Last week, I went to a SF Bay Area ACM Chapter talk by Peter Binfield and Sara Wood of PLoS, which they covered the motivation for establishing the PLoS journals, and talked about some of the challenges of running this operation. PLoS is a non-profit publisher of scientific and medical information that arose out a desire to reduce journal subscription costs to academic libraries.

PLoS publishes six specialized open-source journals, and one additional uber-journal, PLoS-One, that includes everything else. While they refer to themselves in print (and in the talk) as publishers of scientific research literature, they in fact appear to be focused more narrowly on the biomedical literature.

First, a small digression on open-source publishing:

Whereas traditional for-profit publishers use the subscription model for financing their operations, PLoS charges authors a hefty fee ($1,350 to $2,900) to publish their papers, which are still peer-reviewed in the same way as traditional journals. This model is not without its problems, as Michel Beaudouin-Lafon points out in his viewpoint article in CACM. In his analysis, the costs imposed on an organization that publishes extensively exceed the current subscription fees! This is particularly important for fields such as Computer Science with its emphasis on many smaller publications rather than a few larger, multi-author articles typical of the biomedical literature. (I should point out that there are other models for open-source publishing that don’t involve authors’ fees. JoDI, for which I am a Theme Editor,  is one such journal.)

On the other hand, PLoS content is publicly available with a liberal licensing scheme that allows any use with proper attribution. They make their publications and metadata available in a variety of formats, including HTML, PDF, and XML. This means that one can relatively easily crawl their site and collect their articles for presenting through other means.

This open access policy made possible, for example, a recently-released iPad app for browsing PLoS content. The app allows you to browse PLoS articles by PLoS journal and major subject classification (e.g., Chemistry, Hematology, Plant Biology, etc.) and download the articles for reading. Articles are presented as double-column pages, and many include color images. The app implements a page-flipping metaphor that shows animated transitions as you slide your finger left or right across the page. It can also show a list of page thumbnails for moving directly to a specific page.

Other than that, the app is pretty rudimentary: there are no bookmarks, no means of taking notes or saving passages, and no free-form annotation. Furthermore, the PDF files cannot be zoomed in, which makes for strained reading given the rather small font size.

The app also surfaces another oddity about PLoS, one that undermines their claim to represent all of science. Within the PLoS ONE journal, the one that is designed to accommodate the entire breadth of scientific endeavor, articles are classified into over 50 subject categories, only six or seven of which are not some aspect of medicine. Even when you select a subject area such as Computer Science,you still find many articles related to medicine and biology.  What’s even more annoying is that the articles appear to have no subject classification beyond this grouping, making it hard to filter the articles more precisely. Not surprisingly, there are no library science articles (unless perhaps they are hidden in the cell biology subject classification…).

In summary, PLoS represents an interesting alternative to conventional academic publishing, and enables some new kinds of interaction made possible by their commitment to open publication. On the other hand, their funding model is not as conducive to publication, particularly in poorer research areas, as other forms of open publishing. I do hope that if they continue to grow, that they will address some of the issues of breadth and that they will consider the research publication needs of other, less wealthy but more prolific, research areas. I also hope they make it easier to find interesting articles as their collections grow and diversify.


  1. Gene, I’m disappointed. Usually you do your homework better than this.

    Open access != open source. (You can’t GPL an article and you can’t CC-license code.)

    Author fees are not PLoS’s sole source of income.

    The numbers that purport to demonstrate that high-publishing institutions would pay more under an all-open-access regime make precisely the assumptions you demonstrate to be false: that all open-access journals subsist off author fees, and that institutions would pay all author fees.

    PLoS started in biomedicine because (bluntly) that’s where a lot of the grant money is, and grant money (not institutional money) pays for a lot of PLoS publication. I believe they have plans to expand, as they get their feet under them financially and operationally.

    Fair cop on the taxonomy problems, but I hear through the grapevine those are soon to be addressed.

  2. I stand corrected on open access / open source distinction, although based on what Sara Wood was saying, there may be plans to blur the distinction in some of their Hubs as data (and inevitably the associated code) gets integrated into their site.

    I am not blaming them for looking where the money is, but I was a bit put off by their monoculture view what constitutes since, all while talking about their nice shiny Biodiversity Hub.

    The other issue with respect to keeping all your eggs in ONE basket is that it’s difficult for the reader to differentiate among the published materials when everything is in the same venue. Right now, I can tell at a glance what to expect from an article published in (say) ToCHI vs. IP&M just from seeing the journal name. PLoS intends to replace that label with some biobliometric data computed in real time that will allegedly help me understand how important the cited article is. But the problem is two-fold: it forces me to try to interpret data to understand the significance of the article when my task is to read something else, and it doesn’t help categorize the article with respect to its broad area of contribution as the journal title does.

Comments are closed.