00:11
So today is a special episode because we're part of this
like ゆる合同イベント with other scientific podcasts. I know I didn't really like tell you
anything about it, so this is the first time you, Masako, are hearing about this, but basically
several people that I found on Twitter are trying to do like a common theme episode for a specific
week of the month. So the month of May, they have decided that they're going to talk about
論文論文. And that was just like I was looking at it just when I was sharing the links about
this Radiolab episode to you, which is about 論文論文 to a certain extent. So I figured,
hey, you know, it's the perfect timing. I wanted to talk about this anyway. So
yeah, with that, without further ado, so maybe I'll put the link to this episode in a show notes
because I think, yeah, like it's like, you know, they speak really fast English, but I think it's
very interesting topic. And it's very well done podcast. Radiolab has been one of my favorite
podcasts of all time. So it's definitely worth a listen. It's an episode titled Library of
Alexandra. So the episode is about Alexandra Elbakian. And to most people, it's like,
who the hell is that? And it was the same for me. But I think most of us scientists have seen
the website. Have you? No, actually, no. Really? Wow. Okay. It's I've used this so many times in
my life, especially if I'm in between schools, because I took a gap year between undergrad
and grad school. And I was hired as a technician, but I wasn't a student. So I didn't have like the
same access as like the university students did or something. But basically, SciHub, for those of
you who don't know, is a website where if you have, let's say you have an article you want to
read about, and you don't have an access to it, it's behind a paywall. And what you would do is
to go type on your Google search SciHub.com and put the title or DOI of the article that you want
to read. And boom, it's there. And the PDF is there. And it downloads. For the longest time,
I never understood how it worked. It was like magic. And through this episode, I figured out
03:06
actually the mechanism of how this works, which you can figure it out if you listen to it. But I
thought it was quite, quite, it's a very simple conception, right? But very, like, helpful. And it
helped not just me, but like, obviously millions of other people, like, they have some stats, like,
since 2011, SciHub came online, 88 million publications have become available in on this
website. And like, at the peak, peak usage time, like, 90% of paper ever published was on SciFi,
SciHub, which is amazing. All of this, because if you're not part of the university or research
institution, these are behind, like, subscriptions, usually, which costs like, you know, hundreds of
dollars a month. And like, you would think that it's rightfully so because, you know, these are
brand new knowledge, that's like, high quality knowledge, supposedly, right? But what, I guess,
as researchers, we don't really think about is who is getting these money, you know, because as
researchers, we're all getting paid by the publishers, which is bizarre to think about,
right? Because if you are a novelist, like, your publisher will pay you, the writer, in advance,
hey, here's the amount, like, keiaku, my keiaku-kin, right, before the royalty kicks in,
hopefully, you become successful, and then the royalty comes in later. But you get the money
before you write, usually, and usually after you finish writing as well. Whereas like, you know,
none of that is happening in a researcher's, like, we labor our lives away to produce this research,
we get paid by university, but not by publishers. Actually, more often than not, we have to pay
the publishers to let us publish, right? And usually, it's not from like, our pocket money,
it's usually coming from university's money or grant money or something. But it's still like,
it's the producer of knowledge, right? That's paying the circulator of the knowledge that is
a publisher. So it's a bizarre dynamic already. But so like, the big four publishers, like Elsevier,
Sprinter Nature, Wiley, and Taylor & Francis, right? Like, they don't fund the research,
they don't pay the peer reviewers, because it's like, considered our civic duty as researchers
to do this for free. So they pay basically the editors, right? Who is the final judgment call
for what gets published. And then there's like a whole slew of people who are doing that corset
06:00
type stuff, right? Like they just like formatting people. So like, their main function as a publisher
is the quality control aspect of, you know, what our landscape of knowledge looks like.
But like, they get so much money, like the markup for this operation is huge.
Essentially, just being a very good, hopefully, a curator to the knowledge, right? Which is insane
to think about. They're not the ones coming up with the knowledge. They're the one who are
deciding what knowledge is worth circulating. And that's what they get paid on. I mean, sure,
somebody has to do that work, and somebody should get paid doing that. But like, it's a lot of money
for compared to the amount of sort of free labor that researchers do, right?
We sometimes even have to pay for like over $3,000 just to publish one paper.
And that's a lot of money. Yeah.
Right. And like, you know, if you're a successful researcher, if you're having good, you know,
you will be... Yeah, big grants, then that might be, I don't know, tiny, but yeah.
You need to publish multiple times a year. You can't just like publish once every few years,
right? Because your grant money is tied to a certain extent on how productive you are,
and number of publication, even though it's not the end all and be all, and it shouldn't be,
it is one of the biggest measures of productivity.
Yeah, if you don't have a lot of grants, then it's hard to publish papers. But if you don't
publish papers, then you won't get grants. It's a vicious cycle.
So like, it's basically like kind of a biggest like industry
dark side that academia has, I think, this like a weird publication system. So,
SciHub basically disrupts that cycle, right? Like it provides completely from the left field
alternative solutions for decimating the knowledge without going through the publishers.
And I thought, I think the idea was great, right? And I have benefited from them personally as well.
But right now, so I didn't realize that, you know, because it's been a while since I had to
use SciHub last time, I didn't realize that they're under the legal battle right now,
and that SciHub basically stopped updating. And it's still extremely useful, you know,
09:04
like 88 million publication is like more than anyone can read in a lifetime.
But like, in turn, if your field is particularly, you know, two years not old paper is used this
kind of field, then like, you know, maybe this is not a very good deal that they have right now.
But yeah, it's just a very interesting episode, which I highly recommend anyone to
who are remotely interested, intrigued by this conversation to go and listen to it.
She just did it because it was fun. You know, she just did it because it was something that's
like, what's the difference me sending a paper to a friend versus to the entire people in the world.
So I just thought that was very interesting combination of things. But that got me
thinking about just publishing in academic publishing, I guess, in general, and in
particular, like, why we don't get paid to do reviewing? Because yes, editors have the final
call of, you know, like, reviewers don't get to decide whether this gets published or not,
in the end, it's only a recommendation. But it can be a lot of time that eats up in your day,
like I've helped my boss several times, when he had a lot to review that week or something,
he's like, can you just like, read and give me some summaries or like, flag something that looks
fishy. And I've done that before. But I mean, I don't really have a ton of extensive experience
in reviewing other people's article. So yeah, I don't know, like, do you how often do you review
it? Do you do you think it's too much? Or like,
I think it depends on the, I don't know, sometimes I have to review several papers a week. And that's
a lot of, you know, time that I need to spend on other people's science, the, you know, we don't
do science to, to make money. But so I don't... Bad choice.
Yeah, bad choice. Bad decisions that we made. So, and still it's, we can learn so much from
reading papers that were, are not published yet. So it's, it's, yeah, I don't, yeah, I would
like to review papers, but still it's a lot of time and I have to, we have to pay attention to
the details of the methods and, and doing like ad hoc editor is... Oh, so you're an editor as well?
12:07
Not very often, but I'm sometimes asked to do an editor work. And we usually start from
finding reviewers who are available, who have the expertise to review that paper.
So some, well, many actually don't really respond to our requests. And some do, some say,
oh no, I don't have time right now. And find others or... And then finally, after so many,
sending so many emails, we will find someone, several, two or three reviewers who can review
the paper. And then after that, some people are right on time, like, you know, getting the
reviews back. But sometimes they don't. They don't send us the reviews back on time.
Like, can you bail out on reviewing? Like, can you just, can you agree to review and then like,
not do it? Has that ever happened? That has never happened to me before. I'm not sure.
But if, if a reviewer doesn't, you know, send the review back for so long, then probably the
person will be... The paper is stuck in a pipeline, right?
Yeah, yeah. So probably they'll find another reviewer. Yeah. So even doing an editor work,
yeah, will take up a lot of time from you, from doing actual science.
I mean, yeah, like that's, that's what it sounds like. I mean, that's why I like,
it's part of the reason why I don't think I want to become a PI. I mean, I never wanted to,
but one of the reasons is like watching my PI do a lot of admin work, right? Like,
they have to do a lot. My boss has to do a lot of non-science work.
A lot, yeah.
I think he misses being in a lab, but you know, I think he's just become too old and too important
in the field to get away from that kind of duty. So he does it, right? And then in a way, you know,
he, his career was advanced thanks to all these other people who are doing a lot of these civic
duty type of work. So maybe it's his turn and he's okay with it. But I definitely don't think,
like, it's okay if I don't become a professor because I don't, I don't want to do this.
But yeah, it's, it's a lot of work and, and it's a lot of work that we do for free.
And because, because we think that science is important, we think that,
well, I guess we basically agreed to work in this system where publishers and editors get paid a lot
15:07
of money while bulk of the heavy lifting of the work still, you know, hinges on scientists,
right? It's us who do the research. It's us who do the reviewing and, you know, editors can
make a final call and they can arrange a whole bunch of other things. You know, like you said,
you have to arrange reviewers. I'm sure you have to do a lot of other things too. But like, these
are still kind of work where it's basically hinging. It's counting on our goodwill.
Exactly. Yeah. So I feel like sometimes, you know, I, I like doing science and I,
yeah, I don't mind reviewing papers, but still, you know, I feel like I'm a
negishotta kamo. Publishers. We say, okay, I can do that. I can do that. And then, you know,
we spend time and then they will make the profits. We don't get much.
But like, but like the counter argument to that, right, is okay, then how else are you going to
like do a quality control of the knowledge that's going to be publicly available, right? Because we
do need publishers to sort of, you know, do the production and do the, do the other, other
publishers job. And main thing is quality control, right? Because not every research that's ever been
done out there is worth publishing or is not quite ready. And, and somebody needs to make a judgment
call. So that system does need to stay. But what is a system where like scientists are incentivized
to do a good job doing reviewing, right? So that, so that the power is in the scientist's hand,
rather than in the publisher's hand, right? Because right now, it doesn't matter whether
you spend a lot of time reviewing or you take 10 minutes to review, like it's, there's no sort of
real incentive for scientists at the moment, at the minute, it's like really hinging on our goodwill
to do a good job at reviewing, right? And that's sort of one of the reasons why like we are,
we still rely on that publisher functions to do that. So I was just wondering, like, you know,
18:03
what is the alternative? Because this system sucks. Yeah. But do I have a solution?
Nothing that comes to me immediately, you know?
Yeah. I guess one way that people are trying to overcome that is like by doing open peer reviews
and stuff, right? But still open peer review, yeah. But still open peer review doesn't really
counteract the systematic problem that we have with the, that, you know, vicious cycle thing.
That may actually...
Because, you know, someone told me, you know, she doesn't like the
open review process thing, because
she doesn't favor that person, you know? You can show that, you know, I'm on your side, you know?
Right, you can kind of copy too, right? Yeah, yeah, yeah. So that can happen if you expose your name.
Yeah. Yeah. I feel like it's just asking for more politics than it already is. Yeah, exactly.
Yeah, politics is, yeah. That's a fair point. I mean, reviewing, I think, a paper really well
is actually hard. I think it takes a lot of your time and energy to do it, right? And to be critical
but still engage in, you know, research that may or may not be interesting to you. Like, I mean,
the hope is that it's close enough field that you have a thing or two to say about,
and you enjoy reviewing it, but that's not always the case. And yeah, I just kind of wonder what it
means. Like, the one thing I do like about open, the concept, I guess, of open peer review, because
I've never done it myself, is it, I think it incentivize reviewers to actually properly review.
Because now it's now it's attached to your name. Now that's going to be publicly available, because
you know, you know, the, I guess, I don't know, I don't know how much of it is circulating in Japan,
but there's a meme of the reviewer to write the myth of the reviewer to it's always the reviewer
to that's, that's gonna give you shit, it's gonna give you like, Oh, why haven't you considered
these other completely, like different sets of experiment? Yeah, why haven't like, this is badly
21:02
written? This is unusable data? What are they talking about? Like those always come from
reviewer to reviewer one and three. It's like, oh, yeah, this looks fine. It looks like interesting
and topical topics. Novelty. Right. And then like reviewer two, it's always gonna find you the
critical, highly critical, find you find like one tiny little mistakes that you made and be like,
this author is clearly not understanding the whole field. They have no understanding of the
precedent at research. What bothers me the most is when a reviewer is very wrong, or the reviewer
doesn't understand the research content or the methods that we used. And they will just have a
shit to say, yeah, and then they'll say, you know, I'm curious. If this happens, what does this mean?
And why don't you do this? Because I'm curious. But we're not doing the science just to, you know,
just to satisfy your curiosity. Yeah. It's like, well, you're curious. Do you want to
work in this field for 15 more years and find out what that's what is possible? Like,
oh my gosh, yeah. That could be reviewer two. Yeah, that could be reviewer two. It's just
funny. It's somehow magically always lands in that reviewer two. And it's very confusing too,
right? Like when you have three reviewers and two giving you very minor edits, generally overall
positive, thinks that this should go straight to publication. And then reviewer two just like
giving you a hard time and but like your editor wants you to address the feedback from reviewer
two. So you kind of have to do it and then like, you know, wait before you know it half a year
since you thought you could publish it. Yeah. Oh, wow. But it's a lot of problematic stuff. I do
feel like this isn't working anymore. Especially in a field like my field less so because it's
very close to fundamental. But if you're anywhere close to applied field, I feel like the speed of
the reviewing process is actually stopping people or slowing down the research progress. Because
I know that some people who are in like AI research or other things like something that's
very, very like new information coming every day kind of field that's being like delayed because
reviewer two had some stupid curiosity question is like is a major career hindrance. And yeah,
24:02
can imagine that's very annoying. Yeah, gotta do something about it. But I mean, I don't have a
solution. You probably don't either. So this is probably a good place to stop this podcast.
Okay. That's it for the show today. Thanks for listening and find us at Eigo de Science on
Twitter. That is E I G O D E S C I E N C E. See you next time.
