Storm approaching Hull from across Boston Harbor
Image by mccready via Flickr

Recently, Michael Clarke published a memorable post here entitled, “Why Hasn’t Scientific Publishing Been Disrupted Already?” Michael’s arguments were carefully constructed and logical. It’s an excellent post.

And while at first I didn’t realize it, I now think the observations may rest on a different definition of disruption than I’m used to. And as I thought it through a bit more, I could see a different explanation, an alternative interpretation of the same observations.

Michael identified five traits of scientific publishing, using scholarly journals as a proxy. Two of these traits (dissemination and registration) started the whole thing, while three emerged later and inform the value of journals today:

  1. Dissemination
  2. Registration
  3. Validation
  4. Filtration
  5. Designation

Because journals persist, and no alternative has yet supplanted the journal format — primarily because the act of designation is so integral to scholarly career advancement — Michael argued that journals haven’t yet been disrupted, and therefore scientific publishing hasn’t been disrupted.

But disruption isn’t about displacement of the finished goods — it’s about displacement of the incumbents providing the finished goods and their processes for arriving at the finished goods. To use examples from Clayton Christensen’s original book on the topic, steel mini-mills didn’t disrupt by supplanting steel; they disrupted the incumbents by creating a more efficient, more modern, less expensive process for making steel. They did this by starting with the stuff the big companies thought could be ignored (steel bars) and used those meager profits to create an improvement trajectory (of know-how and equipment and markets) that led to the supplanting the incumbents on many fronts. But they still made steel.

Disruption is about blowing apart the provider side, not the consumable or the customer landscape.

Toyota didn’t disrupt GM by making cars obsolete. It accomplished it by making cars in a manner that was more efficient and effective and economical, slowly acquiring marketshare from the bottom-up, gaining customers through segments the high-end incumbents like GM and Chrysler thought were unimportant or insufficiently profitable.

Viewed through this lens, it’s clear that journals aren’t subject to disruption — as long as scholars value them for whatever reason and in whatever format, the “journal” will be around. Journals will go away when they fail to provide value. That won’t be disruption but irrelevance. How they’re made and who makes them — those are activities that can be disrupted.

So the question is really, Is disruption happening on the provider side?

Clearly, the answer is yes.

Look at the library. Before journals went online, libraries were central to the dissemination of journals in academic centers. Some of the savvier librarians have found ways to thrive in the midst of disruptive changes, but with many libraries being converted into education centers or otherwise taken over by other academic departments, it’s clear that this provider of information has been disrupted by the likes of Google and the other major search engines, combined with the online availability of content.

The librarians who are still flourishing have been more willing to take risks, and arguably possess more foresight than some of their peers. This is important.

But what about the publishers? Why haven’t they been turned into scrawny husks of their former selves, with hot new companies sidling up to dance with their dates?

To me, it’s entirely feasible that the main reason scientific publishing hasn’t been disrupted could be attributed to the fact that many of us read Christensen’s book or listened to the disruption gurus or somehow absorbed the lessons of disruptive innovation.

Add this to the fact that we live in a very supportive industry, where competition isn’t as fierce and cut-throat as steel mills or automobile manufacturers, and you can see how disruption could be bent to our collective will, even through osmosis. An academic sense of shared learning permeates scientific publishing, an empiricism blending with cooperation and communication, so an entire culture of publishers has been moving together through these changes.

Basically, I think publishers have been disrupting themselves, making it less likely that disruptors from below could sneak in and surprise them.

From Nature to Science to Elsevier to the American Institute of Physics to the New England Journal of Medicine to Health Affairs to Wolters-Kluwer to Springer to BMJ to JBC to arXiv to PLoS — a broad swath of incumbents of various stripes have been disrupting themselves more or less simultaneously, learning from each other, experimenting avidly.

Now, you could also claim that we’re in a consolidated industry with deep interdependencies on even slower-moving academic institutions, so much so that disruption for us will be more like erosion than earthquake.

But to me, it’s possible this industry learned the lessons Christensen taught, and has acted upon them. More importantly, it’s been able to do it in a manner that allowed nearly everyone to avoid the mistakes of arrogance and incumbency. By so doing, new entrants have emerged, but their effects have been integrated into existing brands and businesses.

From the outside, we may not look that disrupted, but in fact we’ve absorbed the disruptions (so far) rather nicely.

Scientific publishing is changing fundamentally. But instead of waiting for a new entrant to do it, scientific publishing has been disrupting itself.

Reblog this post [with Zemanta]
Kent Anderson

Kent Anderson

Kent Anderson is the CEO of RedLink and RedLink Network, a past-President of SSP, and the founder of the Scholarly Kitchen. He has worked as Publisher at AAAS/Science, CEO/Publisher of JBJS, Inc., a publishing executive at the Massachusetts Medical Society, Publishing Director of the New England Journal of Medicine, and Director of Medical Journals at the American Academy of Pediatrics. Opinions on social media or blogs are his own.


15 Thoughts on "Is It Still Disruption When You’ve Done It Yourself?"

Does this line of thinking leave one horribly exposed to the Black Swan?

How alert was the industry to Open Access? OA seems at this time to have not had the disruptive effect, one might have expected. Perhaps this is cultural – Timo Hannay makes a sharp observation about OA being so wanted by the community that it has had to be mandated… But that perhaps is only a reflection of an instinctive mindset amongst academics to hold onto their carefully created research at all costs. Is that mindset so embedded in a new generation of scholars, ones growing up bathed in a culture of sharing of creativity? OA could explode, just because it hasn’t doesn’t mean it won’t.

Another Black Swan could be open data. If the quality signifier shifts towards the datasets that underpin the scholarly outputs in journals, then we will be living in interesting times I think. Large datasets imply large communities of scholars collaborating to produce them and work with them… What happens if scholars organise around datasets rather than subjects?

One thing I don’t really see much of in scholarly publishing, is the Network Effect (and its best friend Cumulative Advantage). There’s a rich ecosystem of scholarly publishers of all types, all providing value across the market. Compare and contrast with say Web Search. Or Social Networking. Interesting things are happening in the mobile space. Apple may not yet dominate in device penetration, but the market for Mobile apps is theirs. They’ve set all the standards and expectations regardless of the hardware in use. Oh look, there’s Nokia – in trouble.

So, I think we are at high risk of disruption, and it may already be occuring, if not surfacing in quite the dramatic way that it did with the music industry say (but then the music industry isn’t that old really, so their stability might have been an illusion anyway). The question is, how fit are we to deal with it when it hits? Michael’s other post on careers in publishing was also right on the money – If the value lies in data (and I think it does) then we’d better be employing people who ‘get’ data and like to play creatively with it. I don’t think our future lies in subscription fulfillment.

I agree there are risks that journals could become irrelevant, but to me that’s different than journal providers being disrupted. “Disruption” to me has a special meaning around manufacturers and providers — they get disrupted. The things they make are either relevant or irrelevant, valuable or not.

Journals could become irrelevant if an efficient and interesting way of working in large data sets emerges. Journals could become irrelevant if network effects create something that’s more interesting and engaging, relegating them to a non-valuable part of the information economy. But that’s not disruption in the sense I think Christensen used it — that’s failure.

Open access provides a good insight into this. They made journals, so their attempt to disrupt was aimed at the business model of journals. What response proved most effective? Appropriating open access into traditional organizations. So, publishers disrupted themselves rather than acting like incumbents who ignore the threat until it’s too late. That blunted the disruption significantly, and probably has pretty much neutralized it. But it was never aimed at the product (journal) itself. It was about a different way of arriving at the journal.

My point was that when you invoke Christensen and disruption, you’re invoking a particular argument that doesn’t involve the irrelevancy of the end-product. In fact, the end-product is usually improved by the disruption of the manufacturing process, either in price or features (or both). So, we’re really talking apples and oranges here. Will journals become irrelevant and fail? To me, that’s a different question, and doesn’t involve disruption.

Isn’t digital photography replacing chemical photography (essentially – I know you can still do trad photography) disruption in the Christensen sense? If so, it’s semantics surely to suggest the original product still exists? The original product was the film/ camera, the new product is actually the image sensor, off which you can build all sorts of things. It’s still photography.

My point was really about an ecosystem of scholarly outputs. And how fit we are to survive sudden disruption (Your argument is that we are pretty fit isn’t it?)

What if Mendeley or something similar plugs into a wave of socially competant scholars and becomes something that completely enables the five functions you have described in your post (I’d say they could do 1 and 4 now, 2 in theory now, and 3 and 5 if conditions were right). What if it’s Google Wave (I know a scholar or two who are musing whether Google Wave would be a viable medium)? How fit are we? We’ve survived open access so far – agreed. Would we feel confident we could survive “The Budapest Scholarly Networking and Attribution Initiative” of 20??

I think digital photography is a great example of disruption. From the customer perspective, the product is a camera that creates a photo. For years, digital cameras were pretty lame, but then they crossed that 3 megapixel threshold and became mainstream. Some companies (Polaroid comes to mind) ignored this, to their peril. Others (Kodak, Canon, Olympus) embraced it and weren’t disrupted. But people still take photos, print photos, etc. How they’re done, and who makes money off it (HP makes a ton off photo printers, for example, but had no role in photography before) is what changed. But I still take my snaps.

My observation is that scientific publishers may have been adopting disruptive programs fast enough to allow them to avoid disruption from external sources. Instead, it may be that we’ve been disrupting ourselves while looking largely the same to the outside world. Yet, look at the value changes (online is more valuable than print; multimedia is increasingly important; data sets are of more relevance; new payment models have emerged).

My argument is that if Mendeley or something else supplants journals, that’s not disruption in the Christensen sense (strict reading). That’s just plain old stealing business, like cars stole business from stables. So, perhaps next week, I’ll want to consider the questions you’re getting at, and which Michael Clarke and I batted around in email leading up to this post — namely, are journals immune to irrelevance and losing their place in the value chain?

I’m thinking that’s a different question than disruption, in the digital photography or digital music sense. In disruption, the end-product doesn’t fundamentally change, just the ways at arriving at it. Mendeley or something else wouldn’t be perceived by customers as a journal.

Kent – Steel was not disrupted, but the steel mill was. Music was not disrupted, but the CD was. To transpose these examples to scientific publishing, is the journal not similar to the CD and the conventional mill as opposed to steel and music? Wouldn’t steel and music be analogous to articles? (Or even more basic, “the communication of scientific research” in whatever container that communication happen to best fit?)

To borrow David Smith’s example, if Mendeley were to successfully create repositories in which peer-reviewed scientific articles—articles which never appeared in journals—and journals ceased to exist as a result, would that not be a classic market disruption? The articles, like the song or steel, would still be produced and published, but the mode of publication and distribution would shift.

I agree with you that the scholarly publishing industry is a collegial place, open to experimentation, and that culture has certainly helped the industry adapt to the technological changes of the last two decades. But I don’t think that would be enough to forestall a market disruption if dissemination and registration (and even filtration) were the only functions of journals. My post was conceived as a response to Michael Nielsen’s argument that new market entrants will likely disrupt (or supplant) the journal as they are more technologically sophisticated than scientific publishers.

My argument is that the main reason that the industry has not faced a serious disruption event — by which I mean a new publication model that supplants the journal (just as iTunes supplanted the CD) as the primary mode of publishing articles — is that the filtration, validation, and especially dissemination functions of journals are deeply woven into the fabric of the scientific community.

I agree, but I’d still point to the fact that the CD, the cassette, the 8-track, the vinyl LP were all incremental changes to the music buying experience. Before LPs, there were performances, which still have a place in commercial music. What was disruptive around music was the introduction of digital recording technologies, which created new players, and allowed iTunes to emerge. If record companies had been willing to disrupt themselves, Napster, iTunes, etc., would have had much different trajectories and likely wouldn’t have done the damage they’ve done. Instead, they stuck to their guns, surrendered territory they thought wasn’t lucrative (indies, free music, digital devices), and have paid the price.

Ultimately, I don’t disagree with your arguments, but I wanted to see if there was potentially another explanation. In my mind, we’ve been expanding the definition of “journal” so dramatically that it’s become as abstract as you portray it, and then some. Journals are faster, primarily digital, full of new voices, multimedia-rich, data-ready, and so forth. Where is there room left for disruption? Which publisher couldn’t easily absorb (intellectually, economically, practically) any looming disruption because the scientific publishing industry has positioned itself in this way?

I guess from when I first read Christensen’s book and when I gave my first major talk about disruption nearly 10 years ago, I’ve always wondered if the mere presence of his ideas might allow people to inoculate themselves to the dangers. I’m still thinking the scientific publishing industry might have vaccinated itself against most of the current strains of disruption.

That’s no guarantee that there isn’t the disruptive equivalent of H1N1 out there, or that we won’t get hit by a truck.

It should probably be noted that CDs are still the dominant form of music purchased, holding on at 65% of the market to legal downloads at 35%, so technically, the CD is still the primary mode of selling music (though I’m tempted to quote Monty Python here and declare the ever-shifting percentages to be “only a flesh wound”). It would seem to me that the real disruption was not the shift in format (how is the concept of selling through iTunes any different than selling through any other retailer?) but instead the rise of illegal downloading, which has given rise to a generation for whom the concept of paying for music is foreign (music sales dropped another 10% last year). Of course the illegal downloading networks were made possible by the format shift, so perhaps it’s a chicken/egg argument.

In keeping with Kent’s theme of an industry taking on its own disruptor, the VCR might be a good example, going from being seen as the “Boston Strangler” to reaching a point where video sale revenues were higher than box office receipts (though this year saw an interesting reversal).

It is possible that awareness of Christensen’s ideas on disruption can inoculate an organization from disruption (though Christensen cites examples in the disk drive industry where companies knew they were prone to disruption but could do nothing about it). It is also entirely possible that the industry’s embrace of the Web and other potentially disruptive technologies have helped prevent the incursion of entrants from outside the market.

Another interesting possibility is that the Web itself has inoculated the industry. I mean this in two senses:

1. As the Web was built with scientific communication in mind, scientists and scientific publishers were some of the very first make use of it. Up until the Mosaic release in 1993 and the explosion of consumer sites, it was “our” Web. Whereas other industries were blindsided by the Web, we knew it would impact scientific publishing from day 1.

2. Because the Web was built with scientific communications in mind, it resembles scientific publication at a deep level. Hypertext links resemble citations (and indeed, Google’s billion dollar insight was that hypertext linking can be used to calculate a Page Rank just as citations can be used to calculate an impact factor—what is Page Rank if not a more sophisticated impact factor for the Web?). Web pages resemble articles (and are even called “pages”). We imbed images on pages just as typesetters have long embedded figures in articles. The Web’s operating metaphor is, at a fundamental level, a scientific article.

Because of this, it has been relatively easy for scientific publishers to retrofit their products for the Web (it was, after all, made to do that). And because we saw it coming early on, we did it before anyone else had a chance to.

And it might just happen that way again. David Smith mentions Open Data as a potential Black Swan. It could be. But the semantic Web is once again being built with scientific communication in mind. Anyone who has not seen Tim Berners-Lee’s talk at TED 2009, I strongly advise you to make time to do just that. It will be worth your time.

Cracking discussion!
Tim Berners Lee – Let’s recall his original World Wide Web posting:

I’ll extract the pertinent bit;

“The WWW project merges the techniques of information retrieval and hypertext to
make an easy but powerful global information system.

The project started with the philosophy that much academic information should
be freely available to anyone. It aims to allow information sharing within
internationally dispersed teams, and the dissemination of information by
support groups.”

Note the use of the f-word.

We have done a good job of bringing our stuff online, but I guess my concern is this (and what I took out of your careers posting Michael) We were successful to start with as the web was conceived as an electronic metaphor for the page. The problem with digital objects, which is what we deal with now, is that that metaphor falls apart. How well are we equiped to deal with it? A fundemental property of a digital object seems to be its tendency to be duplicated at will, whether we like it or not. There are many other properties, and I’m not sure we understand them well enough to be fit enough (in the evolutionary sense) to survive what coming. Kent,I recall this post:

Have you changed your mind?

Did we suddenly become an industry of agile thinkers?


If Jan 27th brings what I think it is going to, the environment we live in could be very different by the morning of Jan 28th.

I remember that post you reference as well. I actually think they’re somewhat compatible. If disruption has already occurred, and we’ve integrated it into current approaches and experiments, then is the danger abated? Or are we at risk of becoming irrelevant? And maybe that’s the larger point, referred to in another post I published after Michael’s disruption piece — has the disruption happened, and now the battle is for relevancy?

I’ve fallen into the trap of equating irrelevancy to disruption before, and probably will again. But I think this post helped me realign my thinking about exactly what disruption is and isn’t, at least according to Christensen’s theory. And again, maybe we’ve avoided disruption, but can we avoid irrelevancy?

Terrific post, Kent! The whole office here at Health Affairs read it today.

To survive, it is critical that journals look at what it means to serve as an information bridge between author and reader. How do we best serve this relationship in an era of information innovation? This may mean moving beyond a traditional definition of a “journal.”

As part of our major redesign this month at Health Affairs, we thought hard about our reason for being. In doing so, we changed our long-standing tagline from “The Journal of the Health Policy Sphere” to “At the Intersection of Health, Health Care, and Policy.” This intersection – bringing together readers and authors, thought and practice, is our mission – in whatever form it takes.


Comments are closed.