Fillip

The Scan and the Export
Sean Dockray

The scan is an ambivalent image. It oscillates back and forth: between a physical page and a digital file, between one reader and another, between an economy of objects and an economy of data. Scans are failures in terms of quality, neither as “readable” as the original book nor the inevitable ebook, always containing too much visual information or too little.


Technically speaking, it is by scanning that one can make a digital representation of a physical object, such as a book. When a representation of that representation (the image) appears on a digital display device, it hovers like a ghost, one world haunting another. But it is not simply the object asserting itself in the milieu of light, information, and electricity. Much more is encoded in the image: indexes of past readings and the act of scanning itself.


An incomplete inventory of modifications to the book through reading and other typical events in the life of the thing: folded pages, underlines, marginal notes, erasures, personal symbolic systems, coffee spills, signatures, stamps, tears, etc. Intimacy between reader and text marking the pages, suggesting some distant future palimpsest in which the original text has finally given way to a mass of negligible marks.


Whereas the effects of reading are cumulative, the scan is a singular event. Pages are spread and pressed flat against a sheet of glass. The binding stretches, occasionally to the point of breaking. A camera driven by a geared down motor slides slowly down the surface of the page. Slight movement by the person scanning (who is also a scanner; this is a man-machine performance) before the scan is complete produces a slight motion blur, the type goes askew, maybe a finger enters the frame of the image. The glass is rarely covered in its entirety by the book and these windows into the actual room where the scanning is done are ultimately rendered as solid, censored black. After the physical scanning process comes post-production. Software—automated or not—straightens the image, corrects the contrast, crops out the useless bits, sharpens the text, and occasionally even attempts to read it. All of this computation wants to repress any traces of reading and scanning, with the obvious goal of returning to the pure book, or an even more Platonic form.


That purified, originary version of the text might be the e-book. Publishers are occasionally skipping the act of printing altogether and selling the files themselves, such that the words reserved for “well-scanned” books ultimately describe e-books: clean, searchable, small (i.e., file size). Although it is perfectly understandable for a reader to prefer aligned text without smudges or other markings where “paper” is nothing but a pure, bright white, this movement towards the clean has its consequences. Distinguished as a form by the fact that it is produced, distributed, and consumed digitally, the e-book never leaves the factory. 


A minimal gap is, however, created between the file that the producer uses and the one that the consumer uses—imagine the cultural chaos if the typical way of distributing books were as Word documents!—through the process of exporting. Whereas scanning is a complex process and material transformation (which includes exporting at the very end), exporting is merely converting formats. But however minor an act, this conversion is what puts a halt to the writing and turns the file into a product for reading. It is also at this stage that forms of “digital rights management” are applied in order to restrict copying and printing of the file.


Sharing and copying texts is as old as books themselves—actually, one could argue that this is almost a definition of the book—but computers and the Internet have only accelerated this activity. From transcription to tracing to photocopying to scanning, the labour and material costs involved in producing a copy has fallen to nothing in our present digital file situation. Once the scan has generated a digitized version of some kind, say a PDF, it easily replicates and circulates. This is not aberrant behaviour, either, but normative computer use: copy and paste are two of the first choices in any contextual menu. Personal file storage has slowly been migrating onto computer networks, particularly with the growth of mobile devices, so one’s files are not always located on one’s equipment. The act of storing and retrieving shuffles data across machines and state lines. 


A public space is produced when something is shared—which is to say, made public—but this space is not the same everywhere or in all circumstances. When music is played for a room full of people, or rather when all those people are simply sharing the room, something is being made public. Capitalism itself is a massive mechanism for making things public, for appropriating materials, people, and knowledge and subjecting them to its logic. On the other hand, a circulating library, or a library with a reading room, creates a public space around the availability of books and other forms of material knowledge. And even books being sold through shops create a particular kind of public, which is quite different from the public that is formed by bootlegging those same books.


It would appear that publicness is not simply a question of state control or the absence of money. Those categorical definitions offer very little to help think about digital files and their native tendency to replicate and travel across networks. What kinds of public spaces are these, coming into the foreground by an incessant circulation of data?


Two paradigmatic forms of publicness can be described through the lens of the scan and the export, two methods for producing a digital text. Although neither method necessarily results in a file that must be distributed, such files typically are. In the case of the export, the system of distribution tends to be through official, secure digital repositories; limited previews provide a small window into the content, which is ultimately accessible only through the interface of the shopping cart. On the other hand, the scan is created by and moves between individuals, often via improvised and itinerant distribution systems. The scan travels from person to person, like a virus. As long as it passes between people, that common space between them stays alive. That space might be contagious; it might break out into something quite persuasive, an intimate publicness becoming more common.


The scan is an image of a thing and is therefore different from the thing (it is digital, not physical, and it includes indexes of reading and scanning), whereas a copy of the export is essentially identical to the export. Here is one reason there will exist many variations of a scan for a particular text, while there will be one approved version (always a clean one) of the export. A person may hold in his or her possession a scan of a book but, no matter what publishers may claim, the scan will never be the book. Even if one was to inspect two files and find them to be identical in every observable and measurable quality, it may be revealed that these are in fact different after all: one is a legitimate copy and the other is not. Legitimacy in this case has nothing whatsoever to do with internal traits, such as fidelity to the original, but with external ones, namely, records of economic transactions in customer databases.


In practical terms, this means that a digital book must be purchased by every single reader. Unlike the book, which is commonly purchased, read, then handed it off to a friend (who then shares it with another friend and so on until it comes to rest on someone’s bookshelf) the digital book is not transferable, by design and by law. 
If ownership is fundamentally the capacity to give something away, these books are never truly ours. The intimate, transient publics that emerge out of passing a book around are here eclipsed by a singular, more inclusive public in which everyone relates to his or her individual (identical) file.


Recently, with the popularization of digital book readers (a device for another man-machine pairing), the picture of this kind of publicness has come into greater definition. Although a group of people might all possess the same file, they will be viewing that file through their particular readers, which means surprisingly that they might all be seeing something different. With variations built into the device (in resolution, size, colour, display technology) or afforded to the user (perhaps to change font size or other flexible design elements), familiar forms of orientation within the writing disappear as it loses the historical structure of the book and becomes pure, continuous text. For example, page numbers give way to the more abstract concept of a “location” when the file is derived from the export as opposed to the scan, from the text data as opposed to the physical object. The act of reading in a group is also different—“Turn to page 24” is followed by the sound of a race of collective page flipping, while “Go to location 2136” leads to finger taps and caresses on plastic. Factions based on who has the same edition of a book are now replaced by those with people who have the same reading device. 


If historical structures within the book are made abstract then so are those organizing structures outside of the book. In other words, it’s not simply that the book has become the digital book reader, but that the reader now contains the library itself! Public libraries are on the brink of being outmoded; books are either not being acquired or they are moving into deep storage; and physical spaces are being reclaimed as cafes, restaurants, auditoriums, and gift shops. Even the concept of donation is thrown into question: when most public libraries were being initiated a century ago, it was often women’s clubs that donated their collections to establish the institution; it is difficult to imagine a corresponding form of cultural sharing of texts within the legal framework of the export. Instead, publishers might enter into a contract directly with the government to allow access to files from computers within the premises of the library building. This fate seems counter-intuitive, considering the potential for distribution latent in the underlying technology, but even more so when compared to the “traveling libraries” at the turn of the twentieth century, which were literally small boxes that brought books to places without libraries (most often, rural communities).


Many scans, in fact, are made from library books, which are identified through a stamp or a sticker somewhere. (It is not difficult to see how the scan is closely related to the photocopy, such that they are now mutually evolving technologies.) Although it circulates digitally, like the export, the scan is rooted in the object and is never complete. In a basic sense, scanning is slow and time-consuming (photocopies were slow and expensive), and it requires that choices are made about what to focus on. A scan of an entire book is rare—really a labour of love and endurance; instead, scanners excerpt from books, pulling out the most interesting, compelling, difficult-to-find, or useful bits. They skip pages. The scan is partial, subjective. You and I will scan the same book in different ways. An analogy: they are not prints from the same negative, but entirely different photographs of the same subject. Our scans are variations, perhaps competing (if we scanned the same pages from the same edition), but, more likely, functioning in parallel. 


Completists prefer the export, which has a number of advantages from their perspective: the whole book is usually kept intact as one unit, the file; file sizes are smaller because the files are based more on the text than an image; the file is found by searching (the Internet) as opposed to searching through stacks, bookstores, and attics; it is at least theoretically possible to have every file. Each file is complete and the same everywhere, such that there should be no need for variations. At present, there are important examples of where variations do occur, notably efforts to improve metadata, transcode out of proprietary formats, and to strip DRM restrictions. One imagines an imminent future where variations proliferate based on an additive reading—a reader makes highlights, notations, and marginal arguments and then re-distributes the file such that someone’s “reading” of a particular text would generate its own public, the logic of the scan infiltrating the export.