by Mark Garrison, Ph.D.
ITI Faculty Fellow of Faculty Research & Scholarship
October 4, 2018
WHY REFERENCE MANAGEMENT?
The digital transformation of knowledge production and communication has created a paradox: ease and breath of access to research yields to increased incoherence, a lack of systematic organization. The speed with which scholarship and research is produced and disseminated makes it nearly impossible to manage let alone make sense of the store of knowledge without new methods, both technical and social. In this context, reference management is is necessary condition for realizing the potential of the digital transformation taking place and for tamining its sometimes degrading tendencies. It is a means to bring coherence to an unruly landscape of databases, open access journals and datasets, blogs and research institutes publications, books, conference presentations, and government documents, newspapers and magazines.
THREE FUNCTIONS OF REFERENCE MANAGEMENT
Reference management can function in three distinct ways: (1) as a means of organizing information, (2) as a means of collecting information, (3) as a means of collaboratively producing citation-based narrative.
The first function of reference management is that of organization. The organization of information supports the development of intellectual coherence and ease of retrieval. It is no longer efficient or effective to manage references, PDFs, graphics, etc., using a computer’s native folder architecture, or even a spreadsheet. A system for organizing digital information is required. Modern reference management software not only supports a wide array of importable reference types — from books to newspapers to blogs — but also provides tools for organizing these references in user-defined ways by applying tags to references, organizing them into folders, adding notations, and using various metadata such as date of entry into the database or whether or not the reference has a PDF. All of this is searchable. Reference management systems typically have a means for identifying and finding duplicate references, an increasingly common problem as the ease of information retrieval increases and the growth of personal digital libraries surpases personal physical collections of even the most aggressive academic hoarder (my digital library contains more than 7,000 items).
|Why Researchers and Scholars Need Reference Management Software|
Reference management software excels at collecting information and automating the data entry once completed by the researcher or his or her assistant. This second function makes possible the collection of much larger pools of research literature than would be possible using file cabinets. Importing references from a database, Google Scholar, Worldcat or the Library of Congress nearly always brings with it data that can be used in a variety of ways (e.g., abstracts, ISBNs, DOIs) in addition to information required to produce citations (title, author, and so on).
Finally, reference management software streamlines the production of references in academic narratives, and also facilitates collaboration while producing narrative. Common to most reference management software is the ability to “cite as you write”. When this works well, it allows the author (or authors) to focus on producing text, quickly citing as needed, without getting distracted with the citation process or having to go back and figure out what it was one wanted to cite at some later point based on in-text notes to self or other ad-hoc methods. Additionally, much of the challenge of handling citations in journal length or longer manuscripts is handled by the software, eliminating common citation errors such as missing citations or incorrect alphabetization (for styles that require it). Formatting of references and reference lists or bibliographies is also automated, and possibly most important is the ability to switch between citation styles without changing references in the library.
I have been using reference management software since 1993, and as a one-time computer programmer, I find it satisfying to explore software. As a result, I have used and evaluated several reference management programs. Before the turn of the century, I was an avid Endnote user. But as the cost of the program was relatively high, and its usability and stability declined over time, I began exploring other options. Since then I have explored Papers, RefWorks, Mendeley, along with a few other options I don’t quite recall, and most recently and most extensively, Zotero.
While Zotero remains my second choice — it is free, open source software that functions quite well on most platforms — Paperile excels at usability, sharing, organization, searching and most importantly, accuracy of its citation algorithms. Unlike any other software I’ve used, Paperile can accurately produce Chicago-style footnotes! In comparative tests with my students, Paperpile was chosen as more user friendly and reliable than Zotero; the >$40 annual subscription seemed to my students and myself a small price to pay. In short, Paperpile excels at importing references (indicating whether or not a reference found in say Google Scholar is already in your library), integrating with all major databases. Paperpile’s tagging and folder system makes it possible to both group and classify library items in multiple ways — for example, I have a folder named “methodology” that contains items tagged for various projects (e.g., a research methods class, conference paper, book manuscript). Paperpile also has a native PDF markup tool for taking notes, highlighting, etc., and this tool will soon allow collaborative note taking, according to developers. An iOS app is currently being tested. While Zotero and others have many of these functionalities, I’ve just found Paperpile, overall, to be superior. Zotero’s “cite while you write” function requires that one use older file formats when working with Microsoft Word (a less than stable program on Macs), and while offers a variety of sharing options, these are not as easy to use as Paperpile’s. And while there are iOS apps that sync with Zotero such as Papership, the app was in my experience neither reliable nor stable.
It is important to know that Paperpile runs on Google Chrome; no other browser is currently supported although developers suggest this may change (references shared via Paperpile can be viewed using any browser). This means there are no software updates, other than to the Paperpile extension for Chrome. It means that any computer with Chrome can access your Paperpile library (as long as it is connected to the Internet). Paperpile is tightly integrated with Google Drive and stores all reference documents (PDFs or other attached files) in the cloud (Drive) and thus requires a Google account. This cloud-based solution means that your data is “in the cloud” and assuming you use Google’s Backup and Sync software, a copy of what is in Drive can be stored on your machine(s). Unless you have a large library, the 2 gigabytes of storage that Google provides for free will be enough to get you started. (I’ve “upgraded” for $12/year to receive 100 gigabytes of space.)
While developers are also working on Microsoft Word integration, Paperpile currently only works with Google Docs — meaning the “cite while you write” feature is only available when using Docs (there are ways to use Paperpile with other word processing software, but that topic is beyond the scope of this presentation).
Don’t believe what you hear. Google Docs contains all the functionality you need to produce manuscripts for conferences, journals, grants and even books. It can produce mathematical notations. (This document was produced using Google Docs.) My forthcoming book — about 200 pages once printed — is being produced using Google Docs and Paperpile.
HAVE NO ILLUSIONS
Don’t believe what you hear. Computers are not smarter than humans, and no algorithm is going to overcome human error. Reference management software is not a substitute for knowing the reference style required by your discipline or profession. Checking the accuracy of citations requires human intervention; Paperpile will not know if the middle initial of an author is wrong, if the date of publication is accurate, or if “war” should be capitalized. As is the case when evaluating any technology, understanding what reference management software can “do” requires also understanding what it cannot “do”.
For a printable version of this page, please visit this link.