In November 2008 I wrote a blog post called How do you read papers? The blog post was actually about different strategies to find interesting papers, e.g. browsing journal tables of content (TOC), different search strategies, filtering by papers others read, or filtering by experts (e.g. Faculty of 1000). A paper by Duncan Hull et al. published around that time in PLoS Computational Biology (Defrosting the Digital Library: Bibliographic Tools for the Next Generation Web) also talked about finding strategies and the best tools for this.
In this blog post I want to talk about the actual reading of scientific papers that you found with one of the strategies mentioned above. There are some interesting recent developments, and I think we will see some significant changes in how we read papers in 2010.
Holding the printed journal in your hands is probably still the most satisfying reading experience because of professional typesetting and color reproduction. But unless you have a personal journal subscription, it is not convenient as you would have to go to the library to read the paper. Plus, many journals no longer produce a printed version, or the library has only an electronic subscription.
This used to be the most common way to read papers 20 years ago. But the quality of photocopies is usually worse than a printout of an electronic version, and photocopies are far more inconvenient to obtain. Reading photocopied papers will only be necessary for the small number of journals that produce no electronic version, or for older papers.1
This is the way most people read scientific papers today, unless they just want to look up small parts of it. Quality color printers have become affordable, and the reading experience is similar to the printed journal with the added convenience of electronic distribution. Most people use PDF printouts for reading, and later discard the paper copy, sometimes even the same day. This is not only more expensive than reading on an electronic device, but also not very friendly to the environment.2
This approach appears to be very common for reading just parts of a paper, e.g. to look up experimental details, a figure or reference. Most people now store papers as PDF (hopefully with an intelligent program such as Papers) and not the PDF printouts. Looking at the PDF on the screen is therefore often the first step, and then the decision is made whether or not to print out the paper to read it in more detail. Reading PDFs on screen is possible, but not really convenient for longer texts. Screen sizes are often too small for the PDF format (A4 or US letter). Many people don't like the eye strain from looking at a screen for longer periods of time, although this is probably more relevant for reading books rather than reading a scientific paper.
Utopia is a PDF viewer launched in December by the University of Manchester that enhances PDF files of the Semantic Biochemical Journal with interactive content and live linking to web resources. Read more about Utopia at Duncan Hull's blog.
Many mobile devices such as the iPhone can open PDFs and Papers for iPhone makes this process convenient. But PDFs in an A4 or US letter format are almost impossible to read on a small screen.
The ePub format is more suitable for smaller screens found on mobile devices. ePub is usually used for e-Books, but the Open Access publisher Hindawi since 2008 provides papers also in that format. Although ePub is more suitable than PDF for mobile devices, it doesn't solve the problem that figures and tables are simply difficult to show on a small screen. Mobile devices are probably great for reading journal table of contents or the abstract of a paper (and an RSS reader such as NetNewsWire for iPhone is perfect for this), but not fulltext papers.
The screen of an e-Reader uses electronic ink which not only means a much longer battery life, but also a very pleasing reading experience, including reading in direct sunlight. Electronic ink is black & white, which is not a problem for fiction books, but limits the use for scholarly papers (and scientific textbooks). Ideally an e-Reader should have a 10″ screen, similar in size to A4 or US letter paper.
The Kindle DX from Amazon was announced in May 2009 and is currently the most popular e-Reader. The Kindle uses its own file format, but a recent software update now allows the Kindle to open PDF files. Some journals (e.g. the New England Journal of Medicine) offer Kindle subscriptions, but that doesn't include early release articles and no subscriber access to the journal website. A number of similar devices were demonstrated this week at the Consumer Electronics Show in Las Vegas. The Kindle is primarily an e-Book reader, and David Crotty over on the Scholarly Kitchen blog is skeptical about dedicated e-readers, because he thinks that market is just too small.
Unless you have been on a remote island for the last three months, you will know that Apple will announce a tablet computer later this month. There is wide speculation about the technical details, but it looks like this will be a very interesting device for reading scholarly papers. In contrast to e-Readers it will not use electronic ink. This means a shorter battery life, but allows color documents and many other more traditional computer uses. Based on the iPod and iPhone experience, many people think that the iSlate will change the way we use tablet computers. One of them is Kent Anderson (Game Over, Man – Has the Disruption of Publishing Already Occurred?).
Sports Illustrated did a very nice demo of a fictional tablet computer in December, and it is obvious that many of these concepts can also be applied to scholarly publishing.
Most examples mentioned above try to reproduce the experience of reading something printed on paper on an electronic device. An alternative approach would move beyond the traditional format of a paper and rather takes advantage of the electronic medium. And it looks like the web technologies HTML5 and Flash are best suited for this. Cell Press was experimenting with this approach in 2009, and officially launched their Article of the Future with the first 2010 issue of Cell (all papers will be available without subscription for 60 days, you can provide feedback here). The basic idea of the Article of the Future is to break away from the concept of reading a paper from beginning to end, and to make navigation between the different parts of a paper much easier.
Whereas the Article of the Future tries to make navigation with a paper easier, the PLoS article-level metrics help with navigating to related content: citations, blog posts, reader comments, etc. The Notes feature lets registered users highlight text for specific comments – very much what you would do on a printed paper (but with the added benefit that everybody can see this note).
I'm most excited about projects that enhance the scientific paper instead of recreating an exact electronic version of the traditional paper. And HTML is a more promising format than PDF for these approaches. Michael Clarke (with whom I had the pleasure to do a session at SciFoo 2009) reminded us that Tim Berners-Lee invented the WWW in 1991 to facilitate scientific communication (with HTML and navigation both within and between documents as central concepts), but papers and journals have changed surprisingly little in the last 18 years (Why Hasn't Scientific Publishing Been Disrupted Already?).
fn1. Many journals are scanning their older papers and make them available in electronic form, e.g. Nature. The first issue of Nature from 1869 can be seen here.
fn2. We all know that computers haven't brought us the paperless office, but that we all use more paper than 10 years ago.
Mysteries in Reference Lists
On Tuesday the journal PLOS ONE celebrated its 10th anniversary (see blog post by PLOS ONE Editor-in-Chief Jörg Heber and blog post by PLOS ONE Managing Editor Iratxe Puebla and PLOS Advocacy Director Catriona MacCallum). PLOS ONE (and PLOS) ...
The DataCite Technology Stack
DataCite is a DOI registration agency that enables the registration of scholarly content with a persistent identifier (DOI) and metadata. This content can then be searched for, reused, and connected to other scholarly resources. ...
Infrastructure Tips for the Non-Profit Startup
When I started as DataCite Technical Director four months ago, my first post (Fenner, 2015) on this blog was about what I called Data-Driven Development. The post included a lot of ideas on how to approach development and technical infrastructure. ...