Yesterday, Google launched Knol, immediately branded as Google’s answer to Wikipedia. As health care adviser to the project, I’ll say a few words about Knol, but focus on how it – and other forms of electronic self-publishing – may signal the end of medical publishing as we have known it.
First, a word about Knol (the name is short for “a unit of knowledge”). Google’s vision is that providing a tool for people to write about “things that they know” will make the world a better place. Unlike Wikipedia’s anonymous, collaborative writing/editing process, Knols have authors, with names, faces, and reputations. (Authors can choose to have their identity verified, through a cross-check on their credit card or phone records.) Google provides Knolers a tool; authors enter their content and click “publish.” And poof, there it is, on the Web. Users can rate and comment on Knols, send them to friends, and suggest changes. But the author remains the sole owner of the content, able to update and modify it (or remove it) at any time.
That’s where “Search” comes in. There is no real “Knol website” – Knols will appear as results in Google (and presumably other) searches; according to Google, the search position of Knols on their search engine will be determined by the usual Google algorithm, with Knols having no advantage over other information sources (a claim that has been met with some skepticism in the Blogosphere, but one that I believe).
Enter the issue of money, probably the most controversial aspect of Knol. Authors can, at their discretion, choose to allow Google to post ads on their Knols (using Google’s AdSense service). If they do, the authors receive a share of the revenue stream. How much? Nobody outside the Googleplex knows for sure, but the overall revenue will be determined by search popularity and position. So, the thinking goes, experts have an incentive to write terrific Knols and keep them updated in order to maximize their ad revenue.
The Blogosphere is again atwitter with discussion of Knol (for example, see here, here, and here) – much of the discussion surrounding whether Knol is a Wikipedia Killer. Wikipedia, run by a foundation with authors writing presumably for the Good of Mankind (authors and editors are not compensated), is generally characterized as the White Hat in these comparisons. Some observers, though, note that certain Wikipedia entries show clear (but untraceable) corporate influences or other biases, biases that would presumably be more transparent on Knol because of the defined authorship. Most of the reviews of Knol have been pretty positive, and the general sense is that there is probably room for both Knol and Wikipedia to do well. The Internet is a big place.
What was my role in all this? Though there will ultimately be Knols on everything from unclogging a toilet to eclipses (in fact those Knols already exist, here and here), Google thought it would be a good idea to launch the project with a set of Knols on a popular topic already written. They chose health care as that topic, and asked me to develop the list of subjects and recruit the authors, which I did – about 300 subjects in all. (Full disclosure: I received compensation from Google for my work.)
So if you search Google for your favorite health care topic (migraine, or MI, or leukemia, for example), you’re likely to see a Knol – at this point, undoubtedly one that I commissioned – in the search results. The Knols are layperson oriented: I asked authors to write the Knol that they’d want their mother or best friend to read if they had just been diagnosed with the illness. There are also a few Knols on broader medical issues; for example, I wrote Knols on patient safety, quality of care, and hospitalists.
The rationale for this starter set was not only to have an existing set of Knols at the time of public launch, but also to help Google work through all the technical glitches, and at first there were many. There are still a few, but overall the tool works very well and will undoubtedly improve over time.
With that as background, let’s examine how Knol – and other forms of electronic self-publishing such as blogs – may disrupt the traditional world of medical publication.
The disruptive impact of the Internet on the publishing businesses is well appreciated. Newspapers are dying on the vine, venerable textbook publishers are reeling under the competitive pressure of new web-based resources, and most journals have scrambled to make their content available online. But, even as they accommodated the changing preferences of readers, traditional print journals and textbooks (in health care, at least) have not yet been rocked by competition for content. I believe that this is about to change.
Until recently, if I or another author had a research study or a thought piece we wished to disseminate widely, our only avenue was a traditional medical journal. The act of submission required that we relinquish ownership of the intellectual content to the journal’s publisher. For instance, here’s JAMA’s copyright waiver:
“In consideration of the action of the American Medical Association (AMA) in reviewing and editing this submission… I hereby transfer, assign, or otherwise convey all copyright ownership, including any and all rights incidental thereto, exclusively to the AMA, in the event that such work is published by the AMA. …”
That authors voluntarily signed away their intellectual content is remarkable. Think about it: many researchers spend hundreds, even thousands, of hours creating a “product,” and then hand it to a third party without compensation. (Of course, in the United States substantial amounts of research have been funded by tax dollars, with its implied obligation to pursue publication.) In exchange, journals agree to perform peer-review, which – in the case of major journals like the New England Journal and JAMA, with acceptance rates similar to Princeton’s – is likely to end in rejection after a 1-3 month review process. After a few spin cycles (submission, review, rejection, submission elsewhere, review, revision, re-submission…), most articles are accepted – somewhere – and published 4-12 months after that.
I have to believe that the major journals have been thrilled by this arrangement, which led to two salutary economic outcomes. First, individuals and libraries paid good money to subscribe, largely to read and offer this highly vetted content. Second, advertisers, desperate to reach the journals’ large and important audiences, paid premium ad rates. This virtuous cycle (from the publisher’s standpoint) hinged on the willingness of authors to submit content without seeking remuneration.
Why did authors play ball? Because publication in a prominent journal was, and remains, the coin of the academic realm – the ticket to promotion, grants, consulting agreements, and recognition. In other words, authors voluntarily handed over their intellectual property as a loss leader, largely to establish their “brand.” (I don’t discount the importance of altruism in many researchers’ decision-making, but doubt that it, in isolation, would have generated the historical dynamic). So all academics learned to play on this particular field, since it hosted the only game in town.
Contrast that process with my routine when I write an entry for this blog. When I’m done composing, I click “Publish” and it appears online seconds later. I am fortunate to have a reasonably large and enthusiastic readership (thanks!), and I receive lots of feedback and recognition for my postings (my more controversial or topical posts generate more feedback – calls, e-mails, media inquiries – than I have received for anything I’ve ever published in the traditional medical literature, including in the world’s best read journals). For the moment, I have chosen not to advertise on my blog, but the point is that I could (“monetizing” the blog, in the lingo); after paying costs, the revenue would be mine.
But blogging is not without its complexities – one has to figure out the technical aspects and do some marketing. When I decided to start this blog, I asked my friends at the Society of Hospital Medicine and their publishing partner Wiley whether they would consider hosting and promoting it. Luckily, they agreed. So, in my case, a third party took care of the infrastructure and the marketing. Perhaps the journals are safe, since few authors will have access to this kind of support.
But perhaps they are not. As I described earlier, Knol allows authors to post text, pictures, videos, and more, all without breaking a technological sweat or employing a teenager. The author simply creates the content, loads it in, and clicks “Publish” – no peer review, no rejection, no delay, and no relinquishing copyright. And, from the moment they publish their Knol, authors can at least partly realize the monetary value of their content through the advertisement option.
But won’t traditional journal publishing remain the coin of the academic realm, notwithstanding these new dissemination pathways, because of the perceived value of peer review? For now, sure, but we may also witness the democratization of peer review, since readers can comment on and rate Knols. And, rather than looking at a journal’s “Impact Factor” as a proxy for the impact of an article, this blog tracks and reports the number of views of each entry; Knol might ultimately do the same, but for now I can estimate how widely read and linked a Knol is by where it comes up in a Google Search.
In other words, traditional journals’ enviable position as the sole arbiters of the quality and impact of an author’s work may be challenged by web-derived measures of the impact of individual “articles,” such as number of hits, number of links, and reader ratings and comments.
You’re skeptical. And you should be. Having an article peer reviewed by 3 experts is different than having 17 Joe Six-packs (or, if I really crave positive feedback, family members) give it a thumbs up on Knol. I discussed this issue with Udi Manber, Google’s Director of Search Engineering and the brains behind Knol. “Yeah,” he acknowledged, “not all reviews are equal. But we could also rank the reviewers.” How? Lots of ways. For example, reviewers could be ranked by their number of published articles, multiplied by the impact factor of the journals their articles were published in, all determined by an instant PubMed search. Or by the number of visits to their Knols. The head begins to spin with the possibilities.
The point is this: peer review, that most sacred of academic rituals, might ultimately be replaced by real-time rankings by experts and, if you buy the Wisdom of Crowds thing, the masses. Manuscripts might be improved not by month-long editor-mediated back-and-forths between anonymous peer reviewers and authors, but by Wiki-like suggestions directly from readers to authors. And the impact of a work might be determined not by whether it was published in a widely read and cited journal, but by whether the piece itself was widely read, cited, and linked.
Is this progress? I’m not sure. I think the traditional system has generally served us well, and has set a high bar for the quality of published content (at least in the major journals). But, as in other areas in which the Internet is democratizing and transforming commercial relationships (online brokerages or travel sites, for example), Knols and blogs are demonstrating that the technology now exists to “dis-intermediate” the publishing of medical research. With that, traditional journals may need to compete – and not only against other journals – for both readers and authors. Journals may find that they need to begin to compensate authors for their articles, allow authors to retain copyright, or employ other previously unthinkable strategies to remain the preferred choice for authors’ best work.
When I first drafted this article, I chose to send it to two major medical journals rather than publish it on my blog or in a Knol. As an author, I know that the reach of these journals is virtually unmatched, and the peer review process generates an imprimatur that more personalized and democratic venues can’t equal. As a reader, I know that everything I read in these journals is of high quality, largely because of their rigorous review, editorial, and conflict-of-interest processes. As a subscriber to both of these journals, I continue to be more-than-willing to pay for this value.
But how about tomorrow and next year? Like all aspects of the Internet, if faculty, universities, and journals want to preserve the magic of the prior system of peer-reviewed publishing, they will need to adapt and lead in this new world, not just hope that they can hold onto their prior business model in the face of these staggering technological changes.
Oh yeah, both journals turned the piece down.
Robert Wachter is widely regarded as a leading figure in the modern patient safety movement. Together with Dr. Lee Goldman, he coined the term "hospitalist" in an influential 1996 essay in The New England Journal of Medicine. His most recent book, Understanding Patient Safety, (McGraw-Hill, 2008) examines the factors that have contributed to what is often described as "an epidemic" facing American hospitals. His posts appear semi-regularly on THCB and on his own blog "Wachter’s World."