After reading Dan Cohen’s post on Hacking the Academy a week ago, one of his questions has been on my mind: can an algorithm edit a journal? For me, the answer is of course it can. Microsoft Word’s grammar-checker is a good example. It dutifully checks an entire document and offers several options for the author. It is easy to imagine a grammar-checker correcting an article on its own, and that would be a case of an algorithm on its own editing a journal.
Even more interesting to me, however, is one of the deeper issues Dr. Cohen’s question suggests: computers are encroaching on the production of scholarship. Online translators, such as Google Translate, are everyday examples of computers constructing meaningful statements. I know nothing about artificial intelligence, but I imagine scientists have created, or are near to creating, impressive programs that can produce entire documents that look like scholarship. Incidentally, I use the phrase “looks like scholarship,” because I am not exactly sure what a computer’s scholarship would be, except that it might resemble human scholarship. For this post, however, I will avoid tricky issues associated with computers being able to think and use language. Philosophers have devoted so many books to those topics that my humble blog post would not do justice to the academic debate.
Regardless, the fundamental issue I am tackling here remains. Are computers encroaching on scholarship? Yes, they are. More importantly, most scholars probably do not have a sense of just how much processing is going on in their computer. This is especially ironic for those who argue that computers — or new media, if you prefer that term — have had a negative influence on scholarship. As they write their journal articles and book manuscripts, probably hundreds of algorithms are directing their computers to translate what they type on a keyboard into a document.
I offer an example to remind everyone of just how much is going on behind the scenes. A few months ago when I was archiving digital images at the Smithsonian, I stumbled across an interesting file type. It is called “camera raw.” According to what I remember (and the Wikipedia page I used for reference), images in “raw” format are digitally unprocessed and do not look like normal photographs. A digital camera must process the “raw” file data, from which it produces the kinds of digital images with which all of us are familiar.
What surprised me about raw images is that I never thought about how much processing a digital camera does. When I was working with the Smithsonian’s digital photos, I was only paying attention to the images themselves: how clear its subject is, if the image needs to be cropped, if the photo has the right angle, and so forth. If I did think about the processing in the background, either I was impatient about the computer being too slow or I was trying to minimize file sizes. Had I thought more about it, I would have remembered just how strange, for lack of a better word, the information a computer uses to produce images is. In a sense, ultimately, each digital photo is a unique combination of ones and zeroes.
What does my experience with digital images teach me about the influence of computers on scholarship? It makes me question the distinction between people and computers. Even though it may seem obvious that Dr. Cohen created a blog post, upon reflection authorship in new media is complex. No doubt, the medium he used to communicate his words influences the way he writes. Blog posts tend to have a certain style, and this is partly because of technology. It is easy to change blog posts once one has published them. Writing for blogs tends to be more informal than writing for academic journals. I would not argue that Dr. Cohen’s post is as informal as other blogs, which, say, go into the minutiae of their authors’ day to day lives. Still, as I scan through his previous posts, I notice personal details, such as posts about his TED Conference presentation, which separate Dr. Cohen’s blog writing from more formal academic prose.
I am not arguing that it is incoherent to say that people author digital documents. Human beings are obviously part of the process. Instead, I am arguing that we need to reconsider our assumptions about authorship in new media and look for the underlying story, which is probably missing from many debates about the value of new media. We need to think about how the experience and output of scholars changes simply by using a computer.
Keeping with the proper style of blog, I am going to end this post at this point. One of the benefits of publishing something online is that one can revisit and leave open the issues one has raised.