Search Episodes
Listen, Share, & Support
Listen to the latest episode
Subscribe via iTunes
Subscribe via RSS
Become a fan
Follow on Twitter

Support Us:

Please consider making a donation to help make this podcast possible. Any contribution, great or small, helps tremendously!

 
Subscribe to E-Mail Updates

Related Readings
  • Answers for Aristotle: How Science and Philosophy Can Lead Us to A More Meaningful Life
    Answers for Aristotle: How Science and Philosophy Can Lead Us to A More Meaningful Life
    by Massimo Pigliucci
  • Nonsense on Stilts: How to Tell Science from Bunk
    Nonsense on Stilts: How to Tell Science from Bunk
    by Massimo Pigliucci
  • Denying Evolution: Creationism, Scientism, and the Nature of Science
    Denying Evolution: Creationism, Scientism, and the Nature of Science
    by Massimo Pigliucci
Monday
Mar042019

RS 228 - William Gunn and Alex Holcombe on "Is Elsevier helping or hurting scientific progress?"

Release date: March 4th, 2019

William Gunn and Alex Holcombe

In the wake of the University of California's decision to end their contract with Elsevier, the world's largest scientific publisher, a lot of people have been talking about the effect that publishers like Elsevier have on the progress of science. William Gunn, director of scholarly communications for Elsevier, and Alex Holcombe, cognitive scientist and open science advocate, discuss their differing perspectives on the question. The discussion includes: What are scientists' main complaints about Elsevier? What value does Elsevier add? Is the academic publishing market a functioning one? Can Elsevier be a force for innovation?

Links 

William Gunn's website and Twitter

Alex Holcombe's website and Twitter

Psychology in Open Access

Edited by Brent Silk

Music by Miracles of Modern Science

Full Transcripts 

Reader Comments (9)

Gunn states some academic societies would have to expend $30,000 USD per journal article without Elsevier. However, the non profit publisher PLOS ONE produces articles at a cost of $1500 USD, breaks even financially as of 2015, and publishes its peer reviewed fully edited articles under a Creative Commons License that allows anyone in the world to read the article without further charge. So PLOS ONE somehow has 20X the efficiency, and yet publishes under an open access license.

Elsevier has proven itself superfluous and obsolete. Maintaining a paywall limits access to scientific research and hampers progress. Worse yet, for the most part the taxpayers already paid for, and should own, that research.

Researchers should definitely publish government funded research in open access journals.
March 5, 2019 | Unregistered CommenterJameson
I'm disappointed that Julia did ask for a book or article that disagree's with the guests' thinking, especially Hugo!
March 5, 2019 | Unregistered CommenterEran
I really had a great time talking with Julia and Alex for this episode. I wanted to add a few links to things I mentioned and also, because the conversation was fairly wide-ranging, set out the core arguments in a more succinct form here so that the model and the cruxes are more clear to everyone.

In terms of cruxes,
* If I believed that the marginal extra cost of posting an article online was the major cost-driver or source of value, I would believe that journals need to dramatically lower their prices to maintain a constant price/value ratio.
* If I believed that existing open platforms and services such as those mentioned by Alex (PsyOA, Open Science Framework, Publons, Hypothesis, Open Badges, Registered Reports, etc) could replace the combination of for-profit and non-profit publishing and keep their prices from rising significantly, I would believe that those services and the much lower prices they charge are reasonable set-points for the value of publishing and that larger publishers who charge more than that are simply making the excess in profit.
* If I believed that science would be advanced by researchers learning more about how scholarly communications, publishers, and libraries work, then I would believe that researchers should spend less time on their research and more on scholcomm.
* If I believed that non-profit publishers would be able to be sufficiently innovative that they could adapt to changing times & technologies, I would believe that there's little need for commercial publishers.

I don't believe that the marginal cost of posting another article online is a major driver of costs or the major source of value in publishing, because both fixed and variable costs in publishing are large and growing. On variable costs, while it's certainly true that publishers don't have to physically print and ship a physical copy to every reader or every library anymore, they do incur a cost for every article submitted. This cost can be significant, because some journals reject dozens to hundreds of papers for every one they select. Every single one has to be run through the manuscript submission workflow, checked for technical compliance - things like does it have an abstract, does it have a conflict of interest statement, does have an ethics statement, does it include a data availability statement, identifiers to any entities mentioned that have an entity resolver (RRID, Crossref, FundRef, etc) then screened for content - is it a match for the journal - then reviewers have to be recruited, the correspondence between them and the author managed, etc. You get the picture - lots of variable costs before you even get to hosting the article online. Fixed costs are even more relevant to the argument, because they apply to a publisher regardless of how many articles they publish. However, crucially, these fixed costs are largely of the infrastructure variety, which means that some publishers bear more of these costs than others for the simple reason that these things are expensive and necessary. A platform like some of those Alex mentioned which have costs around $500 can do so because they're using infrastructure built and maintained by others. If all publishing were to be done on those platforms, we should expect the costs of those platforms to increase as they have to take on more of the shared maintenance and development. Elsevier, Springer Nature, Wiley etc are major supporters of infrastructure initiatives like Crossref, ORCID, CLOCKSS, etc, of standards bodies like NISO & CONSORT, of supportive groups like the Committee on Publication Ethics, the Research Data Alliance, Scholix, & Force11. There's just so much that goes on behind the scenes to make it possible for researchers to focus on their research yet have access to a clean, well-formatted, accessible, readable historical archive of all of the scholarly literature ever published.

I don't believe that researchers would advance science more by learning more about publishing and how libraries work than by learning more about their research area. When I was a researcher (I studied bone marrow-derived stem cells), I prioritized learning about stem cells, and learning about the mechanics of how all that stuff happens was a lower priority. I might complain about how long it took to get an article published or about the arcane path the library required me to take to get access to certain journals, but for the most part I assumed they were professionals doing their jobs so that I could focus on mine. I would expect most researchers to feel this way, and surveys of researchers indicate this is largely the case. In this situation, it seems like if a researcher wanted to advance science, what they should do is focus on their research, while learning enough about the other stuff to see the value of supporting their library, posting their preprints, and publishing open access. It's reasonable to have an opinion about how much the library spends on publishing contracts, just like it's reasonable to wonder if university IT is getting a good rate from Microsoft or Salesforce or whatever, but only as much as you need to see if the university is following its values and if those values still align with yours. It might sound like I'm putting down the role of advocacy. The best advocates build institutional will for change, but tweeting about how Elsevier is terrible because they make too much money is not the best advocacy. If a researcher feels the to call to advocacy, they should absolutely pursue it, I just want to hear better arguments! That's the whole reason I did this episode - in hopes that we could have better conversations. By the way, most people that I know in publishing are there because they developed, as a researcher, an interest in how the scholcomm system worked and found an opportunity. Elsevier is always looking for these sorts of people and it's an exciting time to be there with all these structural changes finally happening.

You can believe all of the above and still think that it would be better to spend the money on society or non-profit journals, so here's the final crux. I believe that non-profits and for-profits serve separate, complementary roles in scholarly publishing. A non-profit or society will have a mission statement and it will engage in activities that are aligned with that mission. This means that it can do things for which there is no market demand, simply because it believes that these things are good for the field or for research overall, but it also tends to not do certain things that aren't within scope of the mission. For example, it will host conferences for members to discuss their latest research & develop professional skills and it will advice policymakers on areas where research can inform policy. It won't typically invest large sums of capital in developing new technologies, hire huge software teams, or acquire startups to meet emerging market demands. These are the things that for-profits tend to do. Whereas a non-profit tends to limit what it does to what's in scope for its mission, while maintaining low overhead, a for profit doesn't have these limitations & can respond more rapidly to emerging consumer demands.

Now you might say, "fine, but it's still wrong to have a for-profit in academia because libraries have to subscribe to journals and thus demand is inelastic." That used to be true, but these days libraries do cancel subscriptions & researchers can, with relatively little effort, find pretty much any article they want various unofficial places online. Even if that were not true, the negotiation power that universities get as they join together in purchasing consortia gives them strong negotiation leverage to control prices. In this framework, it's clear that both libraries and publishers (for-profit and non) are all partners in how they support the infrastructure of research so that the researcher can do what society expects of them and what they're passionate about - advancing science.
March 5, 2019 | Unregistered CommenterWilliam Gunn
Good discussion, dry topic.
March 5, 2019 | Unregistered CommenterMax
This episode made me extremely frustrated. The Elsevier rep, Gunn, made so many illogical responses and evasions that no one called him out on. He kept mentioning markets. Journal "markets" do not operate as such, and the double-digit profit margins make that clear. Market forces only exist if publishing outlets are fungible, which they are clearly not. Gunn's point about "revealed preference" is a joke. Every academic aims for the highest-prestige (which is 90% determined by historical accident) journal they can and work their way down as they get rejected. I'm now at a university in Hong Kong, and each department literally has a list of journals, ranked by reputation, with constant pressure to keep the number of journals listed to a minimum. Articles not published in these specific journals are literally worth NOTHING at tenure time. It's an extreme situation, but a softer, less-official version of this structure operates elsewhere around the world, including the US.

As someone in the humanities, even many of the points made by the other guest, Alex, don't show the extent of the problem. Alex talks about researchers paying $4000 when they should be paying $500. Humanities researchers generally don't get any grants and don't have even $500 to pay for to provide open-access. There are open-access journals that operate solely on volunteer labor with minor funding (on the order of a few thousand dollars a year per journal, really just in the form of free web hosting) just hosted by individual universities. I'm not against profit margin in general, but in artificial markets without real competition, it's a problem.

Julia might have asked Gunn what he thinks the criteria are for academics choosing what journal to publish in, if he thinks that this is a market situation. I'll tell you, the choice is 95% based on prestige and likelihood of acceptance (the other 5% might be things like their reputation for how fast they reply to submitted articles or reputation of the professionalism of the editorial board). Elsevier controls neither prestige or likelihood of acceptance, and adds no value to these metrics.

Julia might have asked Gunn why Elsevier requires indefinite copyright of the articles they publish. If they add value through services, they can provide those services without requiring copyright, or alternatively only require it for a limited time. Note that for-profit publishers act as if copyright is a natural thing rather than an artificial limitation on free speech, enforced at cost to the taxpayer, with the purpose of encouraging more and better speech overall. In the specific instance of academic research, the writers are already being paid to write, and the benefit journals provide is just editorial selection (as Gunn rightly mentions, attention is a fixed commodity)--but Elsevier is really secondary to that process.

The notion that Elsevier editors add value is just smoke and mirrors. The journal editors are full-time academics who have tiny part-time contracts, not full-time company employees. They would likely provide these editorial services for free as many, many humanities journal editors do, and as they might do when they peer-review articles. (Publishers often provide some small benefit for peer-review, but it's a token). The cost of academic editors to the company is probably a TINY fraction of the companies' overall cost structure, and an even smaller percentage when compared to the journal revenue, but they are 90% of the value Elsevier brings to the journals because they provide the editorial standards that maintain the journal's prestige. Elsevier's full-time employees don't really add or maintain journal prestige--they only add to the production quality of the journal (like typesetting, nice website, etc.).

Consider these foolish points Gunn made (quotes are paraphrased):
- "Only a small percentage of faculty review packets mention impact factor." I'd love to see the study details. If these include humanities packets, then the percentage will be small because humanities have not historically been science-oriented enough to think about journal impact factor, and it is was not even calculated and publicly noted for most humanities journals until recently. AND, the faculty that get to the level where they are applying for tenure will all have published in high-impact journals because if they hadn't, they wouldn't even be applying--they would have been weeded out previously. So all tenure candidates will have high-impact publications, and hence impact factor may not be mentioned as such very frequently. Also, I suspect impact factor is not mentioned as such because it is known to be a poor measure, but it is still influential and operates behind softer phrases like "the best journals" or "high-quality journals".
- "Network analysis of tweets shows that people thank Elsevier Connect." The notion that tweet mentions are a good proxy for whether researchers think Elsevier provides good value for the money is absurd. There could hardly be a larger disconnect between the costless act of thanking Elsevier via tweet and libraries paying millions for research access. As I cannot imagine why research would be thanking Elsevier Connect, I suspect that there is some external incentive at play (I don't use Twitter, so I don't know). Maybe mentioning Elsevier Connect gives researchers a chance to be retweeted by Elsevier's Twitter account.
- "People have been saying that this is the breaking point for years. It's just negotiation and shows there is price feedback." The fact that there have not been cancellations to date shows the monopoly power of Elsevier. NO ONE outside the company thinks Elsevier provides good value for the money. And the fact that there are price negotiations does not mean that there is meaningful price feedback that gets prices even approximately correct. I can tell the emergency room that I can't pay a $40,000 hospital bill and negotiate with them for a lower bill, but that says absolutely nothing about the reasonableness of the cost charged to me.
- "Elsevier is moving in a services direction and thus their interests are becoming more aligned with the researcher." This is only because they are being forced to do so. They are rent-seekers, skimming millions of dollars of profits from (ultimately) taxpayer money because they control access to prestigious publishing venues, when they do nothing to add to or to maintain the prestige of these journals. I wonder - will Gunn identify how much of their revenue and profit come from publishing vs. services? He probably doesn't even know.
- "Elsevier shareholders are pension funds and people that just want stable performance and long-term value." This is so misleading. It doesn't matter where the money comes from. Pensions are managed by investment advisors, who use the exact same metrics they use to evaluate any investments. If Elsevier's profits went from 30%+ to 10%+, they would leave and the company would lose tons of money. The psychological motivation behind investors is a total red herring.

And to respond to some of Gunn's points above:
"If I believed that the marginal extra cost of posting an article online was the major cost-driver or source of value, I would believe that journals need to dramatically lower their prices to maintain a constant price/value ratio."
- A cost-driver is only relevant to the extent that subscription prices are driven by costs, and where costs are proportional to value-added. The double-digit profit margins of Elsevier show that the prices are not determined by costs, but by what institutions will pay to this monopoly publisher, and I suspect most costs are to salaries of employees that don't add value to the end-user (although they add value for the shareholders).

"On variable costs, while it's certainly true that publishers don't have to physically print and ship a physical copy to every reader or every library anymore, they do incur a cost for every article submitted. This cost can be significant, because some journals reject dozens to hundreds of papers for every one they select."
- This important work is done by contracted academic editors. Will you tell us (do you even know) how much of Elsevier's overall cost is salaries to the people who do the actual editorial work? I'll bet it's less than 10% of the cost bar.

"A platform like some of those Alex mentioned which have costs around $500 can do so because they're using infrastructure built and maintained by others. If all publishing were to be done on those platforms, we should expect the costs of those platforms to increase as they have to take on more of the shared maintenance and development."
- Nope. There are tons of platforms, some open-source, that provide internet information-sharing platforms. Again, can you tell us how much of Elsevier's cost is platform development? Very little, I suspect.

"I don't believe that researchers would advance science more by learning more about publishing and how libraries work than by learning more about their research area."
- I agree, which is why open-source platform development should be ideally funded by grants rather than by publishers who pay huge salaries to their top executives and take 30%+ profit margins out of the taxpayer revenue stream.

Non-profits "won't typically invest large sums of capital in developing new technologies, hire huge software teams, or acquire startups to meet emerging market demands. These are the things that for-profits tend to do."
- Sometimes innovation requires lots of capital, sometimes it doesn't. Sometimes innovation produces value for the end-user, and sometimes it simply convinces the end-user of value that doesn't exist. What is Elsevier actually doing -- not in terms of general services, but in terms of basic production of journals that is the majority of their revenue - that is so innovative and valuable that it justifies all of these researchers giving up copyright for their writing?

"Even if that were not true, the negotiation power that universities get as they join together in purchasing consortia gives them strong negotiation leverage to control prices."
No, the very fact that Elsevier has double-digit profit margins shows that libraries have very little power to negotiate. If there was a real competitive market, such margins wouldn't exist.
March 5, 2019 | Unregistered CommenterJonathan
Being bought ie. Mendeley, doesn't mean Elsevier is interested. Purchasing the competition or boutiques isn't a sign of goodwill. Lots of speculation here about market economies... at least Alex' heart is in the right place except for that yoga incident- he's spoken for. Choice of short-term and long-term investment is up to the shareholder regardless of company pitches. UCLA cutting off Elsevier is a big deal for shareholders. A growing liability. I don't associate Elsevier with quality, maybe prestige over integrity, but the service sector is made in India so this talk of outsourcing is the same tune sung by companies forecasting for bankruptcy.
March 8, 2019 | Unregistered CommenterAtieno
weekend work makes a lot of sense. My family goes on a picnic every weekend, hoping to have more time to do the things I like
March 19, 2019 | Unregistered Commentertemple run 2
Thank you for sharing the podcast, I learned a lot from it.
https://192168-l-l.com
August 24, 2019 | Unregistered CommenterVivian
Good Article. Thanks for info which is worth reading it.
September 9, 2019 | Unregistered Commentersupplier

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
All HTML will be escaped. Hyperlinks will be created for URLs automatically.