Going into this course, I didn’t know what to expect, but assumed it would be more about hands-on training in the use of technology; in particular I had coding in mind. And in fact I did learn how to use a browser plug-in called Hypothes.is for annotations, and gained a little familiarity with editing WordPress. But also before the class began, I wondered how much we could learn about a given technology in such a short time. So it made sense when I learned that the course would be more of a seminar and discussion about the “digital landscape” than hands-on training in particular software or apps. I would say this is as close to a takeaway as I can describe from this course.
The starting point was getting us to understand the difference between the Internet and the web. That was helpful for me. I also wrote in my notes that Juan wanted to prepare us to navigate the shifting landscape around publishing, enabling us to see what’s happening and the active forces behind it. This course did provide more of a perspective than a set of skills.
The grading contract was a good incentivizing tool and I definitely was stricter with myself about my engagement with this class than any of the others. The required Hypothes.is annotations on readings were also a good incentive to do the readings. I did notice that when I was reading under time constraints, I skimmed to find points to comment on, rather than skimming to find points indicating the author’s argument. This led to me making comments I didn’t really feel strongly about, but that’s no different from the skimming I did in my undergrad when preparing to write a precis.
Some of my peers gave really excellent and engaging lectures. I would also have liked to hear more from Juan. The early lecture where we learned about the history of the formation of the Internet was excellent, and whenever he allowed himself to interject, he offered interesting and important perspective. In one class discussion many students expressed how little they cared about their personal privacy on the internet. In response Juan tried to clarify the importance of the issue, and this is an example of the type of issues I would have liked more directed conversation about from someone whose work involves thinking them through. That said, I appreciate the trust Juan put in us and the level of engagement and discussion this class facilitated.
If someone asks me what I learned, I would be hard pressed to say anything specific. But I feel like I’m a little better prepared to understand conversations in the media around Amazon and Facebook, AI and machine learning, data and privacy, and the changing business models around the publication of content. Also, because of discussions like the ones we had around print and digital reading, I also feel more aware of personal biases publishers are susceptible to, and how they affect our attitudes toward technologies, the internet, and web apps. This course mainly gave me a little experience in trying things out, and a level of comfort discussing technologies today.
Keeping in mind the readings on marginalia and annotations, and thinking more generally about the life of text online, should audiences be allowed to interact with and shape the text? Does a writer have the right to define who can comment? Should audiences be limited in their online socialization over a text? In “Marginalia and Its Disruptions,” S. Brent Plate examines what it really means when readers write in the margins. He references an article by Laura Miller in which she writes, “Marginalia is a blow struck against the idea that reading is a one-way process, that readers simply open their minds and the great, unmediated thoughts of the author pour in.” Plate adds that marginalia “turn[s] readers into writers, and upset[s] the hierarchy of the author as authority.”
He also references Tim Parks who argues that “We have too much respect for the printed word, too little awareness of the power words hold over us. . . . We overlook glaring incongruities. We are suckers for alliteration, assonance, and rhythm. We rejoice over stories, whether fiction or ‘documentary,’ whose outcomes are flagrantly manipulative, self-serving, or both. Usually both. If a piece of writing manifests the stigmata of literature—symbols, metaphors, unreliable narrators, multiple points of view, structural ambiguities—we afford it unlimited credit.” Parks concludes, “What surprised me most when I first began publishing fiction myself was how much at every level a novelist can get away with.”
These are some of the arguments in favour of encouraging and facilitating marginalia on digital works by enabling comments as well as annotation plug-ins like Hypothes.is. But not everyone agrees. Audrey Watters writes, “I took comments off my websites in 2013 because I was sick of having to wade through threats of sexualized violence in order to host conversations on my ideas. . . . I’ve made this position fairly well known – if you have something to say in response, go ahead and write your own blog post on your own damn site.” I find this position pretty unassailable. I don’t think writers have any obligation whatsoever to allow comments on their work. I think all writers deserve the right to decide, but it becomes clear in light of how women writers and other marginalized groups face harassment issues that can leak into offline threats and aggressions.
There is some anxiety that not allowing comments amounts to some kind of censorship. This is due to a misunderstanding of what censorship means. Being denied access to a specific platform to voice your views is nowhere near the same thing as being persecuted for voicing your views. Historica Canada defines censorship as “the exercise of prior governmental control over what can be printed, published, represented or broadcast.”
One might argue that even if it is not the legal definition of censorship, it’s still bad to ignore others’ viewpoints just because you disagree. But there are still plenty of other reasons not to force authors or publishers to allow comments or annotations on their work.
We can think of an analogy in print. How ridiculous would I sound if I insisted that the Vancouver Sun is censoring me by not publishing every letter I write in? The editors and publishers have a write to decide what is included in their publication. Comments and annotations are an act of publication, and it is no different when it is digital. The person who pays for the publication decides what gets in. As Watters writes, you can always write whatever you want about her work on your own damn blog. That’s what makes it not censorship.
Forcing writers to allow comments directly on their work is not striking a blow in the name of free speech. Especially not in a context where hate speech and threats and trolls are par for the course. Hate speech is not a case of “disagreeing with someone’s opinion.” It’s part of a larger context in which certain groups suffer systematic oppression and violence. That’s why it’s not protected under free speech. Take a look at the comments sections under many news articles involving Indigenous people. You can see why many larger news outlets have disabled comments sections on some or all of their online articles.
At the same time, you can always write a letter to the editor, publish your own free blog, start a social media account, make posters, write to other newspapers or blogs… This saves the author the work of having to read and mediate what could be a bunch of useless and/or hateful comments when all they ever signed up to do was write. To sum up: Declining to publish something is not censorship; hate speech is not free speech; and we are not entitled to access to any platform we want for our opinion. I hope that we as future publishers can see that a writer has no obligation to allow comments or annotations on their work and that not allowing them does not amount to censorship. Public discourse will benefit from leaving the choice up to the writer or publisher on their own platforms and giving writers the chance to opt in or out when they are published on other platforms.
Reading in print is different from reading digitally. One study has shown that students report longer reading times, but also higher multitasking, when they read e-textbooks compared to print ones. Another one found that “students who read texts in print scored significantly better on the reading comprehension test than students who read the texts digitally.” Michael S. Rosenwald says results like these have given rise to concern among neuroscientists that humans are “developing digital brains with new circuits for skimming through the torrent of information online,” and that “[t]his alternative way of reading is competing with traditional deep reading circuitry developed over several millennia.”
There are two claims here: one is that reading behaviour like skimming is taking over from slower, deeper reading, and I don’t disagree. The other claim is that “traditional deep reading circuitry developed over several millennia.” Rosenwald acknowledges that it’s not as simple as this when he writes: “There are no genes for reading like there are for language or vision.” (I’m no expert, so I’ll have to take his word for it that there are in fact genes for language or vision.) However, in the next sentence, he argues that “the brain has adapted to read … spurred by the emergence of Egyptian hieroglyphics, the Phoenician alphabet, Chinese paper and, finally, the Gutenberg press.” In other words, it’s not just that our brains are changing as individuals in response to our reading habits; it’s that humans as a species evolved the “trait” of being able to read in response to our use of specific technology.
I am curious to know which research Rosenwald is using. The implications of viewing reading as an evolutionary adaptation to certain tools are not small. This is an example of a teleological understanding of the history of reading. On this view, science, technology and the human brain march upward together on a linear path toward fulfilling the human goal of the Printed Book. It’s a neat and tidy view, and if we take it, we are more likely to defend print reading as inherently superior over digital technology, at least until we’ve had a few more centuries to evolve our new digital reading trait.
However, human beings did not join hands and agree to begin reading all at once, “several millennia” ago. If this evolutionary view of reading behaviour were correct, then people from cultures where literacy became widespread only a few centuries ago would be centuries “behind” on reading skills, print or digital. But reading is not an evolutionary trait; it is a skill that can be mastered, or lost, in a single generation. Neuroplasticity is different from evolution.
This brings us back to the first claim, that reading behaviour like skimming is taking over from slower, deeper reading. If we are doing more and more of our reading digitally (which we are), and if digital reading gives rise to behaviour like skimming (which it appears to), and if skimming results in less cognition and understanding (as many psychological studies suggest), then we have a problem.
As we attempt to interpret these findings and consider how they should inform our choices as publishers, I think we should be cautious before we draw any sweeping conclusions based on our limited research. A December 2017 article in the Review of Educational Research “Reading on Paper and Digitally: What the Past Decades of Empirical Research Reveal” examined 36 psychological studies on the subject of print vs. digital reading. The authors state that to their knowledge, “this is the only systematic review on the topic of reading in different mediums since 1992 that juxtaposes the contemporary field of reading digitally against the long-established and deep-rooted research on reading in print.” The study leaves me with the impression that researchers have only just begun to investigate the differences between print and online reading, and that, unsurprisingly, further research is needed:
In addition to the aforementioned need for details on textual aspects, there is a need for more clarification regarding individual differences factors and text processing in print or digitally. Simply stated, individual difference factors are the variations or deviations among individuals with regard to the characteristics shown to play a significant role in human learning and development (e.g., working memory, academic ability, gender; Gagné & Glaser, 1987). In the case of reading in print and digitally, individual difference factors such as reading rate, vocabulary knowledge, and topic knowledge have been shown to be particularly pertinent. . . . Surprisingly, very few studies in this review considered such relevant individual difference factors as fluency or topic knowledge as potential explanations for performance outcomes between print and digital reading (Kendeou et al., 2011). Thus, assessing the role of individual differences factors could help clarify patterns in comprehension performance across mediums.
They conclude, carefully, that “medium plays an influential role under certain text or task conditions or for certain readers.” This measured answer makes sense to me. With respect to the question of how publishers should respond to the rise of digital reading, I think we have a responsibility to investigate our biases and personal reactions to these kinds of claims with curiosity, setting aside our personal preferences and assumptions about different reading formats, and thinking about what it is that the readers and writers in our specific areas (genre fiction, children’s literature, Indigenous authors and content, Black literature, etc.) need for their particular “text or task conditions.”
Rosenwald quotes Maryanne Wolf, a cognitive neuroscientist and author of Proust and the Squid: The Story and Science of the Reading Brain (which I have not read, for the record). Wolf worries that “the superficial way we read during the day is affecting us when we have to read with more in-depth processing.” To illustrate this, she describes herself reading a novel: “skimming, picking out key words, organizing my eye movements to generate the most information at the highest speed.” But her reaction to watching herself exercise these (frankly remarkable) skills, the same skills that university education expects of us, was surprising: “I was so disgusted with myself.”
I think this illustrates how odd the panicked conversation about digital reading really is. Yes, reading in different contexts places emphasis on different kinds of skills. If print reading is falling to the wayside, I don’t think publishers need to clutch desperately at it as though it is a thread by which human intellect hangs. We should absolutely study the different affordances of each medium, but not from the point of view that one is inherently better. Wolf points out that “the brain is plastic its whole life span” and is “constantly adapting.” Publishers need only take inspiration from that.
If I were going to use tracking to enhance publishing practice, I would like to use it to address the needs of educational publishing. In my experience in psychology and biology classes in my undergrad, it’s becoming very common for textbooks to come with digital components. In my classes it was usually a website you could log in to and access the text on the web, as well as view other media. There was usually a limited and finicky highlighting and annotation function, too. My experience as a user varied a lot from book to book. I remember finding some textbook’s corresponding sites useful in their content but frustrating to navigate. I think exploring student preferences and consumption behaviour would be a great application of tracking. If I were an educational publisher I would use reader analytics and tracking to specialize in delivering very user-friendly e-textbooks.
One challenge would be that students are required to read their textbooks. This means they don’t have the option to skip passages that are unreadable. The data would show a high engagement rate, but only because the students had no choice but to finish the chapter. Even in cases where they didn’t, this data would be skewed in that it would not be a reliable measure of the readability of the passage.
For that reason, my tracking in education publishing would instead focus on two other areas. The first would be on measuring time. For example, measuring the average length of time it takes students to complete a passage, or how long students are able to focus on a typical textbook before they have to put down their device. This information (which would likely differ between different fields) could be used to tailor the length of sections and chapters so that they are in readable sizes, and to let publishers know which parts need work before they release the next edition. It would also be useful for professors in planning their syllabi.
The second area I would focus on is making the reader analytics software responsive and customizable. The reader would create an account and read the text, and the reader tracking software would become familiar with their particular reading habits. Once the software had analyzed enough of my reading behaviour data, it would be able to tell me how much time to set aside for each particular chapter, when my prime studying time of day is, and how often I need to take a break.
The challenge here would be that the customizable software would function better and better over time as it became familiar with your reading habits. But by the time the software got good at understanding the student, the semester would be over. So maybe my reader analytics software could include a short “training” period where the reader is asked to run through a few pages of different kinds of text, designed to represent the kinds of text common in that particular field. The reader’s habits could then be understood and taken into account by the software a little faster. This is kind of like how Cortana (the Siri-like bot that comes with Windows 10) “learns” my accent and dialect of English by having me read particular phrases out loud.
The reason I would like to focus on educational publishing is that I would rather apply reader analytics to the goal of improving student success and experience than to hyperfocused marketing campaigns. As textbooks today in many fields are a hybrid of print and digital, educational publishers must understand student’s preferences and behaviour and take them into account when planning digital reading experiences.
It has always been the case that data about reader preferences heavily influences marketing, and that successful or unsuccessful marketing in turn influences later book acquisition decisions. AI has massively increased the amount of reader data available, and the degree to which it can be analyzed. But a more direct application of AI to acquisitions is also taking root, and I predict that artificial intelligence will become integrated into manuscript acquisition.
For a concrete example, we can look at Booxby, an app that “reads” and analyzes texts, and offers “A.I. generated data to the four stages of book discovery: manuscript development, acquisition, marketing, and consumer discovery.” Founder Holly Lynn Payne explains that AI can help solve the discovery problem; it can “provide meaningful analytics to inform acquisition decisions, understand a book’s full market potential, and create an effective mechanism to connect books to the readers who would most enjoy them.”
However, as my MPub colleague Taylor McGrath reminded us in a comment in our Hypothes.is group, readers tend to choose books based on personal recommendations, making an AI-driven, Netflix-like service for books unlikely to take hold. I agree, and that’s why I can’t see what it would look like if we used AI to create a “mechanism to connect books to the readers who would most enjoy them.” Payne is overstating the problem. (In fact, I think many readers, myself included, actually enjoy the process of looking for books, in a way that we do not necessarily enjoy shopping for other goods.)
I do think Payne gets it right when she says that AI can “provide meaningful analytics to inform acquisitions decisions” and “understand a book’s full market potential.” It’s acquisitions editors, not readers, who want help choosing books, and that’s where Booxby will shine. On top of providing comps for the books it processes, Booxby also offers “patent-pending Experiential Language Tags that quantify the reader experience.” I have no idea what those are, but if they’re anything like the applications of AI that I’ve been learning about lately, it sounds like a probably imperfect but very powerful tool.
For example, in one of next week’s b-side readings, “A Publisher’s Job is to Provide a Good API for Books,” Hugh McGuire explains how easy it is to use “semantic tagging” to build a smart index for your ebook. Like a conventional index, a smart index can tell you where all instances of John Smith appear in the book; but it can also tell you “where all people appear; where all instances of people named John appear; where all instances of people named Smith appear; that ‘my dear Granny Smith’ is a person and ‘my delicious Granny Smith’ is an apple.” A smart index is what McGuire calls a “semantic map” of the book. (Small Demons is a great illustration of whatthis might look like.)
Semantic mapping is impressive to me in three different ways, which I will explain in order of increasing impressiveness. First, it’s easy to see how this process of semantic mapping is a revolutionary tool for research. Such a tool could let you find a particular person or concept, or even all people or all concepts, in a particular book or collection of books (providing they are appropriately tagged and made available). You could also identify all books (that have been tagged and made available) that contain a reference to a particular event or person or concept. I can’t tell you how this would work but semantic mapping could help you do all of these things at the speed of search.
After semantically mapping many books, this sort of AI application could create categories of these maps, outside of the narrow genres with which humans currently approach books. I don’t know what the categories that emerged would look like, but I’m sure they would be illuminating. We might find a long-neglected category of books that humans had never attended to as such; or to put it another way: we might find a category of book that humans don’t know how to market, which is the exact experience Payne had with her book that led her to create Booxby. The point is, it would definitely be interesting to see books sorted into categories, or genres, based on the way their semantic maps look to an AI application. (I bet it would look a lot like the weirdly specific, yet creepily accurate categories that Netflix recommends to me.)
Now, imagine this process coupled with an AI application that collects data on reader-reported experiences of each of these categories. This data could be measures of sensibilities and emotions that, from the semantic map alone, an algorithm would not know to expect from a particular book (because AI doesn’t have emotions that we know of–yet). These experiential measurements could be straightforward, like the ones taken by the Whichbook application Jesse Savage brought to our attention (happy or sad, beautiful or disturbing). Or they might be more obscure, asking readers to what degree they felt “mournful” at the end of a particular book, how much it reminded them of themselves when they were children, etc.
Of course, we’ve always been able to get this kind of human feedback on particular books, or particular genres of books; or more recently, on books that contain particular content, such as a high frequency of words indicating bodies of water. All of that allowed us to associate certain kinds of reader experiences with certain genres, or certain kinds of content. But this AI application could associate certain kinds of reader experiences with certain kinds of semantic map. This means it could find two books that were likely to make you feel mournful, even if they had absolutely no content or human-created genre in common.
We would then have as data the content of the book, the semantic map of the book, and the experiential map of the book. Add to that the avalanche of consumer behaviour data that is already revolutionizing book discovery, and this would definitely yield some actionable results for acquisitions editors.
They could map their own collections, and make associations between certain kinds of semantic maps and available sales data. They could also map a submitted manuscript to get an idea of the reader experience. They might learn that even though the manuscript seems like a great example of what is selling right now in terms of content or genre, it actually is likely to produce an unpopular reader experience. They might find that a reader experience they thought would be undesirable is doing quite well in terms of sales. Or they could search the slushpile to find the weirdly specific, yet creepily accurate combination of content, genre, author profile, and reader experience they’re looking for. They could semantically map the ocean of self-published manuscripts (whose books were tagged in this manner, and made available) and treat it as a gigantic slushpile. And they could do all this without cracking a single manuscript, without having to summarize the content or squint through a badly edited first draft.
[Edited to add: I’m not saying that I think this is necessarily a good way for acquisitions to be decided; I have doubts, for the same reasons I don’t think a book should be chosen based simply on an author’s sales record.] These are the ways I imagine a combination of AI, tagging, and data-driven marketing will affect acquisitions. My understanding of all of these things is quite limited, but it was a fun experiment, and I’d like to know whether any of you think it sounds useful, dangerous, completely implausible, or utterly obvious.
Full vertical integration of publishing is inevitable, and when the consumer-facing app that effectively marries the Wattpad and Kindle Direct Publishing models achieves scale, many traditional publishing companies will be caught out, and some will fail.
Kitteridge thinks that Wattpad, an app that allows writers to post their own “books” (in what my partner pointed out is a somewhat blog-like format), is particularly well-positioned to “guide writers who are already connected to a sizeable audience through the self-publishing process and [take] a cut.” She points out that this is possible because “much of the work of publishing is so highly commoditised and there’s virtually no cost of entry any more. “
This week our class is taking a close look at what this means for the publishing business. Aleena and Taylor put it this way: “Internet business models for publishing are trending toward favoring the consumer more and more by lowering barriers to access to both publishing platforms and content. Are these business models inherently detrimental to the publishing business?”
I think many of us have similar feelings about this, but the way we answer the question can depend on what conception of the publishing industry we have in mind. If we wanted to measure to what extent the commoditisation of production methods constitutes a threat to traditional print publishing, we would need to know to what extent these commoditised publishing platforms (such as Wattpad or its future incarnation that Kitteridge predicts) are cannibalizing traditional book sales. We would need to see data showing whether the people using Wattpad to access content are doing so instead of, or as well as, buying print books; and whether the audience consuming media produced using these new methods is the same audience as the one that buys print media. Even if the audience is the same, it’s not necessarily the case that the need being filled by traditional publishing is the same as the need being filled by these new platforms. We might be better placed to answer the question if we take a broader conception of the publishing industry, and think of it as encompassing all those involved in the project of “making public.” This allows us to step outside of our point of view and consider the positions of the writer and the consumer, as well as the publisher. From that perspective we can see that no given technology is inherently bad for the industry broadly conceived. From that perspective, the availability of commoditised publishing platforms could be a very helpful development for all involved.
In an October 2017 episode of the podcast a16z, the founder of O’Reilly Media Tim O’Reilly talks about a pattern he observed in the antitrust case against IBM. He says he’s seen
companies which really are platform technologies that enable a whole ecosystem decide at some point to start competing with their ecosystem because there’s just not enough to go around. And at that point they screw up; the ecosystem starts to fall apart, government turns against them, people turn against them.It’s so amazing to see even idealistic companies forget that an ecosystem has to work for all of its participants: you can’t just think about yourself, you can’t just think about users, you actually have to think about your suppliers as well. And if you start competing with them […] the entrepreneurs who are making the services that your ecosystem depends on, go look for one that’s more friendly.
O’Reilly was talking about big corporations, not publishing houses; but the caution against “competing with your ecosystem” struck me as important for publishers to recognize. If we decide that commoditisation of publishing services, which makes things easier for writers and readers, is detrimental to the publishing industry, we put ourselves in a position where we are fighting the ecosystem, fighting readers and writers. Not only that, we are saying that writer and readers are not a part of the publishing industry at all. Looking at commoditisation more neutrally, I wondered: What is driving commoditization? What problem is it solving? And are publishers in any position to make ourselves relevant to that solution? Being as we are at the beginning of this chapter of our careers, we can take the perspective that it is an opportunity, not a detriment. One small example of the problems it is solving is the problem for writers of getting accepted by a publishing house in order to have their work made public. Platforms that facilitate self-publishing solve that one problem. And yes, by self-publishing some authors forego professional editing, marketing, design and promotion services; but most of them would never have had a chance at those things anyway. They would have ended up on the “slush pile” (a term I use with respect). Maybe it will be useful for publishers to think of commoditisation as enabling the slush pile to organize itself, to make itself accessible, not just to readers, but to traditional publishers. We can now see how readers respond to a book by seeing how much positive feedback it has received on services like Wattpad, which will help us decide which titles and authors deserve our attention. If we want to adapt to this new reality, I think publishers will need to embrace these technologies, update their business models, and reassess what services they should be providing; what services readers and writers will need most moving forward. What exactly might this look like? There aren’t a lot of wild success stories to list as examples here. Ebook subscription models like Oyster and Scribd, before they closed down, do illustrate the kind of approach I have in mind. While ebook subscription worked, HarperCollins’ CEO said that it was “very successful in really merchandising and mining the backlist.”
Why did Oyster and Scribd go south? Andrew Albanese and Jim Milliot cite speculations that “the cost of paying for their subscribers’ reading consumption [was] simply exceeding the revenue brought in from monthly subscription fees.” They also point out that “Scribd (which, like Oyster, pays publishers their full retail cut for books read by subscribers) was forced to cut romance and erotica offerings, because romance fans—notoriously voracious readers—were apparently reading far more than expected, driving up Scribd’s payments.” If this is the case, it’s clear that the appetite for this service was there, and it did solve a problem for readers; it was the details of the business model that needed to be worked out.
The important thing is that while it worked, ebook subscription solved problems for different parts of the ecosystem: the reader who wanted convenient access to many books, and the publisher who needed to monetize their backlist. This is the kind of thinking I am trying to encourage in myself in order to understand how to take advantage of the challenges and opportunities provided by commoditisation.
Also, as Juan suggested in our February 19 PUB 802 class, it could be that Amazon is trying to change our idea of what shopping is. After getting used to shopping in a store without queues, maybe we’ll find it even more unbearable to wait in line for our groceries. Amazon could be trying to tip our preferences toward shopping online even more than we already do. Along that line of thinking, Amazon Go is a way for them to collect data on the physical shopping experience; data they would otherwise have to pry from their competitors’ cold, dead hands.
In fact, collecting data is a good enough reason in itself for the move to brick-and-mortar. Data has been called “the oil of the digital era.” Why would an online marketplace care about data from physical marketplaces? Companies need accurate, up-to-date data from all areas of consumer behaviour in order to predict trends and preferences. If you’re selling shoes, you need to know more about the consumer than their current shoe preferences; you need to know who they are, what they will like in six months, and why. I can’t begin to tell you exactly what data they are collecting, but I think Amazon Go is a great example of what Evgeny Morozov calls “data extractivism.” Morozov quotes Andrew Ng saying that “at large [tech] companies, we often launch products not for the revenue but for the data … and we monetise the data through a different product.” Whatever the data is, I think Amazon Go is less of a store than a product; and it’s of a new class of product designed not to collect revenue, but data.
Perhaps most importantly, though, all that data about consumer behaviour and preferences needs to be monetized, often using AI. That AI needs to be tested on a large scale, in a realistic setting, and Amazon Go is the company’s live tech laboratory. Take Amazon’s Echo Look as an example. While most people still prefer to buy clothes in person, Echo Look will, as Juan predicted, teach them to buy online. It takes a picture of you, analyzes it, and makes recommendations; and “it’s always learning.” Amazon Go will always be learning, as well; it’s the perfect lab for Amazon to test new technologies in “computer vision and machine learning.” Vice President Gianna Puerini says, “This technology didn’t exist [before Amazon Go]. It was really advancing the state of the art of computer vision and machine learning.” Amazon Go allows Amazon to test and develop their computer vision and AI technology at a much larger and faster scale. (Especially considering that consumers will take time to adapt to letting a cloud-enabled camera watch us change—although less time than we would have taken in the past).
Conclusion: An example of evolution
I think Amazon’s main reasons for opening a brick-and-mortar grocery store are to collect and monetize data, and to test new technologies. The move into brick-and-mortar is an evolution of their strategy, not a devolution. It’s a strategy that creates for Amazon an opportunity to access up-to-date real-world data, and to test new AI technology with real consumers. These are exactly the things that a big tech firm that runs on data should be thinking about.
In 2012 the Canadian government passed a series of reforms to the Copyright Act, which included a provision that the Act be reviewed every five years. In December 2017 the Canadian government ordered the first such review of the Canadian Copyright Act (Geist, “Copyright”) and we can expect the committee in charge to suggest significant updates and reforms. As a future publisher, I want to identify and test some of the underlying assumptions that inform Canadian copyright law so that I can better assess the coming changes.
Jay Makarenko describes a copyright as “a legal recognition of a person’s natural right of ownership over the things s/he creates.” Given that, this strikes me as the most basic assumption underlying Canadian copyright law:
Assumption 1: An individual has a “natural right of ownership over the things s/he creates” (Makarenko).
This notion of a natural right derives from the concept of droit d’auteur (Younging 57) and is baked into our legal conception of copyright. It explains why copyright is conferred automatically as soon as an idea is given original expression, not conferred by laws. Canadian law assumes this natural right of the individual creator and tries to balance it with “public freedoms” (Younging 58) as well as the many social, cultural, and economic benefits to be gained from a robust and representative public domain.
This view of individual ownership of original expression is not universally accepted. Some view “intellectual goods” as “social property” (Makarenko), arguing that “one individual cannot make the claim ‘it is mine, because I made it.’ The reality is that society … participated in the making of the work and, as such, the work cannot be claimed as private property by one individual” (Makarenko). If this is so, it would seem that individuals have no natural right over our original expressions, because they are not in fact original.
But rejecting the natural rights view in favour of this social property view would not necessarily force us to say that copyright is illegitimate. Makarenko explains another possible justification for copyright is that “copyright and private ownership of intellectual goods are valuable because they will bring great economic and cultural benefits to society.” On this view, creative production is incentivized because creators can sell their creations. This brings me to Assumption 2 underscoring Canadian copyright law:
Neil Gaiman challenges this assumption. He noticed that in “places where I was being pirated, particularly Russia … I was selling more and more books. People were discovering me through [my] being pirated and then they were going out and buying the real books.” Gaiman persuaded his publisher to make his work American Gods available for free for a month, even though it was still selling well; sales through independent bookstores tripled. This suggests that copyright infringement should not necessarily de-incentivize creative innovation. However, Gaiman also mentions that they only measured the impact on sales in independent bookstores, so it is impossible to conclude from this anecdote whether sales overall were affected positively or negatively.
Similarly, Ernesto Van der Sar interprets a study by Professor Tatsuo Tanaka of the Faculty of Economics at Keio University as indicating that “decreased availability of pirated comics doesn’t always help sales. In fact, for comics that no longer release new volumes, the effect is reversed.” He quotes Tanaka saying that “displacement effect is dominant for ongoing comics, and advertisement effect is dominant for completed comics.”
If Van der Sar’s interpretation is correct, then there are cases when copyright is infringed, yet creative expression is still rewarded by increasing exposure and sales. However, as my colleague Sarah Pruys has pointed out (in a private Hypothes.is discussion in my PUB 802 class at Simon Fraser University), Van der Sar takes some editorial liberties with Tanaka’s findings and ends up overstating the claim. So this anecdote, like Gaiman’s, does not constitute a conclusive rebuttal to Assumption 2.
Makarenko describes a rebuttal we can take a little more seriously:
Assumption #3: Copyright stifles the free flow of ideas.
This might happen when a copyright owner hoards their original expression, or limits its distribution by charging a fee, excluding those who cannot afford it. This keeps valuable ideas out of public access, limiting society’s resources of intellectual goods.
The case of orphan works can be said to support this assumption. Orphan works are those for whom “the copyright owner cannot be identified or located” (Harris). The argument goes that if orphan works are protected too zealously, they will be inaccessible, and the public domain will suffer. To prevent this situation, the term of copyright should be short.
It’s worth pointing out that even if you were to find the author of the orphan work, there is no reason to assume you would be granted the right to copy. Also, an orphan work is not protected any longer than any other work. Copyright for works with unknown authors only lasts for “the remainder of the calendar year of the first publication of the work plus 50 years” (“Guide”). This means orphan works might actually enter the public domain significantly earlier than works with a known author.
Let’s say a book is published in 2018. If at any point I want to copy it and can’t find the author, I have to wait until the orphan work enters the public domain in 2068. However, the only case in which a work that is not orphaned would enter the public domain so soon after publication is if the author died in the year of publication. Let’s say the author survived until 2068. Then the work would not enter the public domain until 3018. So it might well be easier to wait out copyright on many orphan works than on published works with known authors.
A work entering the public domain for everyone to copy freely is only one way that society can benefit from it. We can still use the work under fair dealing. And the Copyright Act allows for a license to be issued at the discretion of the Board if they are satisfied that “the applicant has made reasonable efforts to locate the owner of the copyright and that the owner cannot be located” (Sookman).
This does still involve some risk in case the copyright owner shows up within five years to recover royalties. However, in 2012 the then-Conservative Canadian government passed a series of reforms to the Copyright Act. The relevant change is basically a good-faith gesture that drastically lowers the amount of damages you would have to pay if you took a risk on using an orphan work for non-commercial purposes and were later found to have infringed copyright. This change seems fair, but Bill C-11 overall was not well-received by the publishing industry, and it will be interesting to see what comes out of this year’s five-year review.
These three points suggest that copyright protection for orphan works is no less fair than copyright protection for known authors. But that doesn’t mean that copyright terms are fair; that is, they don’t refute Assumption #3.
Several of my colleagues have suggested that creators should be required to frequently extend or renew their copyright in order to prevent orphan works from being withheld from the public domain. I think this would contradict two of copyright’s key characteristics. First, that it is meant to incentivize creative innovation. Cory Doctorow points out that “giving creators more copyright on works they’ve already created doesn’t get them to make new ones, and it reduces the ability of new artists to remix existing works.” If a work has already been created, it has clearly already been incentivized; so why would we require the creator to pay to reproduce their own work?
I think the way you answer this question determines, or is determined by, how you feel about Assumption #1, the second key characteristic of copyright, which is that it is a natural right. If a creator is required to re-purchase the right to their own creation, then copyright is not conferred by nature, but by law. So to call for a copyright term that expires in the author’s lifetime is to say that there is no natural right to copy in the first place; so what is it that the author is extending?
So, if I had to choose between Assumptions #1 and #2 as justifications for copyright, I lean toward #2 and a social conception of intellectual goods. I don’t think the Canadian copyright situation is ready for any radical changes in that direction, but I would like to see the idea reflected more in copyright law.
Younging, Greg. “Gnaritas Nullius (No One’s Knowledge): The Essence of Traditional Knowledge and Its Colonization through Western Legal Regimes.” In Indigenous Editors Circle and Editing Indigenous Manuscripts workshop course pack, Humber College, 2017, Etobicoke, ON.
In “The Information: How the Internet Gets Inside Us,” Adam Gopnik describes three categories of attitudes toward the notion that books no longer matter. I tried to understand the categories, and where I fit in, by looking at the basic claims with which their arguments begin.
Never-Betters and Better-Nevers seem to accept the idea that books no longer matter as either, optimistically, on the brink of becoming true (the situation has never been better); or, pessimistically, at risk of becoming true (it would be better if this situation had never come about). But I don’t think the Ever-Wasers (those who accept the claim and say “’twas ever thus”) necessarily accept the claim at all. They seem more concerned with explaining the phenomenon of “people acting as though books no longer matter.”
So my first point is that I think the first two categories are different from the third. Never-Better and Better-Never are more useful in describing our levels of anxiety about the technological revolution, while the Ever-Waser position is that it is not as new and different a revolution as people think.
The Ever-Waser is more concerned with explaining why people think the book no longer matters. I don’t really have an answer to that, but like the Ever-Waser, I am agnostic about whether technological change can be “good” or “bad.” Gopnik says that “If you’re going to give the printed book, or any other machine-made thing, credit for all the good things that have happened, you have to hold it accountable for the bad stuff, too.” But I (and I suspect Gopnik) think it’s simplistic to give any technology that much credit either way. Human behaviour and activity is heavily influenced by, but not determined by, the technology at hand. I think that makes me an Ever-Waser on that point. I also don’t think the communications / information / technology revolution is unprecedented. It does seem true to me that these types of revolutions happen throughout history. It might be “our” big social revolution, but more in the sense that every age has its perceived big social revolution.
Like the Never-Betters, I believe that “information [is becoming] more free and democratic” and somewhat agree that “news will be made from the bottom up.” This means that stories will be told from a wider array of perspectives which have been previously suppressed. The thing is, stories can be suppressed for good reasons (they are not true, they promote hatred, they are considered unimportant and are) or bad reasons (they are from oppressed groups in society, they counter the narrative that dominant power prefers, they are considered unimportant but are not). So I guess I’m an Ever-Waser on that point.
Like the Better-Nevers, I agree that “books and magazines create private space for minds in ways that twenty-second bursts of information don’t,” but I don’t think that means twenty-second bursts are somehow necessarily inferior. I definitely don’t think that “the world that is coming to an end is superior to the one that is taking its place,” in part because I don’t think it makes sense to say that there is a world that is coming to an end.
So I seem to be an Ever-Waser. I do take the warning to heart, though, about the Romans and the Vandals. Just because it’s happened so many times before doesn’t mean that this time, our fears will also not come true. I also don’t have any answer to the question: “If it was ever thus, how did it ever get to be thus in the first place?”