work A video podcast consists of a live or recorded video with a synchronized soundtrack.108 This can be contrasted to an audio podcast, which consists only of an audio track distributed as a digital audio file, or live audio distributed as a stream. The definition in the Act of a “cinematographic work . . . includes any work expressed by any process analogous to cinematography, whether or not accompanied by a soundtrack”.109 The process of capturing digital video is clearly analogous to cinematography. The definition of a dramatic work in the Act includes “a cinematographic work”.110 Based on these definitions, whether or not a given video podcast is scripted, it is a dramatic work and therefore a “work” for the purposes of the Act.  3.2.1.3. Scripted Podcast: Literary Work Some podcast episodes are based on a script, while many (particularly those with multiple hosts) include un- or semi-scripted conversations.111 Other episodes take the format of interviews.                                                   106 None of the podcasts surveyed for this project or known to the author are solely comprised of musical compositions. An examination of the social and historical facts of the development of podcasting as a genre is beyond the scope of this chapter. I would speculate however that the copyright protected status of recorded music, the largely non-commercial origins of podcasting, and the simultaneous development of distribution for both legal and copyright infringing digital music all likely contributed to podcasting’s development. These may have led podcasting’s development in a different direction than radio broadcasting. It is worth noting, however, that there is no technical reason why the same technology (using RSS to distribute audio files) could not be used to distribute music.  107 See above, s 3.2.1. 108 However, some some content associated with podcasts such as the series or episode art and the logo could be protected as artistic works. 109 Copyright Act, supra note 89, s 2 “cinematographic work”. 110 Ibid, s 2 “dramatic work”. 111 See note 99, above.  29  Whether an audio podcast receives full protection as a copyrighted work may depend on  whether it was based on a script or improvised. If a podcast is based on a script, that script should fall within the definition of a “literary work” and receive the full protection of a copyrighted work. The owner of the copyright therefore has the rights granted under section 3 of the Act (which will be further explored below in section 3.3) including the “sole right to produce or reproduce the work or any substantial part thereof in any material form whatever”.112 An audio recording reproduces the script on which it is based in an audible format. While the recording itself may not be a “literary work”, it would still receive some protection because certain uses of it could constitute infringement of the copyright in the script. It is also protected as a sound recording.113  However, the definition of a “literary work” in the Act and that of “every original literary, dramatic, musical and artistic work” provides little clarification and no obvious place within the categories of works for improvised podcasts. A similar issue is discussed in Gould v Stoddart, which deals with “easygoing conversations” between Carroll, the interviewer, and the pianist Glenn Gould. These conversations, conducted in 1956 for a magazine article, were in some cases tape recorded by Carroll. Excerpts from these conversations were later published in a 1995 book authored by Carroll. Gould’s estate asserted copyright in the transcriptions of these conversations, asserting that the “spoken words are . . . contemplated within the definition of ‘literary work’”. The court of appeal decision held that “Gould did not have a copyright with respect to his oral utterances or in the ‘transcriptions’ of them”. Rather, the interviewer Carroll held “unrestricted                                                  112 Copyright Act, supra note 89. 113 See section 3.5.2, below. 30  copyright” in the “written material” in his book that the appellants were attempting to suppress.114 Precedent cited by the trial court supports the proposition that a person who reports the contents of a speech “is the author of the report and obtains copyright in the report”.  On first blush, this may appear to lead to a suprising conclusion: in the absence of sound recording rights, copyright might be granted to the transcriber of an unscripted podcast episode or segment even if it was done by a third party. However, paragraph 18(1)(b) establishes for the owner of sound recording rights the right to reproduce that recording in any material form. If we assume that a transcription is a “material form”, transcription would be a violation of the sound recording owner’s right.115 Further, the transcription would likely not fulfill the originality requirement under the “skill and judgment” test and so the transcriber would not retain rights in the transcription. 3.2.1.4. Podcast Drama: Dramatic Work  Some podcasts adopt a format similar to a traditional radio play.116 These employ characters, narration, and other trappings of dramatic pieces.117 The Act’s definition of dramatic works which are not cinematographic or a compilation specifies “any piece for recitation, choreographic work or mime, the scenic arrangement or acting form of which is fixed in writing or otherwise”.118 Depending on the facts, a podcast with a dramatic format and a script may fit within this definition of “dramatic work”.119 For unscripted podcasts, the question of fixation is addressed below. However, does the language of the Act regarding dramatic works imply a                                                  114 Gould v Stoddart (CA), supra note 89 at para 23. 115 Copyright Act, supra note 89.  116 See Bottomley, “Revival of Radio Drama”, supra note 32. 117 Ibid. 118 Copyright Act, supra note 89 [emphasis added]. 119 Ibid, s 2 “dramatic work”. 31  different requirement than the general fixation requirement? The language of “fixed . . . otherwise” establishes a broader scope than just the written word; however, the language of “any piece for recitation [or other purposes]” may imply that the fixation has to precede the performance. This may exclude improvised podcasts from the definition of “dramatic work” even if they otherwise resemble such works. 3.2.2. Originality The originality requirement for works is referenced in section 2 of the Copyright Act, which defines “every original literary, dramatic, musical, and artistic work”, and section 5, which states that copyright shall subsist in Canada “in every original literary, dramatic, musical and artistic work” subject to the citizenship/residency requirement.120 The Supreme Court of Canada’s definition of originality for the purposes of the Act in CCH Canadian requires that a work originate from an author who exercises their skill and judgment, and that the work not be a copy.121 The exercise of skill and judgment must be “not so trivial that it could be characterized as a purely mechanical exercise”.122 In the context of podcasts, a show based on an original script would fulfill this requirement. An improvised or conversational podcast would likely also be original. They are necessarily not copied, and the speakers on the podcast would exercise “skill and judgment” in what they choose to talk about if they are trying to make the content entertaining, informative, or otherwise interesting for an audience.                                                   120 Ibid, s 2 “every original literary, dramatic, musical, and artistic work”, s 5. 121 CCH Canadian v Law Society of Upper Canada, 2004 SCC 13 at para 25. 122 Ibid.  32  3.2.3. Fixation Fixation is generally recognized as a requirement for copyright protection. The requirement is not set out in the Copyright Act, but has been recognized by the Supreme Court of Canada as a requirement having a “well settled” definition in copyright law.123 Fixation requires that for something to be copyrightable, it must be expressed in a more or less permanent material form.124 Podcasts are generally distributed as digital audio or video files.125 Some are also broadcast live, on online streaming services such as Twitch or Mixer, or through other means.126 While fixation to a digital file likely fulfills the fixation requirement, live-streamed audio or audio-video may not unless it is simultaneously fixed to a file.  3.2.3.1. Fixation of Unscripted Podcasts As noted above, Gould reveals some difficulty in fitting an “easygoing conversation” into a category of “work” recognized by the Act. However, the Gould case is primarily about fixation. The trial court in Gould states that “for copyright to subsist in a work, it must be expressed in material form and have a more or less permanent endurance”.127 As a result, “a person’s oral statements in a speech, interview, or conversation are not recognized in that form as literary creations and do not attract copyright protection”.128 However, in the case of podcasts which are recorded to digital audio or video files, their “material form” and “more or less permanent endurance” do not seem to be in issue. Other Canadian cases have dealt with media in digital                                                  123 Théberge v Galerie d’Art du Petit Champlain, 2002 SCC 34 at para 25. 124 Gould v Stoddart (SC), supra note 89 at para 34. 125 See section 1.1, above. 126 Ibid. See note 9, above.  127 Gould v Stoddart (SC), supra note 89 at para 34. 128 Ibid, citing Walter v Lane, [1900] AC 539 [emphasis added].  33  formats without finding fixation to be problematic.129 Leaving aside the issue of whether a given podcast is a “work”, recorded podcasts should satisfy the fixation requirement. 3.2.3.2. Fixation of Live-streamed Podcasts This section considers whether a podcast that is streamed live rather than distributed as a recorded file would fulfill the fixation requirement. Relevant to this point, subsection 3(1.1) of the Act states that “[a] work that is communicated in the manner described in paragraph (1)(f) [communication to the public by telecommunication] is fixed even if it is fixed simultaneously with its communication.”130 Therefore if the streaming of a live podcast constitutes communication to the public by telecommunication and it is fixed simultaneously with its communication (that is, the podcast is recorded to a file as it is streamed), then it is fixed.  The definition of “telecommunication” provided in section 2 of the Act is very broad, including “any transmission of signs, signals, writing, images or sounds or intelligence of any nature by wire, radio, visual, optical or other electromagnetic system”. The streaming of audio or video over the Internet undoubtedly fits within this definition as the transmission of sounds or sounds and images over a series of wire, optical, radio, or other electromagnetic systems.131 Streaming on a service such as Twitch likely constitutes “communication to the public” as well, since it is available to anyone with an Internet connection. Finally, regarding simultaneous fixation, if the stream is archived to a digital file by the streaming service or otherwise saved to a file by the                                                  129 See Entertainment Software Association v Society of Composers, Authors, and Music Publishers of Canada, 2012 SCC 34 [ESA v SOCAN]. 130 Copyright Act, supra note 89. 131 See ESA v SOCAN, supra note 129 at paras 28–39 (considering the distinction between a download and a stream, particularly the transient nature of the latter which does not leave the recipient in possession of a copy of the work).  34  creator, this would arguably be sufficient to establish fixation.132 In the event that a podcast was not so “fixed”, it would not be able to be distributed later through a podcasting platform. In the event that the streaming service does not archive the podcast and the creator fails to save their work to a fixed format, fixation may not be established. Where fixation and the other conditions for copyright subsistence mentioned above are satisfied, podcast creators will enjoy economic and moral rights in podcast episodes as works. 3.3. Creators’ Rights in Podcasts The exclusive rights granted to copyright owners capture a wide range of possible commercial activity with respect to works. The first owner of copyright in a work is its author.133 In the event that the author was an employee and the work was made in the course of employment, the employer owns the copyright.134 Copyright can be assigned, but the assignment must be made in writing and signed.135  A work of joint authorship is “produced by the collaboration of two or more authors in which the contribution of one author is not distinct from the contribution of the other author or authors”.136 As applied to podcasts, joint authorship may arise where a podcast has multiple hosts. If each contribute substantially to the podcast, and there is a “common design”, this is likely a case of joint authorship.137 “Common design” would seem likely to arise where the hosts have agreed to record a podcast together. This could also extend to podcast guests or producers, although the                                                  132 “[A] copyright springs into existence as soon as the work is written down or otherwise recorded in some reasonably permanent form (‘fixated’)”: Théberge, supra note 123 at para 25, citing Hugh Laddie, et al., The Modern Law of Copyright and Designs, vol. 1, 3rd ed. (London: Butterworths, 2000). 133 Copyright Act, supra note 89. 134 Ibid, ss 13(1), (3). 135 Ibid, s 13(4). 136 Ibid, s 2 “work of joint authorship”. 137 See Neugebauer v Labieniec, 2009 FC 666.  35  determination would depend upon the extent of their involvement. If a producer made the technical or other arrangements to record the podcast, they may be the first owner of the sound recording rights in the podcast.138 3.3.1. Economic Rights A podcast protected as a work is protected from unauthorized reproduction, performance in public, communication to the public, adaptation, or translation (among other unauthorized actions). These rights are established in subsection 3(1) of the Act, which provides that copyright protection includes the exclusive rights “to produce or reproduce the work or any substantial part thereof in any material form whatever” and “to perform the work or any substantial part thereof in public”.139 In addition, subsection 3(1)(d) protects for literary, dramatic, and musical works the right “to make any sound recording, cinematograph film or other contrivance by means of which the work may be mechanically reproduced or performed”.140   Paragraph 3(1)(f) of the Act grants the exclusive right “in the case of any literary, dramatic, musical or artistic work, to communicate the work to the public by telecommunication”. The definitions of “telecommunication” and “communicate the work to the public” are discussed above in section 3.2.3.2.141 In the case of podcasts, the creator’s exclusive right would cover, for example, streaming a podcast from a recording. Adaptation and translation rights relate to the use of a work as the basis for a work in a different form or language. Paragraph 3(1)(c) provides for an exclusive right “in the case of a novel or                                                  138 See section 3.5.2, below. 139 Copyright Act, supra note 89. 140 Ibid. 141 See above.  36  other non-dramatic work . . . to convert it into a dramatic work, by way of performance in public or otherwise”, while paragraph 3(1)(e) covers “any literary, dramatic, musical or artistic work” and the right “to reproduce, adapt and publicly present the work as a cinematographic work”.142 In the context of podcasts, these rights may arise in the context of adaptation of a podcast for a different medium such as television.143 Paragraph 3(1)(b) applies to dramatic works and covers “conver[sion] . . . into a novel or other non-dramatic work”, while paragraph 3(1)(a) covers the right to “to produce, reproduce, perform or publish any translation of the work”.144 These could apply to transcription of the text of a podcast (in the same language or a different language). The transcribed text would constitute a literary work, allowing the application of paragraph 3(1)(b). The protection under this paragraph would potentially overlap with reproduction, which a transcription could also constitute, but would only apply to podcasts which are considered dramatic works.145  3.3.2. Moral Rights Moral rights are given protection in the Copyright Act but are not included in the Act’s definition of “copyright”.146 Moral rights include the right to the integrity of the work, the right to prevent certain associations with the work, and the right to be recognized as an author (or remain anonymous as the case may be). The rights to integrity protects against a work being “distorted,                                                  142 Copyright Act, supra note 89. 143 Some podcasts have subsequently been adapted for television: see e.g., IFC, “Maron”, online: <www.ifc.com/shows/maron>; IFC, “Comedy Bang Bang”, online: <www.ifc.com/shows/comedy-bang-bang>; John Koblin, “‘Pod Save America’ is Coming to HBO”, The New York Times (February 8, 2018), online: <www.nytimes.com/2018/02/08/business/media/pod-save-america-hbo.html>. 144 Copyright Act, supra note 89. 145 See sections 3.2.1.2, 3.2.1.4, above. 146 Copyright Act, supra note 89, s 2 “copyright”. Rights covering sound recordings, performers’ performances, and communication signals as well as those covering works are included in the definition.  37  mutilated or otherwise modified” or “used in association with a product, service, cause or institution”, if these are done “to the prejudice of the honour or reputation of the author”.147 Moral rights cannot be transferred or assigned except upon the death of the work’s author, but they can be waived.148 In the context of podcasting, examples of a possible infringement of moral rights may include the alteration of an audio clip done to damage the reputation of the creator (such as by making it seem as if they had said something they did not), or the use of podcast audio to imply an endorsement of a product or service by the creator.149  3.4. Users’ Rights in Podcasts The scope of users’ rights has two dimensions for creators: the use they are able to make of other, copyrighted works, and the use that others are able to make of their work.150 The former dimension has been examined in some detail in some legal guides for podcasters.151 This topic is interesting and practically valuable for podcast creators; however, the focus of this project is the extent of copyright in podcasts and its usefulness to creators rather than copyright issues creators might face. As such, this section will focus on how fair dealing rights, one variety of users’ rights, might be applied to podcasts themselves. This should be useful to creators who may be interested to know what use they can make of other podcasts, and the extent to which others can make use of their work.                                                  147 Ibid, s 28.2(1). 148 Ibid, s 14.1(2), (4); s 14.2(2). 149 However it should be noted that many podcast advertisements rely in part on endorsement of products or services. The reputational harm component may not be present in cases where the creator had already associated themselves with that product. Such situations may also engage misappropriation of personality. 150 The private copying and reproduction for private purposes provisions in sections 80 and 29.22 of the Act may also be considered users’ rights. The section 80 rights deal exclusively with musical works: see Copyright Act, supra note 89. As discussed above, few if any podcasts fit the definition of musical works, so these provisions will usually not apply: see section 3.2.1.1, above.  151 See e.g., Vogele & Garlick, supra note 22; Simmons & Kaplan-Myrth, supra note 22.  38  Fair dealing rights apply to podcasts as they do to other copyrighted works. The Supreme Court of Canada has affirmed that fair dealing rights are users’ rights.152 Under sections 29–29.2 of the Act, users have the right to certain uses of copyrighted works without authorization from the copyright owner, provided that the dealing is “fair”. Under section 29, the dealing must fall into one of 5 categories: research, private study, education, parody, or satire.153 Sections 29.1 and 29.2 establish further fair dealing grounds for criticism or review and news reporting with additional attribution requirements.154 Whether a dealing is “fair” is determined with reference to six factors articulated in CCH Canadian: the purpose of the dealing, character of the dealing, amount of the dealing, alternatives to the dealing, nature of the work, and effect of the dealing on the work.155 In the context of podcasting, instances where fair dealing might be invoked could include transcription of part of a podcast for reporting or research, or excerpting podcast audio for criticism (for example if included in a different podcast).156  3.5. Neighbouring Rights As discussed in section 1, a given podcast may not qualify as a “work” under the Copyright Act. However, protection is still available for sound recordings (which all audio-only podcasts would qualify for) and for performers’ performances (which may apply to podcast guests or hosts as performers). The protection offered under the provisions of the Copyright Act dealing with recordings and performances has some overlap with the protection offered to works, but is generally more limited. These rights are often called “neighbouring rights” to distinguish them                                                  152 Alberta (Education) v Canadian Copyright Licensing Agency (Access Copyright), 2012 SCC 37 at para 22; Society of Composers, Authors and Music Publishers of Canada v Bell Canada, 2012 SCC 36 at para 11. 153 Copyright Act, supra note 89. 154 Ibid. 155 CCH Canadian, supra note 121. 156 A fulsome analysis of how fair dealing might arise in the context of podcasting is beyond the scope of this thesis project.  39  from the protection given to works (though the Act defines “copyright” as including performance and recording rights as well).157 As discussed above, scripted podcasts and podcasts including video will likely be considered as works, so neighbouring rights protection is most important for improvised, audio-only podcasts. An improvised, audio-only podcast may fall under performance rights as “an improvisation of a . . . literary work, whether or not the improvised work is based on a pre-existing work” (rights under subsection 15(1)) and will certainly qualify as a sound recording (section 18) as “a recording, fixed in any material form, consisting of sounds, whether or not of a performance of a work”. 3.5.1. Performers’ Rights The performance rights of performers are protected by the Copyright Act, which states that “[p]erformers have rights in their performances that are separate from and independent of the works they are performing.”158 This section will consider whether and under what condition performers’ rights will arise from a podcast. Given the more limited scope of performers’ rights (as opposed to copyright in works) and the likelihood that podcast hosts are authors for the purposes of copyright, this section will consider the particular case of podcast guests, as opposed to hosts, receive performers’ rights protections.                                                   157 See Hagen et al., supra note 92 at 56–57; Copyright Act, supra note 89, s 2 “copyright”. 158 See Hagen et al., supra note 92 at 56–57.  40  3.5.1.1. Podcast Guests as Performers A core feature of the podcast series is its host: many feature one or more regular hosts who are present for every episode.159 Many also include guest appearances, sometimes structured as interviews, sometimes with the guests participating in the podcast segments or discussion in a similar way to the host, and frequently as some mix of the above. For the purposes of this section, the term “guest-performer” will be used to refer specifically to podcast guests who would be recognized as performers, whatever the scope is determined to be.  3.5.1.2. When do Performers’ Rights Arise? The Copyright Act grants rights in a performer’s performance.160 A performer’s performance includes a “an improvisation of a dramatic work, musical work or literary work, whether or not the improvised work is based on a pre-existing work”.161 Based on that wording, this definition may not apply if an improvised podcast does not fall under one of the “works” there listed. As discussed above, improvised podcasts may not qualify as works.162 However, the Act also states that “[p]erformance means any acoustic or visual representation of a work, performer’s performance, sound recording or communication signal, including a representation made by means of any mechanical instrument, radio receiving set or television receiving set”.163 This definition would apply also to the use of “performance” as it applies to a work, but importantly contemplates performer’s performances as not overlapping completely with works. Further, podcasts are acoustic representations of sound recordings, digitally recorded, and as such satisfy                                                  159 Every podcast surveyed for the empirical study in this thesis had one or more hosts. The host or hosts might be the only individual(s) involved in creating a given podcast, though some of the podcasts in the sample credit producers or other support: for example, in the set the series You Must Remember This and This is War (among others) credited production support. 160 Copyright Act, supra note 89, s 2 “performer’s performance”. 161 Ibid. 162 See section 3.2.1, above. 163 Copyright Act, supra note 89, s 2 “performance”. 41  the above-quoted definition of “performance”. Taken together, these definitions seem to indicate that a podcast that did not qualify as a work could still support performers’ rights. 3.5.1.3. What do Performers’ Rights Protect?  Performance rights grant among other things the sole right to, “where the performer authorized a fixation, to reproduce any reproduction of that fixation, if the reproduction being reproduced was made for a purpose other than that for which the performer’s authorization was given”, and if the performance “is fixed in a sound recording that is in the form of a tangible object, to sell or otherwise transfer ownership of the tangible object”.164  As noted above, the question of whether podcast guests ought to be paid for their appearances is a live one in the podcasting community. Performers’ rights may provide valuable bargaining power for guest-performers if they are recognized in law. For example, they may provide legal recourse for the guest-performer where payment is promised but withheld in the absence of a written contract. As described above, some podcast guest appearances are considered as analogous to comedic performances by members of the professional comedy community. To the extent that the performer’s performance right applies in professional comedy, therefore, the considerations which led to the establishment of the right could be similarly applied to podcasting. 3.5.2. Sound Recording Rights Sound recording rights are granted to the maker of a sound recording, defined as “the person by whom the arrangements necessary for the first fixation of the sounds are undertaken”.165 This                                                  164 Copyright Act, supra note 89, ss 15(1)(b)(ii), 15(1.1)(e). These rights are subject to certain “points of attachment” in subsections 15(2)–(2.2) in the Act.  165 Ibid, s 2 “maker”.  42  “include[s] arrangements for entering into contracts with performers, financial arrangements and technical arrangements required for the first fixation of the sounds for a sound recording”.166 In the context of podcasting, a producer may be granted these rights if they make these arrangements. A soundtrack is a sound recording except where it accompanies a motion picture.167 This has two consequences for podcasts: first, that the audio track accompanying a video podcast does not qualify as a “sound recording”, and second that the sound recording rights in a podcast would not be infringed if the audio was used to accompany a motion picture (such as a YouTube video). The first consequence is not likely to be significant for most podcasters given that the video and audio would qualify as a “cinematographic work” as described above, and therefore receive the broader works protection.  However, for any audio podcast which does not fit the definition of a work, this represents a gap in protection. In re:Sound, the Supreme Court of Canada held that when a sound recording accompanies a motion picture, it “does not fall within the definition of ‘sound recording’” even if the recording pre-exists the motion picture. This potentially excludes the rights holder from their rights to reproduction (section 18) and remuneration (section 19). If, therefore, a person were to use podcast audio without authorization as a soundtrack for a video, the creator of said podcast may not have protection against this use under the Canadian Copyright Act unless it also qualified for protection as a work or as a performer’s performance.                                                    166 Ibid, s 2.11. 167 See re:Sound, 2012 SCC 38 at paras 35–36. 43  The soundtrack to a video podcast would not be eligible for sound recording protection: the definition of a sound recording “excludes any soundtrack of a cinematographic work where it accompanies the cinematographic work”. However, the definition of “performer’s performance” includes performance of a dramatic work and recitation or reading of a literary work; therefore scripted and dramatic podcasts would enjoy neighbouring rights protection as a performance in addition to their protection as works (see above). 3.6. Conclusion On the whole, copyright would seem to apply to podcasts much as it does to other media. This further highlights the peculiarity of podcasts’ dominant model of distribution (that is, for free). In contrast to other areas where copyright doesn’t extend in law, podcast creators enjoy certain rights under the current copyright regime. However, the premise of this project is that, unlike in other media contexts, the exclusivity which copyright enables is in many cases not used to create an artificial scarcity to drive sales of podcast works.168 While this chapter has identified a possible barrier to copyright protection as a “work” for improvised podcasts, sound recording rights (which audio podcasts are clearly covered by) give the owner an exclusive reproduction right. There is therefore no obvious relationship between the legal uncertainty identified in this chapter and the absence of a “scarcity model” (as discussed in the Copyright Theory chapter) in the funding and distribution of podcasts. The following chapters will examine how podcasts are in fact funded to work towards a better understanding of why many podcast creators release their work for free.                                                   168 See Chapters 1 (setting out the rationale for this project) and 6 (evaluating this assumption in light of the study results), below. 44  4. Methodology 4.1. Introduction This project will use content analysis methods to generate data on how podcast creators fund their work, as well as the role of copyright in podcasting. As discussed in Chapter 2, copyright provides a way for authors to realize financial support for their work by excluding non-paying users. However, podcasting is a medium in which creators have largely chosen not to restrict access to their works in order to sell them.169 Gauging copyright’s importance to podcasting is one goal of this project. Another goal is to describe the financial support methods podcast creators use. The results of this study will speak to both of these goals, and will inform a discussion in Chapter 6 of how podcasting fits with the abundance models described by Suzor.170 Content analysis methods will be instrumental in generating data relevant to this discussion.171  This study will use content analysis coding methods on freely-available podcast audio and metadata (the “coded material”) to generate data relevant to the project. The study focuses on podcasts which are available on-demand in audio-only format. Limiting the sample to audio-only podcasts will minimize methodological complexity introduced with coding video content or sampling live content.172 Further, on-demand podcasts available as audio appear to constitute the bulk of available podcast content online.                                                  169 See Katz, supra note 1; NYC Media & Entertainment, “New York City, the Podcasting Capital” (2017), online: <https://www1.nyc.gov/assets/mome/pdf/MOME_PODCAST_REPORT_Web_Final.pdf>. 170 See Suzor, supra note 41 at 298–99. 171 However, material limitations on this project prevent the use of multiple coders; as such, the study will be more limited in scope than conventional content analysis and may not qualify as a true content analysis study. See Neuendorf, supra note 6 at 166. See also section 4.4.4, below (describing the limited reliability testing procedures that this study will follow in the absence of multiple coders). 172 See Neuendorf, supra note 6 at 111–13; Chapter 4: Methodology, below. 45  The coded material will be collected from samples of two populations: podcasts associated with podcast networks featured on the iTunes front page, and the top 100 podcast Patreon pages by number of donors. The rationale for picking these two populations will be described at length below. These two populations are associated with two categories of “abundance model” fundraising described by Suzor: advertising and crowdfunding. In brief, podcast networks are relevant to the research objectives of this project because they often offer advertising support to member podcast series, while podcasts with Patreon pages are engaged in crowdfunding. The codes will be tailored to collect data on important aspects of these two models, as well as use (in copyright notices or licenses) or discussion of copyright or other intellectual property. The Content Analysis Guidebook (the Guidebook), written by Kimberley A. Neuendorf, is a leading text in content analysis and will be relied on throughout this chapter to describe the content analysis methods which this study will employ.173 In particular, this study relies heavily on the coding methodology described in the Guidebook. While there are limitations to what this study and content analysis more broadly can reveal, the methodology provides a systematic way to analyze large sets of materials and is in this way particularly appropriate to a study of podcasts.                                                   173 This text has been cited in a number of other content analysis studies. For law-related content analysis studies, see Martin Z.P. Olszynski, “Failed Experiments: An Empirical Assessment of Adaptive Management in Alberta’s Energy Resources Sector” (2017) 50:3 UBC L Rev 697 at 724; Mark A. Hall & Ronald F. Wright, “Systematic Content Analysis of Judicial Opinions” (2008) 96 Cal L Rev 63 at 70. For use of the Guidebook in other content analysis studies, see also Eschenfelder at 1409; Geoffrey Supran & Naomi Oreskes, “Assessing ExxonMobil’s Climate Change Communications (1974–2014)” [2017]:12 Environmental Research Letters 1 at 2, fn 25.  46  4.2. Content Analysis Content analysis is an empirical methodology used to analyze texts or other materials (“messages”) and form generalizable conclusions about a large set of these messages.174 Applying content analysis allows the researcher to summarize these messages in a reasonably objective and reliable way. This is done through coding: formulating a list of message characteristics and counting how many messages in a set display those characteristics.175 Content analysis methods will be particularly valuable for a study of podcasts. Podcast audio and metadata offer a large, easily accessible set of messages. The coding methodologies developed for content analysis provide a systematic way to make use of this large amount of content.176  Content analysis has been used in numerous studies published in law and other academic journals.177 For instance, in a 2008 article, Mark A. Hall and Ronald F. Wright advocated for the use of content analysis of judicial decisions as a standard social science method in legal scholarship.178 Studies of copyright cases include Marcella Favale, Martin Krestchmer, and Paul C. Torreman’s work on the European Court of Justice’s copyright jurisprudence and Barton Beebe’s work on judicial decisions on US fair use provisions.179 Content analysis has also been used in the context of legal and legal interdisciplinary scholarship to study materials other than                                                  174 See Neuendorf, supra note 6 at 1–35. 175 Ibid. 176 Ibid at 23–31. 177 See e.g., Marcella Favale, Martin Kretschmer & Paul C. Torremans, “Is There an EU Copyright Jurisprudence? An Empirical Analysis of the Workings of the European Court of Justice” (2016) 79:1 Modern L Rev 31; Barton Beebe, “An Empirical Study of U.S. Copyright Fair Use Opinions, 1978–2005” (2008) 156 U Pa L Rev 549; Antoni Terra, “Copyright Law and Digital Piracy: An Econometric Global Cross-National Study” (2016) 18 NC JL & Tech 69; Francina Cantatore & Jane Johnston, “Moral Rights: Exploring the Myths, Meanings and Misunderstandings in Australian Copyright Law” (2016) 21 Deakin L Rev 71; Kristin R. Eschenfelder, Robert Glenn Howard & Anuj C. Desai, “Who Posts DeCSS and Why? A Content Analysis of Web Sites Posting DVD Circumvention Software” (2005) 56:13 J Am Soc Info Sci & Tech 1405; Yasmin Morais & Sara Sampson, “A Content Analysis of Chat Transcripts in the Georgetown Law Library” (2010) 29 Leg Ref Services Q 165.  178 Supra note 173. 179 See Favale, Kretschmer & Torremans, supra note 177; Beebe, supra note 177.  47  case law. This includes studies such as Antoni Terra’s on copyright provisions in various countries’ statutes, Francina Cantatore and Jane Johnston’s work on treatment of moral rights in publishing contracts, Kristin R. Eschenfelder, Robert Glenn Howard, and Anuj C. Desai’s article on online rationales for posting DeCSS software (used to bypass copy protection on DVDs), and analysis of chat transcripts from a law library help service by Yasmin Morais and Sara Sampson.180 These examples show the breadth of content to which content analysis methods can be applied.181  4.2.1. Methodological Concerns in Working with Digital Media As noted above, The Content Analysis Guidebook is a leading text in content analysis.182 A chapter of the Guidebook authored by Paul D. Skalski, Kimberly A. Neuendorf, and Julie A. Cajigas (“Skalski et al.”) addresses some special considerations that content analysis researchers should observe when dealing with “interactive media” (including social networking sites as well as content platforms such as YouTube).183 This section will consider the issues Skalski et al. raise as potentially problematic when dealing with digital media, evaluate their relevance to podcasting, and where applicable, describe how this study will deal with them. The primary issue identified by Skalski et al. is that, unlike traditional media forms, interactive or digital media often present a moving target which complicates assumptions about message content and source.184 Digital media are frequently not fixed to a physical form or consistently                                                  180 See Terra, supra note 177; Cantatore & Johnston, supra note 177; Eschenfelder, Howard & Desai, supra note 177; Morais & Sampson, supra note 177. 181 No examples of content analysis on podcasts were found in the course of research for this thesis. 182 See notes 173, 177, above (citing other content analysis studies).  183 Paul D. Skalski, Kimberly A. Neuendorf & Julie A. Cajigas, “Content Analysis in the Interactive Media Age” in Neuendorf, Content Analysis Guidebook, supra note 6. I use the term “digital media” with substantially the same meaning. 184 Skalski, Neuendorf & Cajigas, supra note 183 at 203. 48  available at the same address, and can have their context and content changed after publication. For example, digital media content such as social networking posts or blogs can be edited after the fact, and their context on a web page can differ in presentation depending on user choice (such as whether a user views the page on a mobile device or a desktop computer). Web pages are also frequently accompanied by ads which are programmatically served to the user based on their web history or other factors. These issues make it difficult to accurately describe the context in which messages are presented.  In the context of podcasting, however, these issues are less severe. Podcasts are not as stable as the fixed forms of traditional media content cited by Skalski et al. “such as the newspaper page, the television episode, or the motion picture”, but neither are they as changeable or ephemeral as, for example, social media posts.185 Similar to other digital media, podcast episodes can be changed after the initial upload by replacing the original file, and podcast metadata can be edited. Programmatically inserted advertisements that target users based on location or other factors are used in podcasting and can result in different users being served different ads.186 However, for the most part, podcasts are static (they can be found as long as the podcast feed remains available and the files are hosted) and inert (a podcast episode’s content and metadata are unlikely to change significantly after the episode is published).  The issues of ascertaining content authorship cited by Skalski et al., are also less severe in the context of podcasts than in other forms of digital media. Podcast creators generally retain control of the RSS feed through which they publish their podcasts, so while episodes may be shared                                                  185 Ibid. 186 See e.g., Megaphone, “Megaphone: A Modern Podcasting Platform by Panoply”, online: <megaphone.fm>. For complications introduced to this study by dynamic ads, see section 6.2.1, below.  49  widely, there is usually an authoritative source that can be located.187 Further, research conducted in the course of this project shows that most podcast hosts name themselves and their podcast in the podcast itself, in their own voice. As such, podcast authors are much easier to identify (and harder to conceal or impersonate) than, to take a particularly problematic example from Skalski et al., the authors of meme images.188  4.3. What is Being Studied? By focusing on podcasts which are members of podcast networks and crowdfunded podcasts, this study will draw comparisons that will shed light on each. While the selected podcasts will not necessarily be representative of podcasts as a whole, the study will produce results on two funding models which are important to many podcasts. 4.3.1. Defining the Population: Podcast Networks and Crowdfunded Podcasts This study will focus on two sets of podcast series: one set composed of series associated with podcast networks and another including only series crowdfunded through Patreon. These groups correspond to two major ways for podcasters to financially support their work: advertising (for which podcast networks provide support and sales) and crowdfunding (which platforms such as Patreon facilitate). This project will therefore take a more narrow focus than a study of all podcast series would, instead limiting the population to two subsets of podcasts as a whole.  A narrow focus is justified by the extremely large number of podcasts—podcast directory Blubrry claims to list over 600,000—which would necessitate a large sample size to draw                                                  187 See section 1.1, above (defining “RSS”). 188 Skalski, Neuendorf & Cajigas, supra note 183 at 209.  50  meaningful conclusions.189 The sample that was instead chosen for this study examined (after exclusions) 24 randomly-sampled series of an initial exposure sample of 400 series (6%); to achieve a similar proportion of the Blubrry list would require examining 36,000 series. Further, focusing on two distinctive populations within podcasting rather than the set of podcasts as a whole will provide an opportunity for comparison, and help to characterize the landscape of podcasting more richly.190 This will come at the expense of generalizability, as podcasts not associated with podcast networks and not crowdfunded or crowdfunded through platforms other than Patreon will be excluded. The comparison this approach will allow between podcast network podcasts and crowdfunded podcasts will be valuable. For example, the study compares the use and discussion of copyright between the two populations. Crowdfunded podcasts which provide paying subscribers with exclusive access to bonus episodes would seem to have an incentive to assert the copyright status of their work as a way to exclude non-paying listeners.191 Other concerns may be common to both groups, but to different extents. Podcasters may be concerned about future opportunities for licensing their work for radio play, for example. Whether this manifests more for podcast network podcasts or crowdfunded podcasts is another potentially interesting line of inquiry.                                                   189 See Blubrry, “Blubrry Podcasting”, online: <www.blubrry.com>. Apple Podcasts/iTunes, the most popular podcast platform, does not provide a count of currently listed podcasts. Neuendorf notes that “there is no universally acepted set of criteria for selecting the size of a sample”: Neuendorf, supra note 6 at 90. In the case of podcasts, this methodological difficulty is compounded by uncertainty about the size of the population (i.e., the number of podcast series). 190 Given the large number of podcasts available, it may not be possible to guarantee that enough podcast network podcasts and crowdfunded podcasts would appear in a sample of all podcasts. Since these two populations have been chosen as relevant to the theoretical goals of this project, targeting them directly is appropriate. 191 In the study, bonus episodes were the most common type of bonus content offered to paying subscribers. See section 6.4.1, below. 51  Finally, the choice of podcast networks as one population for this project is supported by the focus of this research. The research question under investigation is concerned with copyright as a mechanism for creators to financially support their work. Some podcasts are associated with existing businesses or institutions. These may be produced as a sort of loss-leader to draw listeners in to another paid or ad-supported product (such as television or online news). In contrast, many podcast networks exist to provide support for podcast creators outside of these traditional media ecosystems. Even where podcast networks are owned by traditional media enterprises, the relationship is different than if the podcasts on the network were podcasts from a particular business or institution. For example, the podcast network Wondery is owned by media giant Fox, but podcasts on that network are branded only with “Wondery” branding. One typical example is Tides of History, a history podcast affiliated with Wondery. The series host, Patrick Wyman, has an academic background and hosted two other podcasts before joining the network. Other series on the network similarly have hosts who are primarily “podcasters”. Contrast this with, as an example, MSNBC podcasts, which include a series hosted by TV personality Rachel Maddow.192 As this study will address how podcast creators generate a financial return on their work, a focus on relatively more independent creators is appropriate. The next section will set out a narrow definition of podcast networks tailored to this focus. 4.3.1.1. Defining “Podcast Network” The concept of a “podcast network” requires some clarification in order to form the basis of this study’s selection process. Podcast networks are a means of mutual support and promotion for                                                  192 Note that there are other examples of podcasters with pre-existing brands from outside of podcasts. For example, Malcolm Gladwell hosted a series that was part of the Panoply network: see Joshua Benton, “Podcast Shakeup! Panoply, iHeartMedia, Stuff, and…Malcolm Gladwell? Are All Making Industry Moves”, NiemanLab (September 13, 2018), online: <www.niemanlab.org/2018/09/podcast-shakeup-panoply-iheartmedia-stuff-and-malcolm-gladwell-are-all-making-industry-moves/>. However, that podcast could not be considered a mere addition or marketing tool to another media property.  52  podcasters, and many provide advertising services as well. While some networks may initiate the development of new podcasts, many instead incorporate pre-existing podcasts which have already built an audience. For example, the comedy podcast network Maximum Fun includes series such as My Brother, My Brother, and Me and the Vancouver-based Stop Podcasting Yourself, both of which were started without a podcast network affiliation.193 Podcast networks are in some respects similar to the blog rings Li & Walejko suggest as a way to resolve sampling and population issues in content analysis studies of blogs.194 Like blog rings, a podcast network may include a number of podcasts which have voluntarily become part of the network. This provides a way to identify a set of series which may be quite loosely related aside from their association with the podcast network.  A definition of podcast networks may include some aspects that would be difficult to use in identification of a “podcast network”. The internal workings of a network will not be as easily ascertained as their outward branding. Two main functions of a podcast network are to provide its members with advertising sales support, and cross-promotion.195 Cross-promotion includes advertisements included in podcast episodes promoting shows on the same network, as well as a brand common to network shows that can include a decal on the image associated with the podcast series or a short audio stamp inserted near the beginning or end of each episode. But determining the extent to which a network provides ad sales or cross-promotion services can be difficult. Cross-promotion will be apparent from the inclusion of branding or advertisements for                                                  193 In some instances, podcasts may leave or switch podcast networks. One of the podcasts in the Patreon population, We Hate Movies, was part of a different podcast network before becoming affiliated with podcast network Headgum. 194 Skalski, Neuendorf & Cajigas, supra note 183 at 220–21, citing Dan Li & Gina Welejko, “Splogs and Abandoned Blogs: The Perils of Sampling Bloggers and Their Blogs” (2008) 11:2 Information, Communication & Society 279. 195 See Simon Owens, “How Podcasters Decide Whether to Join a Podcast Network”, online: <www.simonowens.net/how-podcasters-decide-whether-to-join-a-podcast-network>. 53  other shows on the same network. However, whether a network provides ad sales support may not be apparent from either the appearance or the substance of podcasts on the network, and may not be described on e.g., the podcast network’s website. It may be more straightforward to define what podcast networks are not. Series produced under the banner of a media organization are not members of a “podcast network” in the sense that term is usually used. For example, organizations such as the CBC, BBC, Slate, and others produce podcasts, but these are rarely branded as a “network”. Recognizing this distinction serves the objectives of this project, which include treating podcasts as a new, independent, self-sustaining medium.  The potential for podcasts to bring in revenue may grow as their listenership grows. If this happens, “traditional” media organizations may make them a more central part of their business. (The widespread production of podcasts by traditional media organizations may indicate that this is already starting to take place.) Eventually, these trends may make the distinction between “podcast network” and “media organization which produces podcasts” irrelevant. However, podcast networks are at this time sufficiently distinct to be studied independently, and provide a valuable organizing concept for this project. 4.3.2. Defining the Message Set Constraints on the researcher’s time and resources necessitated some method of selecting podcast episodes within each series selected for this study. The episodes which served as the source for metadata and audio were selected by time period. Performing content analysis on all episodes of each podcast series selected would have been infeasible. The selected series’ iTunes pages 54  display from 5 to 300 episodes, with a median of 146 and mean of 158.2 episodes.196 The message set consists only of freely-available podcasts. The study included episodes published over a three-week period, July 15, 2018 to August 5, 2018. This corresponds to an issue with payment which affected Patreon users and creators across the platform. In brief, a change in how payments were processed by Patreon led to many banking institutions flagging payments as suspicious conduct. This resulted in a large number of declined payments. Many Patreon creators saw a drop in income as a result of this event, evidenced in data drawn from Graphtreon for July and August 2018. By some accounts, however, the initial scale of this problem was worse than is reflected in the data. Affected users could manually resubscribe to Patreon accounts, and many did. Patreon communicated with users about the issue, but many podcasters on the platform felt the need to address the issue as well. Their shows were usually the most convenient venue in which to do so: they took the opportunity to explain the issue to their listeners, to ask them to resubscribe, and to thank those who had already done so.  Basing the selection of episodes around this series of events provided a richer source of data than a random sampling of episodes. The payment issue brought the “business” side of podcasts into salience for creators and listeners of crowdfunded podcasts. Many podcasters took the opportunity to reflect on the crowdfunding system, both on how it enables their work, and on the problems with it. This project is concerned with how financial support of podcasts works, and to what extent the copyright regime facilitates podcast creators earning a living off of their creative                                                  196 Note that the iTunes store pages may not reflect the full length of each series. Two series, All About Android and Hashtag Trending had a very limited number of episodes available on iTunes (10 and 5, respectively), while some series had episode numbering that implied a larger number of episodes than were featured on the iTunes store page. For example, the episodes from series Watch What Crappens and Uhh Yeah Dude were numbered 746–762 and 666–667, respectively. 55  work. The example of these payment issues may begin to show that the work copyright purports to do in supporting creators does not help creators in these contexts.  4.3.3. Why Not Study a Sample of All Podcasts? A content analysis study could take the entire population of podcasts as the population under study, allowing it to make broader claims and inferences about podcasting as a whole. However, drawing on podcast networks and crowdfunded podcasts as the population for this study is a more feasible approach than attempting to survey the entire set of all currently available podcasts. A fulsome survey of all podcasts would have to grapple with an immense number of series across multiple platforms. The initial search process would be complicated by the fact that there is no one comprehensive podcast directory. While Blubrry, mentioned above, claims a half million listed series, submission to that directory is voluntary. Further, some podcasts may be listed on no directory at all, relying instead on direct subscriptions to their RSS feed. For instance, the crowdfunding platform Patreon provides creators the tools to create a private RSS feed only accessible to Patreon donors that can contain either “premium” episodes of an otherwise freely-available podcast, or entire series. One way to resolve this issue would be to focus on a single directory. iTunes/Apple Podcasts is the most popular podcast listening platform. However, its discovery tools leave much to be desired for the empirical researcher: it does not provide a list of all podcasts, nor does it provide a full list for the 16 categories of podcasts listed on the platform. Instead, the interface provides a selection of (likely hand-picked) series, and the option to view more in a category which still does not provide a comprehensive list. It also has an overall “top podcasts” ranking which includes 300 shows. There may be merit in sampling the podcasts that are surfaced by the most popular podcast platform to get a picture of what podcast listeners are likely to see in this 56  context. However, this perspective does not line up with the objectives of this study, which are focused on podcast creators rather than audiences.  In contrast to iTunes/Apple Podcasts, the Blubrry directory is more amenable to random sampling. A list of podcasts can be accessed for each of 17 categories. These lists are displayed in pages of 24 podcasts, and while a total number of pages is not displayed for each list, this could be manually determined with some effort. This could be feasibly sampled by generating random numbers to select a number of podcasts. However, the number of podcasts on the platform (over 600,000) would require a very large sample.197 An initial test of sampling using the Blubrry directory also revealed a freshness problem: 4 of 7 podcasts found through random sampling had not released an episode in 2018. The time period of July–August 2018 would not be available to study for these podcasts. While there may be value in a study looking at older podcasts, the audio for these would not be available if the podcast’s web hosting service has lapsed.198 A methodology which included a large proportion of inactive podcasts would likely need to be shallower: without the podcast audio available, it could only consider the metadata which is preserved on directories.  Staying within the current methodology, an alternative would be to disregard podcasts for which the audio is no longer available and extend the research project to consider the question of why podcasts fail—that is, why podcast creators stop creating. The answer to this question could shed light on the role copyright plays (or doesn’t play) in enabling or hampering creativity in this                                                  197 See Neuendorf, supra note 6 at 90–91 (discussing the determination of sample sizes). 198 Podcast audio, like other data made available online, will become inaccessible if the internet service provider which hosts the material ceases to provide these services. As such, the continued availability of podcast audio usually requires a creator continuing to pay hosting fees, unless the hosting service is free. See R. Anthony Reese, “Super Bowl I, Jazz Radio, and the Glass Menagerie: Copyright, Preservation, and Private Copies” (2017) 51 Akron L Rev 1025 at 1061–63. 57  domain. However, not all podcasts which have not updated recently are necessarily defunct. They may be on hiatus, intentional or otherwise, or they may have moved to a different feed. Those that have actually failed would be difficult to identify, since it cannot be known whether the creator would eventually return to the podcast. The best indicator for failure may in fact be the lapse of podcast hosting, which as noted above would severely curtail the material available to analyze. In short, many complications would be introduced by including a large proportion of inactive podcasts. The question of why podcasts fail may be better explored through a different methodology, such as interviews with (former) podcast creators. 4.4. Study Implementation This research will draw on Neuendorf’s operationalization of content analysis methods.199 It will use three sources for messages: podcast series metadata, podcast episode metadata, and podcast episode audio. Skalski et al. recommend looking to metadata as useful in content analysis.200 Metadata contains content that can be analyzed using content analysis methods, as well as data that can be important for establishing context. With respect to podcasting, metadata includes episode descriptions, publication dates, and episode length and file size. Audio segments will look for “show business” segments (anything to do with running the podcast, particularly anything about financial support for the podcast).  4.4.1. Search Process and Selection Method The podcast series for this study were drawn from two sources: a list of the top 50 podcast Patreon pages and a list of podcast networks compiled based on those featured on the front page of iTunes. For the podcast networks, this included those directly featured as “providers” as well                                                  199 Supra note 174 at 118–25. 200 Ibid at 212 58  as indirectly through one of their member podcasts being featured on the iTunes front page. This is an exposure-based population because it is based on what is essentially a popularity statistic: whether a podcast network was featured on the front page of iTunes, and the number of patrons for Patreon-supported podcasts.201 Why “skim from the top” like this? Both being featured on iTunes and being a top Patreon podcast indicate a degree of success in implementing their respective business models: whether that is attracting paying patrons or getting (ad-supported) podcast episodes visibility. Ad numbers and listener numbers are not usually made public, so it is difficult to say where podcasts/podcast networks fall in terms of popularity and revenue. Comparing those “at the top” at least allows comparison of roughly, relatively similar segments within the respective populations. 4.4.2. Development of Codes The final codes for each of the two sets were generated before coding began. As Neuendorf notes, generating codes before the content is coded is preferable to generating codes simultaneously with the coding process. Entirely a priori coding (i.e., coding at the very outset of a project) is perhaps the most desirable from the perspective of scientific standards. However, developing the codes in the course of the project (but before coding begins) is appropriate in cases where “existing theory [and] research literature cannot give a complete picture of the message pool”.202 As discussed in an earlier chapter, research literature on podcasts is very limited, so it was necessary to first become familiar with the content under study.203 The process of code generation involved listening to podcasts in the population to help provide further                                                  201 See ibid at 79. 202 Neuendorf, supra note 6. 203 Ibid at 102. See section 1.1, above.  59  immersion into the “reality of the messages”.204 This will help to ensure that the coding categories chosen are appropriate. The codes used in this project are attached as an appendix.205  4.4.3. Coding Process The coding process included the audio content of the first and last five minutes of each podcast episode. Where content which satisfied one of the codes (“coded content”) continued beyond the five minute mark at the beginning of an episode, the episode was allowed to continue to play until just after the coded content ended to ensure that it was not followed up by further coded content. Similarly, if coded content was already underway at five minutes from the episode’s end, the coder listened to audio prior to the end mark to find the starting point of the coded content and listened to the content immediately preceding that. For example, a podcast episode in the series The Dollop had 13 minutes of coded content (primarily advertisements) preceding the show content.206 The coder listened to the full length of the segment consisting of advertising and other “show business” and enough of the following segment to be satisfied that it was not coded content. This ensured that the coding process did not miss additional coded content at the beginning or end of a podcast episode but outside the five minute windows at the beginning or end of the audio. (In most cases the non-coded show content began well before five minutes from the beginning of the episode and well after five minutes from the ending.) The codes and variables used for audio content were the same for both the podcast networks and crowdfunded podcast groups. The audio codes were identical across each category and in both podcast groups. These coded for whether the content referenced in the category (such as                                                  204 Neuendorf, supra note 6 at 108. 205 See Appendix A, below. 206 This episode’s ID was DOLL35. See Appendix B, below. 60  reference to a series’ Patreon page) was present at the beginning of an episode (1), end of the episode (2), both the beginning and end (3), or was absent (0). These codes are mutually exclusive. However, the variables are not: the same podcast segment could, for example, reference both the series’ donation page as well as bonus content available to donors.207 The codes and variables used for podcast episode metadata are identical for both the podcast networks and crowdfunded podcast groups. The metadata codes were identical across each category and in both podcast groups. These coded for whether the content referenced in the category (such as advertisement for a non-podcast related product or service) were present in the description metadata (1) or absent (0). As with the audio codes, the metadata codes are mutually exclusive while the variables are not. 4.4.4. Testing Reliability An important aspect of content analysis is testing intercoder reliability: whether, given one set of messages and the same set of codes by which to classify them, different coders will come to the same results. Unfortunately, this project lacks the resources to engage multiple coders, which will mean that intercoder reliability cannot be tested, and the objectivity-intersubjectivity of the study will be questionable. The researcher retested the coding on a sampling of the messages six weeks after the initial coding was completed. Section 6.3, below, contains a reliability analysis of these results. While intra-coder reliability can only test the stability of the results and not their reproducibility or accuracy, this limitation will be minimized by creating codes carefully and                                                  207 See Chapter 5, sections 5.2.1–5.2.2, 5.3.1, below.  61  avoiding ambiguity.208 Further, the data and study materials will be retained to allow for future testing which could incorporate the work of additional coders.                                                    208 See Klaus Krippendorff, Content Analysis: An Introduction to Its Methodology, 2nd ed (Thousand Oaks, CA: Sage Publications, 2004) at 215. 62  5. Results 5.1. Introduction This chapter will review the results of the study conducted for this thesis, which used content analysis methods as described in the Methodology chapter. The purpose of this study was to generate data on the use of crowdfunding, advertising, and copyright in the selected podcast series. Two types of content were analyzed: podcast episode audio and metadata. The metadata and audio content were taken from 95 podcast episodes in two sampled populations: the top 100 podcasting Patreons in the month of July 2018 and podcast series belonging to podcast networks identified from the iTunes front page in September and October 2018. Each episode was assigned a unique identifier to aid with organization of results. The coding was conducted in two rounds. Further information on the reliability of the coding is presented in section 6.3. Where there were discrepancies between the two rounds, the data presented in this chapter assumes that there were no false positives. That is, where only one round of coding found coded content of a specific type (codes 1–3), that result was reported rather than a null result (code 0). As the section on reliability will demonstrate, the number of discrepancies was low, so how discrepancies are resolved will not greatly affect the results. Further, only one coder was used in the coding process, suggesting that the discrepancies may be more attributable to missing coded content rather than differing understandings of what the variable in question means. Section 5.2, below, will present the results from the Patreon podcasts group, while section 5.3 will present the results from the podcast networks group. Subsections under these will cover references to donation pages (5.2.1, 5.3.1), discussion of non-podcasting activities (5.2.3, 5.3.2), 63  and incidence of advertising (5.2.4, 5.3.3) and references to intellectual property (5.2.5, 5.3.5), among other coded content. Section 5.4 will conclude this chapter with the results of a brief survey of the iTunes pages for each podcast series in the data set and the copyright notices or licenses displayed there. 5.2. Crowdfunded (Patreon) Podcasts Group A set of 20 podcast series were sampled from the top 50 podcast series as listed on Graphtreon for the month of July 2018 using a randomly generated sequence of numbers from 1–50.209 Graphtreon is a 3rd party service which collects and presents subscriber and financial data made publicly available by Patreon.210 Of the set of 20 series, 3 were removed from the sample for failing to meet the study’s specifications: two podcast series were non-English language (French and Portuguese), while a third was a video podcast. Two further series had to be removed from the set because their archives were unavailable for study. The remaining 15 podcast series were coded using the process described in chapter 4, section 4.4.3.                                                    209 The random sequence of numbers was generated through an online tool: https://www.random.org/sequences/. The first 20 numbers of the sequence were used to generate the sample by selecting those podcasts as ranked in the Graphtreon top 50. See Graphtreon, supra note 46. These were the first 20 numbers of the generated sequence used for the sampling process: 37, 38, 3, 32, 42, 28, 35, 16, 41, 36, 18, 1, 46, 34, 26, 19, 2, 47, 10, 8. 210 These data are made available for research purposes if the source is attributed. See Graphtreon, “Terms of Service”, online: <graphtreon.com/terms>. 64  Table 1. Sampled Patreon Podcast Series with Exclusions Graphtreon Rank, July 2018 (Selected through random sequence) Podcast Series Title (Patreon Title, where different) Included? (Reason for exclusion) No. of episodes in set 1 Chapo Trap House211 Yes 5 2 Sword and Scale Yes 2 3 Second Captains No (archives unavailable) - 8 The Ralph Report (Ralph Garman) Yes 1 10 The Liturgists Yes 1 16 The Dick Show (Dick Masterson) Yes 3 18 Mike Ward Sous Écoute No (French-language) - 19 #ResistanceLive No (video-only podcast) - 26 The Best Show Yes 3 28 Hello Internet Yes 2 32 The Dollop Yes 3 34 Friends at the Table Yes 3 35 Uhh Yeah Dude Yes 2 36 Rui Unas No (Portuguese-language) - 37 Philosophize This! (stephen west) Yes 1 38 Rob Cesternino No (archives unavailable) - 41 Lore (Aaron Mahnke) Yes 1 42 The Generation Why Podcast Yes 6 46 Talking Simpsons Yes 4 47 Watch What Crappens Yes 16                                                  211 Disclosure: The author is a contributor to this series’ Patreon. This played no role in its selection for this study, which was based on random number selection.  65  A total of 55 episodes were coded from the crowdfunded group. All podcast series surveyed had at least 1 episode in the relevant timeframe (July 15–August 5, 2018), averaging 3.667 episodes and with a median of 3 episodes.  Two of the surveyed episodes did not contain regular podcast content. One of these consisted of a short (1 minute and 1 second) promotion for an upcoming live podcast event.212 The other irregular episode was a ten minute advertisement consisting of a preview for a different podcast series. These two episodes were coded and included in the data set.213 5.2.1. References to Patreon or Other Donation Pages This category coded for references to Patreon or other donation pages associated with the podcast series. For the purposes of this category, references to bonus episodes without explict reference to the donation page were not counted.  Table 2. References to Patreon or Other Donation Pages Podcast Series Title  No. of episodes with reference in audio No. of episodes with reference in metadata No. of episodes in set Chapo Trap House 1 (16.7%) 0 6 Sword and Scale 2 (100%) 0 2 The Ralph Report 1 (100%) 1 (100%) 1 The Liturgists 1 (100%) 0 1 The Dick Show 0 0 3 The Best Show 1 (33.3%) 0 3 Hello Internet 0 2 (100%) 2 The Dollop 0 0 3 Friends at the Table 2 (66.7%) 0 3 Uhh Yeah Dude 0 2 (100%) 2                                                  212 This episode’s ID was SIMP29. 213 This episode’s ID was WHY20. 66  Podcast Series Title  No. of episodes with reference in audio No. of episodes with reference in metadata No. of episodes in set Philosophize This! 1 (100%) 0 1 Lore 0 1 (100%) 1 The Generation Why Podcast 0 0 6 Talking Simpsons 3 (75%) 3 (75%) 4 Watch What Crappens 17 (100%) 7 (43.8%) 17 Total 29 (52.7%) 16 (30.2%) 55 A slim majority (52.7%) of surveyed episodes included at least one reference to the series’ Patreon page. If the outlier Watch What Crappens (which had a relatively high number of episodes) is excluded, the total for references in audio to a donation page drops to 12 (31.6% of 38 episodes). Nine series of fifteen (60%) had at least one reference to donation pages in their coded audio segments. 5.2.2. References to Additional Content Available to Donors This category coded for references to additional content such as bonus podcast episodes available to listeners who donate to the series. This category was understood broadly, to encompass references in passing to bonus content that did not explicitly present the bonus content as an incentive to donate, or explicitly direct the listener to the donation page. (Where the metadata or audio also directed the listener to the series’ donation page, the episode would also been coded for the “References to Patreon or Other Donation Page” category.) 67  Table 3. References to Additional Content for Supporters Podcast Series Title  No. of episodes with reference in audio No. of episodes with reference in metadata No. of episodes in set Chapo Trap House 1 (16.7%) 0 6 Sword and Scale 2 (100%) 0 2 The Ralph Report 1 (100%) 0 1 The Liturgists 0 0 1 The Dick Show 1 (33.3%) 0 3 The Best Show 1 (33.3%) 0 3 Hello Internet 0 0 2 The Dollop 0 0 3 Friends at the Table 2 (66.7%) 0 3 Uhh Yeah Dude 0 0 2 Philosophize This! 0 0 1 Lore 0 1 (100%) 1 The Generation Why Podcast 0 0 6 Talking Simpsons 3 (75%) 0 4 Watch What Crappens 5 (29.4%) 7 (43.8%) 17 Total 16 (29.1%) 8 (14.5%) 55 Nine series (60% of the sample) included some reference to additional content available for listeners who supported the podcast through Patreon in either episode audio or metadata.  5.2.3. Discussion of Non-Podcasting (Secondary) Activities This category coded for references to non-podcasting activity associated with the podcast series, such as merchandise, live events, or books (“secondary activities”).  68  Table 4. Discussion of Non-Podcasting Activities Podcast Series Title  No. of episodes with reference in audio No. of episodes with reference in metadata No. of episodes in set Chapo Trap House 3 (50%) 3 (50%) 6 Sword and Scale 1 (50%) 0 2 The Ralph Report 1 (100%) 0 1 The Liturgists 1 (50%) 0 1 The Dick Show 2 (66.7%) 0 3 The Best Show 0 0 3 Hello Internet 0 1 (50%) 2 The Dollop 3 (100%) 2 (66.7%) 3 Friends at the Table 2 (66.7%) 0 3 Uhh Yeah Dude 0 0 2 Philosophize This! 0 0 1 Lore 1 (100%) 0 1 The Generation Why Podcast 1 (16.7%) 0 6 Talking Simpsons 3 (75%) 1 (25%) 4 Watch What Crappens 11 (68.8%) 8 (50%) 17 Total 29 (52.7%) 15 (27.3%) 55 Examples of secondary activities in the sampled group included the sale of branded clothing (Watch What Crappens), books (Lore, Chapo Trap House), and music featured on the podcast (Friends at the Table), as well as live shows or meet-ups (several series).214 Twelve series (80% of the sample) included at least one example of promotion of non-podcasting activity in episode audio or metadata. 5.2.4. Advertisements This category coded for advertisements for products and services not affiliated with the podcast series, and excluding advertisements for other podcasts. Excluding products and services                                                  214 The series Chapo Trap House, Generation Why, Hello Internet, The Liturgists, Ralph Report, Talking Simpsons, The Dollop, and Watch What Crappens all advertised upcoming live events or meetups.  69  affiliated with the podcast series was necessary to differentiate this category from the discussion of merchandise, live shows, and other non-podcasting activity documented in the previous section. Promotion of other podcast series may come from affiliation with a podcast network or with podcast guests promoting their own series. Differentiating between paid and non-paid advertisements is difficult in this context. On the other hand it is generally clear when a podcaster is advertising a non-podcast product or service that they are being paid to do so. Table 5. Advertisements for Products and Services not Affiliated with a Podcast (Patreon Set) Podcast Series Title  No. of episodes with ads in audio No. of episodes with ads in metadata No. of episodes in set Chapo Trap House 0 0 6 Sword and Scale 0 0 2 The Ralph Report 0 0 1 The Liturgists 0 0 1 The Dick Show 0 0 3 The Best Show 0 0 3 Hello Internet 0 2 (100%) 2 The Dollop 3 (100%) 0 3 Friends at the Table 0 0 3 Uhh Yeah Dude 0 0 2 Philosophize This! 0 0 1 Lore 1 (100%) 1 (100%) 1 The Generation Why Podcast 4 (66.7%) 0 6 Talking Simpsons 0 3 (75%) 4 Watch What Crappens 0 0 17 Total 8 (14.5%) 6 (10.9%) 55 Five series (33.3% of the sample) included some advertisements.  70  5.2.5. Intellectual Property None of the surveyed podcast episodes in this group included license terms in the podcast description. Further, none of the podcast audio surveyed included discussion of intellectual property issues. 5.3. Podcast Networks Group For the reasons discussed in the Methodology chapter, section 4.3.1.1, this study used a restrictive definition of “podcast network”. This definition included excluded any entities whose branding directly associated them with traditional or digital media enterprises or which do not self-describe as “podcast networks” on their websites. A list of providers was compiled based on the iTunes front page on two dates (September 11, 2018 and October 1, 2018).215 This list included providers listed under “Featured Providers” as well as the providers listed for specific podcast series featured on the iTunes front page.  Table 6. List of Podcast Providers Derived from iTunes Front Page Podcast Provider Fits definition of “podcast network”? Included? (Reason for exclusion) ABC News No - American Public Media (APM) No - Audible No - BBC No - Bloomberg No - Boing Boing No - BuzzFeed No - Canadaland Yes Yes Castbox/Studio71 No - CBC No - CBS News No - cnet No -                                                  215 There appeared to be little difference in the podcast networks featured on the iTunes front page on each date. For example, on the iTunes “Featured Provider” list, one provider (Comedy Central) was removed while another (Curiouscast) was added between the September and October surveys. Otherwise, the lists were identical.  71  CNN No - Comedy Central No - Community Foundations of Canada No - Curiouscast Yes No (Content primarily radio shows) Entertainment Weekly No - ESPN No - Gimlet Yes Yes Global News No - Headgum Yes Yes Indiefeed Yes No (No website listing podcast series) IT World Canada Yes Yes LA Times No - Mondo No - NBC News Podcasts No - Nerdist Yes Yes New York Times No - NHL No - NPR No - Panoply Yes Yes PBS No - PRI No - Radio-Canada.ca No - Slate No - Sportsnet No - Stitcher No216 - TED No - The Atlantic No - The New York Times No - The New Yorker No - The Verge No - The Week No - TSN No - tvo No - TWiT.tv Yes Yes Wall Street Journal No - WBUR & the Boston Globe No -                                                  216 Stitcher is unique in that its business does appear to be mostly podcast-based, but is not self-described as a podcast network: see “Stitcher”, online: <https://www.stitcher.com/>. Further, even in the absence of this exclusion criterion this study could not have included Stitcher Premium series which require a paid subscription to access. This fact also reflects that while many podcasts are available for free, there are exceptions. 72  Wizzard Media Yes No (No website listing podcast series) WNYC No - WNYC Studios No - Wondery Yes Yes Two podcast networks were excluded because they lacked a web page listing their associated podcast series. These pages were necessary to generate a list of associated podcast series to sample for the study. One podcast network was excluded because its associated series were largely composed of repackaged radio shows.217  A list of podcast series was developed by reviewing the “series” or “shows” page on each of the eight remaining podcast networks’ web sites. This process found a total of 300 podcast series, which form the population for this part of the study. A sample of this population was created by numbering the podcast series in alphabetical order and generating a sequence of 20 random numbers from 1–300.218                                                  217 This is a justifiable grounds for exclusion because this study was intended to address podcasts. 218 The random sequence of numbers was generated through an online tool: https://www.random.org/sequences/. This was the same process as described in note 209, above. These were the first 20 numbers of the generated sequence used for the sampling process: 170, 92, 196, 10, 15, 253, 110,  31, 299, 25, 43, 73, 177, 200, 28, 122, 262, 74, 242, 173 73  Table 7. Sampled Podcast Network Series Index value (arbitrarily assigned) Podcast Series Title (Podcast Network) No. of episodes in set 10 Accused (Wondery) 0 15 All About Android (TWiT.tv) 3 25 Battle Scars (Panoply) 0 28 Bizarre States (Nerdist) 4 31 Breakdown (Panoply) 0 43 Commons (Canadaland) 0 73 Found (Wondery) 0 74 Foreign Policy: The ER (Panoply) 3 92 Hashtag Trending (IT World Canada) 15 110 Inc. Uncensored (Panoply) 3 122 Lady Lovin’ (Headgum) 3 170 Sandra (Gimlet) 0 173 Secrets, Crimes & Audiotape (Wondery) 0 177 Serial Serial (Panoply) 0 196 TBD with Tina Brown (Wondery) 0 200 Tempest (Panoply) 0 242 This is War (Wondery) 1 253 True Crime All the Time (Wondery) 5 262 Uncivil (Gimlet) 0 299 You Must Remember This (Panoply) 3  Total 40 A slim majority (11/20) of the sampled podcast series released no episodes in the selected period of time (July 15–August 5, 2018). This issue is addressed in detail below in section 6.2.3. Only episodes from the remaining eight series were included in the coding process. While the sample of 20 series included series on each of the 8 podcast networks, Gimlet and Canadaland were unrepresented in the set of 8 podcast series which had episodes in the relevant timeframe. 74  5.3.1. References to Donation Pages This category coded for references to donation or other premium membership pages associated with the podcast series.  Table 8. References to Donation or Membership Pages in Podcast Network Series Podcast Series Title (Podcast Network) No. of episodes with reference in audio No. of episodes with reference in metadata No. of episodes in set All About Android (TWiT.tv) 0 0 3 Bizarre States (Nerdist) 2 (50%) 0 4 Foreign Policy: The Editor’s Roundtable (Panoply) 0 0 3 Hashtag Trending (IT World Canada) 0 0 15 Inc. Uncensored (Panoply) 0 0 3 Lady Lovin’ (Headgum) 0 0 3 This is War (Wondery) 0 0 1 True Crime All the Time (Wondery) 4 (80%) 4 (80%) 5 You Must Remember This (Panoply) 0 0 3 Total 6 (15%) 4 (10%) 40 In the podcast network set of podcast series, only the series True Crime All the Time had a Patreon page. The series Bizarre States included references to a podcast network-wide membership program. These were the only two series which included references to donation or membership pages. 5.3.2. Discussion of Non-Podcasting (Secondary) Activities This category coded for references to non-podcasting activity associated with the podcast series, such as merchandise, live events, or books.   75  Table 9. Discussion of Non-Podcasting Activities in Podcast Network Series Podcast Series Title (Podcast Network) No. of episodes with reference in audio No. of episodes with reference in metadata No. of episodes in set All About Android (TWiT.tv) 0 0 3 Bizarre States (Nerdist) 3 (75%) 1 (25%) 4 Foreign Policy: The Editor’s Roundtable (Panoply) 1 (33.3%) 0 3 Hashtag Trending (IT World Canada) 1 (6.7%) 0 15 Inc. Uncensored (Panoply) 0 0 3 Lady Lovin’ (Headgum) 1 (33.3%) 0 3 This is War (Wondery) 0 0 1 True Crime All the Time (Wondery) 0 0 5 You Must Remember This (Panoply) 3 (100%) 0 3 Total 9 (22.5%) 1 (2.5%) 40 The Bizarre States series included discussion (and audio) of live podcast events, while the You Must Remember This series included references to a book authored by the host of that series.219 Hashtag Trending and Foreign Policy: The Editor’s Roundtable both included references to articles on their respective websites, while an episode of Lady Lovin’ included references to a product line affiliated with one of the hosts. 5.3.3. Advertising This category coded for advertisements for products and services not affiliated with the podcast series, and excluding advertisements for other podcasts. For further elaboration of this code see section 5.2.4, above.                                                  219 Karina Longworth, “Seduction: Sex, Lies, and Stardom in Howard Hughes’s Hollywood” (New York: HarperCollins, 2018).  76  Table 10. Advertising in Podcast Network Series Podcast Series Title (Podcast Network) No. of episodes with ad(s) in audio No. of episodes with ad(s) in metadata No. of episodes in set All About Android (TWiT.tv) 3 (100%) 3 (100%) 3 Bizarre States (Nerdist) 0 0 4 Foreign Policy: The Editor’s Roundtable (Panoply) 1 (33.3%) 0 3 Hashtag Trending (IT World Canada) 0 0 15 Inc. Uncensored (Panoply) 1 (33.3%) 0 3 Lady Lovin’ (Headgum) 3 (100%) 0 3 This is War (Wondery) 1 (100%) 0 1 True Crime All the Time (Wondery) 0 1 (20%) 5 You Must Remember This (Panoply) 0 0 3 Total 9 (22.5%) 4 (10%) 40 Six series (66.7%) included at least one example of advertising, with five series (55.6%) including ads in the show audio and two (22.2%) including ads in the metadata. 5.3.4. Podcast Network Promotion and Cross-Promotion This category coded for references to the podcast network which each podcast series belonged to, as well as any cross-promotion of other podcasts on the same network. 77  Table 11. Podcast Network Promotion and Cross-Promotion Podcast Series Title (Podcast Network) No. of episodes with references to podcast network in audio No. of episodes with references to other shows on podcast network in audio No. of episodes in set All About Android (TWiT.tv) 3 (100%) 0 3 Bizarre States (Nerdist) 4 (100%) 0 4 Foreign Policy: The Editor’s Roundtable (Panoply) 3 (100%) 0 3 Hashtag Trending (IT World Canada) 15 (100%) 0 15 Inc. Uncensored (Panoply) 3 (100%) 0 3 Lady Lovin’ (Headgum) 3 (100%) 0 3 This is War (Wondery) 1 (100%) 0 1 True Crime All the Time (Wondery) 1 (20%) 1 (20%) 5 You Must Remember This (Panoply) 1 (33.3%) 0 3 Total 35 (87.5%) 1 (2.5%) 40 Every series and nearly every episode in this set included a reference to the podcast network to which the series belonged. These came in the form of references to the network voiced by the host, or, commonly, an audio “sting” at the beginning or end of the podcast episode audio. In contrast, only one episode of one series included cross-promotion of a different series on the same network. This episode consisted entirely of an advertisement for a different series, and was, coincidentally, the same advertisement as was featured in one of the Patreon series.220                                                  220 The episode from the podcast network set with the ID TRUE89 consisted of an advertisement for the Wondery series I, Survivor. This was mostly identical to the episode from the Patreon set with the ID WHY20. The difference between these two episodes was the introduction, which in each case was voiced by the regular series host(s).  78  5.3.5. Intellectual Property None of the surveyed podcast episodes in this group included license terms in the podcast description. One episode surveyed included discussion of intellectual property issues in the podcast audio and description; however, this was unrelated to copyright or podcasting.221  5.4. Podcast Series Copyright Notices and Licenses  While there were no license details or copyright notices on any of the individual episodes surveyed, some of the series did include these on their respective iTunes pages. In the desktop version of iTunes, these licenses and notices appeared below the podcast art on the left side of the screen. Table 12. Copyright Notices and Licenses on iTunes Store Pages Podcast Series Title Sample Set Copyright Notice/License (iTunes) The Best Show Patreon © Tom Scharpling Chapo Trap House Patreon © All rights reserved The Dick Show Patreon none Friends at the Table Patreon © Friends at the Table Copyright 2015. All rights reserved. The Generation Why Podcast Patreon © Copyright 2018 The Generation Why Podcast LLC Hello Internet Patreon none The Liturgists Patreon © Copyright 2016 All rights reserved Lore Patreon © 2016 Aaron Mahnke Philosophize This! Patreon none The Ralph Report Patreon © All rights reserved Sword & Scale Patreon © Copyright © Incongruity Media 2013-2018 Talking Simpsons Patreon none The Dollop Patreon © Dave Anthony Uhh Yeah Dude Patreon none Watch What Crappens Patreon none Lady Lovin' Podcast Network © Lady Lovin' Hashtag Trending Podcast Network none Bizarre States Podcast Network © 2014 Nerdist Industries Inc. Uncensored Podcast Network none                                                  221 Episode HASH59 featured a news item about a patent case involving IBM and Groupon. 79  You Must Remember This Podcast Network none All About Android Podcast Network © This work is licensed under a Creative Commons License - Attribution - NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/ This Is War Podcast Network none True Crime All The Time Podcast Network none Foreign Policy: The Editors' Roundtable Podcast Network © Copyright 2018 FP Group All Rights Reserved In total, 13 of 24 series (54.2%) included some copyright information on their iTunes store page. Five (20.8%) asserted “All Rights Reserved”, one (4.2%) was licensed under a Creative Commons attribution license, and seven series (29.2%) were generic notices including only a name and in some cases a year.     80  6. Discussion 6.1. Introduction This chapter will discuss the results of the study described in the previous chapter with reference to the theoretical frame laid out in Chapter 2. This frame is based around contrasting the concept of “abundance models”, in which creators make their work available without relying on excluding non-paying users, with that of the scarcity model, which does rely on exclusion. The scarcity model is central to the influential utilitarian account of copyright. This chapter will consider the extent to which the podcasts surveyed for the study fit into abundance models. While the assumption that many podcasts are available for free was borne out by the study, the results paint a more complicated picture of how “abundance models” might work in practice. Section 6.2 and its subsections below will address several issues which were encountered in the course of the study and in interpreting the results. These issues include the possible omission of some advertisements from the sampled content (section 6.2.1) and the absence of episodes in the relevant timeframe for some of the podcast network series (6.2.3). The reliability of the study results will be evaluated in section 6.3, where two measures of reliability—percentage agreement and Cohen’s kappa—will be introduced. Finally, section 6.4 will consider findings in three broad areas—crowdfunding (6.4.1), advertising (6.4.2), and non-podcasting activities (6.4.3)—and discuss how the results fit with Suzor’s abundance model framework and the assumptions made for this study’s methodology.  81  6.2. Issues 6.2.1. Issues with Detecting the Presence of Dynamic Ads During the initial coding process, segments from one of the coded series were observed to contain references to advertisement breaks that did not occur in the excerpted audio.222 These references indicate that dynamic ads were apparently meant to be included in these segments, but were omitted from the audio.223 While this was observed in only one series, it is possible that dynamic ads were omitted in other series as well where the host(s) did not explicitly introduce an ad break.  A possible reason for these omissions is the way the podcast episodes were accessed: episodes were downloaded to a personal computer, and where possible directly from the URL for the episode audio file included in the RSS feed.224 This method treated the series RSS feed as the authoritative source for each podcast, unless (as in two cases225) the RSS feed did not include episodes going back to the relevant time period. In these cases, the podcast file was downloaded from the podcast series’ web site. This was done for reasons of workflow and to preserve an archive of downloaded episodes. Since the RSS feeds used were the same feeds provided to iTunes, it was assumed that the podcast episodes would be identical to those which could be                                                  222 The episode in which this was observed had the ID INC78.  223 Dynamic ads, discussed briefly in section 4.2.1, above, are not included by the podcast creator when a podcast episode is compiled and published. Rather, they are inserted programmatically when the episode is accessed by the user. See Kurt Kaufer, “The Podcast Ad Playbook: Baked-In Versus Dynamic Insertion Ads Explained”, Forbes Agency Council (December 1, 2017), online: <www.forbes.com/sites/forbesagencycouncil/2017/12/01/the-podcast-ad-playbook-baked-in-versus-dynamic-insertion-ads-explained/#1a84a5d11fe8>. These ads may therefore be targetted based on whatever user data is available to the ad provider, similar to ads on web pages. They therefore present similar difficulties as those described in section 4.2.1, above with respect to the unfixed nature of some online content. 224 “URL” is an abbreviation of “Uniform Resource Locator”. A URL specifies the location of a resource (such as an audio file) on the web. See Web Hypertext Application Technology Working Group, “URL Standard”, online: <https://url.spec.whatwg.org/>. 225 The podcast series All About Android and Hashtag Trending did not include older episodes in their RSS feeds or on their iTunes page.   82  accessed through that platform. To test this assumption, the second round of coding used, wherever possible, podcast episodes downloaded through iTunes.226  The second round of coding confirmed that some episodes had dynamic ads which were not included in the files obtained for the initial round of coding. Five episodes in two series (four in Generation Why and one in Inc. Uncensored) included ads in the second round of coding but not the first. However, one series (Lady Lovin’) displayed the same phenomenon in reverse: the round of coding using iTunes did not detect any ads, while the first round of coding did detect ads. The coder confirmed that the audio was different between the two versions, and that ads were included in the audio used for the first round that were not included in the audio used for the second round.  It is unclear whether this issue was caused by the method by which the podcast episodes were downloaded or by some other factor (such as the approximately 6-week difference in time between downloads). Further, the unique presence of dynamic ads in both the episodes downloaded directly and those downloaded through iTunes means it may not be possible to identify a “definitive” version of a podcast episode. This problem evokes the discussion of digital media as a “moving target” in section 4.2.1, above, and should warrant more attention in future research on podcast advertising.227                                                  226 For this process the coder used iTunes software version 12.9.3.3 running on a Windows 10 PC. Using iTunes was not possible where the iTunes feeds did not include episodes from the relevant time period, or where episodes were missing. This issue precluded using iTunes to recode the series All About Android and Hashtag Trending, and episode TRUE89. 227 See Skalski, Neuendorf & Cajigas, supra note 183 at 203. 83  6.2.2. No Discussion of Patreon Payment Issues None of the series in the Patreon set (or the podcast network set) included discussion of the Patreon payment issues which were described in section 4.3.1 and which informed the choice of timeframe for this study. This finding does not necessarily imply that podcast creators did not inform their audience about these issues, as they may have used other channels of communication to do so. These channels may have included messages sent through the Patreon platform or discussion on bonus episodes unavailable to non-paying listeners, neither of which were available for study. These channels would have the benefit to creators of only going out to paying listeners and not to non-paying listeners who would not have been affected. 6.2.3. Series in the Podcast Network Set Inactive during Relevant Timeframe A majority of the podcast series sampled for the podcast network set included no episodes in the relevant three-week timeframe, July 15, 2018–August 5, 2018. In contrast, every Patreon series had episodes in this timeframe.228 It is not immediately clear from the coded results whether the podcasts which did not release episodes in this timeframe were no longer producing episodes, on hiatus, or published episodes at a low frequency (i.e., less than every three weeks). While the first two of these explanations would be an issue for any study based on a single timeframe, the latter may have been alleviated had the chosen timeframe been longer. However, further research indicated that the three-week timeframe was likely adequate to capture series which were regularly releasing episodes at that time. The table below lists each of the 11 podcast series excluded from the dataset because of the absence of any episodes during the relevant timeframe, and the dates of episodes coming before and after the timeframe.                                                  228 See section 5.2, above. Note however that archives were unavailable for some of the series in the Patreon set and it is possible that these series were also inactive during this time period. 84  Table 13. Recent Updates of Inactive Podcast Network Series (as of January 14, 2019)  Series Title Last Episode before July 15, 2018 Next Episode after August 5, 2018 Additional Notes Accused June 19, 2018 September 24, 2018 (advertisement) No regular episodes after June 19, 2018. Battle Scars December 20, 2017 -  Breakdown April 9, 2018 -  Commons June 19, 2018 October 2, 2018  Found January 23, 2018 (advertisement) November 8, 2017 (regular episode) -  Sandra April 18, 2018 October 17, 2018 (advertisement) Seven episodes of this podcast were released on April 18, 2018. No further regular episodes after this date. Secrets, Crimes & Audiotape December 5, 2017 January 9, 2019 (advertisement) No regular episodes after December 5, 2017. Serial Serial May 30, 2017 -  TBD with Tina Brown - November 27, 2018  Tempest May 7, 2018 -  Uncivil February 7, 2018 (advertisement) December 27, 2017 (regular episode) November 9, 2018  These results indicate that most of the series which were excluded for an absence of episodes during the relevant timeframe were either on a long break or were no longer publishing regular episodes as of July 15, 2018. Only three of the eleven series had regular episodes published after the relevant timeframe as of January 14, 2019. Some of the series published advertisements in their feed, as indicated in the table above. In this context, “advertisement” means that an episode was published on the podcast series feed that was determined to be a promotion for a different series based on the title and description. These advertisements are distinct from regular podcast 85  episodes in that they do not generally contain new show content and are published to direct listeners to an entirely different podcast series.229 (Two of the podcast episodes coded for this study were similar advertisements, and are discussed above in sections 5.2 and 5.3.4.230) Further, one series (TBD with Tina Brown) had no episodes available before the timeframe, and seems to have started in November 2018. The above offers a point of comparison between the podcast network and Patreon sets. The Patreon series all appeared to publish regularly, given the presence of one or more episodes in the relevant timeframe for each series included in the data set. In contrast, several of the podcast network series appear to be limited run or published on a seasonal schedule. For example, the series Sandra included seven episodes released on April 18, 2018 which were numbered 1–7. This implies that the episodes were pre-recorded and released all at once. The Serial Serial podcast, “a podcast about a podcast”, had its release schedule tied to the release of other podcast series with discrete seasons.231 The release schedule for Patreon podcasts may be tied to the way creators on that platform raise funds: since some Patreon pages are configured to charge donors on a monthly basis, the platform creates an expectation of continuous releases of content. 6.2.4. Overlap between Data Sets There was a degree of overlap between the data sets in that some of the series in each set may have qualified for inclusion in the other set had the method of population selection been different. For example, two podcasts in the Patreon podcasts set were members of podcast                                                  229 This is a way for podcast providers to reach a podcast’s audience who may remain subscribed even after the series has concluded or while it is on break. 230 The episodes in question had the IDs WHY20 and TRUE89. See Appendix B. 231 See “The Serial Serial”, iTunes, online: <https://itunes.apple.com/us/podcast/the-serial-serial/id942568327>. This series has episodes on two other podcast series, Serial and its spinoff S-Town.  86  networks identified for the podcast network set.232 Other series in this set alluded to being members of podcast networks that were not included in the podcast network set.233 One series in the podcast network set had a Patreon page, while another had a network-wide membership program.234  This finding is not in substantial conflict with the abundance and scarcity model framework which this project adopts from Suzor.235 The different abundance models Suzor identifies are not necessarily exclusive. For example, several of the podcast series in the Patreon data set included advertising. While the Patreon data set is notionally connected to the abundance model of crowdfunding, this does not mean that creators who rely on crowdfunding cannot also supplement their income from creative work with advertising (a different abundance model). 6.2.5. Inclusion of Podcast Series associated with Other Media Two series in the podcast networks sample were associated with other media publications: Inc. Uncensored (associated with the digital media publication Inc.), and Foreign Policy: The Editor’s Roundtable (associated with the magazine Foreign Policy). In the selection of podcast networks, networks directly associated with a non-podcast media organization were excluded. The reasons for this choice were discussed in sections 4.3.1.1 and 5.3, above. Briefly, the definition of a “podcast network” used for this study emphasizes a reliance on podcasting as an independent business. Further, “providers” on iTunes include many traditional media organizations. As such, a broad exclusion for these providers was justified on theoretical and                                                  232 These series were Generation Why and Sword and Scale (both members of the podcast network Wondery). 233 An episode of the series The Dollop identified itself as being part of the All Things Comedy podcast network, while the series Talking Simpsons and The Liturgists were identified as being part of podcast networks named after the series themselves. 234 This series, Bizarre States, referenced the podcast network Nerdist’s membership program.  235 See section 2.3, above. 87  practical grounds. However, the same exclusion criterion was not applied to individual series, and the two podcast series identified above were included in the coding process. This choice is justifiable for practical and theoretical reasons. As a practical matter, the inclusion of two podcast series to the sample did not entail much extra work, and could be excluded from subsequent analysis if necessary. The list of podcast providers gathered from the iTunes front page included over 50 providers, the majority of which were clearly associated with existing traditional or digital media publications or other organizations.236 An alternative to applying a blanket exclusion to those providers was to evaluate each on a case-by-case basis using some criteria for “podcast network”. While it is possible that some of these providers would have met these criteria, case-by-case evaluation would have been time consuming and (depending on the criteria) potentially error-prone.  With respect to the theoretical aims of this project, the concept of a “podcast network” is a different concept than “podcast affiliated with a podcast network”. Taking a stringent approach to classifying “podcast network” to identify providers which are close to the core of that concept does not necessitate using the same criteria at the level of individual podcasts. The information necessary to determine the precise relationship between an individual podcast and the podcast network it belongs to (such as financial support and arrangements around advertising) may not be publicly-available. Where series on a podcast network are associated with other media ventures as well, this uncertainty is compounded. Restricting the scope of “podcast network” served the theoretical aims of this project because it limited the population to series associated with providers that are first and foremost in the business of podcasts. Excluding podcast network                                                  236 See Table 6, above. 88  series which are also associated with other media would require assumptions about those series’ relationships that would need to be justified further. 6.3. Reliability As noted above in section 4.4.4, resource constraints meant that this study could not be tested by multiple coders. Testing inter-coder reliability is an important part of content analysis studies, and this study’s lack of inter-coder testing likely places it outside the bounds of a strict definition of “content analysis study”.237 However, the re-coding process allows the evaluation of intra-coder reliability; that is, the consistency of results between coding of the same content by the same coder after some time (in this case, six weeks) had passed.238 This section will evaluate the intra-coder reliability of the study based on the results of the two coding processes. 6.3.1 Measures of Reliability This section will evaluate the percent agreement and Cohen’s kappa for the intra-coder reliability of this study, using the results of the two coding processes. “Percent agreement” is a measure of whether “coders agree on the precise values assigned to cases on a given variable” and appropriate for studies for “categorical” measures where “each pair of coded values is either a hit or a miss.”239 In this study, podcast audio and metadata were coded for the presence or absence of certain types of content. The codes used indicated whether the coded content was present at the beginning or end of the episode audio, both, or neither. This measure is therefore clearly categorical, as the reported codes between the first and second coding either matched or did not.                                                   237 See Neuendorf, supra note 6 at 166–67. 238 See ibid at 165. 239 See ibid at 174. 89  Other measures for testing inter-coder reliability of content analysis studies take into account the possibility of chance agreement—the extent to which agreements between two coders could be attributed to mere chance. Correcting for chance also has the advantage over percentage agreement of allowing comparison of observed agreement between studies.240 Although Neuendorf notes that there is some disagreement about the effect of chance on coding, the Content Analysis Guidebook recommends presenting percent agreement paired with a measure that takes chance agreement into account, such as Cohen’s kappa.241 This measure typically ranges from “.00 (agreement at chance level) to 1.00 (perfect agreement)” with values below .00 indicating a level of agreement less than would be attributable to chance.242 Neuendorf further notes that Cohen’s kappa is reported to be the “most widely used reliability coefficient (after simple percent agreement)” despite concerns about its stability and interpretability.243  Noting that the Guidebook does not provide guidance on whether the same considerations are applicable for testing intra-coder reliability as for inter-coder reliability, this study will employ percent agreement along with Cohen’s kappa for the sake of comparability. Future research on this data could expand to other reliability measures. 6.3.2. Calculating Reliability for This Study The formula to calculate percent agreement as described by Kimberley Neuendorf is:                                                   240 See ibid at 176. 241 Ibid at 175. 242 Ibid at 176. 243 Ibid at 177.  90  Figure 1. Formula for Percent Agreement244 PAO = A ÷ n In this formula, “PAO stands for proportion agreement, observed, A is the number of agreements between two coders, and n is the total number of cases the two coders have coded for the test” as well as the maximum number of agreements possible.245 The audio re-coding showed 21 discrepancies across 19 episodes (of 93 total episodes coded in both sets).246 There were 731 items coded in the first round of coding and 855 items coded in the second round.247 Since only the 731 items were coded in both rounds, this is the proper value to use for n in the above formula. A is equal to the number of agreements, or the difference between the number of possible agreements and the number of discrepancies (731 ˗ 21 = 710). The percent agreement is therefore calculated as: Figure 2. Percent Agreement: Coded Audio PAO = 710 ÷ 731 ≈ 0.971 The re-coded metadata showed only three discrepancies across 744 items. Figure 3. Percent Agreement: Coded Metadata PAO = 741 ÷ 744 ≈ 0.996                                                  244 Ibid at 174. 245 Ibid. 246 This excludes discrepancies which could be accounted for by the dynamic ads issue considered in section 6.2.1, above. 247 The difference between these is accounted for by two additional episodes which were discovered in the course of the recode and the addition of podcast network codes to the Patreon set.   91  The calculation of Cohen’s kappa is somewhat more complex than the calculation of percent agreement and will be presented here only in abridged form. The formula is as follows: Figure 4. Formula for Cohen’s kappa248 PAO-PAE1- PAE As in the formula for percent agreement, PAO stands for the percent agreement as observed. PAE stands for the percent agreement expected based on chance. For Cohen’s kappa, this value is calculated based on the products of marginal totals for each coder (or in this case, each round of coding) for each code.249 For the coded audio, Cohen’s kappa was calculated as 0.908, while for the metadata kappa was calculated as 0.972.  6.3.3. Calculating Reliability for Study Variables Percent agreement and Cohen’s kappa can also be calculated across different variables; in this case the different types of content which were coded for. (The full calculations of Cohen’s kappa for each variable are omitted.) Table 14. Percent Agreement and Cohen’s kappa for Audio Variables Variable Percent Agreement Cohen’s kappa Audio mentions Patreon or other donation page 91/93 ≈ 0.978 ≈ 0.958 Audio discussion of Patreon payment issues 93/93 = 1.000 = 1 Advertisements for non-podcast related products or services* 83/85 ≈ 0.978 ≈ 0.871 Audio mentions podcast episodes for sale 93/93 = 1.000 = 1 Audio mentions bonus content for donors 89/93 ≈ 0.957 ≈ 0.887                                                  248 Neuendorf, supra note 6 at 176. 249 See ibid at 181. 92  Audio advertises other money-generating activity by creator (such as live shows, merch) 83/93 ≈ 0.892 ≈ 0.800 Audio mentions copyright or IP issue 93/93 = 1.000 = 1 Audio mentions podcast network** 37/40 = 0.925 ≈ 0.875 Audio mentions other show(s) on podcast network** 40/40 = 1.000 = 1 * For the variable relating to advertising, episodes which displayed differences that could be attributed to the dynamic ads issue discussed in section 6.2.1 were omitted from the reliability calculations. ** The variables related to podcast networks were only coded for the podcast network set in the first round of coding. As such, reliability could only be tested for the episodes in that set. All variables for podcast audio showed a high degree of percent agreement, with only one variable (advertising secondary activity) falling below 0.900. The values for Cohen’s kappa were also high, with all but the secondary activity variable above 0.870. The relatively low reliability of the secondary activity variable may indicate that the description of that variable should be refined for future research to improve the stability of that measure. Four variables achieved perfect reliability; however, these variables had very few coded instances in the dataset or none at all.250  For the coded metadata, every variable achieved perfect reliability (PAO = kappa = 1.000) except for the secondary activity variable (PAO = 0.968; kappa = 0.894). As this variable was also the most problematic for the coded audio, this reinforces the need to refine that variable in any future research.                                                  250 See sections 5.2.5, 5.3.4, and 5.3.5, above. 93  Ultimately, while a high degree of intra-coder reliability was observed across both metadata and audio coding, for each variable and in the full dataset, without at least a second coder these results do not satisfy the methodological requirements of a content analysis study. Future extension of this research could code the same dataset with a second coder to ensure methodological rigor. This extension should also include other calculations of reliability which make up for the shortcomings of Cohen’s kappa, as discussed above.251 6.4. Analysis of Findings on Fundraising in Terms of Abundance and Scarcity Models  The two sets of podcast series were chosen to emphasize two ways of funding creative work in the absence of using copyright to create scarcity in order to sell copies of works. The following sections address the study’s findings with respect to three alternative ways of realizing economic returns through creative work: crowdfunding, advertising, and promoting secondary activities such as merchandising or live events. Based on the evidence provided by the study, these sections will assess how each method fits with the concept of abundance models, or if they exhibit some aspects of scarcity models. 6.4.1. Crowdfunding A majority (52.7%) of the series in the Patreon set included at least one reference to the series’ Patreon donation page. This indicates the importance of Patreon fundraising for podcast creators in this group. Notably, a larger majority (60% of the sample) referenced some bonus content available exclusively to paying subscribers. This bonus content largely consisted of additional podcast episodes. This may indicate that creators in this group find it useful to reward subscribers for their support in order to attract new subscribers. Since such content is available                                                  251 See Neuendorf, supra note 6 at 175–78. 94  exclusively to paying subscribers, this section will consider whether this practice constitutes a scarcity model in another guise.  A defining characteristic of the “scarcity model” is excluding users who do not pay from accessing works. The practice of providing bonus content to subscribers (the “premium content model”) certainly appears to be an example of this. However, there are significant differences between this model and the traditional sale of copies model. First, under the former model, subscribers pay for access to works, rather than for copies. As such, a subscriber to a podcast’s Patreon may have access to a lengthy back catalogue of premium episodes.252 Second, while some of the surveyed series restricted access to older podcast episodes, all had at least some episodes available for free.253 Over the chosen three-week time period, series which were included in the Patreon set had released an average of 3.667 episodes and a median of 3 episodes, with an average episode duration of 1h 23min 15s. This represents a substantial amount of content that podcast listeners have access to without paying. While these podcasts may not entirely adhere to the promise of abundance models, neither do they entirely exclude unpaying audiences.  The above two points support a third: it is not clear that the premium content model depends on copyright in the same way traditional scarcity-based sale of copies models do. As described above in sections 2.2 and 2.4, copyright’s exclusive reproduction right makes it possible for                                                  252 See e.g., Patreon, “Chapo Trap House”, online: <www.patreon.com/chapotraphouse/posts> (showing a $5 donation tier including “access to weekly premium episodes and entire premium back catalogue”); Patreon, “Sword and Scale”, online: <www.patreon.com/swordandscale/posts?tag=Plus> (showing “Sword and Scale Plus” episodes available for donors at $5 donation tier). The similarity to streaming platforms such as Netflix or Spotify, which also offer access to a library of works rather than copies, is also notable. 253 As noted in section 5.2, two series from the Patreon set did not have freely-available archives of episodes released during the relevant timeframe. See Table 1, above.  95  creators and other copyright owners to sell copies of works at a cost higher than the marginal cost of production without being undercut by competition. Asserting the reproduction right might be useful to a podcast creator whose premium episodes are shared with non-subscribers.254 However, the first line of defense is technological: through Patreon, creators can provide paying subscribers with a unique RSS feed to download premium episodes.255 Copyright may be valuable to podcast creators in the event of some future dispute, but—as suggested by this study—may do little to support the everyday business of making a financial return on creative work. With these differences from the traditional scarcity model in mind, however, it is still clear that some exclusion is taking place. The finding that many podcasts offer exclusive bonus content to subscribers makes it difficult to say that these podcasts are operating purely on abundance models, for which the defining feature is not excluding non-paying users. Further research could investigate whether “bonus” content in this context is seen as qualitatively different by creators or users from the free content offered by podcasters. Another question may be to what extent paid bonus content might be shared with non-paying users, and what measures if any creators take to prevent this practice. More information on these topics would help to better understand the practical application of abundance models, and better characterize the qualified form of these models that the forms of crowdfunding observed in this study seem to use.                                                  254 This study’s focus was on free episodes and as such found no evidence of this, but future research could investigate the assertion of copyright with respect to subscriber-only podcast episodes. 255 It should be noted that the circumvention of technological protection measures is addressed in sections 41–41.21 of the Copyright Act and is an infringement of copyright except under certain circumstances: see supra note 89. Whether a private RSS feed is a technological protection measure for the purposes of the Act and whether certain actions (such as sharing the feed with unauthorized persons) would constitute circumvention is a question beyond the scope of this project. Future research could explore the use (or non-use) of these and other “digital locks” in podcasting, and contrast their use to protect traditional media distributed online with their possible irrelevance to “born-digital” media based on abundance models. 96  6.4.2. Advertising Five series in the Patreon set (33.3%) and six series in the podcast network set (66.7%) included some advertisements. A higher incidence of advertising in the podcast network set was expected because that set was chosen for podcast networks’ support for advertising; however, a substantial number of Patreon podcasts also contained advertising for products or services unrelated to the podcast itself. As every podcast in the Patreon by definition engaged in crowdfunding as well, this demonstrates that crowdfunding and advertising are not incompatible as funding models.  One limitation of the methodology was that the coded audio portions of each episode were generally limited to the first and final five minutes. As some podcasts also feature ads mid-episode, this limitation may have resulted in the incidence of advertising being underestimated. (The issue discussed in section 6.2.1 with respect to undercounting of dynamic ads may also have contributed to this.) 6.4.3. Secondary Activities Secondary activities, as described above in sections 5.2.3 and 5.3.2, include other activity associated with a podcast that is either qualitatively different from podcasting (such as books or merchandise) or which goes beyond merely releasing an episode (such as live podcast events). These activities are often paid, though recorded versions of live events are sometimes released on the unpaid podcast feed.256 As such, they may provide further financial support to podcast creators. A large majority of series in the Patreon set (80%) included at least one example of promotion of secondary activity in episode audio or metadata. Five series in the podcast network                                                  256 See note 9, above (listing episodes in the sample set which consisted of recorded live events). 97  set (55.6%) also referenced secondary activity. Examples of secondary activity included published books, merchandise, live events, and meetups. With respect to the abundance and scarcity model framework, some secondary activities include traditional media that may operate on scarcity models. Traditionally published books are a clear example in this data set of secondary activity which operates on a scarcity model.257 Other podcast series have been adapted for television.258 Podcasting may in these circumstances serve as a springboard into traditional media, although further research might examine whether podcasters who undertake these activities leave podcasting behind. A further question that remains unanswered by the data from this study is to what extent secondary activities provide financial support to the podcast creators who undertake them. Future research could explore the motivations behind these activities, such as through interviews with podcast creators.                                                    257 See e.g. Chapo Trap House, The Chapo Guide to Revolution: A Manifesto against Logic, Facts, and Reason (New York: Touchstone, 2018).  258 See e.g., IFC, “Maron”, supra note 143; IFC, “Comedy Bang Bang”, supra note 143; Koblin, supra note 143. Whether television in fact operates on “scarcity models” as Suzor describes them is beyond the scope of this project, but it is worth noting that broadcast, cable, and satellite television are also not tied to a “sale of copies” model (though the latter two do rely on exclusion of non-paying consumers). 98  7. Conclusion This chapter will discuss how the results of the study address the research question and their implications for copyright law more broadly. This chapter will also pose some questions which future research in podcasting and copyright could address. As established in the Introduction chapter, the research question for this thesis was: How do podcast creators make a financial return on their work while using free distribution (i.e. not excluding non-paying users), and what role if any does copyright play in realizing these financial returns? While the findings from the study provide some answers to this question, future research could further investigate unexplored aspects of this question as well as the relationship between podcasting and copyright more broadly.  All of the surveyed series had free episodes available (affirming that many creators in this area are using free distribution). However, the assumption that creators are not relying on excluding users to make a financial return from their work was complicated by the finding that some series had further bonus episodes that users had to pay to access. As discussed in Chapter 6, this practice can be distinguished from the sale of copies model, as it is based on subscriptions and still has creators offering a large amount of content for free. However, this model cannot be said to entirely avoid the exclusionary aspects of scarcity models.259 Most series employed crowdfunding or advertising, which was to be expected based on the criteria for selecting the podcasts, but is nevertheless important to answer the question of how podcast creators make a financial return. There was no indication that any of the series in the sample sold copies of episodes.                                                   259 See section 6.4.1, above. 99  Despite existing copyright law protecting podcasts (as described in Chapter 3) the study showed little evidence of the use of copyright by podcast creators. As noted above, there were no direct mentions of copyright in the episode audio and episode metadata, and no licenses mentioned in episode metadata. At the level of individual series, copyright notices or licenses appeared for a small majority of the surveyed podcasts.260 These notices were visible on the series iTunes page in an out-of-the-way location under the podcast art and details. These findings should not be taken to mean that copyright is playing no role in podcasting, but at least indicate that copyright is not something that is made very prominent for listeners of this medium. These findings also indicate a need for further research into the role or roles copyright might be playing that are not apparent from the publicly available materials used for this study.  As discussed in an earlier chapter, crowdfunding and advertising do not have the same relationship with copyright that the sale of copies model does.261 While creative work funded through advertising or crowdfunding and released for free may be protected by copyright (as in the case of podcasts), creators of these works do not have to rely on copyright to exclude non-paying users in order to fund their work.262 This study does not indicate why creators are choosing to make their work available for free. The study does suggest, however, that crowdfunding and advertising are in use by podcast creators, and that these creators offer much of their work for free. But if abundance models work for podcasts, this complicates the utilitarian theory proposition that “copyright incentivizes creativity” and suggests that copyright is insufficient as the only or primary creative policy lever. Questions which remain unanswered by this study include: are crowdfunding and advertising models sufficient to meet creators’ needs to                                                  260 See section 5.4, above. 261 See section 2.4.3, above. 262 See Suzor, supra note 41. 100  make a return on their work? Do creators feel that they are receiving just compensation for their work through these models? Are there other issues with crowdfunding and advertising models that this study was not able to uncover? Further research could employ interviews with podcast creators to help answer these questions and provide a more detailed picture of how creators relate to copyright in this area. I argue that interpreting this study’s results in the abundance framework elaborated by Suzor gives a richer description of the differences between distribution choices than focusing solely on utilitarian concepts of incentives. However, this study has not fully answered the question: what is copyright doing for podcasts? It is clear enough that copyright is not supporting a scarcity model based around the sale of copies for at least a substantial number of podcasts. Several of the crowdfunded series surveyed offer bonus podcast episodes for paying subscribers, however, which seems to satisfy the main criterion of scarcity models (exclusion of non-paying users). There are also clear similarities between this bonus content model and premium or “freemium” subscription models such as those used by Spotify or Netflix.263 More work could be done on how these models differ with respect to incentives and use of copyright, particularly as these services are seen by some as the future of content distribution.264 This thesis has taken a different starting point from earlier legal scholarship on podcasting by emphasizing the positive aspects of copyright law. That is, rather than starting from the question “how does copyright law limit what podcast creators can do?” I have instead asked “what does copyright law do for podcast creators?” This question should arise naturally from the influential                                                  263 “Freemium” models offer a lower tier of service to non-paying users which may have fewer features or advertisements. 264 See Bottomley, “Podcasting: A Decade in the Life”, supra note 7 at 165. 101  utilitarian account of copyright which sees it as an incentive for creators and publishers. What my study found was that many podcast creators realize a financial return from their work using methods such as crowdfunding and advertising which do not require excluding non-paying users, and which therefore have a more ambiguous relationship with copyright.265  Podcasting is important to consider in this context because it is a form of creative expression which has been made possible and widely accessible—both for potential creators and users—by ubiquitous digital technologies. The disconnect with traditional funding models for creative production based around exclusion and scarcity may in turn be a product of the digital environment in which podcasting was created. Perhaps podcasting was able to thrive without scarcity because it was born unencumbered by those traditional funding models, and with a freedom to experiment enabled by low-cost online distribution and accessible digital technologies. In podcasting, we might see elements of a possible future for other forms of creativity, both existing and as-yet unimagined. Critically, the role copyright will play in encouraging this new creativity and rewarding its creators remains ambiguous.                                                     265 These methods may still be compatible with excluding non-paying users in some cases, however: see section 6.4.1, above. 102  Bibliography LEGISLATION Copyright Act, RSC 1985, c C-42. JURISPRUDENCE Alberta (Education) v Canadian Copyright Licensing Agency (Access Copyright), 2012 SCC 37. CCH Canadian v Law Society of Upper Canada, 2004 SCC 13. Cinar Corporation v Robinson, 2013 SCC 73. Entertainment Software Association v Society of Composers, Authors, and Music Publishers of Canada, 2012 SCC 34. Gould Estate v Stoddart Publishing (1996), 30 OR (3d) 520 (SC). Gould Estate v Stoddart Publishing (1998), 39 OR (3d) 55 (CA). Neugebauer v Labieniec, 2009 FC 666. re:Sound, 2012 SCC 38. Society of Composers, Authors and Music Publishers of Canada v Bell Canada, 2012 SCC 36. Théberge v Galerie d’Art du Petit Champlain, 2002 SCC 34. Walter v Lane, [1900] AC 539. 103  SECONDARY MATERIAL: MONOGRAPHS Hagen, Gregory; Graham Reynolds, Cameron Hutchison, Teresa Scassa, David Lametti & Margaret Ann Wilkinson. Canadian Intellectual Property Law: Cases and Materials, 2nd ed (Toronto: Emond Montgomery, 2018). Krippendorff, Klaus. Content Analysis: An Introduction to Its Methodology, 2nd ed (Thousand Oaks, CA: Sage Publications, 2004). Neuendorf, Kimberley A. The Content Analysis Guidebook, 2nd ed (Thousand Oaks, CA: Sage Publications, 2017). Shavell, Steven. Foundations of Economic Analysis of Law (Cambridge, Mass: Belknap Press, 2004). SECONDARY MATERIAL: ARTICLES Balganesh, Shyamkrishna. “Foreseeability and Copyright Incentives” (2009) 122:6 Harv L Rev 1569. Barrett, Jared. “Podcasting Pop Songs?: Licensing Concerns with Podcasts that Contain Mainstream Music” (2006) 3 Shidler JL Com & Tech 3. Beebe, Barton. “An Empirical Study of U.S. Copyright Fair Use Opinions, 1978–2005” (2008) 156 U Pa L Rev 549. Berry, Richard. “A Golden Age of Podcasting? Evaluating Serial in the Context of Podcast Histories” (2015) 22:2 J Radio & Audio Media 170. 104  Boling, Kelli S. & Kevin Hull. “Undisclosed Information—Serial is My Favorite Murder: Examining Motivations in the True Crime Podcast Audience” (2018) 25:1 J Radio & Audio Media 92. Bottomley, Andrew J. “Podcasting: A Decade in the Life of a ‘New’ Audio Medium” (2015) 22:2 J Radio & Audio Media 164. ———. “Podcasting, Welcome to Night Vale, and the Revival of Radio Drama” (2015) 22:2 J Radio & Audio Media 179. Bracha, Oren & Talha Syed, “Beyond Efficiency: Consequence-Sensitive Theories of Copyright” (2014) 29 Berkeley Tech LJ 229. Cantatore, Francina & Jane Johnston. “Moral Rights: Exploring the Myths, Meanings and Misunderstandings in Australian Copyright Law” (2016) 21 Deakin L Rev 71.  Carter, Edward L. & Scott Lunt. “Podcasting and Copyright: The Impact of Regulation on New Communication Technologies” (2006) 22 Santa Clara Computer & High Tech LJ 187.  Cohen, Julie E. “Copyright as Property in the Post-Industrial Economy: A Research Agenda” [2011] Wis L Rev 141. Cwynar, Christopher. “More than a ‘VCR for Radio’: The CBC, the Radio 3 Podcast, and the Uses of an Emerging Medium” (2015) 22:2 J Radio & Audio Media 190. Eschenfelder, Kristin R.; Robert Glenn Howard & Anuj C. Desai. “Who Posts DeCSS and Why? A Content Analysis of Web Sites Posting DVD Circumvention Software” (2005) 56:13 J Am Soc Info Sci & Tech 1405. 105  Favale, Marcella; Martin Kretschmer & Paul C. Torremans. “Is There an EU Copyright Jurisprudence? An Empirical Analysis of the Workings of the European Court of Justice” (2016) 79:1 Modern L Rev 31. Florini, Sarah. “The Podcast ‘Chitlin’ Circuit’: Black Podcasters, Alternative Media, and Audio Enclaves” (2015) 22:2 J Radio & Audio Media 209. Gervais, Daniel J. “The Purpose of Copyright Law in Canada” (2005) 2 U Ottawa L & Tech J 315. Golob, Brandon. “Un-Making a Murderer: New Media’s Impact on (Potential) Wrongful Conviction Cases” (2017) 54 Cal W L Rev 137. Kapczynski, Amy. “Intellectual Property’s Leviathan” (2014) L & Contemp Probs 131. Lang, Michael N. “The Regulation of Shrink-Wrapped Radio: Implications of Copyright on Podcasting” (2006) 14 CommLaw Conspectus 463. Li, Dan & Gina Welejko. “Splogs and Abandoned Blogs: The Perils of Sampling Bloggers and Their Blogs” (2008) 11:2 Information, Communication & Society 279. Marder, Nancy S. “Batson v. Kentucky: Reflections Inspired by a Podcast” (2016) 105 Ky LJ 621. Markman, Kris M. “Everything Old is New Again: Podcasting as Radio’s Revival” (2015) 22:2 J Radio & Audio Media 240. Markman, Kris M. & Caroline E Sawyer. “Why Pod? Further Explorations of the Motivations for Independent Podcasting” (2014) 21:1 J Radio & Audio Media 20. 106  Meserko, Vincent M. “Going Mental: Podcasting, Authenticity, and Artist-Fan Identification on Paul Gilmartin’s Mental Illness Happy Hour” (2014) 58:3 J Broadcast & Electronic Media 456. Morais, Yasmin & Sara Sampson. “A Content Analysis of Chat Transcripts in the Georgetown Law Library” (2010) 29 Leg Ref Services Q 165. Olszynski, Martin Z.P. “Failed Experiments: An Empirical Assessment of Adaptive Management in Alberta’s Energy Resources Sector” (2017) 50:3 UBC L Rev 697 at 724. Reese, R. Anthony. “Super Bowl I, Jazz Radio, and the Glass Menagerie: Copyright, Preservation, and Private Copies” (2017) 51 Akron L Rev 1025. Shaver, Lea. “Copyright and Inequality” (2014) 92 Wash UL Rev 117. Sterne, Jonathan; Jeremy Morris, Michael Brendan Baker & Ariana Moscote Freire. “The Politics of Podcasting” (2008) 13 Fibreculture J, online: <http://thirteen.fibreculturejournal.org/fcj-087-the-politics-of-podcasting/>. Supran, Geoffrey & Naomi Oreskes. “Assessing ExxonMobil’s Climate Change Communications (1974–2014)” [2017]:12 Environmental Research Letters 1. Suzor, Nicolas. “Access, Progress, and Fairness: Rethinking Exclusivity in Copyright” (2013) 15 Vand J Ent & Tech L 297. Terra, Antoni. “Copyright Law and Digital Piracy: An Econometric Global Cross-National Study” (2016) 18 NC JL & Tech 69.  107  Appendices Appendix A: Example Code Sheet ID POD## Description (metadata)  License in podcast description # Description mentions Patreon or other donation page # Description mentions Patreon payment issues # Description mentions podcast episodes for sale # Description mentions bonus episodes for donors # Description advertises other $-generating activity by creator (such as live shows, merch) # Description mentions copyright or IP issue # Advertisements for non-podcast related products or services # Notes  Audio  Audio mentions Patreon or other donation page # Audio discussion of Patreon payment issues # Advertisements for non-podcast related products or services # Audio mentions podcast episodes for sale # Audio mentions bonus content for donors # Audio advertises other $-generating activity by creator (such as live shows, merch) # Audio mentions copyright or IP issue # Audio mentions podcast network* # Audio mentions other show(s) on podcast network* # Notes  Codes: Audio  0 = none  1 = start  2 = end  3 = both  Codes: Metadata  0 = absent  1 = present  *  These codes were included only for the podcast network set in the initial round of coding. However, for the second round they were added to the codes for the Patreon set as well, after it became apparent that several Patreon series were also part of podcast networks. See section 6.2.4, above.108  Appendix B: Episode IDs, Series, Sources, Dates, and Titles ID Title Series Date Population URI Source TBS1 MICHAEL KUPPERMAN! MEL FROM NEWBRIDGE MEWS! MORE! The Best Show 01-Aug-18 Patreon https://traffic.libsyn.com/secure/thebestshow/TBS_073118.mp3?dest-id=232563 TBS2 SCOTT THOMPSON! DERRICK BECKLES! BASKETBALL GREG! The Best Show 17-Jul-18 Patreon https://traffic.libsyn.com/secure/thebestshow/TBS_180717.mp3?dest-id=232563 TBS3 NIGHT OF A THOUSAND PARDOS! BETTY'S HUSBAND NICK! CLARE O'KANE! WHIP TALK! WHITE MYSTERY! MORE! The Best Show 24-Jul-18 Patreon https://traffic.libsyn.com/secure/thebestshow/TBS_180724.mp3?dest-id=232563 CHA4 Episode 228 - Supreme Clientele feat. Elon Musk* and Ashley Feinberg (7/15/18) Chapo Trap House 15-Jul-18 Patreon feeds.soundcloud.com/stream/472019733-chapo-trap-house-episode-228-supreme-clientele-feat-elon-musk-and-ashley-feinberg-71518.mp3 CHA5 Episode 230 - A Colossal Wreck (7/22/18) Chapo Trap House 22-Jul-18 Patreon feeds.soundcloud.com/stream/475395969-chapo-trap-house-episode-230-a-colossal-wreck-72218.mp3 CHA6 Episode 232 - America, You Sexy Bigfoot! (7/30/18) Chapo Trap House 30-Jul-18 Patreon feeds.soundcloud.com/stream/479027394-chapo-trap-house-episode-232-america-you-sexy-bigfoot-73018.mp3 CHA7 Episode 234 - Congrat's (8/5/18) Chapo Trap House 05-Aug-18 Patreon feeds.soundcloud.com/stream/481660929-chapo-trap-house-episode-234-congrats-8518.mp3 CHA8 Bonus: SEX! Now that we have your attention: Work Chapo Trap House 17-Jul-18 Patreon feeds.soundcloud.com/stream/472910568-chapo-trap-house-bonus-sex-now-that-we-have-your-attention-work.mp3 109  DICK9 Episode 111 - Dick on Moon Lasers The Dick Show 17-Jul-18 Patreon https://traffic.libsyn.com/secure/thedickshow/The_Dick_Show_111.mp3 DICK10 Episode 112 - Dick on The Bride Catcher The Dick Show 24-Jul-18 Patreon https://traffic.libsyn.com/secure/thedickshow/The_Dick_Show_112.mp3 DICK11 Episode 113 - Dick on Green Eggs and Body Dysmorphia The Dick Show 31-Jul-18 Patreon https://traffic.libsyn.com/secure/thedickshow/The_Dick_Show_113.mp3 FATT12 Twilight Mirage 61: The Restitution of All Things Pt. 2 Friends at the Table 19-Jul-18 Patreon http://traffic.libsyn.com/friendsatthetable/FatT200TM61.mp3?dest-id=550849 FATT13 Twilight Mirage 62: The Feast of Patina Friends at the Table 27-Jul-18 Patreon http://traffic.libsyn.com/friendsatthetable/FatT201TM62.mp3?dest-id=550849 FATT14 Twilight Mirage 63: Guaranteed Events, Or: An Accounting of the Time When We Built the Machine Friends at the Table 02-Aug-18 Patreon http://traffic.libsyn.com/friendsatthetable/FatT202TM63.mp3?dest-id=550849 WHY15 Andrew Gosden - 288 The Generation Why Podcast 15-Jul-18 Patreon https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/75276295-c84b-4c39-9368-db794df9a0f7.mp3 WHY16 Isiah Fowler - 289 The Generation Why Podcast 22-Jul-18 Patreon https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/c9bab6af-55f8-484b-acb6-9d27264f7526.mp3 WHY17 Jens Soering - 290 The Generation Why Podcast 29-Jul-18 Patreon https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/044253fd-08db-41de-9e78-ef0b08d16b17.mp3 WHY18 Travis Walton - 291 The Generation Why Podcast 01-Aug-18 Patreon https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/2279a751-e8f1-4b1f-bf21-5c2b88df809d.mp3 WHY19 Murder of Larry McNabney - 292 The Generation Why Podcast 05-Aug-18 Patreon https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/d110  5f8bfc3-87cd-4f9d-be0d-6703073a636c.mp3 WHY20 Preview - I, Survivor The Generation Why Podcast 04-Aug-18 Patreon https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/8cc26015-bc36-4df4-8a8e-a4dd81b69f4f.mp3 HELLO21 A Recent Hello Internet Hello Internet 20-Jul-18 Patreon http://traffic.libsyn.com/hellointernet/ARecentHelloInternet.mp3 HELLO22 H.I. #106: Water on Mars Hello Internet 31-Jul-18 Patreon http://traffic.libsyn.com/hellointernet/106.mp3 LITU23 Man The Liturgists 26-Jul-18 Patreon https://mcdn.podbean.com/mf/web/tguypt/Man_-_The_Liturgists_Podcast.mp3 LORE24 Episode 91: Beneath the Surface Lore 23-Jul-18 Patreon https://traffic.libsyn.com/secure/lorepodcast/Lore91.mp3 PHIL25 Episode #120 ... Logical Positivism Philosophize This! 27-Jul-18 Patreon https://traffic.libsyn.com/secure/philosophizethis/Logical_Positivism.mp3 RLPH26 THE RALPH REPORT 135 - Monday, July 30th, 2018 The Ralph Report 30-Jul-18 Patreon http://media.blubrry.com/the_ralph_report/http://feeds.soundcloud.com/stream/478562757-theralphreport-the-ralph-report-135-monday-july-30th-2018.mp3 SAS27 Episode 118 Sword & Scale 22-Jul-18 Patreon https://content.production.cdn.art19.com/episodes/daac5766-9a47-4989-a9b6-d94736c8c255/c168a90a7cf561b332f8e564ad44e7299c5cc30b333ed66bc68509870fd497ec38184f6c9538273941663b006500ef6079ad88d810c716da8bde55e6b9941650/SAS118_AD.mp3 111  SAS28 Episode 119 Sword & Scale 05-Aug-18 Patreon https://content.production.cdn.art19.com/episodes/eb3490fc-819d-4c14-8371-f217f5d753ac/71636428ab27bad3f7c6aa6574197713f2f615869e970aaf116d22fd6c67d8b39c45e10ea1b2a4fd19bd36b9a1442e0b895e288d6a05bb17233abeb18c967a2f/SAS119.mp3 SIMP29 See Talking Simpsons Live Saturday, July 28th at PianoFight in San Francisco! Talking Simpsons 15-Jul-18 Patreon https://www.podtrac.com/pts/redirect.mp3/traffic.libsyn.com/secure/talkingsimpsons/See_Talking_Simpsons_Live_Saturday_July_28th_at_Pianofight_in_San_Francisco.mp3 SIMP30 Talking Simpsons - Treehouse Of Horror VII With Louis Peitzman Talking Simpsons 18-Jul-18 Patreon https://www.podtrac.com/pts/redirect.mp3/traffic.libsyn.com/secure/talkingsimpsons/TalkingSimpsons_S0801.mp3 SIMP31 Talking Simpsons - You Only Move Twice With Allie Goertz & Julia Prescott Talking Simpsons 25-Jul-18 Patreon https://www.podtrac.com/pts/redirect.mp3/traffic.libsyn.com/secure/talkingsimpsons/TalkingSimpsons_S0802.mp3 SIMP32 Talking Simpsons - The Homer They Fall With Kat Bailey Talking Simpsons 01-Aug-18 Patreon https://www.podtrac.com/pts/redirect.mp3/traffic.libsyn.com/secure/talkingsimpsons/TalkingSimpsons_S0803.mp3 DOLL33 338 - Abolitionist Benjamin Lay The Dollop 16-Jul-18 Patreon https://traffic.libsyn.com/secure/thedollop/Women_and_Transportation.mp3?dest-id=139738 DOLL34 337 - Bonfils, Tammen, and The Denver Post (Live) The Dollop 24-Jul-18 Patreon https://traffic.libsyn.com/secure/thedollop/Bonfils_Tammen_and_The_Denver_Post.mp3?dest-id=139738 112  DOLL35 336 - Women and Transportation The Dollop 30-Jul-18 Patreon https://traffic.libsyn.com/secure/thedollop/Benjamin_Lay.mp3?dest-id=139738 UYD36 Episode 666 Jul 17th 2018 Uhh Yeah Dude 17-Jul-18 Patreon http://traffic.libsyn.com/uhhyeahdude/ep666beastmode.mp3 UYD37 Episode 667 Jul 17th 2018 Uhh Yeah Dude 17-Jul-18 Patreon http://traffic.libsyn.com/uhhyeahdude/ep667spanishflyandhighkarate.mp3 WWC38 #747 RHOP: Crepes of Wrath Watch What Crappens 16-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-747rhop-crepesofwrath/media.mp3 WWC39 #748 RHOC: Roped Into A New Season Watch What Crappens 17-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-748rhoc-ropedintoanewseason/media.mp3 WWC40 #749 BelowDeckMed: Patient Zero Patience Watch What Crappens 18-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-749belowdeckmed-patientzeropatience/media.mp3 WWC41 #750 RHONY: CartagenaAAAS KWEEN! - Live from Philadelphia Watch What Crappens 20-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-750rhony-cartagenaaaaskween--livefromphiladelphia/media.mp3 WWC42 #751 SouthernCharm: You Reap What You Sew (in 3D!) - Live from Philadelphia Watch What Crappens 21-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-751southerncharm-youreapwhatyousew-in3d--livefromphiladelphia/media.mp3 WWC43 #752 RHOP: French Toasted Watch What Crappens 23-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-752rhop-frenchtoasted/media.mp3 WWC44 #753 RHOC: Much a Deuteronomy About Nothing Watch What Crappens 24-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-753rhoc-113  muchadeuteronomyaboutnothing/media.mp3 WWC45 #754 Below Deck Med: Superstorm Sandy Watch What Crappens 25-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-754belowdeckmed-superstormsandy/media.mp3 WWC46 #755 SouthernCharm:  Hi Ashley Watch What Crappens 27-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-755southerncharm-hiashley/media.mp3 WWC47 #756 RHONY: A Tale of Two Biddies Watch What Crappens 28-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-756rhony-ataleoftwobiddies/media.mp3 WWC48 #757 RHOP: It's Not Delivery, It's Di-Robyn-o Watch What Crappens 31-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-757rhop-itsnotdelivery-itsdi-robyn-o/media.mp3 WWC49 #758 RHOC: Mexico-Conspirators Watch What Crappens 31-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-758rhoc-mexico-conspirators/media.mp3 WWC50 #759 BelowDeckMed: Sandy Has Coke Zero Tolerance Watch What Crappens 01-Aug-18 Patreon https://media.acast.com/watchwhatcrappens/-759belowdeckmed-sandyhascokezerotolerance/media.mp3 WWC51 #760 RHONY: Woah-tanic Watch What Crappens 02-Aug-18 Patreon https://media.acast.com/watchwhatcrappens/-760rhony-woah-tanic/media.mp3 WWC52 #761 Shahs of Sunset: Persians in the Attic w/ Guest Lara Shoenhals Watch What Crappens 03-Aug-18 Patreon https://media.acast.com/watchwhatcrappens/-761shahsofsunset-persiansintheatticw-guestlarashoenhals/media.mp3 WWC53 #762 Bonus: Two Judgey Girls. Well, One Watch What Crappens 05-Aug-18 Patreon https://media.acast.com/watchwhatcrappens/-762bonus-114  twojudgeygirls.well-one/media.mp3 LADY54 #145: We're Always Learning Lady Lovin' 16-Jul-18 Podcast Network https://warp.gumballcdn.com/v1/mp3?uuid=acce077622&show=248455a223&mp3=https://rss.art19.com/episodes/796e6d13-fea7-4cf8-a52d-3f9fbd810d32.mp3 LADY55 #147: Social Media Behavior with Atlanta de Cadenet Taylor Lady Lovin' 30-Jul-18 Podcast Network https://warp.gumballcdn.com/v1/mp3?uuid=acce077622&show=248455a223&mp3=https://rss.art19.com/episodes/6f353492-9c38-4caf-8c94-cbad1e7d61b8.mp3 LADY56 #146: Dating Apps Lady Lovin' 23-Jul-18 Podcast Network https://warp.gumballcdn.com/v1/mp3?uuid=acce077622&show=248455a223&mp3=https://rss.art19.com/episodes/835bafe0-d96c-41ae-80ea-40f6f4ee7156.mp3 HASH57 Hashtag Trending – Amazon Prime Day; Microsoft Inspire; Dr. Alexa gives bad advice  Hashtag Trending 16-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-amazon-prime-day-microsoft-inspire-dr-alexa-gives-bad-advice/407120 HASH58 Hashtag Trending – 12 Russians charged with hacking US election, Jeff Bezos is richer than you, Amazon Prime Day worker strikes  Hashtag Trending 17-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-12-russians-charged-with-hacking-us-election-jeff-bezos-is-richer-than-you-amazon-prime-day-worker-strikes/407151 HASH59 Hashtag Trending – Increased tech use could lead to ADHD in teens, IBM sues Groupon, human-machine mind control may be a reality Hashtag Trending 18-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-increased-tech-use-could-lead-to-adhd-in-teens-ibm-sues-115  groupon-human-machine-mind-control-may-be-a-reality/407215 HASH60  Hashtag Trending – Google is fined billions by the EU, Amazon Prime Day breaks records, Nest CEO steps down  Hashtag Trending 19-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-google-is-fined-billions-by-the-eu-amazon-prime-day-breaks-records-nest-ceo-steps-down/407263 HASH61 Hashtag Trending – Apple could be the first to a trillion, tech leaders pledge to not develop killer AI, new i9 Macbook Pro is too hot to function Hashtag Trending 20-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-apple-could-be-the-first-to-a-trillion-tech-leaders-pledge-to-not-develop-killer-ai-new-i9-macbook-pro-is-too-hot-to-function/407319 HASH62 Hashtag Trending – Google Cloud Next, Google partners with US tech giants for data transfer project, Google’s plans for a trans-Atlantic cable  Hashtag Trending 23-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-google-cloud-next-google-partners-with-us-tech-giants-for-data-transfer-project-googles-plans-for-a-trans-atlantic-cable/407365 HASH63 Hashtag Trending – Pinterest ad sales nearing $1 billion USD; Uber/Lyft ban driver for filming riders; Tesla asking suppliers for refunds Hashtag Trending 24-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-pinterest-ad-sales-nearing-1-billion-usd-uber-lyft-ban-driver-for-filming-riders-tesla-asking-suppliers-for-refunds/407370 HASH64 Hashtag Trending – Toronto beats San Francisco in tech job creation; London taxi drivers sue Uber for £1 billion; We interrupt your Netflix binge with this emergency alert  Hashtag Trending 25-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-toronto-beats-san-francisco-in-tech-job-creation-london-taxi-drivers-sue-uber-for-1-billion-we-interrupt-your-netflix-116  binge-with-this-emergency-alert/407421 HASH65 Hashtag Trending – Robot dogs; Airbnb’s priciest cities; and museums want more nudity on Facebook Hashtag Trending 26-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-robot-dogs-may-soon-be-taking-over-our-homes-bloomberg-names-the-most-expensive-airbnb-cities-and-museums-think-facebook-needs-more-nudity/407460 HASH66 Hashtag Trending – Amazon’s facial recognition software fails a test; Trump accuses Twitter of bias; Facebook loses billions in value Hashtag Trending 27-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-amazons-facial-recognition-software-fails-a-test-trump-accuses-twitter-of-bias-facebook-loses-billions-in-value/407502 HASH67 Hashtag Trending – Amazon HQ2 frontrunners  Hashtag Trending 30-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-amazon-hq2-frontrunners/407535 HASH68 Hashtag Trending – Uber drivers declared employees; newest Gmail update puts 1.4B at risk; Venmo’s default setting is public Hashtag Trending 31-Jul-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-uber-drivers-declared-employees-newest-gmail-update-puts-1-4b-at-risk-venmos-default-setting-is-public/407378 HASH69 Hashtag Trending – Why airborne Wi-Fi isn’t free; New Zealand gov’t teaching entrepreneurship with video games; Russians have hacked U.S. utilities Hashtag Trending 01-Aug-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-why-airborne-wi-fi-isnt-free-new-zealand-govt-teaching-entrepreneurship-with-video-games-russians-have-hacked-u-s-utilities/407505 117  HASH70 Hashtag Trending – Rising sea levels endanger buried fibre; Samsung teams up with Nasa; Best Buy shares its secrets  Hashtag Trending 02-Aug-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-rising-sea-levels-endanger-buried-fibre-samsung-teams-up-with-nasa-best-buy-shares-its-secrets/407536 HASH71 Hashtag Trending: Car makers exposed; Should Amazon replace libraries?; is thinner better? Hashtag Trending 03-Aug-18 Podcast Network https://www.itworldcanada.com/article/hashtag-trending-unsecured-data-exposes-top-carmakers-forbes-thinks-amazon-should-replace-libraries-is-thinner-actually-better/407537 BIZR72 MAD POOPER STRIKES CLOSE TO HOME Bizarre States 19-Jul-18 Podcast Network http://traffic.libsyn.com/spookyshit/bs199_poop.mp3?dest-id=203202 BIZR73 Burrito Baby Bizarre States 26-Jul-18 Podcast Network http://traffic.libsyn.com/spookyshit/bs200_burrito1.mp3?dest-id=203202 BIZR74 Can't Please Anybody Bizarre States 02-Aug-18 Podcast Network http://traffic.libsyn.com/spookyshit/bs201_cant_please_anybody.mp3?dest-id=203202 BIZR75 BONUS Live from SDCC 2018 with Michael McMillian Bizarre States 03-Aug-18 Podcast Network http://traffic.libsyn.com/spookyshit/bsbonus_sdcc_2018.mp3?dest-id=203202 INC76 #178 Silicon Valley’s Hidden Talent: Making People Hate Its Inventions Inc. Uncensored 20-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY1913273309.mp3 INC77 #179 The Secret Behind Sweetgreen’s Explosive Growth Inc. Uncensored 27-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY9282029708.mp3?updated=1532647900 INC78 #180 Rest In Peace, MoviePass Inc. Uncensored 03-Aug-18 Podcast Network https://traffic.megaphone.fm/PPY9038651778.mp3?updated=1533307473 YMRT79 123: Roscoe "Fatty" Arbuckle and Virginia Rappe (Fake News: Fact Checking Hollywood Babylon Episode 3) You Must Remember This 17-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY5359120957.mp3 118  YMRT80 124: William Desmond Taylor (Fake News: Fact Checking Hollywood Babylon Episode 4) You Must Remember This 24-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY3255897330.mp3 YMRT81 125: Mabel Normand (Fake News: Fact Checking Hollywood Babylon Episode 5) You Must Remember This 31-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY6553867733.mp3 ANDR82 All About Android 378: Hold on to Your Phones All About Android 17-Jul-18 Podcast Network https://www.podtrac.com/pts/redirect.mp3/cdn.twit.tv/audio/aaa/aaa0378/aaa0378.mp3 ANDR83 All About Android 379: The Age Of Material All About Android 24-Jul-18 Podcast Network https://www.podtrac.com/pts/redirect.mp3/cdn.twit.tv/audio/aaa/aaa0379/aaa0379.mp3 ANDR84 All About Android 380: The Mockadile Industrial Complex All About Android 31-Jul-18 Podcast Network https://www.podtrac.com/pts/redirect.mp3/cdn.twit.tv/audio/aaa/aaa0380/aaa0380.mp3 WAR85 12 | Baskett This Is War 25-Jul-18 Podcast Network https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/d2e8cf32-4a14-4eb6-afd0-5a9eb8ce7b5d.mp3 TRUE86 Ep87 - Melvin Ignatow True Crime All The Time 16-Jul-18 Podcast Network https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/c2981cd0-e137-428e-939f-f2d7ddcbe982.mp3 TRUE87 Ep88 - Westley Allan Dodd True Crime All The Time 23-Jul-18 Podcast Network https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/a3455e8b-ef2c-408c-8416-b2f34a4ce221.mp3 TRUE88 Ep89 - Chevie Kehoe True Crime All The Time 30-Jul-18 Podcast Network https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/fc25a04a-1b7c-4ea8-8870-3a5e7c504271.mp3 TRUE89 Introducing I, Survivor True Crime All The Time 01-Aug-18 Podcast Network https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/f3e96502-477e-49f2-ab68-63dafee0ff98.mp3 TRUE90 Ep90 - Robert Garrow True Crime All The Time 05-Aug-18 Podcast Network https://dts.podtrac.com/redirect.mp3/rss.art19.com/episodes/8119  66fe4d8-87b3-4bde-bd2e-968303ff8efa.mp3 FPER91 The Woman Who Defied Iran Foreign Policy: The Editors' Roundtable 20-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY2936380336.mp3 FPER92 ‘It Could Have Led to a Nuclear War in the Middle East’ Foreign Policy: The Editors' Roundtable 27-Jul-18 Podcast Network https://traffic.megaphone.fm/PPY4801533330.mp3 FPER93 The Ghost of Smoot-Hawley Foreign Policy: The Editors' Roundtable 03-Aug-18 Podcast Network https://traffic.megaphone.fm/PPY4863512143.mp3 CHA94 UNLOCKED episode 225 - The Poster's Crusade (7/6/18) Chapo Trap House 16-Jul-18 Patreon https://dts.podtrac.com/redirect.mp3/feeds.soundcloud.com/stream/472509036-chapo-trap-house-unlocked-episode-225-the-posters-crusade-7618.mp3 WWC95 #746 Southern Charm: Winter Ball and Chain - Live From Kansas City Watch What Crappens 15-Jul-18 Patreon https://media.acast.com/watchwhatcrappens/-746southerncharm-winterballandchain-livefromkansascity/media.mp3  * The series indicated with an asterisk in the “Population” column were present in both populations. “Patreon*” indicates that the series was selected as part of the Patreon set, but was a member podcast of one of the selected podcast networks. “Podcast Network*” indicates that the series was selected as part of the Podcast Network set, but also had an associated Patreon page in the top 100 used for the Patreon set. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0378227/manifest

Comment

Related Items