There is much discussion these days about the future of scholarly publishing. Much of this surrounds the value of traditional publishers. When challenged those publishers point to the value and potential value they create. Here is Elsevier responding to a recent boycott led by mathematician Tim Gowers:
And we invest a lot in infrastructure, the tags and metadata attached to each article that makes it discoverable by other researchers through search engines, and that links papers together through citations and subject matter. All of that has changed the way research is done today and makes it more efficient. That’s the added value that we bring.
One of those elements of added value is the format for the published article itself. Publishers are so confident that this adds value that they permit working paper versions — prior to getting the publisher magic touch — reside freely on the web. To be sure, articles are typeset and tables and figures are cleaned up to look good on paper. But does all that make it better for those who are looking for knowledge?
Publishers know there is more potential there. Elsevier has launched its “Article of the Future” experiment to show what digitisation might make possible. Here, for instance, is their representation of a future mathematics article. The main text looks like a working paper. It isn’t a pdf but a webpage and the equations are rendered in latex. And it looks awful. Indeed, for reasons that are perplexing it is hard to read. Now I’m assuming stuff like that can be fixed but having it out there doesn’t inspire confidence.
But let’s focus on other elements that they have put in. First, they have an interactive element to allow you to play with a graph of a particular formula. That looks like a good feature to have available to readers. Second, they have included a video abstract. This could be a good thing but it shows one of the authors in front of a blackboard. This might be useful, it is hard to say. I have to admit that he did look the part. But I can imagine that seminars could be embedded here and that such things may be of use to readers. (Some academics have taken it upon themselves to provide such materials on their own websites, here is Glen Weyl). In another prototype, there are videos all through the article. Third, there are hyperlinks everywhere. The most useful of these link in to Elsevier’s database for references.
The things Elsevier are trying to do here are sensible from an adding value perspective. But they augment print and are still fundamentally based in it. The problem is that as the technology for sharing information changes we can refocus on what we should really care about. Print was a repository of knowledge. It allowed access and catered to the person who would spend time with an article. The additions Elsevier proposes are all about spending more time with the content. But I would argue that that is a narrow view of scholarly communication.
If you are like me, when you look at a scholarly article, most of the time, you want to spend as little time with it as possible. You want to look at it, see if it is relevant and get out. Better still you might want to find what you are looking for quickly. The more context you are required to have, the worse the experience is. Now, to be sure, there are occasions where you want as much as an article can give you. Invariably, print versions come up short there with appendices moved elsewhere to save on print and little extra content like PowerPoint presentations and even video thoughts from the authors.
But how can you cater to those who need to access knowledge efficiently versus those who need to access it deeply?
Here I am going to present my approach to doing that. It is focussed on reading and so I am imagining reading articles on a tablet. But I want it to be efficient. To this end, I took an old paper of mine published in Economics Letters in 1996. The idea was to find something short but also mathematical. If you have access, here is the article as currently represented on Elsevier and here is what the printed version looks like.
What I did was take the text and reformatted it using iBooks Author. If you have an iPad, click on this link and open the attachment in iBooks. If you don’t have an iPad you can see a little of what I have done with this pdf version. Notice that that version is already more readable than the published version. If you don’t have an iPad you can watch the following that gives a run down of the interesting features.
To be sure, those features are three-fold. First, you can easily adjust the font size for easy reading and you can scroll through the article very quickly rather than by confined to pages. Second, the idea is that proofs are things that are for in-depth reading while other stuff is not. So in portrait mode, the article is presented in a light form but as you turn it to landscape you get the full thing at the point you are at. You can hide proofs, literature reviews and all manner of other stuff that are secondary to the knowledge but often embedded in the article and require the reader to sort through. Also, on the proof front, I presented the proof as a PowerPoint as well that allows you to work through it. These are often better than textual proofs as they allow you to present steps and build through. There is much more that can be done there but the point is iBooks makes it easy to embed these things and call them up.
Finally, and this is the main point: I could do this myself. The author is the best person to think about how to present the material in a paper. We take so much away from authors in the whole editing for print process and this harms scholarly communication. These tools allow authors to put in the enhancements as they see fit and, indeed, compel them to think more about the reader. I know it did with this almost 20 year old article of mine and I have to admit I didn’t go nearly far enough there. Imagine thinking about that when presenting current research.
You might wonder: how long did all this take? Well, I did it initially with iBooks Author 1.0 and was exploring as much as writing. So it took about 6 hours. Now that I know what I am doing it would take me about 4 hours for a regular length article. That is not much for so much greatly improving the experience for the readers of your work. When you spend years on a paper, 4 hours making it easier to read doesn’t seem much of an ask. If iBooks was optimised for this, it would take even less time.
This is just a start but I think it shows that there is real potential in using new tools to enhance scholarly communication. But the key is to put the reader front and centre and to ensure that authors can more directly communicate and represent knowledge to readers. We also need to ensure that readers who need to access knowledge efficiently are catered for. On that score, this extends beyond scholarly publishing. I recommend this talk by Craig Mod presented at the Tools for Change in Publishing conference that argues for a ‘sub-compact’ approach to magazine publishing that will focus on readers.
Finally, traditional publishers allow scholars to post non-published versions of articles for free and open to all. Wouldn’t it be something if those non-published articles were, in fact, the ones people preferred to access over what the published versions are? Nothing would shake up the traditional market power of those publishers quicker.