Bookkeeping is one of those things you never explicitly learn in grad school, but it makes up a good percentage of being a scholar. There's just a lot of stuff to keep track of—experiments, data, notes, documents, programs, scribbled-upon napkins and envelopes, and, most relevant to Papers 1.8, journal articles. For a while, my filing system was so bad that it was easier to walk to the library and make a fresh copy of a scholarly treatise than it was to sort through the piles of paper in my office. And I was not alone.

The advent of electronic access to journals and inter-library loans via PDF, along with the growing availability of high-speed document scanners, has permanently changed the way scholars collect and organize literature. My academic life has gone from piles of papers on my desk to folders full of PDF files on my hard drive. On the Mac, Spotlight can help locate specific documents, but real organization requires a little more structure: enter Papers 1.8, an article organization program from Mekentosj.

Mekentosj is a software operation started by two Dutch graduate students, Alexander Griekspoor (Mek) and Tom Groothuis (Tosj), in the late 1990s. Now PhDs, they continue to dedicate themselves to producing high-quality scientific software for the Mac.

More than organizing

Papers is one of a group of organization/meta-organization software packages for managing collections of things--think iTunes  ( ), Delicious Library  ( ), iPhoto  ( ), or Yep, which is for collecting PDFs. But Papers goes beyond the basics of traditional organizing via a plug-in architecture that supports importing, exporting, and searching references and papers from specific scientific and scholarly databases.

The Papers interface provides a clean, easy-to-use workflow for searching, sorting, downloading, and reading PDF articles. Its Web interface is built on top of Mac OS X's WebKit open source browser engine, so you can also use it as a browser.

Papers lets you import and export reference databases to a variety of formats, such as BibTeX, Endnote  ( ), and Microsoft Word 2008's  ( ) reference format. This makes handling your existing bibliographic databases relatively easy.

The real strength of Papers lies in its ability to quickly search for articles and match them to their bibliographic information using a wide variety of online databases. Papers bills itself as "your personal library of science" and skillfully covers the traditional scientific databases, with access to PubMed, IEEE Xplore, ACM, and the Web of Science. But with its recent update it adds facilities that make it relevant and useful even to non-scientific scholarship.

Broad appeal

Papers already offered access to Google Scholar and Books, which are useful to researchers outside of the sciences; recently added interfaces to Project Muse and JSTOR greatly increase the program's breadth of appeal to students and instructors in other disciplines. Users who access their institution's database services remotely can configure Papers to use EXProxy. This works well most of the time, but occasionally becomes problematic when users need to access items without using this proxy server.

Papers also includes an extensive help system that integrates with its online forums and provides screencasts illustrating several of the program's most useful features.

Papers isn't a reference manager per se, but it doesn't pretend to be. For scientific writers who use the TeX and LaTeX typesetting systems, the BibTeX integration is quite good. I use it regularly to build reference databases for applications like TeXShop and BibDesk.

Each revision of Papers brings improvements in reference management--the Word 2008 reference support newly introduced in version 1.8 is a good example—but Papers does not yet have the incredible flexibility of EndNote or BibTeX. However, despite those programs' document and library management features, the aesthetics and ergonomics of Papers makes it far more desirable for reading, searching, and sorting one's collection, as that is its primary function.

Papers lets you search your own database and the various Internet databases simultaneously, and that is a really wonderful feature. Wondering what the author of the paper you are reading now has published recently? Pop up a pane and see. What other articles does this journal have in this issue? Same thing. As a system for searching and sorting, the program's document matching feature works very well. Getting search field information consistent across all search engines is a difficult problem by itself, but Papers handles the task deftly.

Differences in style

There are two main downsides to Papers, and I felt both keenly due to my own particular work style. Neither are deal breakers.

I had grown somewhat reliant on the DevonThink ( ) "find other documents like this one" capability to search my database for articles that were semantically related to a target document. For example, if I were reading a paper on shape perception, DevonThink could find other papers that seemed to match the subject matter. The DevonThink implementation was somewhat primitive and frequently generated false alarms. However, the opening of the Latent Semantic Analysis framework in recent versions of Mac OS X (a system for calculating the semantic similarity of text) will no doubt appeal to the Mekentosj developers for inclusion in future versions of Papers.

The other small hurdle for me is the need to synchronize my references database and PDFs across multiple machines. I work on a laptop a lot, my desktop a bit, and a machine in the lab sometimes, and I like to add papers and search for references on all of them. I currently use the open source Unison to do a lot of laptop-to-desktop syncing and it works well with Papers too.

Macworld's buying advice

Papers 1.8 is a great program for organizing academic file cabinets. It makes searching and organizing papers for reading and referencing a near-enjoyable task. And because it's so efficient, I might have to hit the gym a little more often now; my frequent walks to the library have almost vanished and, as an academic, that's often about the extent of the exercise I get.

[Flip Phillips is a professor of psychology and neuroscience at Skidmore College.]