 Okay, I think we'll get started now. So thanks everybody for coming to this session. I'm Mackenzie Smith, and I'm the University Librarian at UC Davis. And I'm joined here today by Ivy Anderson, who's the interim executive director and many other things at the California Digital Library. So we're the PIs for this Pay It Forward project, and what we're doing today is giving you a sort of findings-to-date update on this project. So I'll walk through what the project's doing in a minute for those of you who haven't heard about it before, but I did want to say first that this session is being recorded, so you can listen to it again if you want to later, and the recording will end at the end of the presentation, so don't worry about, you know, pointed questions. Those won't be included. And we have deliberately left time at the end for a discussion, which we're really hoping to have with you about some of these results and what you think we should focus on and what you'd like to hear as the project wraps up in the next couple of months. With that, and the stragglers, there are seats up front here and there, and so let me just, without any further ado, dive in. This is a project, fairly brief and intensive project, funded by the Mellon Foundation, that as asking the question, can a large-gilt conversion across the board to open access, funded via article processing charges or gold open access, be viable and financially sustainable for large North American research-intensive institutions like mine? We were very concerned about that question as we saw certain things coming out of other parts of the country, so that's what motivated the project, and we'll talk a little more about that too. So, the particular context that we're operating in is, I'm sure, known to all of you, and that is that in Europe and the UK, the model of gold open access is pretty much taken hold to a very high degree. Ivy here just got back from the Berlin conference. I'm sure some of you were there too. This is happening very rapidly, and they're not even talking about green over there. Meanwhile, we have a number of policies either in place in this country or coming, which are mandating green open access, and so we are on a collision course, right? The two largest publishing parts of the world are heading in different directions that compete with each other in many ways, so that was one concern. At the same time, in the global context, we're seeing big changes in the funding of research and research publishing, so this is Ivy's slide. I'm not going to get into it in great detail, but the point is that global R&D is shifting the funding of it to the developing world in particular. So the percentage of research funding that comes from the US is dropping while the percentage from Asia is increasing. That is beginning to show up in publishing patterns, so that a percentage of publications coming out of the US and the EU is gradually decreasing as Asian and other developing parts of the world increase. So that would have a huge impact on that sort of global financial picture over the next few decades, and we have an increasing, yes, so we have an increasing emphasis at particular universities on green open access, so UC is one of many institutions in the US that have a policy in place that require all faculty and now all employees. So UC is one of the universities of California to deposit their scholarly articles in e-scholarship green. So look at it from my point of view. I'm the head of a library in the UC system, and I am paying a huge amount of money for subscriptions to journals. I am paying to help subsidize the green infrastructure that we have in e-scholarship, and I'm providing quite a large sum of money to cover open access fees for authors that want to publish in plus one or very many other gold open access journals, and this is clearly not sustainable. I can't pay three times for the same article. So this is a very personal issue for me and one that I think I share with many other people who are trying to make very difficult financial decisions for the future of scholarly journals in our institutions. So we designed a project that has interesting characteristics. So we have four university partners. The University of California, of course, is the primary partner, but we chose three other partners who are large, diversified, and have open access policies in place. So their faculty were familiar with the concept and you'll see why in a minute. We then partnered with some industrial information resource companies, notably Thomson Reuters and Elsevier, for access to the Web of Science and Scopus Databases. So we could look at bibliographic or bibliometric data and ALPSP to get access to a large number of publishers. And we've done two phases of this project, a qualitative study and then a quantitative study, and both of those are sort of coming to an end now. So we'll be able to share preliminary results. So let me start with the just a quick note of the partners. So these are the people involved in the project, and I won't go through it all except to say that we did select a number of consultants to work with us who have great expertise in their area. So for example, Carol Tenepere is our partner on the quantitative, sorry, the qualitative side of the project. And David Solomon, both Christopher Bjork and Mark McCabe, are all working with us on the quantitative side. And as we publish our results, we'll be asking members of the community who have this kind of expertise to also sort of validate and see what they think about our methodology. So the deliverables for this project are really two things. One is a financial model that depicts what the emerging APC model will cost large research institutions under a variety of reagoriously modeled scenarios. And so for that, Ivy will be describing the kind of data we've collected. But basically we're looking at how much libraries are paying today to get access to this kind of thing versus what our authors are publishing today across a variety of disciplines. This was a very important aspect of the project that we not just focus on STEM or in particular life science disciplines, which is where the bulk of this kind of publishing is going on, but really look across all disciplines. And a methodology that any institution can apply itself to its own situation. What we're trying to understand is what level of APC would be sustainable under a few different scenarios across different disciplines and how might we redistribute those costs within institutions. So you hear more about that as we go along. Timeline. So we started last January and it's a 14 month project. So this is kind of a very key moment in the project. We have two months left to finish up our analysis of the data. So we have enough data now to start to make some speculations, but not enough to be absolutely sure about our numbers. So what you're seeing today is kind of ranges and best guesses, but we're getting some pretty solid evidence. So let's go through that. First, I'll talk about the surveys we did. So we started with Carol's team talking to people and we did a bunch of focus groups at the four partner institutions, talked to 77 faculty postdocs and graduate students in just about every discipline, just to kind of gauge their knowledge of and attitudes about open access, scholarly publishing, who should pay, how much they should pay, and all of those kinds of things. And the purpose of those was to inform a survey that was done and we got a very good response rate. So the survey was only run at actually three institutions. You see only two campuses, University of British Columbia and Ohio State, and we got a very strong response rate, which just indicates how interested everybody is in this topic. They all wanted to weigh in. So as you can see, it was about evenly split between faculty and grad students, postdocs who kind of sometimes classify themselves as faculty, sometimes as grad students, they're interesting. And then in every discipline, and interestingly, if you look at one of the things we've been analyzing is the impact of co-authorship on an APC-based model and the majority of survey respondents do publish with at least three to five authors. So there are exceptions, of course, in the humanities it tends to be one or two, but the vast majority had multiple co-authors. So I'm not going to go through the results of the survey in depth. We will be publishing the raw results and much more detailed analysis of them, but here are some of the takeaways that Carol Tenebier had. She saw a range of perspectives. I think people who came to these sessions and then answered the survey probably know something about open access. They've heard about it. So she did see pretty dramatic extremes of people who were either religiously in favor of it or religiously opposed. But she felt most people were squarely in the middle. And a lot of senior faculty said, I'm already doing this. I've been posting my articles on my website for years. I don't really understand what the problem is. Almost all of them support open access as a philosophy. As a reader and as a moral good, they're like, yeah, this is great. As an author, they felt quite differently. And they also felt that they weren't necessarily the right people to ask because they have access to everything they need. These are people at highly privileged universities. We already subscribed to the vast majority of the literature, so they don't really feel the problem personally. And as you would expect, arts and humanities and social science faculty were less supportive of open access on the whole. Concerns were where the APC funding would come from, obviously, that richer nations may dominate publication. It's a constant concern we hear. Of course, that's a problem today. They can't subscribe to journals either in those countries. Potential for APC price increases. In other words, if we're having problems now with the subscription model, won't this be worse? Predatory vanity publishing, lack of transparency. APC is too high. Again, Carol found some really interesting evidence that most authors just don't have a clue what publishers do or why things cost what they cost. They really don't. And when they see an APC, so part of the survey was if the funding was coming from here, how much would you pay? If it was coming from here, how much would you pay? And they're like, why are there these fees? Isn't publishing online free? So clearly a lot of education needed to explain to authors that publishing isn't free, but that there are dials within that cost that we can turn. The library's role they thought would be good in negotiating institutional publishing licenses, much as we do now, and handling the administrative side of it. So the transactional work of paying the fees. Ambivalence. So the faculty did kind of get the point that if we have this scholarship repository, and they're all posting things on their websites anyway, why do we need gold also, right? There's a very, as I just said, there's a very clear signal that they don't understand what these APCs are paying for or why they are what they are. Just no idea. They believe that quality is still going to cost money. So they equate low cost with low quality. They think that this is a very complex problem and it's not going to get solved overnight. We really need to look at different disciplines. So what we're finding in our data is that the STEM fields, as you would expect, life science, health sciences, way further ahead in having solid evidence about how this works than in fields like the humanities and social sciences where there really aren't a lot of open access journals and there's just not that much data yet. And as I already said, they kind of understand that they're really not looking at it as a reader or a class of people. They're not the ones who will benefit the most from this shift because they're already getting access to it. And this was something that shouldn't surprise anybody but it's hard evidence of something we've suspected which is that when authors who know everything there is to know about open access rank, what matters to them when they're choosing a journal to publish in? This is their rank order, okay? Quality and reputation of the journal and fit with the scope of the journal were number one with a bullet. And then there's a long tail that who reads the journal, the impact factor of the journal, whether or not the paper would be accepted, these kinds of things matter. Last but not least is open access. And this was across the board, right? Very strong signal. So the implications are that quality is still the most important thing to authors, particularly at our institutions. But the good news, I guess, in a way as we thought about it is that if those journals were open access, that wouldn't stop people from publishing in them, right? So if their society journal that they've known and loved for decades suddenly switched to gold and charged an APC and they thought they could afford it and it was fair, it would not stop them. So we find that generally positive. Another thing that we found is that, as I said, there were some questions in there about your willingness to pay depending on who was paying. So there were some options like your grant is paying, your library is paying, or you yourself are paying. And what we saw is that there's a lot of price discrimination going on. So if the library is paying or even, well, so if the library is paying, they'll pay a lot of money, okay? Really high. If their grants are paying, they'll pay kind of a lot of money, right? Because it's really not their money. But when it was them paying, the number dropped dramatically. And that was really important because we need to know that if we can somehow involve authors in the publishing decision, they do have incentives to economize and that could have some dramatic effect on the overall model. So we are looking at behavioral models now for how we could think about an APC Gold World. And we're just starting to get into that phase of the project now. But at least we have evidence that that is something that they are sensitive to. So with that, I'm going to turn it over to Ivy to talk about the quantitative data that we've collected. And then we'll get into the discussion. Do you want the speaker? Thanks, Mackenzie. So first, I'd like to come back to one of the earlier slides about the disconnect between the North American and European approaches. Because as Mackenzie mentioned, I just gave a very similar talk at the Berlin 12 conference, which was all about conversion to open access. And I said something very similar, you know, that there's green approaches in North America and increasing gold in other parts of the country. World. Well, excuse me, yes, the world, the European continent. And of course, the first correction was my UK colleague said, well, UK is in Europe, so you shouldn't be saying Europe slash UK. So I will correct that in version number two. But the other comment was from folks in that European Commission who came up to me afterwards and said, well, you know, it's really not that black and white or it's really not that green and gold that, you know, in Europe. And I see, you know, John nodding his head here that there's plenty of green open access support in Europe as well. But there's also a lot of discussion about green as an archiving strategy in Europe where there's a lot of focus on repositories to capture national output and institutional output. But that said, there still is a lot more activity on the gold APC front in Europe. So I think the basic insight is correct, but I wanted to add a little bit more nuance to that statement. And China, what you told me. Yes, right. So a little bit on the quantitative data. So this is just some background on the bibliometric data that we've collected. As Mackenzie mentioned, we have four major partner institutions, Harvard, Ohio State, University of British Columbia and UC. UC is actually 10 institutions, so we really have 13 institutions in the project. But we're looking at both UC as a whole and UC as individual institutions, which provides some also interesting valence to the analysis. This slide is about corresponding authorship rates, and so you can see that depending on whether you're looking at Scopus data or Web of Science data, the rate of corresponding authorship is somewhere in the 50 to 60 percent range. And that's a pretty important finding when you're talking about funding of gold open access because we tend to think about having to pay for all of our output. Well, in fact, the way this particular realm of gold OA is developing, one isn't typically paying for all of one's own institution's own output. The payment is usually tied to corresponding authorship. So it's pretty important to know what percentage of time your institution is a corresponding author in a given discipline or overall. And these findings pretty much mirror a lot of the analysis that folks like the Max Planck Society in Germany have done and some others as well that that's pretty standard that that rate in the 50 to 60 percent seems to be a fairly common percentage. So overall, this is how the publication data that we've captured breaks down from a disciplinary perspective about 73,000 publications across all of our partners in 2013. The period of our study is 2009 to 2013. So that's the latest year of data that we're working with. And you can see that most of the publication is in the hard sciences, particularly the biomedical sciences, arts and humanities, a tiny sliver, business and economics, a small bit of publishing. There's some complexities around how you count journals that are in multidisciplinary or how do you count things like plus one. So there's a certain amount of imprecision in our ability to analyze this data, but by and large we've tried to, we did a lot of work trying to map, do some subject mapping on the front end so that we could feel somewhat confident about using reliable subject categories. But in any event, most of the publishing is happening in these fields that, as you'll see in a minute, do have grant funding and so forth. And so that also has something to do with as we look at this kind of model taking hold in the area, in the fields and disciplines where there's more skepticism about our more reluctance to move into that mode of publishing. It's also a much smaller percentage of our publishing output overall. So we're also looking at data by article type. One of the things that we would like to be able to analyze, and I will say we're a little bit challenged right now. We've collected some gross data, but we haven't really drilled into the financial implications of these different kinds of publications. But we're looking at the dark part of those bars are the core research articles. And then we're also looking at proceedings papers because proceedings may or may not have fees associated with them. And we want to understand what happens in fields like engineering and computer science where that's the bulk of the publishing. And then we're also looking at review articles which may or may not be funded in the same way. Very interesting to see that at least according to the data that we've been able to capture, Scopus is capturing more proceedings and review article output than Web of Science. So some interesting differences there. But you can see that research articles are, depending on which database we're looking at, Scopus or Web of Science, 80 to 90% of the publishing output. What about grant funding? In the models that we're thinking about, we're trying to think about, where does grant funding come into play when you're funding open access publishing? So it's pretty important to understand what percentage of publishing output is funded by a grant. So the way that we've mined this data is by looking at the grant acknowledgement statements that are in Web of Science. Web of Science does a very good job or has done a pretty good job since about 2009 or so of capturing grant acknowledgement data. Scopus is beginning to do that but not in a very structured way. So this data is all from Web of Science. And as you can see, the percentages are fairly consistent across these major research institutions in the 60 to 70% range of grant funded articles that acknowledge a grant. So we're translating that as grant funded. But looking at that again by discipline, arts and humanities on this chart shows no grant funding. That's probably not exactly true but Web of Science does not capture any grant information for arts and humanities publications. So that's probably not exactly correct but it's not a major area of grant funding. But you can see the kinds of disciplines where grant funding is extremely high chemistry. Again, earth science is a little bit surprisingly, physics and astronomy medicine. Medicine looks a little bit low there. I'm not sure what's going on there but it may be a fact function of the way that we've categorized the output as well. But you can see that grant funding is pretty high in the hard sciences and the biomedical disciplines in particular. So what does this all mean for our project? Well, we're trying to look at what would it cost our institutions to convert to gold open access and how would that work if grant funding were or were not part of the equation. So one version of the analysis that we've done recently and this is where we still, even though we're close to the end of the project, we actually still have a lot of work to do in this area. So what we're going to be talking about here is still, I have to say, pretty preliminary analysis. But one of the ways that we've analyzed this is to say, well, what would be a break-even point for our institutions given what our subscription spend is now, what level of APCs could we support? And let me just preface that by saying, and I'm not sure if I have a slide on this. I might have a slide on this in a bit but I'm not sure actually. So let me just say one of the premises of this project was that we would be able to analyze APCs across disciplines and we thought we would find different APCs in different disciplines. We may still be able to get there but right now the analysis that we've been able to do has not really revealed statistically significant differences in APC levels. And that may be because the large publishers are publishing in all sorts of disciplines and they tend to have fairly consistent pricing policies across their portfolio of journals. There aren't that many Gold Open Access journals or APCs in areas like social science, for example, that aren't coming out of the large commercial publishers that have a more consistent pricing. So we're not seeing a lot of differentiation in across discipline, but that was initially the idea was that we would be able to map that. But so what we've done is looked at the APCs that we've analyzed and we've broken them out. Let me see how to explain this. What we've done is taken the subscription spend at our institutions and converted that into what average APC would each of our institutions be able to afford given their subscription spend. And the volume of publishing. And the volume of publishing. Right, obviously that's the other variable there. We don't have any slides about subscription spend and I don't actually have the numbers in my head for what all of our partner universities are paying for journals. It's probably what everyone wants to know. Well, what are we really spending on these journals? I will share one data point though which I think we found pretty surprising. And that is we all talk about how much of our collections budget is going towards journal publication. But what we did in terms of collecting subscription payment data, financial data, we mapped all of that to the data set of the journals that we were analyzing. So we looked at, we pulled all of the journals that are published or that are, excuse me, indexed in Web of Science and Scopus and we gave those lists to our library partners and we said, tell us what you're spending for those journals. And of course we also identified a lot of the big packages that wouldn't show up on a journals list but where we know those journals are found in big packages. So we did a lot of work to bound the data set for journals to the bibliometric data that we were capturing. And it is most of the journals that we're all paying for. It excludes things like newspapers and trade journals and all sorts of serial publications that are not scholarly academic journals. When we looked at the subscription spend, those scholarly academic journals were about, I want to say, 40 to 50 percent of the serials budget. It was a much lower percentage than everyone tends to assume and float as, you know, this enormous part of our budget that's going towards these expenditures. So we're, we work really hard with our partners to try and pull as accurate funding data as we could. We're not sure that we got every, you know, every penny of data is pretty hard as you can imagine to pull that kind of data out of your library systems and your acquisition systems. So it may not be perfect, but the percentage of spend is not quite as high as you might expect. So the next slides were sure a little bit complicated to show our break-even points. And what we did was we assigned a break-even point for every partner institution based on their publication output keyed to corresponding authorship rates. And if the break-even point is high, that is to say the institution could actually fund a very high APC. That's showing up as green in the next chart. If the break-even point is low, that means that the APC has to be really, really low for that institution to be able to afford gold open access, and that's showing up as red in the chart. So I think there's something missing here, but in any event it's missing, so we won't go into it. So what you see here is that this, so now is where I have to talk about University of California because all of our institutions are research-intensive institutions, but when you break out the University of California at the campus level, some of them are much smaller than others. So University of California at Merced, for example, a young university, small university, not a very publication-intensive university yet. And some of our other campuses are in that category as well. And so those are probably the universities that are showing up in that green bar where they could actually afford a very high APC because their publication output is low. The average APCs, this is the slide that's missing. Somehow we're missing the slide, which I really apologize for, which was a chart showing how we derive the average APCs. So there are two sets of averages that we've highlighted here as average APCs. The first one, that figure of $1907,907, that is coming from the European payment data. So in Europe, Germany, the UK, some of the other countries in Europe have done a lot of work over the last year to capture and make available very transparent data sets about the open access APCs that they're spending in Europe and the UK, Europe. So we've looked at all of that data and looked at that both by subject and in general. And for full OA journals, excluding hybrid journals, the average APC that's being spent in Europe with some currency, obviously conversion, is about $1,900. That next figure, $1,775. That's a calculation that we did to take... So there's another fairly recent data set that Heather Morrison in Canada created where she and her colleagues researched current APCs across all of the gold journals that they could identify in the DOAJ. And so that's very current APC data. It dates from at least 2014, maybe 2015, APC list prices. So this is not what some universities spend. This is what publishers list on their websites as list APC prices. We took that data set and we mapped it to our publication output. So one of the things that we felt was important, we're talking about major research intensive institutions. Well, we don't necessarily publish... Our publishing output doesn't necessarily show up in the... what one might think of as lower quality journals. The center of gravity tends to be at the higher end of the publication scale. So a lot of publications in very high-end publications. So we felt it was important to look at average APCs, not globally, but mapped to the publication output of our institutions. And when we did that mapping, the average APC that we came up with was $1,775. So again, this is simply full gold OA journals. So this does not take into account hybrid APCs, which is one reason why these averages are below $2,000 perhaps. But we're using that right now as something of a bellwether because we think those are important benchmarks. So the green bar is anything above that. It's going to be quite affordable for most of our institutions. Down in the red zone, so under $1,000, some of the largest universities in our partner set are going to have a difficult time breaking even, if you will, on APCs if something like in the $1,800 to $2,000 range is in fact a reasonable benchmark. So you can think of obviously Harvard, Berkeley, some of the other UC institutions are going to be in that lower end of the scale. So there are some challenges there. Now this is assuming that the library is paying for everything, for the grant funding in this particular chart. If we add grant funding into the mix, then it starts to look much, much more sustainable. So in this version of the analysis, we're assuming that if there's a grant, the grant is paying for all of the full APC. So it's kind of, it's binary in this case, if there's a grant acknowledgement statement, a grant is paying for that article, otherwise the institution is paying. So again, that's a fairly rough cut at the analysis. But in this analysis, sustainability seems much, much more achievable. So you can see that yes, some of our larger institutions are in the yellow zone here, but yellow zone is a lot better than the red zone. So it just gives you a sense of how important grant funding may be to achieving sustainability for Gold Open Access, which and typically grant funding is being deployed in supporting Gold Open Access. So the other thing that we're working on here is some work on what, we're really interested in some sort of multi-stakeholder funding in developing scenarios. So one of our ideas is that we're going to develop scenarios for how one might combine funding from various sources to provide a sustainable and feasible environment for Gold Open Access to flourish. And one of the things that we would like to introduce in that kind of scenario is some sort of market dynamic. So Mackenzie mentioned, well, if an author is involved, they have to pay out of pocket or out of some discretionary funds that they have. They're going to be much more price sensitive than we as the libraries are. We know this. This is why we're in the situation that we're in. I think even in the case of grant funding, many faculty also indicated great reluctance to spend money from their grants. There's a willingness to do that, and grants are being used to fund this. But there's also sort of a market pressure there because a PI is going to be very sensitive to how much of their grant funding they're going to allocate to publications when there are other things that are competing for those dollars. So I would encourage you not to look at the details on this slide because this was an early version of something, and we're doing a lot of work in this area right now. This is probably not the current version of the model that we're thinking of, but we're trying to develop a model in which the library would pay up to some amount, and then some form of discretionary funding would also kick in, whether it's coming directly from the author, whether it's coming out of a grant, and how we combine those components is something that we're doing a lot of work on right now. We also have another scenario which is sort of institution pays all, single payer system, and we'll look at that also to see how that might play out. Some of the offset deals, for example, that are being negotiated in Europe are predicated on that model. The institution is funding the library subscription, the subscription purchases now. You just flip that payment, and it becomes an open access payment, and then it becomes a budget neutral, if you will, which was a very controversial concept of the Berlin meeting here, probably there. So we will look at sort of the single payer model and how that might work, but we're really more interested in a model that might have some market dynamics associated with it, some dynamics that would help to keep pressure on APCs and bring them down. So we've also done some modeling, which you're not seeing here, on the cost to publish. So you saw the APC, average APCs that we came out with, but we've also talked with nontraditional, if you will, publishers, some of the newer publishers, and we've done a fair amount of analysis on what does it really cost to publish, and we're going to bring that into our project as well to try and talk about how cost effective can Gold at open access publishing be, as opposed to what does it cost today? And the averages that you saw there are what does it cost today, but again, we're hoping that we can talk about how one might push those APCs lower or the overall cost lower. So we have to do more work on refining all of our data, both the APC data, the library expenditure data, the publication output data. We do want to do some modeling growth over time, modeling of growth over time. So the reason we took a five year period, 2009 to 2013, was so we could look at, particularly at publication output and how that was what that growth looked like over time, as well as the growth in library budgets or actually the fluctuation in library budgets is probably more like it. And then again, iterate and develop our funding scenarios to see if we can identify some potential financial flows that might be attractive. We are taking some of the analytical data that we've developed and we're working on, and this is something that will probably not be an output of this grant, it might be a follow on grant, but we will have an initial version of a calculation tool. So this is some sort of a spreadsheet or maybe some more complex tool, but whereby any institution could input its publication output data, other aspects of the model and sort of pick and choose what elements of the model you want to input into that tool and let that translate out for you what the cost might be under various scenarios. So we think that that could actually be very valuable if we can develop a little tool that would let you do that. And then we have to write a report and that report is due in June. So Mackenzie thinks that we're going to be done in... The money runs out. Then there's the report. Except for my data analysis. He's still on staff. So I would imagine around June is going to be the timeframe for our report or something like that. And we are going to publish all the data that's publishable. Some of the data is, as you can imagine, proprietary. We're not going to publish what Harvard spends or Berkley spends on subscriptions, particularly with particular publishers. But we will publish whatever data we can, including with the bibliometric partners. We determined with them on the front end that we could make in an aggregated form the bibliometric data that we have available. So that's it. Thanks. Let me just conclude with a few comments. We've spent a whole lot of time looking at these APCs, and I know they have in Europe, too. But one thing we hear anecdotally from a lot of publishers is that those are fairly arbitrary now. With very few exceptions, they're not based on rigorous cost modeling on the publisher's side. They're just trying to understand what will fly. And then the publishers tend to kind of flock together. So a lot of them say, well, I'm charging what he's charging. So some publishers have said quite explicitly that the APCs they're charging now are way too low. Way too low for what they actually are going to need if we switch the model. And then you've got a lot of new publishing companies whose APCs are much lower than those averages that you just saw. So really, on the cost modeling side, we did the best we could. But the cost of publishing is not something that we understand particularly well. And so let me just throw out a few other things. Some of these publishers who said the fees are too low also said that we're really not taking into account surplus. The nonprofit publishers don't necessarily need a surplus, even though some of them have one. But the commercial publishers are going to expect a surplus of some. And by surplus, I mean profit. So how much you need to take that into account is something that we haven't really tackled. So that is an interesting question. And it has to do with this viability. We're telling authors, well, we're not going to take away the journals that you love. But those journals can charge pretty much whatever they want, even in an APC world if there's no competition. And the other thing some of these publishers have pointed out is that right now they get a lot of money from non-academic subscribers, corporations, government agencies who are not authoring anything at all, but they're subscribing. So kind of getting back to our original premise, they're saying that a lot of the revenue they get from those sectors are going to shift over to research universities and the authors if we don't somehow take those into account. And so they're trying to figure out if there's some way to capture the value to those parts of the world in addition to the authors so that it doesn't entirely fall on authors to cover cost plus surplus, right? So these are the kinds of discussions that we're starting to have now, which is good because before, you know, it was sort of a black art what was going into this. And now we can begin to talk about what costs really are. And I've had interesting conversations even here today with some publishers who are starting to think about disaggregating what they do and charging for different parts of it, discretionarily, if that's a word, so that an author who now has a little bit of money to play with could say, okay, I want to be in this journal, but I'm not so concerned about the copy editing. I did a good job of that. And that's a terrible example, but you kind of get the idea, right, that you can pull out different components of the workflow and see if, in fact, bundling those together is driving costs up unnecessarily. So what we're really hoping to do is not get the data 100% accurate permanently because APCs are in flux, but to come up with a model that you can have and start to plug your data into, you know, what am I spending on journals? How many articles do my authors publish? What journals are they publishing in? How much grant funding do they have? And then start to think about a multi-payer model that could work, right? Because we all know that just having a library's pay probably isn't sustainable in this other model, at least for the majority of the partners. But a multi-payer scenario has a lot of promise. And some of the people involved in this project, like Mark McCabe, the economist, says, until you get authors involved in the decision, we are never going to get out of the situation we're in now. They have to be involved and they have to have some reason to care what it costs. And they could still decide, you know what? I do want to get published in nature because I'll get tenure for that. And then we have to make it possible for them to be able to afford that. So we're not saying that every journal has to cost the same in the end, right? Some journals may well be worth more than other journals. We just have to make sure that authors can afford to pay the fees when they're lucky enough to get into those journals and that it doesn't get so out of balance that one journal is charging, you know, $100,000 an article or anything like that. So now we can start to have that debate. And with that, I think I'll stop. I'll thank you for your attention and we can have a discussion. Thanks.