Wednesday, March 6, 2019

Subscription Publishing is the Hotel Industry. Science needs Airbnb.

[This is a copy of my blogpost, previously on thespectroscope.com.]


I just re-read “Dragging scientific publishing into the 21st century” (Razib Khan, Laurie Goodman, David Mittelman. Genome Biology 2014, 15:556). It ends with:
Should publishers continue to do business-as-usual, then they will (and frankly should) become dinosaurs, while younger, more innovative and more robust communication venues take the lead. The publishing industry as it is constituted today is doomed to extinction. That isn’t an ‘if’. It’s a ‘when’.
The more I deal with subscription publishers, the more I am convinced that they are trapped in a business model that basically dooms them to the extinction predicted in the above article. The publishers know what is coming, but they are almost powerless to do anything except resist the change. That’s because, as I recently wrote, their business model is intrinsically anti-Science. Moreover, just as Uber to the taxicab industry and Airbnb to the hotels, the organizations trying to innovate in scholarly communication are directly threatening the very existence of the Corporate Subscription Publisher.
Science publishing today is tedious and demoralizing. It’s a war that leaves the authors tired and angry. The system we have wastes precious research dollars; it delays and undermines science. It’s begging for change, and luckily, much is happening to improve it. Let’s take a look at the major areas of innovation in scholarly communication and the actors involved (spoiler: 1 of the 25 mentions below is from a publisher; 13 are startups, 7 non-profits, 1 government, and 3 corporate non-publisher). To me, in no particular order, the key ingredients for fixing the mess we have include:
  • Open access
  • Open peer review
  • Post-publication reviews/discussion/annotation
  • Improved discovery of research knowledge
  • UX upgrade for reading, writing and submitting research
  • Article-level metrics
  • Instant publishing and sharing
-------------------------------------------
Open Access
Governments, funding agencies, PLOS, Librarians, universities, several non-profits (this category is somewhat unfair to the subscription publishers, so I won’t include it in the totals)
Open peer review
Post-publication reviews/discussion/annotation
Improved discovery of research knowledge
UX upgrade for reading, writing and submitting
Article-level metrics
Instant publishing and sharing
Reproducibility
So my formal prediction is that the innovation will come from the organizations above. Some publishers will die. Some will evolve by buying the startups. Some of the publishers will be bought by the startups of today and tomorrow. But publishers have not innovated truly for centuries, they are not innovating today and they won’t. That’s okay. We shouldn’t blame the publishers for failing to improve science communication any more than we should blame Hilton for not starting Airbnb. But I must admit that working hard to make these publishers extinct feels very very good.
------------------ 
Note 1: I wrote this in part for the new group on Science Communication & Publishing, started by David Mittelman, one of the authors of the above article. Please join that group and contribute to the discussion.
Note 2: Like a research review article, I know this post will invariably upset the people that I should have and didn’t mention. This is not an exhaustive list. I apologize deeply in advance. Please make a comment below and I will try to fill in the gaps.
Note 3: I know there is a rather liberal sprinkling of protocols.io up there. But also kind of silly to take out our platform from the areas it is designed to improve. I excluded the extra protocols.io mentions from the counts.
-----------------
The standard problem with categories is that you always end up with elements that are outside of the sets you defined. This is especially true when dealing with innovation. My copout solution is the "other great companies" below.
Other Innovators
  • Research Square (not only doesn't fit into the above - the organization type itself is unique: for-benefit corporation; go read more on them, they incubated fanstastic solutions for science communication)
  • Rubriq (for-benefit, supported by Research Square above)
--------------------
[comments below]


It seems to me that arxiv.org should be on your list as well.

I am particularly enthusiastic for uploading preprints (I am told that the cost of the Arxiv server and web interface is about 10-20 USD per paper -- I know it is more at the bioRxiv) and publishing non-anonymous highlights in overlay "journals/blogs" to emphasize the best research. I think such a system can resolve most of the fundamental problems and save billions from the pockets of conventional publishers. It is perhaps the second best system after all scientists reading all papers, which is clearly not possible. The preprint servers are already functioning, and we need more active scientists to highlight research that they find interesting. I see examples of such highlights as well but we need more, much more.

Reply 1
There's also MyScienceWork, ScienceOpen, Riffyn, LabGuru, Authorea, Zenodo, & experiment.com, just to name a few particularly promising ones off the top of my head.

Large corporations mostly know they struggle with innovation. Few large companies have successfully re-invented themselves because this takes lots of risk and lots of failure. When there is success, it's often a case of timing rather than strategy or skill. The few that have (Apple, Amazon), have a culture of taking risks that goes all the way to the top and they still get beaten up in the press for it: https://www.google.com/search?q=is+amazon+profitable+yet Somehow "let's make less profit for a little while" doesn't go over well in shareholder meetings.

So the role of a large company is to recognize innovation when it does happen externally and embrace it, rather than try to shut it down. To their credit, Elsevier is trying to do this with Axon: http://elsevieraxon.com and Macmillan is trying to do this with Digital Science.

So it's not that they just don't understand and are actively hostile. There are many good people there who want to do the right thing (along with some hostile ones too), but mainly there's just not a culture of taking risks in these large traditional organizations. That's really what it comes down to.

Taxi companies didn't have apps before Uber, and publishers weren't really interested in social networks before Mendeley (as anyone who worked on Connotea can tell you). Now they do and now they are. Stay tuned, because we live in interesting times.

- William Gunn
Reply 3
Hi Lenny, Thanks for this post! Agreed that innovation in scholarly publishing is absolutely key to companies surviving. Authors are demanding more from publishers, as they should be. Just wanted to chime in with a few publishers you might want to add under open peer review (and other categories):

PeerJ: They offer authors the option to publish their review history alongside their article. Many of their reviewers also sign their reports.

BioMed Central: All the medical journals in the BMC series publish the peer review history with the articles. BMC's journal GigaScience also practices open peer review.

Royal Society: Their new open access journal Open Science will offer optional open peer review. They will also be implementing post-publication comments and ALMs.

MDPI: Their journal Life offers optional open peer review.

eLife: They publish the decision letter along with author responses (if authors agree), and also have an interesting practice where reviewers meet online to discuss their reviews. As an aside, check out some of their innovations in viewing content, like eLife Lens.

ScienceOpen: They implement open post-publication peer review, and are doing many other innovative things, like free online collaboration tools.

It's important to mention too that all of the above journals are open access. I'm sure I've forgotten some, but hope all this info helps!
Reply 1
Cool list! I'll put in a plug to add Research Square. :) We've got peer review untethered from a particular journal (through Rubriq) and I am hearing a lot about article-level metrics, discovery, and post-publication discussion/review through JournalGuide. Plus the language editing/translation stuff to make it all go a bit smoother (through AJE). https://www.researchsquare.com
ALL COMMENTS (9)

No comments:

Post a Comment