18 January 2010

The journal article... of the future!

ResearchBlogging.orgElsevier is trying a new format for research articles here and they are, somewhat pretentiously, calling it the “article of the future.” It’s always risky to attach “future” to these sorts of things, as it tends to get people bitching that they don’t have their flying cars and personal jet packs yet.

As far as I can tell, Elsevier’s experiment seems to be confined to one of its big flagship journals, Cell, for now. An example is here, which I like because it has brains (click to enlarge).

What problems of the traditional research paper are they trying to solve? An editorial says:

(W)e have worked to develop an online format that breaks free from the restraints of paper and allows each reader to create a personalized path through the article's content based on his or her own interests and needs.

That’s supposed to be a problem? Of all the various media that I use, print is one of the easiest to create a personalized path. It’s called, “turning the page.” I guess the editors at Cell never read murder mysteries, so never experienced that thrill of knowing that you can go right to the end of the book find out who the killer is, without all that tedious investigative mucking about in the middle.

Let’s say I want to create a “personalized path” and read the introduction and discussion – very common thing for researchers to do. Notice the number of tabs at the top? I have to click, scroll through the introduction, scroll back up to the top of the page, locate the tab, click again, and start scrolling down again. Clicking through all the tabs feels like a lot of pointless work. Further, click one of the main tabs, and you are often presented with two more sub-tabs, one for the text, and one for the references.

The constant breaking down of the article into little pieces reminds me of David Pogue’s jab at wizards in his talk here (fast forward to 7 minutes and 50 seconds).

I much prefer the single page approach. I start at the top, and can just keep scrolling down. HighWire Press journals even provide a little shortcut menu at the top of each section, making it easier for me to jump to the next section I want to read.

Tabbed and hyperlinked navigation through the Introduction, Results, Figures, Experimental Procedures, and Discussion allows subject-area experts to quickly access in-depth information on a particular experiment while providing more general readers an opportunity to absorb the conceptual insights without being overwhelmed by additional details.

Let the skimmers be lazy while still providing details that experts demand? A good abstract does that. Again, this feels like a solution being offered to a non-existent problem.

(A) film strip of thumbnails for all of the figures... allows a reader to rapidly scan through the data and then connect from an individual figure to the related textual discussion of the findings.

This is under a tab called “Data,” rather than “Figures,” since it can include spreadsheets and the like. I’m neutral to this. It neither seems to add value or to be annoying.

The Results tab lets the reader view a zoomable figure, the legend, and associated Results text easily on a single screen.

Keeping the text and relevant figure visible side by side at the same time is a excellent idea, but it’s hamstrung here with a horribly clunky execution. The results break down into three skinny columns: one for text, one for the figure legend, and on for the figure caption. The text column is so skinny, it forces me to do more near-constant scrolling. I want to read, not scroll. Reading is made even harder because paragraphs aren’t separated by either spaces or indents.

Highlights and a Graphical Abstract on the landing page of each article complement the traditional Summary text and promote article browsing by creating a visual summary and bullet points that easily convey the main take-home message of the paper.

Cell may be able to get good mileage out of the “graphic abstract,” because so many of their papers are all about cellular and biochemical pathways, so it’s easy to make a flowchart summarizing results. I’m not sure that this is something that could be used by other journals. Perhaps a better way to think about it is to think in terms of general “entry points”: images that can provide a way in for the reader. They may not have to be summaries, but something relevant like a picture of an organism, a field site, or a juicy pull quote.

(T)he online display fully integrates supplemental information including multimedia content within the context of the main article and facilitates more fluid navigation between the two.

Then why is it called “supplemental” at all? If it’s incorporated into the paper in the same way, present the data democratically. I suppose that it needs to be labeled as such because it’s material that won’t go into the print edition.

The treatment of references is one of the nicest additions. You can reorder them alphabetically or chronologically, and there’s a little sparkline at the top of the main reference section showing how the references are distributed by year of publication. You can also click to see how the article is cited in the text, which could be very useful in identifying interpretation problems, and so on. Very well done, and really does point a way forward in using references that I haven’t seen before.

Not showing the digital object identifiers (DOI) for articles in the references is nearly inexcusable, though. I don’t want to have to go through PubMed to get to an article, I want to go directly to the article.

There are audio and video clips, given the cutesy names of “Paper Flicks” and “Paper Clips.” One example of video I’ve looked at is a basically a mundane PowerPoint talk dumped to video. This does not interest me, personally. Video could be helpful if the actual material is highly visual, like movies of cells dividing or something.

They’ve also added a comment section. If the experience at other journals is anything to go by, they will sit and gather dust, for the most part. No rating system, though.

Finally, I like that they are offering a single PDF file that rolls in all the supplemental material. Very logical and sensible. I’ve often had to go digging for supplemental material, only to be confronted with multiple PDF files for me to keep track of.

My current analysis is much like my first impression: I’m unconvinced. The approach feels like someone with a hammer looking for a nail to hit: “We have these tools, so are there any problems we can solve with them?”

Apart from the ability to do a little impromptu data mining in the reference sections, and a convenient PDF, this “article of the future” feels needlessly complicated. It’s a little too early to tell, but I wonder if the “article of the future” might end up looking like predictions from old popular science and mechanics magazines.

“In the future, we’ll play polo... on motorboats!”

Hat tip to Dr. Doyenne.


Marcus E. 2010. 2010: A Publishing Odyssey Cell 140(1): 9. DOI: 10.1016/j.cell.2009.12.048

No comments: