The BMJ's website scales up
BMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7138.1109 (Published 11 April 1998) Cite this as: BMJ 1998;316:1109
Now it provides free access to full text
Three years ago, it was hard to find a medical journal on the internet. Now most have websites, providing selections from their paper journals in electronic form. This week the BMJ joins the Lancet and a host of specialist journals in taking the obvious next step: providing the full text of the paper journal online. Soon most other medical journals interested in their long term survival will follow suit.
If “surfing the net” has become the defining catchphrase of the age then the wave we caught three years ago has turned out to be a tsunami. Since the world wide web was first used commercially the number of websites has grown exponentially. When the BMJ launched its original website in May 1995 there were fewer than 20 000 other websites. Last month the total was 2.1m.1 The number of people online has also been growing exponentially—to an estimated 107m.2 The widespread adoption of webTV should increase this figure by at least an order of magnitude.
The BMJ embraced the world wide web so avidly because it looked like fun and offered an almost miraculous escape from the limitations of paper publication. Costly and cumbersome, the printing presses and binding lines take 30 hours to churn out the 117 000 copies of the paper BMJ each week. The Royal Mail takes another day or two to deliver copies to most addresses in the United Kingdom. Further afield, the delays rapidly escalate—four days to get journals to continental Europe, two to three weeks to Australia. The resources we devote to increasing topicality are wasted for most of our non-UK readers.
By comparison, the electronic journal is available to all countries with world wide web access at 00.01 (GMT or BST) on the Friday preceding the cover date. The geographical reach is a marketer's dream—we are attracting hundreds of readers to the electronic journal from countries with only a handful of paper subscribers. Some 40% of the 20 000 visitors to our website each week “rarely or never see the paper journal.”3
Unlike the paper journal, the website suffers no constraints on space. This allows us to satisfy the previously irreconcilable demands of readers (for brevity) and authors and other researchers (for detail). We have already begun to publish shorter versions of articles in the paper journal, while posting more extended accounts on the website and are considering augmenting these further. In the case of research articles, that might mean including protocols, fuller descriptions of methods, raw data and the computer programs used to analyse them, and full documentation of the peer review process.
Freed of limitations on space, we need no longer limit ourselves to publishing only a third of the letters we receive, four to five months after the paper to which they refer. Within the next week or two, comments boxes will be appended to each article on the website, allowing readers to email their comments back to us. We intend to post these on the website within 48 hours. Locating responses to articles currently entails thumbing through a stack of journals; on the website all responses will be linked electronically to the relevant article.
Web technology allows all articles on the website—and not just letters—to be found easily. A “search engine” can scour the entire site in seconds for a word, bibliographic citation, or issue date. Our searchable full text archive extends back to July 1997; by the end of this year it will run from January 1994. Readers will be able to read any article online or to print out near perfect copies of the paper version. The irritations of missing or damaged BMJs in libraries, the expense and delays of using document delivery services, disappear when users can be assured of the existence of a full text archive. Gone is the pile of journals in offices and studies.
We don't expect readers to stop reading their paper journals, and we envisage paper and electronic versions of the BMJ running in tandem for the foreseeable future. Reading words on a page is much easier than on screen. Louis Rossetto, publisher of Wired, asks audiences to imagine living in a world of computer monitors and then coming across paper for the first time. “Wow, this is fantastic!” he suspects would be the response. “It's cheap, it's light, it's tough; I can roll it up and stick it in my pocket; I can read it on the beach, in the bath, and in bed. I can jot notes on it, tear off the bits I want: this is epoch shattering!”
Few who have trawled through the massive printed volumes of Index Medicus could muster up this level of enthusiasm for paper, and for the past year we have helped visitors to our website locate studies in other journals by providing them with direct access to Medline. New features devised by the National Library of Medicine allow us to extend this service. Readers will now be able to link directly from references appearing in BMJ articles to their Medline abstract.
While bringing ever more information to the desk top we cannot provide users with any more time to digest it. So our new site includes features to maximise the efficiency of users' visits. We currently email the journal's table of contents to 10 000 people, but from next week users will be able to opt for only articles on topics that they have nominated. The alerting email will allow links back to a page that includes not only the reference to the article in question but also details of all articles from the BMJ on the topic, information on relevant books and specialist journals published by the BMJ Publishing Group, and jobs from our classified supplement in the appropriate specialty. In time, we hope to link to trusted resources off site. All visitors to the site will have access to these collected resources.
Initially, our website was updated once a week, coinciding with publication of the paper journal. Over the past year we have begun to post more material between issues, and this trend is likely to continue with news and letters. Some journals are posting stories to their websites as soon as they are publishable—that is, after acceptance and technical editing. Following their lead would reduce by 2-4 months the current delay between acceptance and eventual publication in the BMJ. The idea of the journal as a collection of articles that rolls off the presses each week gives way to the idea of an archive, which is continually updated. So when is the moment of publication? Lawyers say, the moment when something is made public, whatever the medium.
Three years ago, we thought of the website as providing a taster for the paper version. We published the table of contents, abstracts, and the occasional full text article. Since then we have added more full text sections and have begun to publish some material exclusively on the web. Soon the paper journal will be a pared down version of only the most important material that has appeared on the website.
The camera moved
Although this account may suggest we have a clear idea of where we're heading, we haven't. We are poised at the beginning of a profound shift in how information is disseminated but our “paper mindset” blinkers us to the possibilities of the new medium. 4 5 This is usually the case with a new medium, which is handled like old media until its unique properties are recognised and exploited. The first film directors' idea of cinema was to set up a stationary camera in front of a stage play, recording it from the point of view of a single theatregoer. And then around 1908, something occurred of crucial importance to the history of cinema: the camera moved. Eventually, that led to close ups, tracking shots, reaction shots, and the action shown from different points of view—all those features that differentiate a cinematic from a theatrical experience. The internet awaits its camera moving moment.
But before the electronic BMJ assumes much more importance in our plans we will have to find a business model to support it. Even without the substantial costs of paper, print, binding, and postage the journal still costs millions of pounds to produce. Our current strategy is to make the electronic journal as attractive as possible so that readers will eventually be prepared to pay for content. And if we can attract a large enough audience then we hope that advertisers will follow. Certainly, some American journals have attracted large sums in sponsorship from the pharmaceutical industry.
The 2% of the world's population who are online may be suffering from an information glut, but what of the 75% who have yet to hear a telephone dial tone, let alone get online? Subbiah Arunachalam, India's premier information scientist, says that the internet will widen the information gap between the developed and the developing world before it reduces it. Poor connections mean that even those with internet access “must spend hours downloading material that would take only minutes for those in the developed world with the best access” (p 1116).6
Paradoxically, the world wide web might do more to level the playing fields for the information poor than any number of out of date medical journals and books sent to the developing world. Medical libraries in these countries seem far more likely to acquire computers with internet access than they do to fill their shelves with a critical mass of current information. Several publishers—including the BMJ Publishing Group—are already thinking of how they might provide the developing world with access to their journals. A few lines of computer programming could allow free (or heavily discounted) access to computers from selected countries and, unlike discounts on paper subscriptions, the gesture would cost publishers nothing.
Those of us from whose tongues the words “world wide web” trip most easily have the greatest obligation to ensure that this wonderful new creation lives up to its name.