This episode of rationally speaking is brought to you by Stripe Stripe builds economic infrastructure for the Internet. Their tools help online businesses with everything from incorporation and getting started to handling marketplace payments to preventing fraud. Stripe's Culture puts a special emphasis on rigorous thinking and intellectual curiosity. So if you enjoy podcasts like this one and you're interested in what Stripe does, I'd recommend you check them out. They're always hiring. Learn more at Stripe Dotcom.

Welcome to, rationally speaking, the podcast, where we explore the borderlands between reason and nonsense. I'm your host, Julia Gillard, and I'm trying something a little different for this episode of rationally speaking. I've been wanting for a while to try having two guests on the show with at least somewhat different perspectives on a topic and then have a joint discussion where we try to map out where do we agree, where do we disagree and why. The tough thing to do, even offline, as I know from having tried it many times.

But I think it's a worthy thing to to attempt. And I'm excited about the topic and the two guests today. So the topic is what role major scientific publishers like Elsevier play in advancing scientific progress. Are they helping or are they holding back progress? And I'm especially interested in open science, which we've talked about on the show before. For example, I did an episode with Brian Nosek, who's the head of the Centre for Open Science, and we've talked about the importance of being able to share data and methods in other episodes.

The Smain era and Uri Simon and major publishers like Elsevier have been coming under fire for holding back the progress of science in general and of open science. So we're going to talk about why and how justified is that. Now, let me introduce my guests I have with me in the studio first, William Gunn, who has a background in biomedical science and is currently the director of scholarly communications for Elsevier, which is the largest scientific science publisher. They publish thousands of journals.

William, thank you for being here. Thanks for having me. And then joining us by Skype is Alex Holcomb, who is a professor of cognitive science at the University of Sydney and active in the open science movement. Alex chairs an organisation called Psychology and Open Access that provides resources to people looking to start their own open access journals. It's a psyops. Why? Craig Alex, welcome to rationally speaking.

Thank you, Julie. I'm excited to be here.

So just for common knowledge, whenever I like moderate's debate or disagreement, I like to emphasize that our goal, what I think is like the most productive approach is to aim, to explain rather than persuade, like explain your model, ask questions about the other person's model and so on, rather than, you know, trying to make persuasive arguments, which I follow both of you on Twitter. I know you're very reasonable people and I see you do this all the time.

So almost goes without saying. But again, just for common knowledge. So let's start by laying out the general contours of this disagreement as it exists, you know, among the scientific community, not just sort of between you two. I'll have Alex go first and summarise this. And then, William, you can give your take after that. So, Alex, what would you say are the main conflicts between that, you know, you or people in the open science movement see between large publishers like Elsevier and the interests of the scientific community or scientific progress?

Yeah, the conflicts I see are largely based on certain companies like Elsevier having come from a more traditional older system that I now think is is outdated and can now be Internet publishing. With Internet publishing, things can be done much more efficiently.

So actually, right before this podcast, I was chatting with a PhD student here at the University of California where I'm visiting. And it had happened that we had come up in her yoga class this week and she was asked what had made her happy today. And this was the other day when University California cancelled their subscriptions elsewhere and said, well, the university California canceling this elsewhere made me happy today. And of course, the yoga class was like, you know what subscriptions which Elsevier.

Right. And so then she said they asked her to explain and she said something like, well, what I would say basically said, well, the taxpayer ultimately, by supporting university, paying for research grants to universities, they support the work that we do as scientists. I get my salary and Julia Budesonide, who was telling me about the yoga class. She receives a graduate stipend through government support of universities. And using that money, we get paid to make scientific discoveries and also paid to write up the results into an article.

And under the traditional system that's used largely by several other publishing companies. After we write up that article, the article then gets reviewed at a journal by other scientists, typically at other universities, who are also paid to do so as part of their university work, ultimately by students or governments. And then after the article comes back with comments from reviewers, I then revise my article in response to the criticism by the reviewers. But then finally, at the last step, when the article when I've refined all the wording of the article and so on, you might imagine I would just then post this finished work on the Internet for Sydney's citizens and university researchers, whoever is relevant to download.

But. Instead, what we do is we give the copyright to a large multinational corporation, often like Elsevier, who then publishes it in their journal and then charges everyone an annual subscription fee to access it. So many universities, major universities are currently paying Ellisville, for example, million dollars a year for access to this content that those universities themselves already paid for to both create and review. So with the business, that traditional business model being based on restriction of knowledge of dissemination because they're making money by considering how much they can charge for people to access, that brings it in conflict with the modern open science movement in which we want to have knowledge disseminated as far as wide as possible.

Yeah. So I'll I'll just first note that the timing of this episode is fortuitous or not fortuitous in some sense, depending on your perspective. So a couple of days ago, the the University of California school system just just ended their contract with Elsevier, which is what Alex was talking about, where we're not really going to talk about that in the episode because that William is not part of the negotiations team at Elsevier and he can't really comment on it. But it is kind of the background context against which this conversation is happening.

So before William, before you jump in, I just want to get a sense of the categories of the complaints, so to speak. Would you say I mean, I guess either of you can answer this. What do you say that the complaints fall into one, the category of sort of unfairness, like unfair pricing or unfair restriction of access on the one hand, and then to like holding back progress, like the the practices of Elsevier and other publishers are like slowing down scientific progress or making the quality of science worse.

Do those seem like the two categories? I don't like to I'm not as not being a moral theorist, I hesitate to make judgments about fairness. OK, we're living in a more, you know, market economy. So I'm happy to pay for for things. So to me, the issue is not that a company like Elsevier is going to charge as much as they can to maximize, you know, value for their shareholders. Instead, I just see that there's a more efficient way of doing things that will not result in as much money being sucked out of universities.

So one. So that's the way I would reframe it as opposed to fairness, just a more efficient system. And then, yes, the restriction of knowledge for your other category to me does fall under the category of holding back science because greater dissemination helps researchers more easily. Kind of finds and remix as well as you know, just read the articles and the science can progress more rapidly.

Right now, William, you've been very patient. Can you give us an overview of what you see as the main value that's being added by publishers like Elsevier?

Well, the value question is a really interesting one because of my background, having been a researcher and wrote papers and peer reviewed papers, now working for a publisher where I see the other side of it and I work with a lot of the editors, I can bring a number of different perspectives to bear on this question. And there are some perspectives from which the question itself is just, you know, an absurd one. And then there are some perspectives from which it's fairly reasonable.

So on from one perspective, just to start with, the one of the absurd perspectives that not as many people, I think have access to, if you think of just Elsevier itself, employing 3000 editors, all the people who are top experts in their field, specifically, you know, for the purpose of distilling the information that's out there into a really useful high signal to noise ratio form for their colleagues to consume. Of course, that's valuable. But I also see the perspective that Alex was outlining where, you know, hey, we have the Internet and we should be able to exchange information a little bit more easily.

And, you know, why is it still such an expensive thing to do? So the way that I have bridged these two is kind of with, you know, a modification of the of the old saying that, you know, information wants to be to be free. Well, yeah. And then also wants to be expensive. Right. So the missing piece, I think for a lot of people when they come to this conversation is I think that that perspective where they don't see the value of all of the work that the editors do behind the scenes, not just in selecting really the best, most relevant research, but making sure that it adheres to high ethical standards, making sure that that the the data and the code they're supposed to be made available are actually made available and, you know, making and then just making sure that the language is of, you know, reasonably good quality.

All these kinds of things go on behind the scenes. And the fact that they work so well is almost detrimental because people have come to take them for granted. And so the question really is, why can't we do a lot better at disseminating research when we do have the the Internet? You see a little bit of some of the some of the challenges in what's been happening with with Facebook and other social media outlets these days. There's a lot of motivation for people to spread misinformation.

And when there's when there are any gatekeepers, it's very easy for that stuff to spread. And and as we know, people will make their decisions about what to believe based on, you know, things that are going on in their life and their perspective and and how it makes them feel. You know, we might like to believe that everyone is a is a perfect rationalist, but in fact, that's not necessarily the case. And so when it comes to something as critical as scientific research where, you know, there are literally lives hanging in the balance, we want to make sure that the signal is as high quality as it can be, so that when you are a researcher and you're, you know, developing your your theories and you're building your knowledge on what has come before, that you're starting with the most high quality signal that you can.

So I think that's really the most succinct way of putting it, is it's talking about this refining and purifying process. Now, a lot of people will say, well, you know, that's that's fine, but that could be done and, you know, through a non-profit society. And that's kind of where you get some of this fairness that comes in, because I do see a lot of that. People don't understand why there needs to be a financial interest in this at all, and just to to briefly discuss that.

Well, yes, of course you can. But the but in fact, the cost structure of some of the society journalists are quite a lot higher than the cost structure of of Elsevier, the A-S, the publishers of one of the large, largest scientific journals, science.

They estimated that I would cost them 30000 dollars per article, which is if they went to an all author pays model, which is quite a lot higher than Elsevier. Some of the other commercial publishers charge. And in fact, a lot of societies use Elsevier far to host their journalists because of that cost structure. So I can see the argument why someone would say that research is being held back because not everyone can can access it. But when you dig into that and ask what does it really mean, you find that that's really an edge case where someone can't get access.

Do you want to ask any questions or comment on the the question of value added so far?

Yeah. Agree with will that. The services that he mentioned are really important, so will you think you mentioned things that editors do and you referring to professional editors or you referring to it to most journals? The editors are ultimately employees of universities, but a lot of them get like sort of an honorarium or small, like five to ten thousand dollars a year from some publishers to do the sort of things they do to coordinate the peer review. And I agree with Williams that there's a lot of value there.

But for these particular examples you cited, William, like science charging thirty thousand dollars per article, I think the cost really is orthogonal to the need for these services. So there's other journal models, open access models that don't cost nearly as much, but you still get the same kind of quality services, in my experience that you're referring to, for example. And they're not some not just some kind of not for profit organizations that you might have suggested or would have more trouble handling this.

For example, Ubiquiti, Pidgey, Scholastica, they're all for profit companies, but they're companies that started in the last 10 or 15 years. So by starting from Slate, they were able to start, you know, Internet centric. They weren't tied to the legacy business model of owning the intellectual property and then charging, you know, a subscription fee for us to access it. Instead, it's based on a fee for service model. So the universities or the research funders, they pay each of these publishers a fee to provide those services that William was referring to.

And then the cost ends up coming out as well, less than 1000 dollars per article for less than five hundred dollars per article. I think for each of those three publishers and their each publishing several dozen journals, that, in my view, is just as high quality as the kinds of things William was referring to that seemed like it might be a crux here of like empirically.

Is it true that the journals that are edited by Elsevier are, you know, the standards are actually higher quality, the results are higher quality than the the journals that are edited by, you know, lower cost providers that. Well, would you say that that is your perspective?

So, yeah, there's there's two different perspectives on this. I, I would I think that there's a place for price differentiation in the market and there are going to be some higher quality offerings and they're going to be some that are more DIY and not quite as professionalized. But it's an interesting question to talk about because you're in a market economy, it's you're not deciding top down like what services should be available and what they should be priced at. You know, you you have options and choice out there for consumers and people just kind of self sort into these things.

But in academia, the question gets really funny because we're talking about, you know, empirical facts. You should be able to assess, you know, whether a fact is true or not based on the quality of the evidence presented. And so there's two things in the conversation get mixed together. You know, the helpful for our for or understanding. If we could try to keep separate the the concept of the quality of, you know, are the epistemic status of the facts reported and the quality of the publication process, because I think it's right, as Alex noted, to think about it through the framework of fee for service.

And in fact, a lot of what Elsevier is talking about these days referred to themselves as information analytics provider, is explicitly moving in the direction of casting themselves as a service provider.

Can I ask while we're talking about the price compared to the value add, it seems to me like it must be true that some significant percentage of the price Elsevier charges comes from the bundling of journals that, like universities or other institutions, can't buy a subscription to a single journal. They're offered sort of package deals like, you know, cable or something, which I assume must increase the price because journals, you know, otherwise would just want to buy, you know, pick and choose.

But instead, they also have to pay for other journals. And I don't know how to estimate what percentage of the price that's responsible for, but it seems like, you know, it must be significant. And B, it doesn't seem like it's adding value. Does that seem right?

Yeah, I think you're you're right to point out that there are strategies that Elsevier employees to keep its its value and its price high. And the strategies aren't very popular. I mean, I don't watch TV, but I do remember some resentment back when you you couldn't get an Internet. Cable, Internet access connection without paying for the whole, like, you know, TV and telephone bundle. Are they dramatically, you know, we cut the price of one and so you might as well get it all.

So I get like, yes, it is a business decision, which is, you know, kind of ruthless capitalist in one sense. But at the same time, you have to the moderating influence on that, I think, is that you see nowadays, you know, whole consortia and, you know, and over in the EU, whole countries getting together and negotiating in mass against against a company. So there's there's definitely a lot of aggressive pushing from both ends to try to find the optimal price overall.

Can I ask, Alex, is your if you imagine a world in which there are a bunch of for profit journals, but they're not bundled together like, let's say elsewhere, still offers all of these journal subscriptions, but you can buy them individually.

How much better is that world, Julie, if you put your finger on like one aspect that as William, you know, admitted like Elsevier uses as a business practice to, you know, jack up the price. I mean, there that's one sort of anti-competitive practice to obscure kind of transparency in the market, such as it is. But it's not a functioning market anyway, but other which I can get into. But other aspects are, for example, that all the contracts that Elsevier signs with universities, they traditionally put in a secrecy clause so that the university librarians can't even admit to their own.

Can't even I can't even disclose to their own researchers how much they're paying for the subscriptions to Elsevier journals. And it's only through Freedom of Information Act requests that we've been able to find that out in the U.K. and America and Australia. So in another further furthering anti-competitive practice, you know, as William saying, you know, somewhat ruthless business practices are that academic editors that sign up to join Elsevier journals, they often have a clause in their editorial contract that they can't I don't know what the legal clauses for something, you know, on the long lines of working against the interests of Elsevier.

In other words, you can't work on starting another journal in that field because, you know, we sort of own you sort of thing. So, I mean, to put it crudely so but I think these kind of sort of anti-competitive or sort of ruthless capitalist practices, sure. They further jack up the price by some amount that Elsevier and other publishers are able to charge. But the real meat of what has contributed the most to this 37 percent profit margins, operating profits that Elsevier has had over the years, as opposed to, for example, Apple that comes in around twenty nine percent, Google around twenty two percent.

You know, that should be a big warning sign in itself, that Elsevier, by packaging, you know, it shows there's something wrong with the business model. And to me, that's what's driving up. The price that they own are intellectual property and thus they're selling it back to us for an annual fee. Now, Julia, you mentioned for profit journals, but I would put it differently. I'm happy with for profit organizations, companies, corporations being involved, providing the services that lead to the journal publication.

But I don't like the idea of for profit journals if that means they own the journal and thus they have a monopoly on that kind of important research or community that all the researchers are trying to publish. And if they own the content of those journals so that their incentive is to restrict access and then charge the most subscription they can and said we'd rather just pay publishers a fee to publish the thing. And then that's our contractual relationship fee for the service.

We can look for different publishers. We can switch to a new publisher. If we're not happy with the services, we find a cheaper one. No one owned, no corporation owns or journalists.

Yeah, and that's really the whole whole idea behind the author pays open access where you're essentially you're putting the the financial burden to run the services and to maintain the corpus on the paper. It's just a different way of paying for the same thing. And and so the author pays open access.

Are you talking about some journals under the Elsevier umbrella? Yeah. The author who submits the article pays Elsevier in exchange for Elsevier making the article available to the public.

That's the way the the author pays model, sometimes referred to as the gold open access model works. You you collect a payment upfront as opposed to charging suppliers and do that's that's a pretty common structure across the industry.

We know how common like what roughly what percentage of articles on journals are that model?

I mean, roughly the amount of open access on a per article basis is going to be in the neighborhood of seven percent, which is not a. A whole lot in, you know, in relative terms, but in absolute terms, the that puts Elsevier as one of the top publishers of open access content, which not a lot of people realize, and it is also growing 25 percent since last year. I just do do want to make one little correction.

Just, you know, I'm sort of obliged to do so about the the profit margin numbers. In fact, elsewhere doesn't report profit margins at the level of divisions. They have an operating margin, the actual numbers somewhere around 23 percent. But that number gets out there and he gets mentioned a lot. So I just wanted to make that correction.

And that's that's part of the, you know, obscure this part of the difficulty in knowing the true cost, which is, again, the problem with having the subscription model that we're not paying directly for a service. You know, Elsevier and other corporations, you know, they don't break down their business enough for any outsider to find out what it's actually costing them. So while the operating profit was 37 percent, I think by 2013, 2014 for the corporation as a whole, nobody outside of Elsevier knows for the Journal, you know, actual production costs, how much it costs.

Yeah, they have there have been various reports on the costs of open access publishing and, you know, like I said, a Big Society journal like Science, it can be very expensive. The fact that a lot of societies use for hosting shows that we have a little more competitive cost structure than some in that in that respect. But I think it's the whole idea of open access makes a whole lot of sense in that you have a lot more clarity over what the actual cost is.

You know, there's you don't necessarily get away from from the bundling necessarily, because what ends up happening is that the individual authors don't want to pay individually for their articles. So that ends up being paid on a consortium level where a buying consortium gets a, you know, a block grant and they pay for the members within that consortium. So the ideal, you know, is atomized, kind of like you said, where each person is making their own individual payment.

But the actual in actual practice, the way that looks is very similar, is that the library still has a chunk of its money that goes to a publisher for the same stuff, really just kind of calling it something different.

Alex, would the open access model that William was describing, how much of your concerns would that address like, let's say the number one up from seven percent to, you know, 70 percent or something? Is that, like, cool? You know, now everything's like mostly good or no.

So, yeah, it's a little bit of a tricky point because a lot of those now in general go into pure open access with something like that model, whether its author pays or whether it's in my I prefer a system under which the funders would pay for journal platforms, that then any scholar could then publish their articles without paying any fees. So that way researchers with or without funding have equal opportunity. And that might sound a little bit idealistic, but when you think about the Internet platforms that are, you know, potentially there or Internet platforms that are already there, most of the cost, you know, is actually in the set up.

And then there's overhead and, you know, updating it each year and like updating all those Web standards and so on, all the database operations and stuff. But that's like a cost that applies to the whole platform. So then when you get to the marginal cost, the periodical cost, it comes, you know, ends up being very, very low once you put a large scale system in place that's really functioning at scale. So that's why funders who have, you know, supported these broader platforms.

For example, the over 100 university libraries that have banded together to fund the Open Library of the Humanities that publishes several journals. Well, the Open Library, the humanities is a scholarly organization. They pay a fee to a for profit publisher to help assist with the platform. But that for profit publisher does not own the copyright to the articles, unlike in most cases. So that's an open access model that doesn't require authors to pay individual fees. And I like those companies that are using 100 percent that business model.

Now, when we talk about, to your point, the case of Elsevier presently publishing seven percent, but that the percentage is going to go up, well, that's great. Although the the difficulty for me there is that most of those articles are being published in hybrid journals. So rather than being pure open access journals, even the seven percent, they're they're being published largely, although not exclusively. There's Elsevier also publishes a bunch of fully open access journals, but a lot of them are being published in hybrid journals, meaning that they're both subscription and open access journals.

And when the author or someone else, the funder pays a fee that makes that individual article open access, which is great for that individual article. But what it doesn't do is it doesn't it's still supporting a subscription journal. So it doesn't bring us closer to that future that there's a consensus of research universities and funders around the world that we want to get to, which is to have all the articles in a journal be open access.

But couldn't you get there by convincing individual authors that they should make their articles open access? And then, you know, even if the Journal was still technically hybrid, then in practice they would be open.

Yeah, it's tough, though, because, you know, plenty of researchers don't have the funding to spare. So I don't envision in that kind of incremental system us reaching a 73 percent or, you know, ninety five percent open access for those subscription journals and then even, you know. And so in the meantime, I mean, Elsevier, their incentive is to just keep maintaining the subscription model because that's where you can earn the most profits. I mean, that's what I like about the newer companies, they don't have the privilege, maybe they would be, you know, just as exorbitant in the fees they charge, if they could, to deliver more value to their shareholders.

But they don't happen to be in the position that Elsevier has of owning all of these legacy journals that researchers are competing to get into. Well, so Elsevier, a specific incentive, but that's not true of all publishers, but true of many. The multinational is to they got to, you know, defend to their shareholders. Well, right now, we're earning whatever kind of more than 20 percent operating margin, and that's mostly on the back of subscriptions or charging a large author processing charge averaged four thousand dollars for the Wellcome Trust for the articles they paid for last year, four thousand dollars per article for the charges that William Cohen is referring to.

And so it's that's that means that because of those prestigious journals, they're able to charge a much higher fee than the actual cost.

When you look at the newer publishers that are charging closer to five hundred dollars per article.

But so wouldn't you still have the same problem of convincing journals to go full open access even without Elsevier in the picture? What? Yeah, how else if you're making that worse.

Well, OK, so so so for an individual researcher like myself, you know, we're not trying to make money and we just want as many people to read articles as possible. So we'd like to make it open access if we can. However, we're trapped. We're trapped in the system of in order to get promoted, in order to get the next grant so I can write more articles, I need to publish the most prestigious journal that I can.

And those journals happen to be most of them, subscription based. So while, for example, Elsevier likes to talk about, you know, author choice and the reason that authors are publishing subscription journals because they chose to do so, but it's not that choice is unfortunately rather constrained. You know, real freedom means having a realistic alternative that will maintain your career.

And but is it is it Elsevier is false that the choices are constrained or is it just sort of historical caused by the historical accident or not?

As you know, is the function of the printing press used to be that in order to get something published? Well, you know, someone had to make a big investment. The printer who owns the printing press in order to decide to run, you know, 10000 copies of this thing. And so you had to make, you know, make them a big concession. Like, I'll I'll give you the copyright to my thing. And then and then the publisher would just take care of it.

But and so that's the way, you know, most journals work and so does elsewheres able to own the copyright and and now, you know, charge larger fee than the service is actually worth.

So, you know that that is true. That is a historical fact that, you know, there was a reason that we're you know, we took this kind of approach to where we were. We are you know, it was a path dependent process to get to to the state that we're currently in. And, you know, if you think about how to shift this equilibrium, you have to consider some of some of it being path dependent like it was this historical legacy.

But we've been past that. Publishers have been fully digital for decades now. And so there are things that are there are forces that are at play that are fairly separate from that historical legacy. And that would remain in any new equilibrium that we found herself in. Some of those forces are like the brand value, like we spoke a little bit about before, about, you know, how do you if the truth of a fact is something that, you know, can be ascertained and that is a thing that's independent from the value of the quality of the production of the journal.

Right. So you're still going to have this disconnect where you how do you know what papers you should be reading as an academic? And you're still going to have a need for some sort of of a of a proxy to help you guide? How are you going to allocate your attention? Because you've only got a finite amount of attention no matter what publishing system you use. And so we right now, we use Journal Brand as a proxy for that.

And, you know, if something is published in The Lancet or in cell or something, it's probably going through a very rigorous review, the editors there or top experts in their field. And they had their pick of literally thousands of other articles they could have published and they chose yours.

So it seems to me that that a question under dispute here is whether Elsevier is is contributing to that or whether Elsevier is just benefiting from that, having acquired the journals. Does that seem right? Yeah, yeah. I think it's reasonable to query. Like, to what extent is is elsewhere like contributing to that or just benefiting from, you know, the historical legacy. And I think, you know, we could we could kind of get into the weeds all day talking about different business models.

But, you know, fundamentally, it really does come down to this question of of choice.

Do you think authors do actually have a meaningful amount of choice in contrast to Alex's view that they don't?

Well, so they do have a meaning. No amount of choice in most cases, but in most cases also it's they're not they're not feeling constrained. Like I feel I feel it's important to point out just in, you know, to like set the context a little bit here, that if you take all of the tweets mentioning Elsevier and you do a network analysis of those the tweets that people are, you know, happiest about, the ones that have the highest page rank in that network that people are retweeting and getting.

Congratulations for our tweets where people are saying finally got my article published, you know, in this journal. Thank you all of your connect, you know, which is which is surprising to me because from my research or perspective, I thought people who care what the publisher is, they care what their journal is, they care about their field. But they're actually like there's a tremendous amount of positivity out there about the work that Elsevier is doing and the value that they're adding to the whole process.

So we hear debating, you know, specifically to what extent is elsewhere, you know, helping or hindering. Really all you have to do is, is look at the review of preference of of people in the marketplace where they do have quite a lot of different choices available to them. The overall overall kind of idea here we may want to dig into this a little bit is, you know, how free that choice is to to put some empirical facts on the table.

There was an analysis done by the Scorecard Lab one, Pablo Alperin and some others where they looked at the content of faculty tenure and review packets and they looked to see kind of what phrases were mentioned, more what things people were using to assess their faculty for promotion and what they found. There was a surprising finding was that the impact factor of the journal, which is a way of measuring, a bad way of measuring the importance of a journal, because it's just basically an average of how many things are published, divided by how many citations were there.

So it's you know, it's a terrible metric. But that used to have this status of, oh, you know, researchers have no choice because we have to publish in this high impact factor journals because it's important for a career. And in fact, when you look at the data that is mentioned in like single digit percentages of tenure and few packets. And so I think it's probably similar when it comes to the choice that people have in in making where they would like to have their things published.

It's sort of like a common thing that people put out there as like or we don't really have this choice. But in fact, there there's a lot more maybe a lot more choice than they realize. Alex, you want to comment on that?

Yeah. Yeah. Well, actually, the student before this podcast was telling me about her yoga class and other student in that same little coffee. I was having coffee. She was saying in response to some very improbable that you just I know I know.

Conversation with people that are so wrong. I couldn't help but brag that I was going to be appearing on the rationalistic podcast.

And they asked why. And so Elsevier came up and then this other student, she said, well, you know, in my in my field, there's not really much choice available besides Elsevier, like the leading journals, I think it was maybe personality science, which all seem to be elsewhere in my particular field. We do have one a couple of excellent open access journals, but that's fully open access journals, but that's very spotty. And the traditional journal that goes back in history that as people sort of traditionally venerated because of its age, because of the self reinforcing cycle of, well, you know, the the great scientists from the previous generation, they published there.

And so that's where we want to publish. So yet this inertia with journals, it's very hard to displace so that I think that speaks to why. And as William pointed out, this kind of prestige hierarchy has been maintained and Elsevier was very smart. So decades ago, there was an executive there who went around and had the foresight to buy up lots of these journals, such that Elsevier now is often does own the most prestigious journals that are the most prestigious, not because necessarily because of Elsevier, because they often started, you know, been that way for decades, but because of their age.

And but to me, I want to you know, it is difficult if under the current system to figure out the value of things. So another end round, around, round and round around the system that provides open access immediately is the posting of preprint so that we separate out more cleanly the services that William and I are discussing and into a point where we just since we want to make our since it was the taxpayer that did pay for my research, just make it available on the Internet, I'll do that, will post it to what we call pre-print server.

And then all the value that publishers might be adding can happen after that.

Yeah, I think that's great. It's a great model. And several people, Elsevier, agree. In fact, that's why we. SSR, in which is a pre-print repository and has since expanded that into from the social sciences, into biology and chemistry, and it's a great model and Elsevier is incentivized as an organization now to make that happen. But coming back to the point about revealed preference, like if if there aren't options and there is a large desire for there to be options, then what is the disconnect?

Why aren't you know, is it true that there is a demand or is it more actually true that maybe the reason that it's not there is that there's not that much of a demand? I think it's an interesting empirical question that we don't know the answer to. As a company, you put things out there and you see if it is successful as a product and you know, it survives or fails based on whether or not people take it up. Well, there are also monopolies.

Yes, there are. And that was the piece I was going to raise.

So since you mentioned it, let me just let Alex I'm glad you mentioned monopolies, because I view that the journal ownership as not being these different journals that you can choose among, although as my student mentioned in her field, she doesn't seem to have an alternative to Elsevier. And and for example, in the review article reviewed journals that I would publish in in my field, there doesn't seem to be any alternative to Elsevier, but let's put that aside and talk about just individual journals.

Assuming that you are in a field where you have choices besides Elsevier, the journals are you can think of them as monopolies because they're not a commodity, they're not substitutable goods. It's not a situation where because of the prestige hierarchy, as a researcher, my goal is to publish. I mean, my goal, if I want to, you know, speed my career as much as possible, is to publish in the most prestigious journal. And so people are willing to do anything for that.

It doesn't matter what the price differential is. In fact, most researchers don't have any idea what the differences in prices among different journals.

And even if they did that weird disconnect between who is making the decision about which journals to subscribe to or which journals to exactly the librarians, you know, paying for the subscription.

And even if they did know the price and they're not and as you say, they're not even paying for the price. You know, to them, their incentive is I want to get promoted, I want to get the next grant so I can do my life saving science. So I want to have my CV looking as good as it can. So I'm going to first publish, you know, with the most prestigious journal and the most prestigious journals to be the most prestigious journals, for the most part for many years because of the self reinforcing cycle that everything that, you know, you just want to be part of the club that the most prestigious journals last year were part of.

So I don't see it's not a functioning market at all. The the people who are deciding where the resources go to those researchers who are sending articles to journals, they don't know what the price is of those journals. And those journals are not substitutable. So the economist Ted Bergström, the University of California, San Santa Barbara, he describes it as an inelastic demand situation. So so his analysis is that Elsevier and other large traditional publishers have been able to consistently increase the subscription prices to these outsized profit levels because the economic technical term being a demand is inelastic, meaning that it almost doesn't matter what you charge.

These people want it so badly that they'll pay whatever.

Well, clearly, the context that we're having this conversation in where, you know, we're very clearly getting the signal that, you know, actually, no, there there is a lot of price feedback, you know, and there is a whole lot of price feedback that's currently happening, you know, right now, this very moment. So I don't think that that analysis is quite correct. But what it sounds to me like you're saying is that there's a bit of a principal agent problem here.

And I think that's that's probably true, that there is a disconnect. So the opposite that most researchers won't just want to do their research. That's their priority. Right. And they do the the publishing and the applying for grants and other stuff because they have to. But really, if they, you know, could be left alone and have all the resources they need to do their their work, that, you know, you wouldn't hear from them again, that's that's just what they would be doing until they, you know, finally had something wonderful that they wanted to come up and share with the world.

So I would really like to get more to that kind of place where we where we give the researchers the freedom. And I've worked for quite some quite a lot of work on helping to get more efficient allocation of the grants to of the grant money so that researchers can spend more of their time doing research and less of their time doing dealing with the administrative stuff.

I agree with you, William, that it's changing, is changing, but it's taken a long time and you've seen these mass cancellations. In Germany, potentially in Holland for a while in California. When you say mass cancellations, Elsevier dispositions. Yeah, yeah. Institutionally in their contract. Yeah, yeah. So the University of California system being one, but in other countries, it's often many more universities, you know, together bargaining as a as a large consortium with Elsevier and then ultimately saying we're going to cancel our subscriptions in mass.

But that's only happened in the last five to ten years that. Yes, yeah. I agree with you in that this price feedback has started. But but unfortunately, it's come to the point where, well, not too late, but at a point where already the prices had reached the breaking point. You didn't see that before. And if you look at it historically, I think what's happened is it's just taken so long to educate researchers because the universities are not going to they don't want to stick their head out there and, you know, cancel the subscriptions of the research as if the research is going to go into revolt.

It took the education of researchers to understand the nature of the business model and realize, although, yes, a lot of these Elsevier journals the University of California canceled are the top journals in their field. They can now get behind the university administration and saying, you know, no more. These price rises have just gone on too far. And the value that Elsevier is providing is just not worth it.

Wouldn't you have been saying that for years? Like for 20 years now, people have been saying, oh, we've reached the breaking point, reached the breaking point. And, you know, I don't know like how to answer that in the context of a market economy other than like, you know, let things to happen. But you're absolutely right that these cancellation conversations, I'll say, because in a lot of cases, the negotiations are still going on.

They're just and Elsevier is continuing to provide access, you know, for example, to the University of California while the conversations are going on. But they're framed as, oh, this is you know, this is the end. It's finally it's the cancellation when that's actually not the case. It's really just a negotiation tactic that happens more these days.

Well, can I ask if you think I feel like I remember seeing you on Twitter, say something about the interests of Elsevier being aligned with openness. And I wanted to ask about that because it doesn't seem obvious to me that they would be aligned like it. I don't know. I mean, one one thing that seems like evidence against that is that Elsevier has campaigned against openness in the past, like they awarded bills that have, you know, tried to restrict access or tried to prevent federal funding agencies from from requiring research that was federally funded to be made open access.

Oh, yeah. And if you want to see all of that stuff, you can go to the Wikipedia page and read about, you know, the like people of color.

Yeah. You know, read about the things that happened. You know, for the most part, 10-15, you know, are longer years ago what Elsevier has been doing more recently. And, you know, like the reason I brought up the piece about research, you just want to be in their lab doing their research is because, like, they're not paying attention to this stuff. The the concept or the model that people have in their heads of Elsevier, I think is still that 10 years ago, model Elsevier, in recent years, they've they bought Manderlay, which was a tool that was one of the employed early employees to help researchers to find and share and collaborate on research.

They've bought a pre-print repository as this are in. They bought a publication platform and a university services suite of tools called B Press over there in Berkeley and a number of things. So all of those are the things that they're moving in the services direction. And since they're moving in the services direction, their interests are becoming more aligned, I think, with that of the researcher because they're providing researchers services directly. So there's the publishing concept of Elsevier as a publisher, which is where a lot of people still think, and then there's what Elsevier actually is now, which is this information services business.

But they've become that over recent years. And people people just kind of need to update a little bit on what has been happening. Interesting.

OK, I'll tell you how much I have not updated. So the on so I haven't seen your tweet that Julia was referring to, but I did see elsewheres statement in response to the University of California subscription cancellation. And the first sentence of that statement was Elsevier has first and foremost the interests of researchers and students at heart, which I felt it just sort of insulting my intelligence. I mean, come on.

I mean, I feel that about every business. Well, OK, let's make it look something like. Yeah, and we should most of us don't really care about like. Yes. But look like let's say Facebook over connecting the world, OK. At least it's a vague kind of feel good statement. They're not trying to tell us. Oh yeah. Our interests are exactly aligned with, you know, the public relations way that you should be dealing with a situation where, you know, you're trying to maximize profits is not.

But, you know, just saying, oh, no one's going to believe this. First and foremost, the interests of researchers and students at heart. As a corporation, we know you're responsible to your shareholders. They have their own interests. Researchers and students have somewhat different interests. So so this gets a little bit into the long term versus short term perspective.

And I'm sorry, I didn't mean to cut you off there, but I think I've heard a version of this, the same argument over and over again. And so what I just, you know, jump to the to the end, if you don't mind. So and is absolutely true. I mentioned, you know, I had this kind of dual perspective thing that I can bring into play and I can totally get why you say, oh, this is ridiculous.

You know, nobody's going to believe this. They're all about their shareholders. But like Elsevier is owned, you know, who are the shareholders of of Elsevier, their pension funds. There are these very slow, stable funds that just want something that has, you know, metronomic performance in the market. And that's and that's basically what Elsevier has done, has been, you know, like a couple percent, you know, increase a year, steadily, year after year after year.

And that's who the shareholders are. So they're very much have a long term perspective. And they want these kinds of things like this. They want us to be aligned with the transformation of the industry and not just being, you know, a publisher and being one among many things that the library deals with, but with being a company where we have so many touch points across the research lifecycle, not just at the point of publication, but at the point of figuring out what you should be reading, figuring out what kind of experiments you'd be doing, hosting your data, collaborating as you're doing your work, discussing things that the preprint level, all of those kinds of things, which is, you know, the core of open science.

Right. All of that stuff is right directly square where Elsevier is going as a company. So from the inside, it makes sense to talk about that. And that's why I'm trying to bring out this dual perspective thing and say I get how you would not get that, not knowing what I know, but knowing what I know about the like, what's actually going on. It really does ring true to me that that's what they're they are, in fact, trying to do.

So in the interest of time, I kind of want to summarize what seems to me to be the crux of the disagreement. And then I guess each of you can briefly comment on my summary and then we'll probably have to wrap up. OK, that sounds good.

William, you've talked about the value added by Elsevier, both in terms of, you know, just look at the value we're providing like letters to all these good things, increasing the signal to noise ratio and enforcing standards. And you can see this, presumably, you would say in the just look at the quality of the journals produced by Elsevier versus journals that aren't. And and then you can also see you claim that those valuable services are being valued by people because of revealed preference, like people are in fact choosing to submit their articles to Altieri and they're choosing to purchase subscription subscriptions.

And then as I see it, the argument that Alex is making is that it's not that no value is being provided, but that if you unbundled everything and basically like allowed institutions to purchase subscriptions, you know, to a single journalist and not whole packages of journals. And if you separated out all the different services you're providing and and I guess didn't didn't own the copyright, but instead, just like provided all the value that you say you're providing and allow people to purchase it separately, that the the landscape would look very different and that Elsevier wouldn't be wouldn't be charging as much and that the purchasing decisions people would make would look will look pretty different than they currently do.

Does that. That seem like a good summary to you guys? Well, the one piece that I was trying to put out there was we didn't get a chance to talk too much about the about the innovation side of things. And, you know, maybe we can talk a bit more about that and you can cut some other stuff or I don't know how you want to do that. But I do think it's a it's a it's important to say that, you know, there are there are examples of different models that are out there.

I really want to see in this unbundled world if things would be more innovative or less, because we've had we've had archive for around four decades now. And it still looks basically kind of the same way. It always has.

You know, we've archive is just the archive dog. It's a place where you can finish this post their prints. And they've been doing that for for literally decades. And the field of physics is far from the most innovative side of things. And so I really do think that the innovation point of like, is it actually going to make things better or worse is an important thing to think about, because there's a lot of reason to believe that a company with a profit motive and an incentive and like customers that it needs to please, which is why the the services model is an important I really wanted to get that perspective moving away from a publisher to more of a service writer, because that makes them need to do this kind of innovation.

It makes them more answerable. And I think it's going to resolve a lot of like the problems people have, a lot of the issues that people bring up over and over, I think are going to get resolved in that in this way. And maybe I'm too much of an optimist to to think so. I certainly speak with enough cynics that I was like, oh, well, we've been asking for this for years and now we're trying to co-opt it.

Michael, what do you want? Do you want us to not do it or to do it? Because, you know, so it's a little bit of of an issue where some people just don't like for profit in general. And you have to be able to distinguish whether it's like we don't like for profit and elsewhere is an example of for for profit. And so we come up with slightly orthogonal reasons that we don't like Alcivar. We're really at the heart of it is that we just don't like for profit enterprise.

There is some of that. But who can be more innovative than someone who has, you know, lots of money, who has a trusted, respected brand and and is actually showing, I believe, empirical evidence that they are innovating in the way that the market wants.

Alex, do you have a reaction to my attempt to summarize the disagreements and or to to William's take on the innovation point?

Yeah, we'll start with that. It's great. That sounds like him. And I agree. And it's great to hear William say that he's seeing a possible transition to a service provider as being the future. And we've both seen from the complexity of this conversation, as Julie, as you pointed out, unbundling is really going to help expose any value that's being provided. So that's really what I've been arguing for, is that the current business model is just a bit inverted where I mean, it's kind of like Julia Williams, she became pregnant, maybe more likely, Julia, like if you if you then went to an obstetrician and the obstetrician said, well, I'll deliver your baby for you, but what you need to do is let me adopt the baby and then I'll lease the baby back to you for an annual fee.

So in other words, it's an ownership model where they have your creation, their baby. In my case, some of my babies are, you know, scientific discoveries that I want to just broadcast the world. But they end up the publisher ends up owning it and leasing it, you know, back to to me and my universities and my my colleagues. So, yeah.

Whereas in practice, what we want to do is share our babies with as many people as possible. Has that OK?

Well, that's what analogy stops, Alex. But one very last question. What's the name of the dog that was chiming in throughout the show?

Oh, you could you hear. Yeah, no, that's great. I say, oh, you could hear Hugo the dog Hugo.

OK, no, I wasn't complaining. I just wanted his name. Oh, I think the headphones I'm wearing are so so I guess that I never even heard Hugo bark.

He was great. I thought he made some very good points. Well, guys, thanks again. I really appreciate it. And and yeah, it's great having you both. Thanks, William.

Yeah, it was really nice to have this back and forth and yeah. Some points of agreement.

And so like I said in the email, I'm so glad to be able to, like, sit down and have a, you know, a mature, rational discussion that that goes somewhere that isn't, you know, that's like, you know, an experience where you have other people chiming in with their points.

You know, you're talking about are you? Twitter, Facebook, Twitter, and so tweet Twitter on a podcast.

All right, well, this concludes another episode of rationally speaking. Join us next time for more explorations on the borderland between reason and nonsense.