Scientific Publishing, Academic Peer Review & the Business of Science | Michael Eisen | 17

Full episode transcripts & show notes at: https://www.nickjikomes.com/

Nick Jikomes    
Professor Michael Eisen, and how are you? Good, how are you? Good. Where are you calling in from? And can you tell people who you are and what you do?

Michael Eisen  2:31  
I am calling from Berkeley, California. I am a professor at the University of California here in the department of molecular cell biology and the Department of Integrative Biology. I run a research lab there I study, mostly study fly development and fly genetics and genomics to do a little bit of work on mind controlling parasites. I also spend a lot of my time working on trying to fix the dysfunctional system of science publishing, I, you know, been doing this for about 25 years, I've been heavily involved in the open access publishing movement, and more recently in efforts to kind of change the way that peer review is organized and carried out by scientists.

Nick Jikomes  3:20  
So in academic science, the currency is really publications, you're trying to produce research so that you can produce publications so that you can get more funding to keep doing your research. And obviously, that involves sending in papers to journals, and the journals published the work so even if you're a non scientist, you may have heard of the big journals, nature, science sell things like this, these are the the structures that contain all of the scientific work that's out there. Can you just explain at a very high level in very broad terms, what is a scientific journal? And what role are they meant to play in an idealized scientific process?

Michael Eisen  3:59  
Right? So first of all, just to give people a sense of the world that we're operating in, there are something like 50,000 different scientific journals. The you know, the scientific journal in its purest form is a an organization of scientists really who come together to collectively assess the work submitted by their colleagues for their for their assessment, and to decide, you know, both whether they are, you know, scientifically valid to do that sort of classic version of peer review where you, you comment on the the actual science and your colleagues work, but also to decide whether or not they're interesting and important enough to, to to warrant the time of their, you know, the people in their community to read so scientific journals are, you know, they're the mechanism by which scientists communicate with each other. And there they are. There they place where Kind of two parts of peer review take place the part where we, we decide whether Sciences is done right. And the part where we decide, you know who who's going to be interested in the work and how important we think it is to them.

Nick Jikomes  5:13  
50,000 scientific journals, eventually I will come to how that whole sector works and how it is that journals start and what goes on there. But first, I think it'll be useful for people that haven't been a part of academic science to understand exactly how it works, and I want to start with scientific funding. So how does a typical academic research scientist at a major university get funding? Where does that money come from? And what is it generally used for in the lab?

Michael Eisen  5:44  
Okay. So it obviously there's no single answer to that question. So but but in the US, the vast majority of scientists who work in academic research environments, you know, their salary is paid for by the university, at least in part, so I'm paid in part to teach, and I'm paid in part to run a research lab, but then the labs are, are almost always run by external grants that you have to apply for. So you know, in my my departments, the typical grants would come either from the National Institutes of Health or the National Science Foundation, who are the major public funders of research in the US. And those, you know, they have too many programs to keep track of, but the basic idea of them is you submit a proposal to the organization, it's reviewed by your colleagues and by people at the at the funding agency. And they make some set of decisions about what what of the many different proposals that are sent to them, which ones are going to get funded, and to, you know, to what, you know, to what level of money and for how long and those can vary from small grants that are in the 10s of 1000s of dollars that are meant for small research projects to, you know, the biggest federal grants are millions of dollars and involve lots and lots of people. So it, it's a little hard to describe that world in any consistent way. There's also a, you know, a fairly decent number of private funders, you know, private nonprofit funders. So my lab, for example, is funded almost entirely by the Howard Hughes Medical Institute, which is in a very big, private foundation that was set up by Howard Hughes with the large fraction of his wealth, and they fund 300 or So scientists across the country, kind of paying our salaries and funding our labs and fairly generous way to to do work that is kind of broadly related to medical to medical research. And then there's, you know, there's dozens of other private funders that that have, you know, varying amounts of money to support to support research like that, then there's a smattering of private funding and, you know, company funding in an academic research labs, but that's a small minority of what goes on.

Nick Jikomes  8:07  
So a professor at a major university is running a lab, they also likely have some teaching responsibility, the university is paying you in part for teaching and in part for creating new knowledge through your research efforts. And so let's say you're a PA running a lab and you get a big NIH grant, that money then comes in and what happens to it? What's the first thing that happens to it? Is it all going right into the lab to do the research?

Michael Eisen  8:34  
Um, no, it's a very elaborate system that I think very few people fully understand. So if I get a grant for a million dollars from the NIH, the NIH actually cuts a check to the university for something like $1.6 million obviously spread out over over time they don't give it all to them immediately. But and you know, you know say the million dollars goes to my lab and and a lot you know, for paying salaries of people in my lab and you know, most of my expenses in the lab our salaries and reagents, you know, things we need to do experiments So, but the majority of that being salaries, and then and then the university takes some money on top of that, but I think at my institution, it's something like 56% or something like that. On top of the money that I get in my lab as what they call overhead and the overhead is goes to support all the things that universities do that they don't directly bill me for so I don't get a bill for you know, the people who maintain them the you know, my building I don't get a bill for the for the library. I don't get a bill for you know, I mean I get bills for all sorts of little things, but but some major infrastructure things at the university, just keeping the buildings up to shape and building new buildings and just running the place. That is that is covered by the the overhead. And it's a huge amount of money for the universities, a large fraction of the operating budget of universities comes from these kind of overhead on federal grants to the point where even a place like the University of California, I think I saw a stat recently where they get more money from the federal government than they do from the state of California. So it's, you know, it's a major, it's a kind of a major part of the operation of the university, especially a big research university, like University California, that the faculty, you know, bring in research grants. And via those research grants, we got to fund operations to university.

Nick Jikomes  10:42  
And so I used to be in academia, and now I'm in the private sector. I'm wondering if there's any analogies we can draw between compensation in your world versus the private sector. So let's say you've got two or three professors at a university, one of them brings in a half million dollar grant, one of them brings in a million dollar grant, one of them brings in a $10 million grant, if you were in the private sector, and you were doing sales, say your competition would be proportional to the money that you bring in, is there anything like that in your world,

Michael Eisen  11:12  
um, there's not supposed to be so are our and, and, you know, again, like, a lot of things, it's complicated, but, you know, our salaries at a public institution like University, California are not supposed to be tethered to the income we bring in for the university, it's not, not considered conducive to our educational mission. And so, you know, um, faculty salaries that you see are set on kind of a scale and, but, but there's always, there's a little bit of fudge factor. And, you know, the reality is our salaries are not explicitly tied to our, to the grants we bring in. However, you know, people get offers from other universities, especially private universities that have fewer constraints on what they can pay people, and they, they, you know, they can, if you bring in a lot of grants and incentives, University has an incentive to offer you, you know, various and sundry perks, both financial and non financial, to try to lure you to that university. So I think it is definitely the case that, that, you know, on average, if not directly, and explicitly, faculty salaries are, are a function of how much money they bring in for the university. And that's, you know, why, for example, I think in, you know, well funded departments where the grants are big, the faculty salaries are generally higher than in, you know, like, our history department where they don't have that kind of, kind of, you know, Lou art faculty get offered tons of money. There are some circumstances where the feedback is more direct, I think, some places like medical schools don't actually often don't pay their faculty salaries at all out of the university's budget, they're entirely raised by grants. And they're, you know, depending on where the funding comes from, there's different constraints on what, what kind of, you're allowed to effectively pay yourself out of your own grant. So I think it is definitely, overall the case at universities that the more grant money you bring in, the more your salary will go up in a court, although I think by and large universities try to not make that calculation, quite that quite that explicit.

Nick Jikomes  13:34  
I see. So there's a bunch of scientists with a bunch of labs across the country, at major universities, they're bringing in money. for research purposes, a large chunk of that goes to the university just to fund University stuff, building buildings, and so forth. a chunk of that then goes to your lab, which is primarily going to pay your graduate students in your postdocs, the scientists actually in the lab doing the work. And a good chunk of that is going to buying all of the stuff, the reagents and the equipment, you need to actually do the experiments. So you do the experiments, and you create a story. So you have to write a paper, can you talk to people a little bit about what goes into actually writing a paper set, let's say it's a pretty good body of work that you're going to send to a high impact journal, or that one might send to a high impact journal, how long and I know there's no a lot of my questions aren't gonna have one answer, but let's just kind of sketch the contours for people. How, how long is a normal research process, thinking in months, years, many years?

Michael Eisen  14:36  
I mean, I would say the typical, I can speak for myself and probably fairly representative, at least a certain, you know, portion of the scientific world. You know, a prototypical paper that comes out of a prototypical research lab is probably, you know, several years of time to Three years of time of the primary scientists who carried out the work, usually there's a single point person who is then has worked with others in, in varying degrees on a project like like that. So, you know, I've never done a full accounting of this, but I would suspect that the typical paper from my lab is probably, you know, four, four ish years of time from the people who were carrying that workout. And, you know, it varies, we've

had some that came, you know, where, from idea to paper was extremely quick and didn't didn't have that duration, and I have others that were the fuse is even slower burning than that, that take years, they take decades to kind of fully mature, but in terms of the total amount of labor and effort and time that was spent on a paper, it's probably, you know, on that order four or five person years worth of time, and you know, some of that time is not directly reflected in the data that's in the paper, since every every project is in, you know, has a long period where you're kind of figuring out what works, what doesn't work, what's interesting, what's not interesting. And then, you know, once your thinking has crystallized around what the right experiments to do are and how to carry them out and analyze them, then there is a much more intense period during which you're collecting data that you are actually now thinking about publishing. I think oftentimes, at the beginning, you're doing experiments that are really for yourself, and for learning your way around the subject, at least, at least for us. And then once you've done that, you kind of put it, put two and two together, and you're like, Okay, this is, this is the topic, this is what's interesting. And now I know what to do, I know how to do the experiments, right? And you do them. And probably the period of time between, you know, when that happens is like, you know, that can happen quickly, once you've got everything in place. And then, you know, we collect the data over the course of several months, and then several months of analyzing it and putting it together. And then the process of writing a paper this is, you know, I would say, I don't know what your experience was, but but, but that is probably the most difficult thing for everybody in science in the sense that, you know, you are trained and directly and indirectly, in doing research, like you learn how to carry out experiments in the lab that someone shows you how to do it, and you learn how to analyze data and think about science by reading papers and talking to but but we actually have very little, it's a very hard thing to teach in the sense because it's very, it's just one of those you kind of have to do it learn by doing Yeah, right. And, and so, you know, everybody has a different style, I mean, I, I try to encourage, like, Alright, figure out what your story is in your head, write it down, but this isn't gonna be in the paper, this is for your own your own metrics, and then make figures you know, the unit the unit of currency in a paper if the paper is the unit of currency and science, by and large the unit of currency in a paper is the figures, they let they capture your story, they tell the story, they they display the results in the way you want them to be absorbed by the readers, they're kind of instrumental in in setting out what you're up to. And I think the two the two things that are you know, they're also kind of the easiest things to make materially because you know, people have writer's block but they don't really have figure block, right like because you can play around with it either it's good thing that people don't seem to have a you know, you you make graphs and whatever you know, we don't do it in Excel anymore, but we only make we write code to make pictures and and you can play around with those quite easily until you come up with something that makes some sense and you make you make I mean I literally been doing this the last week I make the first figure and I then I go and make the second one and the second one makes me realize that the first one needs some tweaking and then I make the second one in concert with the first one then I'll make the third one and I do the same thing over again. And so it's a it's a bit you know, then you've once you've got kind of the figures laid out, you've got a story and then you start to wrap text around it to explain what the figures telling you why you did it this way and not everybody works that way. You know, there's there's as many ways of constructing a paper as there are, you know, you know, different you know, ways of thinking about them, but it's, you know, it's a very, it can be a very long drawn out process to do that to do that. Well because, you know, this is the You know, just forget about all the infrastructure of publishing and the way it actually happens with this is what you're in science for, its to tell its to its, you know, to me, like the distinction between science as a way of asking questions. And science as an endeavor is sharing your information, right? Like, I can be a scientist, if I'm locked in a cave, and no one ever knows what I'm doing. And I'm applying the scientific method, I have a hypothesis, I test my hypothesis, I, I learned something, and I do it again and again, and again, that, that science, but it's not science, unless I, you know, describe it in a way that other people can learn from it, they can build on it, they can benefit from, from what I've done, they can tell me where they think I've made mistakes, that we can learn together how to do science better. And so, you know, you can conceive of mine, you know, we could make YouTube videos about our experiments or something. So it's not, it's not like science has to have papers, it's just that that is the

you know, by and large, still the, the way that we record the things that we think matter about what we've done, what our ideas were, how we did our experiments, what we found, and what we think it means, and so those are the elements of science, right? methods, data, ideas, results, conclusions, those are the basic elements of science, and, you know, we share them in papers. And so they are, you know, you know, forgetting us for a second the, the whole career infrastructure that's, that's, that's attached to them. And that it, you know, it really is true that like, the scientific literature, which is the collective body of papers that have been published by scientists, since they started writing papers, you know, the scientific journal is, was born in the 17th century, it's been around for a long time, and, you know, collectively, there's probably been, you know, on the order of 100 million papers written maybe slightly less than that, but, but something like that science papers that many science papers have been produced, at that collective body of information is just, it's like one of humanity's greatest, greatest creations. And so it's, it's, you know, it's an it's an amazing thing. I mean, it's so many problems and how we do it, that you can sometimes get lost in the details. But in the, in its collective value, it's, you know, so much of our lives are improved and, and our intellectual life well, as well as our material lives by things that are in the scientific literature that, you know, all these COVID vaccines, for example, right. They're born from the scientific literature, and many, many, you know, direct ways. And so I think it is useful to sort of step back and appreciate that, you know, that thing that scientists have created over time is a pretty, it's, it's pretty cool, amazing.

Nick Jikomes  23:00  
Yeah, it is really, it is really amazing, you've got this giant corpus of information, each one of these things is taking, on average, some number of years to do. And for those that are not have never done this, it's not a casual few years, it's a very involved and dedicated few years for each one of these things. Yeah, I mean,

Michael Eisen  23:19  
I don't, you know, I don't, again, I can't speak for everybody, and I don't even sure this is the best, the best way to function and as a human, but like, you know, the site that most the papers that I have done the work on, like, you know, obviously, as a faculty member, now I do less than less work on individual papers, compared to the students and postdocs to guide them out. But back when I was in grad school, and a postdoc, you know, that the, though the year, the year is, the year is that went into making those papers that I published, where we're, you know, those were years of a lot of work and effort, toil. So that this is this is, you know, it's it's Yeah, it's it's an immense amount of human physical time, labor time, and mental energy that goes into each of those. Each of those works so.

Nick Jikomes  24:12  
So you've got some number of people that have put in that time and that effort, they've got a story that they think is really good, they've got results that they think are really exciting, so exciting that they want to send it to a top tier journal. What is that process? Like when you're submitting a paper? Let's caricature that for people.

Michael Eisen  24:30  
So I think it's probably worth before doing that. Just describe what when you say talk to your journal, I think yeah, it's worth understanding the structure. And in a little bit of the history of this, I mean, back when, you know, back at the beginning of when there were science journals, there was just science. They weren't differentiated, right? There was the Royal Society in the UK. That was scientists, every scientist whether you were a botanist, or, you know, or, you know, microscopist or, you know, I don't know what other scientists You want to think about back then. But like every, you were a scientist, and you went to the Royal Society, and you shared your results. And so you know, you might hear a talk by somebody, you know, attaching electrodes to a frog's leg. And then the next day, you'd hear someone, you know, looking at the layers of, of, of rockin mountain, they were exploring, right, like it was science was a single thing. Over time, as the, it became a profession where people, you know, specialized, as more and more scientists got involved, there started to be more and more specialized venues for sharing your work, it started to be true that, like, if you were describing a new species of plant, and your travels to South America, you're your audience, like the people who were, you know, looking at, at the layers of dirt and coal mines in, in Finland, born can be interested in the plant you're describing from Brazil. And so there, it started to be a differentiation of, of, of, you know, science into little pods of people who had like interests. And, you know, back in the day, when the means of conveying your ideas show, again, the Royal Society, you should just have meetings, you just show up on Thursday night, or whatever it was, and like, someone would read it, like, read their results, and there wasn't a journal, because you were all there, and you made her talk, right? Like zoom. And, and, but of course, you know, people couldn't come some week and or they lived in far north of Scotland, or whatever. And so they started as printing became cheap enough to do they started to print, you know, reports and send them around. And so that was, you know, for most of the history of science, that was an expensive thing to do, right? It costs money to typeset print mail. issue. And so journals, really, in their current form, were born in the 19th century, as as to try to

define audiences of scientists, and in some cases, non scientists who are interested in the same type of stuff. So you would have a, you know, there were some general science journals, the same ones that exist today, came into existence in the 19th century science and nature, they were meant to be, you know, for the most interesting stuff that like, the kind of geology that you should care about, if you're a botanist, or that right, like a species that's discovered that's so interesting that even somebody who's you know, you know, just, you know, inventing, you know, new ways of, of manufacturing metals in the lab, right, there were these general science journals that were meant to capture the most exciting stuff. And then there were a growing number of specialty journals that were the Journal of botany The Journal of mal ecology, all these things, right. And they were partly in response to the just the growing scale of science and partly to the fact that the economics only makes sense if you print and mail things to people who are willing to pay for that person piece of information, and a botanist isn't going to subscribe to a Malakal journal. And so the the there is, you know, what was born out of a kind of natural, you know, it was born out of a natural kind of segmentation of the industry, it became over the course of the 20th century, especially in the latter half of the 20th century, a way of measuring your success as a scientist, it became it became a particular feather in your cap if the botany work you were doing didn't just get into the botany journals, it got into the general science journals, because that meant that what you did was of sufficient import that everybody was interested in it and, and so that system became hardened over time. And, and it particularly in the, from the 1960s onwards, when science funding exploded, science grew as an endeavor. And it it stopped being true that that the people who were kind of deciding who should get funded or hired in departments were experts in every area of science they were covering and so there was a little bit of like, well, if the you know, if nature says this work is interesting, we're going to believe it's interesting because the people that nature know what they're doing, that was the assumption at least and and then that so that notion, that that there is a kind of relationship between the size of the audience of a, of a journal, and how important the work it publishes, has become kind of systematized in science. So it's not an accident that the top tier journals, the ones that people aspire the most To get into where most people aspire to get into, are the ones that have the least specific names, science nature cell, those are about as, as generic statements about, about things that they contain, as you might imagine, right? Science is probably the most generic right? Nature maybe it's a subset of science but it's it's still like very big and sell. You know, everything has cells all life has cells, so it's alright, I'm not gonna I'm gonna piss off the people who who think viruses are alive, but right, so there is this hierarchy that exists in people's minds, where, where the more the more generic the journal, the more, the more difficult it is to get into. And therefore, the journal started to behave that way. So science nature cell made them sell, because they were attractive, they, they became very difficult to get into, right, they're like, you know, the Harvard of, of publishing, right? They become appealing because of their name. So everybody wants to get into them, and now they can become really exclusive. And, and so,

you know, I want to be clear, I think this system is terrible. I think it's broke a broken system, it's bad for science, but it is the way that people operate. So it is useful to understand it that that the the, the ability to get your work into those journals is taken it you know, in some it figuratively and in some parts of the world literally as a measure of your worth as a scientist, because if it is sort of equating equating the assessment that the journals do have what is going to be of interest to a very wide audience with, with science that is important. And you know, you can pick apart that, that notion, quite effectively, I think, but that is the pervasive idea in science. So, so when you ask the question, what happens, what happens is, if I have a paper that I think, you know, even potentially belongs, you know, I will say, I don't do this. So like this, I'm not describing this for my lab, this is we shun the system, in many ways. So I did to describe this generically. But, you know, the way it works is you pick a journal that, you know, you think your paper belongs, and then usually, there's a little bit of aspiration involved, right? Because you're rewarded, you know, that like, just like with college admissions, there's like, people have safe schools, you don't, you don't submit your paper to the safe journal, you try, most people try to, you know, go higher than they think the paper might be if they were betting where it would end up. And so, you know, a lot of people send papers to nature science, so they reject, I don't know where the exact numbers is, but they reject probably 98%, or something like the papers that gets submitted to them. Most of them, they reject, just at first blush, some editor there, reads them and says, You know, I just don't think this is going to MIT gonna, gonna make it, and they just say thank you politely declined to push, the ones that they think are have potential, they then send out to peer review. And, and, you know, different journals operate differently, they all have editors who do that first level of screening. Some of those editors are professionals, usually PhDs, who decided that they were more interested in science communication than in doing bench science. And, you know, usually, usually these are pretty, very broad minded people who can cover large areas of science and sort of just enjoy the process of reading about the latest discoveries more than they did, making them themselves. And so they you know, there's either professionals who do that or active scientists who do that a lot of debate in the world are which of these is better? I think they're both they both have their pluses and minuses so But in either case, an editor looks at the paper and just make some initial decision about whether it potentially could fly at the journal. And then and then if they think it can, they send it out to people who they consider to be experts in the field to assess the actual science in the paper. So

Nick Jikomes  34:19  
gotcha. So the editors Oh is a scientifically trained person and may or may not be a practicing scientist today.

Michael Eisen  34:25  
That's right. So it depends on the journal and some journals, there's a mixture and whatever but you know, these are people who, who have been active scientists in their in their lives. And I'd say the vast majority of them have written papers, often many papers, so they have some experience on the author side, as well as on the consumer side of, of science. And they're, you know, they're not, depending on the journal, they usually don't act as directly as a peer reviewer. They're kind of overseeing the process. But they then you know, they they use either Their own knowledge of the field or some database of reviewers to pick reviewers who they think are appropriate for the, you know, to assess the science and not usually mean somebody who is in the field, you know, does does research, like the research that's described in the paper so that they can assess the, you know, the methods and the details in a way that that usually requires having some specialized knowledge of the of the field and the context of the work? And then, you know, they are, they're asked to do that in a couple of weeks, it usually takes them longer. And then they write a report that is partly a communication to the, to the authors describing what they think about the work and partly a communication to the journal describing whether or not they think it, you know, belongs in the journal using their knowledge of other papers that have typically been published in the journal. And then, and, and often some, some description of what if it doesn't already belong in that journal, what it would need in order to get there. And so that's the typical components of a peer review are direct comments on the science for the author's editorial comments on the work for the journal and kind of a roadmap in their own minds for for, for how to bridge the gap

Nick Jikomes  36:31  
if there is, and what's a typical number of peer reviewers that would look at one paper,

Michael Eisen  36:38  
you know, that the typical number is three? For no reason other than you need someone to break a tie? I think it's the right, why three? I mean, you know, I think generally, we feel like one isn't enough, because, you know, there's often different views. And, you know, two or three are probably equally useful in some sense. But, you know, it's good to have a third person in the room to write anyway, it doesn't always happen. I mean, the journal I run, sometimes it's two, sometimes it's four, you know, we often ask more people than we need, just because the, the median answer is no. And so because people are busy, and there's 2 million papers published every year, which means 6 million reviews have to be written. And it's actually more than that, since papers get rejected from one journal, and then go to another one. And so it's probably 10 million, tried to estimate it, somewhere between five and 10 million reviews are written every year of papers. So if you're in a field that publishes a lot of papers, and you're generally regarded as a good person in the field, whose judgment is valued by the journals, you know, I get review request, many review requests every day, just to give some perspective on this problem. So So, yeah, so that that three is aspirational. It doesn't always happen. But I'd say by and large, it's, that's what you get.

Nick Jikomes  38:03  
And so, you know, there's this incredible number of papers being submitted and reviewed every year, you just said that, you know, you're getting requests to review these every single day. Can you give people a sense for how much of your professional time you spend reviewing papers as a peer reviewer? And who is that is are you getting paid to do that by your university? And or is the journal actually paying you because you're doing it on their behalf?

Michael Eisen  38:30  
Yeah. So um, the? That's a tricky question. So the, the answer, I, again, right now, because I'm running a journal, I don't actually do a lot of time peer reviewing, because anyway, because I have other things to do. But I'm the, I would say, back when I wasn't doing that I would probably peer review, I'll probably on average of paper a week, maybe a paper every week and a half or something like that. So so something between 30 and 60 papers a year would be typical for the number of papers, I would peer review. In a part of that, that's probably more than typical, but not by much and, and you know, each of those papers probably takes, you know, even if everything's smooth, it's hours of work to review a paper to read it carefully, to construct, you know, to write a review that's thoughtful and warrant, you know, like, you don't want to send someone just a cursory you know, review of their work, or you want to show that you've read it and you want to do them, right. And so, yeah, yeah, so the question of who's paying for that, so, in very few cases, journals, the journals pay for that directly. So like, there are some fields where this happens like economics, I think they pay reviewers, nominal fee, but even if they paid a nominal fee would be less than, you know, then what? I'm being paid for my time by the university. Right? So it, you know, it is considered a professional responsibility in the sense that I included on my CV that I review journals as part of my assessment by the university. And so, um, yeah, like, in some sense, I'm being paid by the university. To do this as part of by role as a scientist. It's not really explicit, though. And this is something that,

you know, that comes up a lot, like, oftentimes graduate students and postdocs are kind of looped into the review process by their professors. And, you know, it's a good way of learning how to write papers is to help to review them. And so part of that to training exercise, but it's also, you know, they're not they're definitely not being paid for it. So it's a very fuzzy thing in this industry as like, you know, because because the industry, you know, what the publishers actually do? Why do we need journals? Right? So one big question that, that I think is that I asked all the time is, why do we need journals at all? What do they do? Right? They used to print things and mail them, we obviously don't need that anymore, we have the internet, right? Like, you're going to post this video on YouTube, you're not going to make CDs and mail them to everybody who's interested in the video, right? So the mechanisms of dissemination have changed since the internet came along. And, you know, so the actual physical dissemination of work is no longer really a thing. So what then did, you know, if you distill out what journals do, they have two main functions, right, they oversee the peer review process. And they oversee a process of transforming whatever, you know, poorly formatted. document, I sent them in the first place into something that looks pretty, that too, is becoming more of an automated task. I mean, honestly, like, it's mostly farmed out to companies that specialize in converting you trash the Word documents into pretty web content. And so the real thing that journals do is oversee peer review. And so you can ask the question, like, you know, the scientific community is basically subsidizing that endeavor by paying the salaries of all the people who do the work. And, and so one of the big questions is just, you know, what are the kind of terms of that relationship between the scientific community and the journals? What should science writ large, and scientists individually expect to get back for that, for that, for the fact that they are actually the really the intellectually valuable work, but the value add, as you would say, in any industry is, is the peer review, right? That the content that comes in, if we just took the content that authors sent us and publish them online, that would be a business, it would cost some money and whatever, but it would have no value add would be easy to quantify, because it's just paying, paying XML chop shop to turn turn documents into web content, right? that's a that's a commodity, the intellectual value added, which is, you know, comes in two pieces, it comes in work changing and being improved notionally by having someone else read it, comment on it, point out where the flaws and shortcomings are even just where they didn't understand something to help them communicate their work better. So that's, that's part of it. And the other part is this annotation that comes with putting it into a journal, which is, you know, it's valuable to me as an author and valuable to the reader to know that this particular group of scientists thought this work was worthy of my time, how valuable it is, and can we do it in a different way is another question but, but they, you know, those two pieces of value are based on information that's being provided by scientists in the vast majority of cases for free, or at least in a, in a way where they're being paid for their time, external to the, to the to the industry. And so the fact that that's where the value comes from, and yet, journals, journal publishing is, you know, an incredibly lucrative business and,

and it's Comcast, so let me let me finish just for people describing the process because I think it's worth knowing after you get the peer reviews, usually the editor looks at them and sometimes in consultation with the reviewers, and comes up with a decision and the decisions are complicated. Sometimes it's, hey, we'd love it, we'll publish it. It's rare for that to happen on first, first run through, almost always they reviewers in the editor have something to say back they'll say, Yeah, well, it's great, except, you know, this one experiment could be presented in a different way. Or maybe you could do this other control or did you think of this bla bla bla bla bla, usually, even when their their judgment is favorable, there's some asks from the journal to the authors to modify the manuscript or modified analysis are in sometimes do more, do more experiments. And, and you know, the more, the more attractive the journal is to the authors, the more difficult it is to get in, the more the journal feels it can ask for things from the author. So a very high impact journal will, is feels in a pretty comfortable position to say that, yeah, this paper is fine. But if you did this experiment of this experiment, this experiment, now it would belong in our journal, the authors have an incentive to do that. And so at a journal that's less that's more specialized doesn't have the cachet, the motivation, the author's motivation to do that is lower, and so the asks are usually smaller. And so you know, and then, you know, even at journals, even the journals that have decided, you know, once they've decided to peer review paper, they will often reject it, because either flaws were found in the science, that's typically why flaws are found in the science that can't be easily addressed by a few additional experiments or analyses, or, you know, what looked like an interesting and important results didn't turn out to be because the, the data are believable, but don't, reviewers fail, they don't really support the broader, bigger conclusions of the authors made. So that's what typically big white papers get rejected. And those rejected papers, we've looked many times at different journals, they almost all end up getting published somewhere. Because, you know, either because different reviewers see in a different way, or because the different journal has different editorial kind of policies and goals. So you know, it, there's a little bit of ping pong that takes place both back and forth, one journal, but also between other journals, and then, you know, then they get published. And that's where the 2 million number comes from. Every year, about 2 million papers get published in this big collection of 50,000 journals. And you can do the math and realize that most of the journals are obviously not publishing a huge number of papers, right? The average number of papers published per the for the average journal is less than one per year. So per week, I mean, and yeah, less than one per week. And so, you know, some obviously, some publish a lot more, and so some published fewer. So while there are 50,000 journals, it's a very long tail of, of journals that don't publish a ton of content.

Nick Jikomes  47:24  
Okay, so just to sort of regurgitate some of that, you're a scientist, you're working in a lab somewhere, you've been funded through grants money, you've spent probably a few years doing work and crafting a story. you've submitted that to a journal, it's now taking months, potentially even longer than that to go through the entire peer review process, depending on if you have to do more experiments or anything like that. And then it finally gets published, we've talked a little bit about the really, really large volume of papers that get published. And you mentioned that you mentioned two things that are interesting one, the biggest value add for what's going on here is the intellectual work done in the peer review process. So the scientists reviewing the work of their peers, the journals have basically crowdsource that right there, they're shipping that work out to all the other scientists, because that's what peer review is, you also mentioned that it's a very lucrative business. So where, where does that revenue come from?

Michael Eisen  48:26  
Right, so so this is kind of the elephant in the room here, which is that in the pre internet era, journals, there was kind of a logic to the way this industry worked in that you know, it costs money to print and mail and pay a journal around the world. And so since the costs, largely scaled with the number of copies you produce, not entirely there's fixed costs involved in peer review and things like that. But you know, a large fraction of the cost at least were, were were scaled with copies of the paper sent. And the the the standard business model was a subscription one where if you wanted this content, you would subscribe to the journal. And that was in many journals done by individuals like even I'm just old enough that the beginning of my career like we're still we're getting print journals. And so for some period of time I subscribe to the print journals I liked the most like, you know, I subscribe to the journal genetics because I love it into until long after I could get it online, just because I like getting a tactile, physical version. So most scientists was subscribed to a few journals, usually through a scientific society. But then the bulk of that, you know, that the that started to become less and less efficient over time. And so the bulk of that subscription, kind of, you know, revenue was what's coming from library. So library started, you know, live braze took over the task of kind of providing access to journals, especially the ones that people didn't read on a daily basis. Like, you know, most scientists still, I think, get science nature, individual subscriptions to them to read the science, and because they also have a new section that they like, and then, you know, I don't know how many people still get print issues of other things, probably fewer, fewer. But back, sort of when I was starting, you know, typical scientists would get, you know, four or five subscriptions to four or five journals that would be in their office, the kind of things that they looked at all the time. But obviously, I, you know, I will, when I'm doing a research project, I will be drawing on work that's published in dozens of journals. And so it's not practical for me to subscribe to every journal for every field I'm interested in. And so that work over time got shunted over to the libraries, who were obvious, natural places to do this. And the industry developed a kind of business model where libraries paid more and more for their subscriptions. Because they, you know, the institutions knew they were sharing them with larger and larger number of people. So it's typical to have like, to be much, it used to be typical to be much cheaper to subscribe to a journal as an individual than as an institution. So increasingly, the publisher started to see the libraries as their customers. And they, you know, and but because they were few, you know, there are fewer libraries than there are scientists, they had to start charging more and more, because the libraries were where the, you know, they're so you know, not their sole, but probably their predominant source of income. And that started to become a problem. Even before the internet came along, it started to be a major economic strain on journal on libraries to, to subscribe to journals, especially because as printing got cheaper, more journals were born, because the journal saw in libraries, library, University Libraries, especially had especially big research libraries have a mission. Like, you know, they're scientists. What if I, if I started to do a project, and my project is on, you know, the snails of Bolivia, I want the Bolivian you know, Mallika ology journal, right? And so So scientists came to expect, especially a big research universities that the library would have everything. And so it cut, the industry started to get a little out of whack, even as early as the 60s, when, when there was, you know, journals were born, essentially, to cater to the library market instead of individual scientists. And part of the reason the journal industry has grown so big is because of the centralized kind of purchasing of subscription.

Nick Jikomes  52:53  
So is that so is it fair to say? So when you say that libraries are the primary customer, you're typically referring to a university library? Yeah. So you go to university, whether you're a scientist or just an undergrad, or whatever, you have access to all of this knowledge now through the university, universities paying essentially for a bundle of journals the same way that like bundling is so common just in our ordinary lives, now

Michael Eisen  53:17  
you buy the bundling is a relatively new innovation in that industry, it came kind of late to the play to the field. But yes, that's what they do now, like, right, and it's okay, so so it's useful to make to draw a line in the sand here between the before and after the internet, right. So as the, you know, even before the internet came along, there were big cracks in this model, because there were too many journals, they weren't charging higher and higher prices, their customer, their primary customer was not really operating in a free market, because, you know, the University of California library felt a responsibility to subscribe to every journal out there. And if they canceled a journal, they ran the risk of some, you know, professor who, in general is in a more prominent position within the university than the library and complaining that I can't believe they canceled their subscription to this journal. And so there was a lot of pressure on universities to subscribe to everything, a situation that publishers took advantage of, they took advantage of it to, to spawn new journals, and to increase their prices. And, you know, since the 60s, the cost of science journals has been going up at a rate greater than inflation. It's been very consistent. It seemed like right so but but but but importantly, at its core, there was still a logic to the system, right? Like you were still disseminating things in print, you still needed to buy access to a physical copy, even if they were charging too much for it. And at least the that you had to do something like that it made sense to send copies of every scientific paper to every library and this was a mechanism to do it, even if it wasn't the most economically sane and efficient one. But as soon as the internet came along, all the logic of that just completely evaporated, right, it stopped being expensive to send an extra copy to anybody, right? It's, you know, the marginal cost of downloading a PDF of a, of a paper is close to zero. And so, so um, but because the publishers were operating in this incredibly lucrative business environment where, you know, their profit margins were workstream, even before the internet came along, they were they were profiting the big publishers, there was a lot of consolidation in the industry, smaller journals got bought up by big publishers, who again, went to the universities with these big deals and said, we'll give you a 75% discount on every one of our titles, if you buy them all at once, which, of course, is great for the publishers, because they get this huge check, encourages them to buy more journals, because now they can justify a higher price. So

Nick Jikomes  56:01  
not just, it's not just a bunch of individual academic societies with their journal, it's large publishers that are creating bigger and bigger libraries of journals. And then

Michael Eisen  56:11  
that's right, the whole pay, you know, they would in fact, in many cases, just say, we'll take over your journal, and we'll give it cut you a check every month or a year for something. And the society now had a guaranteed revenue stream and no expenses, because they were, it was a great deal for the society. It's like it, it was, it was good for everybody except science. And so and the reason it wasn't good for science was when the internet came along. And we should have just like there should have instantly been a YouTube for science, right? We're like, I have a paper, I post that paper and anything that happens subsequent to it. Especially because most of the works being done by scientists, anything that happened subsequent to it is, you know, happens in the open and it's paid for by science, right? The libraries could have taken their subscription budgets and thrown them out the window and paid for all these costs easily, there was more than enough money. But that didn't happen. I that didn't happen. Because in the meantime, it became true that like your worth, as a scientist was measured in the journals that you published it. And so the publishers could basically just dictate whatever they whatever terms they wanted to science, because the individual scientists weren't going to change things because their careers they felt their careers. I mean, you can quibble with whether or not this is true, but they certainly feel that their careers are dependent on publishing in the right journals. And so they, you know, the journals have no incentive to change anything, scientists are still sending them their papers. And now the journals can say, Yes, sure, we don't have to mail you a copy of the journaling. But we still have to pay if you don't pay, you don't get access. And so the journals were like, pioneers and paywalls. They were like, you know, within months of the birth of the modern internet, the first modern paywalls were erected by, by science publishers, trying to like continue a subscription model. In a world where the technology like it no longer made sense. And so I always like to joke that if, if the internet had been invented before the printing press, right, right, like, it sounds crazy, but there's a real particular reason why that that you can imagine alternative histories where digital technology takes takes takes off before print. And you wouldn't not have invented the system. This way, you would have invented a system where there's a common central place where people post all their work, and it gets peer reviewed. On top of that, like the logic of that, there became a huge mismatch between the economic model and the structure of the industry and the technology of the moment. It's a it's as anachronistic as anything around. And it's fueled. It's fueled in part by profit motive of the publishers who make tons of money, but it's really fueled by the fact that scientists, you know, we basically continue to judge scientists, you know, they publish, yeah, we judge them in 2021, in exactly the same way we judge them in 1994. And that the internet has had, you know, close to zero impact on on that. So,

Nick Jikomes  59:25  
you have the there's a screenshot that people love to share on social media, where there's some article written in one of the big journals titled something like the growing and accessibility of science, and then it's behind. Yeah, of course, like, it's

Michael Eisen  59:39  
like, and it's like, it's, it's like, it's, it's, that happens all the time. There's not just one science thing, right? So it's, it's, you know, like, that's the way it works. And it's, it's not like it really doesn't have to be that way. And I think that's the critical You know, that's the critical thing. Like we have So just so people understand this, this is a huge industry revenue for science publishing is in excess of $10 billion a year. That's more than $5,000 per paper that they publish. And their profit margins are, are, you know, as good, you know, they would make apple jealous. Like, like, if you look at the list of the most profitable corporations, it's, you know, software companies and science publishers dominate, because

Nick Jikomes  1:00:30  
who are the big publishers just so people have a sense of the Yeah,

Michael Eisen  1:00:33  
I mean, the biggest is Elsevier, who is a Dutch publisher who has been around in some form for many centuries. And, you know, we're, they were innovators in this, buy up other journals and sell big packages to universities, which help them grow. They're also a big book publisher, as often that's the case with these, these science publishers there, they are the biggest by far of the publishers then the next two are widely and, and then Springer, who also owns nature. So Springer nature, those are the big companies, there still are a fairly large number of independent societies that operate their own journals. So science the other so nature is obviously owned by Springer nature, cell is owned by Elsevier, it was not started by Elsevier, but was started by an individual scientist who then sub sold it to Elsevier. And science is still operated by the American Academy for the Advancement of Science and American Association to the best of science. And they, they are not for profit, but you would be hard pressed to tell that from their business practices because they look you know, they're there they function, the society is funded in large part by selling subscriptions to science. So it's, you know, there there are still are nonprofits, but it's a shrinking part of the industry. And really, these these commercial players dominate the dominators. We know that you see, you see how had drawn a little bit of a line in the sand against Elsevier, unfortunately, they retreated back across it recently, but they had they had refused to pay Elsevier demand. So for about 18 months, we didn't have access to Elsevier journals, didn't have legal access to Elsevier journals. And, you know, you notice there's a lot they publish a lot of stuff. The point like, you know, day to day, it's a day to day actions of science, you you touch, probably touch all three of those big publishers.

Nick Jikomes  1:02:41  
Okay, well, let me let me try and summarize some of the key points here that we've touched on. So far, you've got this big legacy industry, which for historical reasons, has come to take the shape that it has, you've got these big publishers that have sort of corralled a bunch of journals together, so that they could sell access to all of that information to universities. And by virtue of its nature, a university wants to provide all of that information to its students, and its faculty, and so forth. So you've got kind of a captive audience, which allows these publishers to have the business models that they have, yes, the peer review process itself, the single arguably most valuable part of getting something from written in a Word document to publish out in the world, is done by scientist, for other scientists, that's what peer review is, that's Kwazii, paid for maybe in by the universities, because it's supposed to be part of your job. But it's this duty driven thing that everyone feels rightly an obligation to do. So one of the biggest single expenses for a journal in theory, is actually not there, because they don't actually have to pay someone to do that work. And that's why in part, their profit margins are so high that expenses are not arguably where they should be.

Michael Eisen  1:04:00  
That's right. It's just like, it's like a professional football team if someone else paid for all the athletes, like let's say the NCAA is a good example. Right? Like, like, you know, they're the NCAA is probably a perfect example of like an organization that that, you know, effectively exploits free labor. You know, the least we can say that peer reviewers and science journals don't get concussions and the injuries and stuff but, but it's the same model, right? If you're, if if the actual vast majority of your labor is free, and you have a customer base that will pay huge amounts of money for your product, you're running a pretty darn good business, I think in Business School. Like if you came to them with a case study that says hey, I'm going to start a business where we're not going to pay anybody for the stuff we're buying, but we're gonna have a fixed captive market that will that will absorb ever escalating prices to pay for our product, they would fail you because it's ridiculous, right? But of course, it's it's it's The function that that is, that is what this industry is. And, you know, I should be clear, I don't blame the people who run these companies for operating this way, right? It's sciences fault that that it exists, we could change it tomorrow if we wanted to, like, there's nothing, there's nothing, it there's nothing essential, they provide no essential activity other than organizing peer review, which is something that, you know, could be reconstructed relatively easily from from the ashes of the publishing industry. So it exists, it would go away entirely if we didn't have this, like linkage between careers and journal titles. So yeah,

Nick Jikomes  1:05:40  
I want to I want to talk about that next I, you know, I can just tell people from my own experience, when I was a graduate student, like, no one would walk around and say, explicitly, usually, I only want to publish in nature and sell because they're the best, like, but but you kinda like, but people did

Michael Eisen  1:05:59  
it. That's it's like you don't even they don't say it, because you don't have to say it. Right. Like, that's part of the problem is it's it's, it's so it's so kind of embedded in the system. Right? It's, it's just like that, like, it's not that they don't think it's that you don't even have to say it, because everybody just assumes,

Nick Jikomes  1:06:18  
yeah, absolutely. So people truly believe that, you know, the nature in the cell paper is a better indicator of your quality as a scientist. Now, one of the natural questions here is, is it? So is it in the sense that people get hired to better positions at top tier universities? If they have those publications? Or is that an illusion to some extent?

Michael Eisen  1:06:39  
Yeah, I mean, so again, I think it's really important to differentiate between different places here, there are parts of the world, in some countries in Europe, some countries in Asia, in particular, where it literally is true, like you get a score, and the score is based on where you publish, and each journal has a value attached to it. Sometimes it is literally this impact factor. So people's, like we've used this term before. So just to tell people what it is, the impact factor of a journal is a really easy thing to understand. It is the number of impact factor for a journal in 2020 is the number of times the papers published in 2018. And 2019, were cited in 2020. It's just like a direct ratio of its citations divided by divided by the number of papers of published so it's, it's Sorry, it's, it's like, it's a very crude way of characterizing the influence of the journal right, it's basically says, If you publish a lot of papers that everybody cites, you get a higher score, in some parts of the world, that score is directly use, if you get, you know, I forget, I tried to avoid knowing about impact factors, but it's hard to avoid, because sometimes they send you solicitations to their impact factor calculated to four decimal places. So let's say, you know, journal, like nature is probably in the 30s. And, and, you know, that means that the average paper is cited 30 times and I think I got the formula wrong, it's like, your x citations and the following two years or something, I don't even care,

Nick Jikomes  1:08:12  
but it's a number, it's related to how often you're gonna get cited.

Michael Eisen  1:08:16  
Right. And, and it, it is used just to calculate your value as a scientist, okay? Now, the US by and large, does not do that explicitly. So, you know, here here, the the journal influence is a little bit less is a little bit less, is less direct, but in some ways more insidious, because it isn't clear what you have to do to achieve the goals of the system. It is certainly the case that if you were to look at jobs at, say, the top 100 research universities, who's getting faculty jobs at those places, they are highly enriched for people who publish in these top Impact Factor journals. And not just science, nature and sell but also their spawn. There's sell and sell in nature particular have spawned sub sub journals for the papers rejected from the parent journal, and so forth. And so there's a there's a lot of fat going around. And the so it is certainly true that that there is a correlation. I think it is probably also true that all things being equal between two scientists, like if they literally had exactly the same profile that whatever people were looking for the same field the same stuff, that the person who got their paper into a high impact journal, compared to the person who did it would benefit. However, I think I think the extent to which other factors play a role in this is deeply is underappreciated in the sense that, you know, it is also true that The people are getting those jobs and those top 100 universities, trained, you know, did their postdocs at, you know, 20 places, right? Like, people who do their postdocs at Harvard, MIT, Columbia, Berkeley, University of Michigan University, Wisconsin, you know, with UCLA University of Washington, right? There's a list of like, you know, biggest well funded most prominent research universities and they are trainees tend to get the jobs at most other places even a bigger foot, you know, they are trainees get jobs at a bigger footprint than those universities have. So that has a huge influence, the field you work in as a huge influence, the your results themselves have a huge influence, right? Like, the fact that it got into a high impact journal isn't an accident, it reflects that the community as a whole likes that work, whether they're right or wrong, sort of doesn't matter. It's it's the there is a you know, the same standards are being applied, kind of uniform, not uniformly, but fairly regularly across the field. So I think it's certainly true that like, the type of science, that tends to get you a job at a, at a top university, and the type of pedigree, academic pedigree that gets you that job is the same science and pedigree that gets you papers in these high impact journals. So they it is one indicator of success within this system. The idea that, however, that if you got rid of that indicator that the system, would it fundamentally perpetuate as it is, is probably false, too, in the sense that like, you know, we, you know, if if, if universities were forced to stop looking at journals, as their, you know, in their faculty evaluates, let's just say we passed a, you know, a ballot initiative in California that said that, you could not look at journal titles, when you were looking evaluating candidates for faculty positions, we probably wouldn't change our hiring all that much. Because the same, the other things come in to just hold a little bit more sway. And since they're all highly correlated, you end up making the same decision. So that's both good and bad. To me, that means in the good sense, it means that we could get rid of the system without destroying science, like, we could we could save $10 billion a year, do something, right? We didn't talk about it. But let's like one of the big problems of that whole financial structure is that very few people on earth have access to all the scientific papers, right? Like, like, if only libraries are subscribing to these journals that only people affiliated with universities whose libraries are doing that, which usually means which universities in the developed world, basically, even within the US, like, like, like, for a while my mom was a teacher at George Washington University and DC, it's hardly a tiny wallflower place. But there are lots of papers she didn't have access to, because their library couldn't afford it. Whereas UCS good. So.

So a major problem with the current system is it cost $10 billion, and we don't even buy access for the whole world with that $10 billion, we buy access for just a few, tiny fraction of even scientists at elite universities, not to mention, like, basically, no teachers have access to, to research articles, few physicians do patients when they are trying to research a disease or something don't so it, there's a huge inefficiency in the industry, we could get rid of it tomorrow. Right? It's unnecessary for the functioning of science, we could get rid of it tomorrow, and we would be able to replace it with something better. That would be cheaper and fair, whatever. Um, some people like think that like have an idealistic view that you get rid of the journals, you've you've solved the problem. But it's a little bit hydro like, though, in the sense that like, journals are a problem journal access as a problem, journal structure is a problem. The inefficiencies of peer review are problem. But that, you know, they don't, they don't seem like it's both a good and a bad side that they don't immediately solve all the other problems that exist in science, because those are more deeply entrenched then, than just journals.

Nick Jikomes  1:14:11  
So I want to talk about open access journals and what you're doing at Ely, for example. So just to sum up part of this, it sounds like in theory, there's journals like the ones we've described, that are behind paywalls. There's also these things called open access journals that we'll talk about. And if all scientists tomorrow just decided we don't want to play within the certain the system we've been playing and we're all just going to publish to open access journals. Effectively nothing would change in terms of the science being produced. And you could get things out from behind paywalls but for sociological reasons, people do not choose to do that. Correct. So

Michael Eisen  1:14:50  
you know, this is that precisely that insight was what drove our us to start to open access in the first place which was the internet came along Honestly, up to that point, I hadn't even bothered to think about journal, like access. So it was just like I went to the, you know, I was a grad student at Harvard, I had access to everything I ever wanted, and a lot of stuff I've never thought about, didn't care about, right, like the university was paying for me to have that privilege. And so I, you know, I started my postdoc, you know, coincidentally right at the dawn of the modern internet. And I did it out out in California, at Stanford where, like, you know, internet was dominating the way that everybody was thinking. And it you know, it just sort of became quickly obvious to, you know, initially my advisor, Pat was sort of the first one to kind of really start railing about this, but just how dumb it was that journals were, were not taking advantage, you know, taking advantage of this technology to make their contents free, like the The, the, you know, in a very literal sense, the internet was invented to help scientists communicate, right? Like those early nodes, you've seen, sometimes pictures of the early wiring of the ARPANET. That whole purpose was to help scientists now, they were defense scientists working on defense science thing, so it wasn't meant to be shared with the public, but it was meant to help them communicate with each other. And a lot of the early technological development in the internet was explicitly about university to university communication to allow scientists to share information with each other. So for the actual public internet to come along, and have one of the first things that happened on it was, you know, the, the, the, the organizations who had been entrusted with control over the scientific literature or publishers, to use the internet to, to perpetuate the access problems of the previous technology was just was just nuts. And so, you know, after you know, we our initial view was that we should just just tear the whole system down, don't even try to think of a new way to fund it just start from scratch. And, you know, ironically, the physicists had already been doing this. physic physics since the early 90s, before the modern internet came along, had this thing called the, you know, the proof, they had a preprint server called archive, it actually originally was called xx xx.la nl.gov. That was their first website, which obviously, as the true modern internet came along how to change its name, but the, the X in their original right, it's kind of a funny story, they visit the famous physicist who invented it, Paul Ginsberg, you know, physicists had been mailing copies of their papers to each other, their preprints, not their published journal papers, they had been mailing fifth papers to each other, not to everybody, but to all the, you know, colleagues at prominent places that they cared about. You can talk about the unfairness of that system if you want to, but like they would mail papers to people they thought would be interested in reading them, or that they wanted to read them. And every physicist would have like a, you know, cabinet of papers that their colleagues mailed to them. That was basically their library of papers that was just an informal network of shared preprints that your colleagues were sharing. So along come the early internet, the pre, the pre public internet to the one that mostly we had access to it universities in the late 80s, and 90s. And they just said, Well, why why do this by mailing physical copies, you have an electronic copy on your computer, just mail it to your friends. Okay, so they started doing that. And then someone said, Why mail it to your friends just posted in this common server. So Paul Ginsburg, who was at Los Alamos at the time, just created a site, which again, there was already a www.la nl.gov. So this was xx, xx, le and algo lexicographically. Next, and, and it just, they just did it, and they continue to do it, the physicists just post their papers on, on archive To this day, and you can get them for free and read them. And it's totally fine. They have not collapsed as the site. So we felt in biomedicine, which is, which is much bigger, much more expensive journals and so forth, we should just do the same. And we,

you know, convinced the NIH that they should do it. The big funding agency was got lots of money to do these kind of things, and has a library, the National Library of Medicine whose mission is to make scientific knowledge available to the public. You can if you happen to be in Bethesda, Maryland, which is where I grew up, so I know about this, you can just go into the National Library of Medicine and get any journal you ever want, read it for free and whatever. So they have that mission. We felt it was just an obvious continuation of that mission to do that electronically. The NIH put forward a proposal to do that, and it was killed by publishers Scientific societies and and published this is in the late 1990s. How did they kill it? The the standard way you kill things, if you are in a position of power, you get your Congress person to say that they will not fund this. Right. So the NIH who is obviously dependent on annual budgets from Congress was told you can't do that we will strip your funding for that for that. Not the whole funding, but we will not fund this activity, as you have to do them, presumably,

Nick Jikomes  1:20:29  
the big publishers used the money that they have to pay lobbyists to,

Michael Eisen  1:20:34  
presumably, they did. There's a paper trail of this. So they paid lobbyists, and they lobbied themselves. And, you know, part of the problem is, one of the reasons we're kind of stuck in this world is that the scientific societies are viewed by Congress as representing science, and AI, in most cases, they do a good job, like they're good at advocating for funding, they're very strong on promoting the use of science and public policy. It's not like they, they can be very good and useful voices for for science. Unfortunately, their revenue stream is built on subscription journals. And so when this policy got put forward, the scientific societies who had a lot of credibility in Congress, speaking for science went to them and said, No, this is gonna, this is terrible, it will kill us. Right. And it was implied that killing scientific societies would be bad for science, which is, I don't think the case but but so they were they were effective in convincing Congress that, that this proposal was was would be disastrous, and so it didn't happen and what instead,

you know, instead, we tried to figure out a way to accomplish the goal of, of removing paywalls to scientific journals that didn't, didn't require government action. And so what we did instead was kind of, not just us, there were other published, you know, other, you know, startups and people in the industry kind of latched on to this model, which is to say that, let's stop treating scientific knowledge as something you own. Right? Right. Now, the reason journals can sell papers is that the authors assign them copyright. they assign them copyright as a part of the process of publishing your paper, I used to do that, too, because I want to care about copyright, I'm not gonna sell my work, I might as well give it to the journal, it's fine, right? That's true before the internet comes along, because I had no use for the copyright. Nobody else did, either, like no one else was going to like, produce a version of nature that they would print themselves and mailed everybody, it just didn't make sense. Internet comes along, traditions continued, you're still assigning copyright to those publishers. But now they're using it to restrict who can get access to their content, which is obviously what copyright, it's used for the internet. And so our view was, well, let's let's not have a model that's based on journals making money or covering their costs, and whatever, by taking ownership of the content and selling it back to interested parties. Let's have a model where you just did whatever expenses, the journal incurs in the process of producing an article, let's just pay them those expenses and a profit, whatever we think the sensible business is, and, and in exchange for paying them upfront, they don't own the content, the public baths. So it's a different service. It's transforming it from a content model into a service model. And the obvious benefit is that there's no longer any paywalls that the scientific literature could be free. And lots of arguments from us. But also lots of studies from people have said, In addition, it's a more it's a more efficient market, it'll be cheaper, whether that's true or not, it sort of doesn't matter. It like even if you just took that $10 billion we spent today and just continued to pay all the publishers but you did it under open access, a different agreement, a different way of transferring the funds, they would keep all their money, but all the content would be free. It's such an it's such a frustrating inefficiency in the world that that there is a path here that where everybody gets what they want, right, the publishers can still get paid, the public can get access to the content authors can still publish in journals. And just like our, our inability to bridge the kind of incentives that exist in each of those communities, to get people to actually do the right thing collectively. But that's what open access is open access is a model where publishers say we'll publish your work it costs us some money to we have to have staff who deal with peer review and find the peer reviewers and convert the papers. It's not a it's it's it's a you know, it's a very heavy touch kind of process. So it costs some money, but you pay us you pay us when you publish a paper. And then we don't own it, you do it for the public does. And you know, in exchange for this work we're doing, all we ask is that you license your work to the public domain and let people let people use it for free. And so we started Initially, I started a publisher called Bloss Public Library of Science, which was, which still exists in this big publisher in the field. And that was instantiating, this open access model along with now, now a publisher that's been folded into the Springer universe called biomed. Central, which was also kind of trying to develop and promote this open access model. He liked the journal I run now is another open access journal loose, you know, he's trying to do other things like, like, like, we're trying to go to the next step of reform publishing, which is rethinking how peer review happens. But

Nick Jikomes  1:25:50  
yeah, I did want to talk about that. So the actual peer review process, what are, if any, the major problems with how that is typically conducted today, in terms of, like, conflicts of interest that different scientists might have?

Michael Eisen  1:26:05  
Yeah, so that's certainly a problem. I mean, I think I think, you know, it's a tricky business, if you think about it, that, like, the ideal person to review a paper is someone who's doing exactly the same experiments, right. And so, um, you know, I, you know, a lot of a lot of a lot of scientists, I'd say, the vast majority of scientists are able to dissociate, they're, you know, largely, everybody's got hidden biases in these things. So I don't want to pretend everybody, anybody can ever be unbiased this way, but like, I think the vast majority of scientists are able to, you know, largely act is fair, and reviewers of even their competitors work because, you know, it's just like, you don't go into this business, mostly to be to be a jerk, so and so. So, you know, I would say that by and large, that part of the process is okay, like that people review article, things fairly, I obviously, we can do a better job, because there's always conflicts of interest. And they're not always clear. And part of that there is this weird, you know, this lack of transparency in the system is problematic. And it's hard, it's tricky thing to overcome, because forcing people to be to identify themselves, for example, in peer review, it tends to exclude people who are in marginalized positions within the field, like the, the only people who feel really comfortable trashing their colleagues work are people who have tenured positions, right. So so there's a problem. But I would say that that actually isn't the big problem here. Like, I think peer review is biased. But the real problem is, and like, given that there are aspects of peer review that are biased, and I'd say even more, more problematic, that, that even when everybody is operating in completely in good faith, even when the right people in terms of expertise, and knowledge and breadth of thinking are chosen to peer review paper, even when they do that work diligently, they read the paper carefully. They don't just do it in the last two hours of some flight where they're about to fall asleep, which is often what happened, they do it diligently and they do it fairly and honestly, and and and constructively. their decision, their judgment is being reduced to a binary assessment of the value of the work that lives without work forever. both good and bad, right. So if it I would say the biggest problem in this peer review is that it happens only once at one fixed point in time by by a small prescribed group of people. And and that it isn't a problem that they do it. I think, right? peer review is very valuable, the peer reviewer spend a lot of time doing things, that's a good thing. It's a good thing when someone reads their colleagues work and comments on the problem is that we greatly greatly greatly greatly overvalue that, that that

that decision or in particular the way the decision is rendered such that a paper that was published in Nature and six months later is you know is generally thought by the community to be wrong. Not not bad not that they did the experiments bad or something just that the conclusions are wrong or that it wasn't as interesting as they thought it was or whatever. like everything's on the up and up nobody cheated. Nobody missed something. It's just it's just quite often, things that seemed really important are quickly realized to be unimportant. That paper remains a nature paper forever because the the assessment at that one moment was that it belonged in nature. Conversely, you know, you see this all the time when people like make really big discoveries or win Nobel Prizes or something, like, if you look at CRISPR, right, like the coolest biological innovation in the last decade, like it, that the history of CRISPR is littered with papers whose significance was missed by the people reviewing them. And, and so those papers published in, you know, yogurt journals, and, like literally like, like, a lot of that work came out of Dannon. And, you know, like, the a lot of really, really, really important science, its significance, it's not even that it was missed, it's just, it wasn't apparent at the time. And so, it forever those papers remain obscure. And, you know, if you look like CRISPR is a good example like, like, you know, um, you know, I have nothing but immense amount of respect for for my colleague, Jennifer Doudna, at Berkeley and manual serpenti. A with whom she won the Nobel Prize for this. But there were, you know, there were other scientists, whose work kind of suffered in peer review, that probably would have been credited with discovering CRISPR had the world gone in different publishing function different way. And that's bad. I mean, it's not no discredit to the people who were, whose work has been instrumental in getting this to the forefront. But I do think that, that we just, it's this it's this historical legacy of the fact that you had to make a decision, if you were going to print a journal, you had to make a decision at the time you were choosing to print the Journal of what's in and what's out. Right, that was a sensible decision to make. We continue to render it now as as a decision long after it's stopped making sense. I mean, it's, it's, there's literally no reason for us to attach a single permanent mark on every paper and not revisit it. Right? We don't revisit it. We don't let other people participate. Right? Like, why if a paper is so important, surely, if a paper really is so important that it warrants like going into, like being one of the handful of papers that appears in the elite journals every year, surely other people could review it, too. And see if they come up with the same conclusion, right? Like this. It's this. And again, it's not the journals fault, it's our fault for only valuing that, right? If I, if I started publishing a list of papers that were in, published in Nature, and shown to be wrong, a year later, no one would care, because the authors of that paper, have a nature paper, and it's on their CV, and the fact that Mike's list of bad nature papers, includes their paper, either they're not going to care. Similarly, if I like said, Hey, here's 20 papers from 2015 that were missed, and you should pay attention to them, I might make a good Twitter stream, but it would take won't show up on people's CVS and it won't affect their career path in the right way. So it's, it's that I would say more than anything that that like this, like fact that we only do peer review, once we do it on a small scale it we don't come back and revisit it, which not like an important paper that's published in Nature. Right, it's probably read by 1000s of people. A large fraction of those people spend enough time thinking about it, that they have used the paper that would be worth sharing with the community, we have no way of capturing that, we have no way of rendering that you see, you know, you see quite sometimes a paper has been published for minutes, and you realize people realize it's wrong, because they find fatal flaws. And if that were missed in peer review, it doesn't matter. And it doesn't matter because the the all the value is placed on the journal title. And that's the kind of core that's the core problem that that needs to be fixed in my mind.

Nick Jikomes  1:34:05  
How often are the peer reviews published alongside with papers that get accepted

Michael Eisen  1:34:14  
that's growing so he life so he life which is the journal I am editor and chief of now, um, you know, it was founded on this idea of being much more transparent about peer review by Randy schekman, who was the person who founded it kind of really was, you know, push this notion that that, you know, we're not on the internet, like it's not wasting ink space to publish the peer review. So the peer reviews in life are published, but only for papers that are ultimately accepted. And, you know, it's, it's, um, they are available if you go to any life paper, you can actually see the pair of us and actually students find them very valuable because it's kind of an insight into the process. They're not really used that much though. Like that. Like, again, it's like kind of an irony that that, you know, he life which was founded as a journal to try to avoid using journal titles as a way of indicating things, it's now become a journal title that people use to indicate things. And I think part of the way you can tell that is that nobody really cares what the reviewer says, like, once it's any life, either because they feel like they can judge the paper themselves or because they don't, they just all they care about is that it was a neat life paper, by and large, those aren't looked at, this is something we're trying to change now. Like, I think, you know, part of the part of the problem is there is not a lot of visibility attached to peer review. And also part of the problem is this usually have usually published after the papers already been published, when it serves much more of a, I think, an important role, but it serves a role more role and transparency than in than in. And, you know, they're not written to be used in a way. So we've we've kind of implemented a new policy at life recently, in which, first of all, I should, we should come back to the preprint thing. Finally, biologists are using preprints took 20 years, but the system that was killed by publishers has now started to take off. And I would say, you know, we're we're definitely on the right part of the trajectory in terms of preprint adoption, that, that it's easy to imagine a world where all papers are posted as preprints. First,

Nick Jikomes  1:36:22  
even though we already talked about it somewhat, can you just define for those who are unfamiliar, what is a preprint?

Michael Eisen  1:36:27  
So preprint. So, so I'm a preprint is that is a paper that the authors share prior to peer review, as the easiest way to think about it, that it's a manuscript that the authors have written that they feel is worthy of sharing. And it's that, you know, it's something that they post there is now a place, you can post those many places, but no field specific place, but if you're a biologist working in sort of any area of biology, you can post your paper on bio archive. And, and, you know, it's a very low touch process, I have a PDF of my paper, I submit it, they look at it, basically to make sure that it is legitimate work of science, that it's not just an opinion piece, it's not porn, it's not, you know, it's not a prank, really crank science, and it's not dangerous, like, you know, it's like you can't publish a recipe to synthesize COVID in your basement or something. And so, so I can post something there, and usually within 24 hours, it's up, and then everybody can access it, it's completely free, you can download the contents and use them. But they're on peer reviewed. And so, you know, authors, you're basically either going on your own judgement of the work or you're relying on the author's as a credibility to know that it's at least it's it's good. So you know, there's a little bit of a of a buyer beware kind of attitude around preprints. Although I should also say that same buyer beware attitude should exist around peer reviewed papers, because peer review is only three people. It's not, it's not perfection, so. So preprints exist, a lot of scientists are posting preprints, they're visible, like, you know, there's no like you can anybody can look at them, there are a lot of Twitter activity around preprints. A lot of scientists are now primarily using preprints as their means of communication, of reading papers. And certainly we post preprints of everything we publish, and a lot of scientists are doing. So what we're trying to do now in life is to just actually start reviewing preprints, to say, Alright, we saw, like, if preprints take off, we solved the access problem. We solved it, not by converting existing publishers to an open access model, but by just having everybody share the work for free on a preprint server. Okay, so that's one thing. The second thing is it if you start to attach reviews to preprints instead of publishing papers in your journal, and claiming like that, that's the final act. Now, now you can get past this problem of peer review being a static object. So if I if he like peer reviews of paper and on bio archive, and posts are the results of our assessments, and some other journal or some other scientist wants to come along and review that paper, they can do so too. And the community can make their own judgment of which of which of these assessments is more valid, or they can the assessments can even talk to each other at some level that we can start to develop a little bit more of a robust culture of assessment around preprints not without its potential problems, like we don't want, you know, science to become any more of a popularity contest. And it already is, it's, you know, you know, the downside of public commenting on the internet is well, well known to you to the universe. So it needs some level of, you know, it needs to be run by grownups or at least bye bye run by responsible kids is probably the best way to do it. And, you know, so it's it's not it's not a trivial system to create and to think about, but it's definitely to be, I think would be better. And so you know, because journals by, by and large are where peer review takes place, because journals have an incentive to a financial incentive to just keep publishing either as open access journals, because they're making their money that way or as subscription journals, there hasn't been as much activity around reviewing preprint says there, I think there should have been, so we're trying to take advantage of our position to do that.

Nick Jikomes  1:40:37  
So one of the big lessons of the internet for me in my lifetime has been that when someone in some position of power or some power structure locks the front door to something, technology will eventually be used to get in through the back door or the side door. And so you told the story earlier of how some of the big publishers use lobbying to lock the front door, so to speak, and maintain their dominance. One of the things I want to talk about there related to that is this new thing called Sai hub. So what is Sai hub? How does it work? And how have journals and universities been responding to it?

Michael Eisen  1:41:15  
Yeah, so sigh hub, um, it's really simple. It's a it's a bunch of computer programmers who got frustrated with their inability to access the scientific literature. You know, they were in Eastern Europe, where were these access problems are huge, right? I think, you know, the countries and the research universities in Eastern Europe are probably in the worst position of all because they're not they're not small enough to be there, you know, they're not poor, the countries are poor enough in the universities aren't small enough to qualify for free subscriptions, right, so they don't get free access to journals. Like if you're in a really poor country, where the publishers have no hope of ever generating revenue, they unlock their firewalls because it cost them nothing, or the paywall. But, you know, if you're in Kazakhstan, or Ukraine or, you know, wherever these people were, not always entirely, they're not always entirely forthcoming about this, which is good for them. Um, you know, you, you and your want to do science, like you just don't have access to the stuff like it's just almost impossible, right? It's, it's impossible to be a functioning scientist in a country or university situation where you don't have access to the literature, it's like literally cripples not just your ability to do science, but but by definition to cripple science, right? Like, like, if you believe in science, as a thing that makes society go better, you have to believe that the scientists everywhere in the world who are not able to get access to the scientific literature, and therefore doing either worse, or less effective or slower science that that's hurting us all. So people behind Sai hub faced this problem and being adept at using the internet to obtain content obtained content they got by unclear means the access to the universities that had access to all of this content and downloaded it not entirely clear how or what means they use to obtain it, it's presumably not legal but but that's between them and their carts. And, and, you know, made it available, you can go to Sai hub, which is unfortunately at an ever moving URL, because the forces of darkness catch up to them. But if you go to Sai hub, and you have a you found a paper that you want to read, it's to visit a great for searching the literature, it's great for if I have if I find a paper through some other means Google or PubMed or something that I can't get access to. I can just go to Sai hub and get on all admit like I've used it up all the time for doing science when Elsevier and my access to Elsevier content was blocked by the access screen. So it's, it's great. It's probably not a long term solution because the publishers, you know, when it just existed on the margins of the Internet, and nobody would go to it because it was in Cyrillic, and they were afraid that they were going to get viruses or whatever. It didn't really cause a threat to the publishers. But I think one of the reasons why the Elsevier thing with UC was less of a big deal for UC scientists was that we could get content on sign up. And and I think the publishers have started to realize that it's a threat and they have used the fact that some of his practices are do not conform to current copyright law to To get it blocked, you know, they have two strategies, one to prevent them from getting access to the content, they need to perpetuate it, which they seem to actually do have done effectively recently. It's it's harder and harder to get recent content on site out because they've sort of made it more difficult. And seconds, the second they've blocked. Like there are some countries where you just can't get like the internet as they blocked access to sign up via my internet block. So yeah,

Nick Jikomes  1:45:26  
so you mentioned that you've use I have other scientists have you side hub? Is this common? And if someone out there just a regular citizen is using Sai hub? Are they doing anything legally wrong? They can get in trouble for

Michael Eisen  1:45:39  
I'm not a lawyer. So I'm not going to give anybody legal advice. But I think I think the answer is, you know, it's complicated. let you know. So I have talked to lawyers who will, you know, can come up with a next legal framework for why using Sai hub is illegal. But you know, I think we know like, you don't have to look very far on the internet to know if you get pocket like, if you get access to pirated content you are participating in in piracy. So I think it is probably true that it is illegal to both it's certainly illegal to take the content and make it available and it's presumably illegal to get that content, although there are some exceptions under fair use that some lawyers will argue apply here. Also, it's not exactly clear who has to initiate that legal action, right? Like there has to be a copyright complaint can't just be like the police can't just arrest you for using someone else's copyrighted content unless they have a complaint that you're using illegal. I don't think I don't know. I'm not a lawyer. I don't take any of this advice as legal advice. But the the like look, this is the problem with SEO is it exists in its even its best manifestation on an on a dubious legal footing. And, and and I think it is on 100% solid moral footing. I think there is absolutely nothing wrong with sight of the publishing industry is what's wrong and Sai hub is. It's like trying to fix a major, not just a theoretical problem, it's an actual problem for the world that science is being impeded by journal paywalls. So I think sign up, I love it. Completely righteous thing to do. But I think it'd be a mistake for us in the scientific community to think we can rely on it forever. Because it it's it's legal status is such that it is, you know, it's it's just one, you know, one effective legal maneuver away from being inaccessible to everybody. And so I think, I hope that it has not created a sense of complacency among scientists that this backdoor will always be there for us to get access to content. And so, so I think it is important that the scientific community recognize that anybody who's used SEO, should just demand that we should take advantage of what the internet was invented for and make Sai hub legal. Right? Like Sai hub should exist, there should be a place where you can go to download every paper ever written for free. It, the fact that it is illegal, is is our fault. It's a pathology of science, it should exist. And it could exist it like like, since since the internet came along, science publishing has received revenues of in excess of $200 billion. Right? That's just publishing that's not doing the science. That's just $200 billion to publish articles where almost all the labor was done for free by the community. It's like, couldn't Can you imagine that we could have invented and paid for and sustained a better system. Think about if we just invested that $200 billion in an endowment for science publishing, we'd be fine. We'd have like huge revenue every year that would cover the costs of, of everything we ever have to do in science publishing. So it's really sy up should exist, it should be immortalized as a legal premium. That's that really that's my dream, even honestly, like day one, when I was involved in this publishing stuff, like literally like 1997. We had a meeting with people at the Stanford libraries where we asked for caught we asked for some of the content they had on there. The they had a publishing operation, internet publishing operation within the Stanford libraries and we wanted to use their content in our research, not publish it, just use it in our research. They said no, and that kind of triggered a lot of our work at least an open access. Almost the first thing we thought about doing was kind of creating kind of like Napster for scientific papers, right? It's just that, like living in the US, we were afraid of the consequences of setting up an illegal server. And so we didn't do it, but we should have because it would have made science better would have been worth would have been worth the legal peril. I think.

Nick Jikomes  1:50:28  
So in your own lab, do you have a policy of only submitting work to certain types of journals and is that

Michael Eisen  1:50:35  
since since the lab started, we get its policy, everybody knows when they joined the lab, we only submit open access journals since since the absolute beginning. You know, I will say it was not the case when I was a postdoc in graduate students. So this only happened after open access became a thing. But you know, my lab started basically just when we were starting Public Library Science, so so every paper we've ever published has been, you know, where we, you know, as you know, like, sometimes you're on papers where you don't make the decision about where it's published, because you wasn't really your paper, you were just helping someone else. you're collaborating. So every paper where I've been the senior author and has come from my lab has been in an open access journal or for made open access by some means. And so, yeah, so that's just we do that. It's not, it's not unheard of, but it's not common, you know, a lot of people publish an open access journals, but they do so in a more opportunistic, you know, just like paper by paper away. Right, like they it you know, I think most people appreciate that it's valuable, but sometimes, you know, I think our lab makes the choice that we would rather make everything we publish open, then publish in nature. And I understand, kind of understand why other people don't make calculus. So it's, it's, it's, you know, it's so I would say, disappointingly, few people have that have that stance, more and more people have the will preprint everything stance, which, you know, honestly accomplishes the same goal, right, preprints accomplish the same goal, which is to make the content freely available, they're not as disruptive in some sense of the economics of publishing, but in the long run, they could be more disruptive like I've ever been post preprints, the existing publishing industry cannot continue in its current form, because no library, we that's acting in a rational manner is going to subscribe to by content that can be obtained for free, and so legally obtained for free. So it's, I think, you know, I think the current, the current manifestation of people saying, Oh, I'm just gonna post preprints of everything, it accomplishes the goal, it it, you know, there are some questions about whether that system will continue, because obviously, publishers could say, we won't review your content and publish it, if you post a preprint. They do in some, some circumstances, for now, public opinion is very strongly against that. And I think that they would have trouble implementing such a policy, but if their businesses really start to get threatened, which at the moment, they're not, you know, there aren't enough preprints that libraries can say, you know, what, we're just not subscribing to your channel. So but it's, I've always been afraid that that, that that will happen. And so one of the reasons that I think it's so important to get pure have you built up around preprints is, you know, there needs to be a different poll, than just journals, we need to make it so that journals are not the thing, the only thing that people can, can, can aspire to do to have their work assess. So and that's the ideal world. To me, the ideal world is everybody posts, a preprint. Peer Review happens around preprints. And there are no germs.

Nick Jikomes  1:53:49  
Well, Mike, thank you for your time. I think we need to wrap up here soon. I think this is this is an important area, for many reasons. But you know, the vast majority of people that are not scientists, I think, have very little or no idea how this world works. I even think a lot of people in science, at least at the level of graduate students don't

Michael Eisen  1:54:08  
actually I think most like most scientists just focus on their work. They don't think about the they don't think about like, and I'm sure this is true for me and all sorts of things like, you know, I like the you know, the part of the problem is that, that this is such a like, integral part of the system of science that, that you don't question it so so like the vast majority of scientists think about journals only in terms of what it does for their careers. And they don't think about the system of publishing, they don't think about the economics, they don't think about how it could be done better. It's just the way it the way it goes. So yeah, so I agree with you completely. It's a topic that you know, like all this and other topics that are like about the functioning of science, most scientists don't spend a lot of time thinking about so

Nick Jikomes  1:54:59  
any final thoughts you want to leave people with related to this whole area

Michael Eisen  1:55:05  
only just to pay attention to what happens over the next couple years around preprints. Because I think that this is the critical moment that, you know, it's it's possible that, you know, three years from now, the momentum for everybody posting preprints will have just accelerated and at some point, we'll reach a tipping point where your work just isn't getting any attention if it's not posted as a preprint. And when that happens, or hopefully before that happens, there will also develop a culture of, of commenting around preprints and peer review of preprints that is functioning robust and better than the existing one. And at that point in time, journals, I mean, they may not cease to exist, but they will cease to hold their sway over the scientific community. So you know, if you're interested in this topic, I think the place to look is around what happens with preprints and preprint review and, and just in general, whether the, the current trajectory for the increasing use of and an even reliance on preprints continues because if it does, then things things are really going to change for the better finally.

Nick Jikomes  1:56:13  
All right, well, Dr. Michael eizan, thank you for your time and good day. Thanks, you, too.