Peer Review

188 112 6 years ago

simonkafan• 6 years ago

Peer review is broken. More than once I've learned that papers get easier through peer review if you cite the paper of a reviewer (of course you don't know in advance who reviews it so it's basically luck). I often got comments about missing related work "from relevant authors".

_emacsomancer_2• 6 years ago

I'm not sure that's exactly the reason it's broken. Usually an editor will give a paper to people who are specialists on your topic, and they will have thus likely written papers related to the topic as well.

One major issue is that, because of this, they will sometimes have a vested interest (conscious or not) in blocking the publication of your paper if it shows flaws in their own.

jhrmnn• 6 years ago

Many journals I'm familiar with allow you to suggest referees, as well as suggest which researchers should be avoided as referees. If I'm publishing something that is competing in some sense with the work of other people, I will usually suggest to the editor to leave them out from the peer review process.

goostavos• 6 years ago

The quickest path towards realize peer review means almost nothing is... reading peer reviewed papers.

Of course, take my semi-anonymous random internet opinion with a grain of salt (I've dropped out of college not once, but _twice_, so what do I know?). However, I'm moderately good at computer science topics, and have spent the last year reading and implementing papers in detail in pursuit of a fairly niche topic, and, what I was dismayed to find, is that most papers are absolute, complete and utter garbage. Like, honesty embarrassing. And that's before you even get into the fact that _you can't reproduce their results_.

Encountering this has been something of a crisis in my entire belief system, to be honest. Although, maybe I've just been exposed to what "real" science looks like: a big, gross, often wrong mess of assumptions about the world that takes a long, long time to correct.

sideshowb4• 6 years ago

I suspect if you think this of most papers you've either stumbled into a really bad cluster, or you've missed something about the purpose of the papers you've read.

I would agree that some papers are terrible though, I've seen a few I would never have let through review including in the very top journals.

I like the Plos mega journal approach but we can't do away with peer review altogether otherwise what separates science from the conspiracy posts that circulate on Facebook ...

sideshowb4• 6 years ago

"From relevant authors" is such a cowardly review. Either give me the citation to the work you want me to cite or keep quiet.

So often as a reviewer I have avoided asking for citation of my own work, even if highly relevant, as I perceive that to be an abuse of power.

I've also started a comedic collection of terrible reviews I've received, though it took me 15 years to gain the confidence to know when these are an indication of the reviewer's failings, not mine.

Veen• 6 years ago

often as a reviewer I have avoided asking for citation of my own work, even if highly relevant, as I perceive that to be an abuse of power.

That's noble, but isn't it counter-productive? If you're reviewing a paper whose readers would ultimately benefit from a citation of your work, then would the paper not be improved by citing it?

chrisco2553• 6 years ago

Any ideas for how the system might work differently? Could peer review learn anything from the open source community?

WoodenChair• 6 years ago

Could peer review learn anything from the open source community?

I think that's a great question. Imagine papers with pull requests, issues, and additional experiments contributed from anyone with relevant experience. And imagine those experiments being open data and up for third-party verification before publication. That could be powerful.

chrisco2553• 6 years ago

Yes, and a reputation system built in but still democratic enough that any modern-day "Einstein" could break through.

Vinnl4• 6 years ago

An article detailing this and similar ideas can be found here, for those interested: https://doi.org/10.12688/f1000research.12037.3

pdonis14• 6 years ago

The ostensible purpose of peer review is filtering: making sure that the signal to noise ratio in published journals remains high enough for the journals to be useful to other scientists.

That logic probably made sense back when journals were the primary means of communicating scientific results. But the Internet has changed the situation drastically. Scientists don't need to wait for a journal to publish things; they just go to arxiv.org. Nor do they need to wait for a journal to filter things through peer review; they have plenty of other ways of judging papers for themselves.

So at this point I don't think we need a "system" at all for what peer review used to accomplish.

rland• 6 years ago

What other ways are you referring to?

cycomanic2• 6 years ago

I can tell you that this system would make many things even worse. The problem is that so much gets published so it's very hard to follow everything, so peer review is a good first filter to reduce numbers.

If we move to a system where everyone just publishes on e.g. Arxiv what happens is that people will just follow certain researchers that they deem good (you can already see that happening in some fields). So you have cabals developing who only read each others articles and it becomes very hard for new researchers to break into.

hitekker• 6 years ago

This paper proposes a return to editorial review: https://sci-hub.tw/https://doi.org/10.1007/s11017-012-9233-1

cycomanic2• 6 years ago

I think a first step is to make reviews public, in particular also rejects. In particular the big journals like science and nature seem to rely on a very small set of reviewers and often papers get reviewed by nonexperts, we had a paper rejected when a reviewer contradicted scientific fact that is in textbooks. When we protested the decision they just send it back to the same reviewer who stayed with his claim. I think if these reviews would be public the journals would get much clearer feedback on how bad some of the reviewers are.

probably_wrong• 6 years ago

I think a first step is to make reviews public, in particular also rejects.

As a reviewer, I have no objections towards my reviews going public. However, I have often had to tell authors that their paper is dumb (not in those terms, obviously), and I'd rather keep those reviews private. Being rejected is hard enough without the entire world knowing how much your paper sucks.

Of course, there are terrible reviewers too. But IMHO that's better solved via better area chairs than with a Twitter mob. I would rather not being 4chan into the review process.

DoreenMichele5• 6 years ago

From the article:

Peer review grew up in a world where there were many fewer people engaging in science than today. Typically an editor would know everyone in the world who had contributed to the field in the past, and would have enough time to understand the ideas of each new entrant to the field as they started to submit papers. It relied on personal connections and deep and thoughtful understanding.

That has changed just due to the scale of the scientific endeavor today, and is no longer possible in that form.

I don't know how you fix that. But it points to why peer review worked well at one time and became some kind of gold standard and why it's failing to be some kind of gold standard today.

I think you have to parse what was going right about the system that created peer review in order to have any hope of parsing how to replace it at scale for a world that's grown vastly larger in terms of sheer population numbers.

barry-cotter4• 6 years ago

Yes, peer review should be an informal process that happens after publication, not a formal one that turns a 40 page paper into a 100 page one and takes six years.

Many fields have developed work arounds to peer review in the form of working papers, preprints and conferences. That this is necessary suggests that peer review be burned to the ground and the ashes sown with salt.

https://www.stat.cmu.edu/~larry/Peer-Review.pdf

The refereeing process is very noisy, time consuming and arbitrary. We should be dissem- inating our research as widely as possible. Instead, we let two or three referees stand in between our work and the rest of our field. I think that most people are so used to our system, that they reflexively defend it when it is criticized. The purpose of doing research is to create new knowledge. This knowledge is useless unless it is disseminated. Refereeing is an impediment to dissemination.

...

Some will argue that refereeing provides quality control. This is an illusion. Plenty of bad papers get published and plenty of good papers get rejected. Many think that the stamp of approval by having a paper accepted by the refereeing process is crucial for maintaining the integrity of the field. This attitude treats a field as if it is a priesthood with a set of infallible, wise elders deciding what is good and what is bad. It is also like a guild, which protects itself by making it harder for outsiders to compete with insiders.

etrautmann• 6 years ago

You’re creating a false dichotomy. It’s all about maximizing signal to noise, and expert reviewers are absolutely required (at some point in the process) regardless of whether they’re perfect.

esfandia2• 6 years ago

A couple of ideas I've been mulling over for a while:

- the quicker but less rigorous way: set things up like a forum like HackerNews, let the crowd-sourced upvote and comment system work its magic; make the karma/reputation points act as a proxy for the quality and relevance of the article, its authors, as well as the commenters.

- the slower way: let articles "bubble" up from smaller but more specialized workshops toward more generalized and prestigious conferences and journals. Have the technical committee recommend the best articles to those more reputable venues, along with whatever it is the authors need to work on to re-submit the extended version at that "better" venue. This is kind of already the case with workshops or conferences that have an associated Special Issue of a journal; I'm advocating for a generalization of the practice to all venues and more than a 2-level hierarchy. The reputation of the lower venues could be backpropagated by the number of papers they successfully "bubble" up, PageRank style, so that would be the incentive the committees would have in recommending their papers to the other venues.

randcraw• 6 years ago

Your option 2 might work if the bubbling process is somehow made informed and fair, but option 1 is an invitation to promote charismatic loudmouths. Using pure populism as the means for seeking truth clearly doesn’t work very well, as the rising tide of of beloved trolls in media and politics attests.

Vinnl4• 6 years ago

Currently, peer review fulfils three distinct functions, to the detriment of all of them:

1. Acting as a filter for what research is worth reading.

2. Providing input to the author on how to improve their work.

3. Influence the author's professional evaluation.

The first is the most blunt instrument: when something is not considered worth reading, it doesn't get published at all. This is problematic in the cases where the peer reviewers misjudge, or when an article is evaluated on more than its academic merit, e.g. whether it's "groundbreaking" enough for a particular journal.

The second is useful, but made less so due to the power imbalance - an author cannot judge for themselves whether the feedback makes sense in the context of their work. This is especially problematic when someone has to take their work to multiple journals and/or gets conflicting feedback.

And that power imbalance is the result of the third point: it can be more important to the author to get that stamp of approval by the reviewers than to meaningfully contribute to science, because that's the only way they can stay in academia.

Ideally we'd split them up, but that does require aligning the academic community - which is a hard problem, especially considering there are strong vested interests in the status quo.

(Disclosure: I do volunteer for a project (https://plaudit.pub) aimed at splitting evaluation from publication, because I think it's an important issue.)

j88439h84• 6 years ago

Interesting list!

albertzeyer2• 6 years ago

Plaudit is nothing more than like an upvote, right? I don't think this is really helpful. We already have a very similar metric, which is citation count. Which is definitely a useful metric.

At first, I thought that Plaudit adds some post peer review platform, or discussion forum, for each paper. Similar to OpenReview but for any paper. I think this might actually be a great idea.

I think that https://fermatslibrary.com/librarian actually provides a similar functionality. But it has not seen much adoption.

jkhdigital4• 6 years ago

As another comment implies, there is a fourth function that has been saddled on the process against its original intent: provide a “stamp of approval” for journalists to write misleading articles about “peer-reviewed research”

timkam• 6 years ago

I think we can learn from the OS community, but this requires changing the incentive system of academia. Two ideas:

* Research papers should be living documents that are properly versioned; updates should get reviewed again. This obviously means that incremental improvements to papers and reviewing need to result in something tangible for one's CV. The versioning makes sure one can still point to a specific revision of a paper. Of course, updates should consider the implications of changes carefully (as is the case for software). For example, I might get familiar with a paper over the course of a month, and then use it as the cornerstone of one of my own works. Then, I want that the paper is stable and does not constantly mess with important formal definitions et cetera. Otherwise, people won't be able to agree on the implications the paper has for the community.

* It might be better to make review and reviewers fully visible and move to a more interactive review process. For obvious reasons, practically "double-blind" review does not work, because if you're a visible member of your community and submit a paper, most people can guess that it's yours. If we would move from "one-shot" (or: one-shot plus rebuttal) reviews to more interactive, public, non-anonymous discussions, we would get rid of at least some of the social back-channeling that is only available to the bigshots in the community.

Of course, I don't think we will get there quickly.

sytelus3• 6 years ago

One issue is that reviewers themselves don’t get rated. They also don’t have any incentives to do quality reviews. I think StackOverflow model is the only one I know of which is able to scale this process. It’s not without flaws but it’s content usability and quality is extremely high than any other collaborative filtering model out there. Academic reviews can perhaps be scaled with similar model. Imagine each paper is question and each review is answer. People vote for which answers and questions were useful and everyone accumulates reputation that in turn can be used for career development. You can see everyone’s questions, answers and votes. People with enough karma gets moderator privileges to be able to knock off bad content. People will then have incentive to do the right thing, improve quality and avoid favoritism.

BurningFrog3• 6 years ago

Science did well before radical ideas could be voted down by peer review.

Would Einstein and Darwin have made it through peer review? I doubt it.

What current day people like that are held up by their "peers"? I fear we'll never know.

DoreenMichele5• 6 years ago

Would Einstein and Darwin have made it through peer review? I doubt it.

Einstein worked in a patent office for years, unable to get a job in academia like he wanted. There were two attempts to prove his Theory of Relativity. They occurred years apart because it involved photographing a total eclipse of the sun and while eclipses typically happen every six months in pairs (one solar and one lunar, though sometimes there are three in a row), most of them are partial eclipses, not total eclipses. It had to be a total eclipse to prove his theory.

The first attempt to make the photograph failed. There were only two teams and it was during World War I. One was stopped or arrested or something because they were assumed to be spies. The other failed due to rain.

After this failure, Einstein realized there was an error in his math and redid his formula. Had the first attempt at taking photographs succeeded, we might have never heard of the man because the world might have decided he was a quack since his math didn't hold up, so, "obviously," his theory was bunk.

He probably had one and only one shot at this unusual peer review process.

By the time they made the second attempt, it was something like six or eight teams of photographers trying to get a photograph of the total eclipse of the sun. Only one succeeded. I think the rest were again rained out.

Nicely, it was a team that had also tried several years earlier that actually succeeded.

Once there was photographic evidence supporting his theory, he became world famous overnight and then was finally able to get a job in academia, IIRC.

What current day people like that are held up by their "peers"? I fear we'll never know.

Me. But just saying that is likely to get me mocked. No one is trying to prove I know anything. No one wants to peer review (or otherwise check) my ideas. People mostly just want me to shut up and go away.

pdonis14• 6 years ago

Einstein worked in a patent office for years, unable to get a job in academia like he wanted.

Actually, while the patent office was his day job, he was working on his Ph.D. when he published his classic 1905 papers including the one on Special Relativity. He got the Ph. D that same year, and got his first academic appointment three years later, in 1908. He didn't start working on General Relativity in earnest until about 1911 (although he had the initial ideas as early as 1907).

There were two attempts to prove his Theory of Relativity.

More precisely, two attempts to verify one particular classic prediction of General Relativity, that of bending of light by the Sun. (The second, in 1919, btw, is now generally agreed to not have actually been a valid proof: the observations weren't actually good enough to see the effect when you take into account the appropriate error bars. But they were accepted at the time, probably in large part because of the influence of Eddington, who spearheaded the process.)

A second classic prediction, the extra precession of the perihelion of Mercury, was already known in the 19th century, but was a mystery to physicists until General Relativity explained it.

The third classic prediction of General Relativity, gravitational redshift, had to wait until the 1950s before reasonably accurate experiments were done to confirm it.

Once there was photographic evidence supporting his theory, he became world famous overnight and then was finally able to get a job in academia, IIRC.

As noted above, he actually started working in academia well before the 1919 eclipse expeditions. But the results of those expeditions were indeed what made him world famous.

pdonis14• 6 years ago

After this failure, Einstein realized there was an error in his math and redid his formula. Had the first attempt at taking photographs succeeded, we might have never heard of the man because the world might have decided he was a quack since his math didn't hold up, so, "obviously," his theory was bunk

While the failure of the first expedition might have been a factor, it was certainly not the only one. Einstein did not like a number of things about that version of his theory and probably would have revisited them anyway; his writings during this period show him going back and forth between liking the theory and thinking it's obviously wrong and needs to be patched up.

The key thing that appears to have gotten Einstein on the right path to the final 1915 version of General Relativity was a visit with David Hilbert in Gottingen in the summer of 1915. Ironically, it also got Hilbert started on a path that led to the same final field equation that Einstein found, but by a different route (a much more elegant one mathematically), and which has spawned a long-running dispute about priority [1] (though as far as I can tell, not one that either Einstein or Hilbert ever participated in--AFAIK Hilbert always credited Einstein for the main physical insights that led to General Relativity).

[1] https://en.wikipedia.org/wiki/Relativity_priority_dispute#Ge...

pdonis14• 6 years ago

Would Einstein and Darwin have made it through peer review? I doubt it.

Einstein did make it through peer review; his classic 1905 papers were published in Annalen der Physik, the most prestigious physics journal in the world at that time, after being refereed and accepted.

Darwin didn't technically go through "peer review" since he published his results in a book, not a scientific journal. But when he did so, what we now call scientific journals were extremely rare.

BurningFrog3• 6 years ago

Peer review, in the modern sense, only became common about 40 years ago. In Einstein's time, the journal publisher decided on their own what to publish.

Or so I'm told. I know that when I went through a pretty serious physics education in the early 80s, I never heard about peer review.

btrettel• 6 years ago

From what I've read the vast majority of Einstein's work was not peer reviewed.

http://michaelnielsen.org/blog/three-myths-about-scientific-...

How many of Einstein’s 300 plus papers were peer reviewed? According to the physicist and historian of science Daniel Kennefick, it may well be that only a single paper of Einstein’s was ever subject to peer review.

barry-cotter4• 6 years ago

Einstein was never published in a peer reviewed journal. Annalen der Physik had editorial review. The only time one of his articles was sent out for peer review he withdrew it from the journal in question.

For instance, the Annalen der Physik, in which Einstein published his four famous papers in 1905, did not subject those papers to the same review process. The journal had a remarkably high acceptance rate (of about 90-95%). The identifiable editors were making the final decisions about what to publish. It is the storied editor Max Planck who described his editorial philosophy as:
To shun much more the reproach of having suppressed strange opinions than that of having been too gentle in evaluating them.

...

We (Mr. Rosen and I) had sent you our manuscript for publication and had not authorised you to show it to specialists before it is printed. I see no reason to address the – in any case erroneous – comments of your anonymous expert. On the basis of this incident I prefer to publish the paper elsewhere.

—- Einstein

https://theconversation.com/hate-the-peer-review-process-ein...

glofish• 6 years ago

in Einstein's time there was very little peer review,

there is a famous anecdote there, the first time that his paper was actually peer-reviewed (perhaps in the 50s) and sent back to him with instructions on what to change he wrote back something along the following lines:

"there must be a mistake there, I sent in this paper for publication, not to get an opinion on how to improve it"

scottlocklin• 6 years ago

Editorial review is not peer review; and it's arguably the system we should be moving towards. Preprint servers mean you could appoint yourself as an editor and more or less DiY. The problem is, the way the journal system is set up; nobody wants to pay for this, and it would be career suicide to do it for free.

cambalache• 6 years ago

Would Einstein and Darwin have made it through peer review? I doubt it.

What are you talking about? Einstein first outstanding contributions (in his 1905 Wunderjahr) were all published in Annalen der Physik , a peer-reviewed journal. Even years after becoming a worldwide-known name, he received rejections (which he was not too much fond of). Einstein was always part of the conventional publication system of physics,

Darwin , and Wallace original papers were read before the Linnean society in London and were published in its Zoology journal. You can read it here: http://darwin-online.org.uk/content/frameset?itemID=F350&vie...

In the history of modern science , let's say from mid 19th century on almost the entirety of solid scientific contributions have been published via the regular channels.

BurningFrog3• 6 years ago

No journal had anything like the modern peer review system in 1905. That system is maybe 40 years old. You can think of it as a response to the vastly bigger volume and specialization of papers.

Back then, if the editor thought the paper was decent, it got published.

Perhaps the biggest problem with peer review is that your peers are often also your competitors/rivals. Imagine if each Apple product had to be approved by Google, and vice versa.

rrss• 6 years ago

Einstein sent his papers to Planck, and Planck, as an editor of Annalen der Physik, decided to publish them in the journal.

Here's what Einstein had to say about the modern peer-review process in which a draft is reviewed by anonymous referees:

We (Mr. Rosen and I) had sent you our manuscript for publication and had not authorized you to show it to specialists before it is printed. I see no reason to address the—in any case erroneous—comments of your anonymous expert. On the basis of this incident I prefer to publish the paper elsewhere.

https://physicstoday.scitation.org/doi/10.1063/1.2117822

On the other hand, Einstein was really into what we might today call post-publication peer review. A significant chunk of Einstein's publications were reviews of other scientists' publications.

gwern• 6 years ago

Darwin sorta did go through peer review, in one instance, giving us the immortal peer reviewer revise-and-resubmit suggestion that Darwin rewrite On The Origin of Species to focus on pigeon fancying, because "Every body is interested in pigeons."

https://en.wikipedia.org/wiki/Whitwell_Elwin#Works

haltingproblem3• 6 years ago

Rodney Brooks subsumption architecture revolutionized robotics. Back in the 90s you could replicate his work with Attila and Genghis and recreate hexapod robots for under $1000 which was an pittance for a robot.

He was at MIT when he submitted those papers. The churlishness of those reviews is staggering. What if he had given up and gone on to something else?

p1esk• 6 years ago

What if he had given up and gone on to something else?

Maybe he would have accomplished even more?

haltingproblem3• 6 years ago

What would have happened if he had moved on to another field? Subsumption architecture would eventually have been discovered by someone else, standing on the shoulders of giants and eventuality of all discoveries etc.

Or it might have taken another 50-100 years. Fundamentally unknowable.

buboard2• 6 years ago

What if he had given up and gone on to something else?

Science isn't supposed to be easy and it's the faithful ones that thrive. If anything, i wish reviewers were more critical of stuff before it's published.

gilbetron• 6 years ago

Why shouldn't science be easier? We should work at making the doing of science as easy as possible. Getting ideas out there is a boon, suppressing them is of no value. Great ideas should percolate up because they work well, not because someone that has never tested them doesn't like the author. There is no moral superiority in making ideas harder to propagate. Science itself is tough enough as is.

INGELRII• 6 years ago

There are two types of peer review.

1. Pre-publication peer review Reviewers examine the work before it's presented to the rest of the people.

2. Post-publication peer review Paper on an online forum where others can read and comment and cite them. Papers can be accepted into journals and go trough peer-review later.

Post-publication peer review does not mean that authors can't use pre-publication peer review. Scientists can (and should be encouraged) to ask their colleagues to review papers before posting online to spot errors.

Some fields have long tradition to publishing and circulating working papers, manuscripts, drafts, posters etc. years before they submit the final paper for publication and pre-publication peer-review. That works as well.

ChrisMarshallNY• 6 years ago

I think peer review is quite important.

Note the word “peer,” though.

I would not peer review papers on robotics or AI, as those are not my forté. I could review papers on some generalizations of software engineering or the Swift programming language.

Editing is a different matter. My mother was a scientific editor, and absolutely brutal. She edited a book I wrote on writing software for a certain specialty of Web site, and it was quite humbling. She knew nothing at all about the technology, yet provided an immensely valuable service.

The book is long past it’s “sell-by” date, yet stands as maybe the only “perfect” artifact of my life, thanks to her.

Then there’s comment threads in venues like this one, which some might equate to “peer review.”

There was an old Dilbert comic on Peer review. I don’t think it can be found online, but was in “Build a Better Life by Stealing Office Supplies.”

It was entitled “The Joy of Feedback,” and probably applies to comment threads.

higeorge132• 6 years ago

I don't consider myself an expert in computer networks research but i still have 10 accepted (and ~20 rejected) papers with ~100 citations in journals, conferences and workshops within my 3 years working on R&D. Hence my comments below derive from my non-academical and more development-oriented background.

Peer review is completely broken nowadays for the following reasons:

-Reviewers might not have relevant expertise to judge whether a paper is good or not. I am not talking about top-notch conferences, mainly lower-tier ones and workshops. I have seen a lot of professors pushing paper reviews to their PhDs, or even students. I was also invited to review a paper as soon as i submitted my first one (and it was rejected btw)! In another workshop i was invited to review papers just because i was a colleague of one of the organisers.

-Old-school academia. This might not apply to all fields or all academia, but i have had good papers rejected because they didn't have a lot of math or simulations! My paper was examining an actual implementation of an SDN (software defined networking) protocol or a strategy with platforms and orchestrators which require weeks to setup and implement (Opendaylight, Openstack, OpenMANO, etc) with actual experiments on real users' video activity, yet to be rejected because i didn't provide any simulation. Jeez, novelty does not come only from theory, somebody has to implement and test these things in actual machines..

-Politics. I won't say much on this aspect, other than a colleague of mine had 1000 citations within the same time just because there was a "mafia" (not my words but his) of professors, lecturers, PhDs reviewing and accepting all the group's papers because they were overloading them with each other's references and co-inviting them to each other's conferences and workshops.

JohnStrangeII• 6 years ago

I'm in philosophy, though personally working in formal philosophy, so this is quite different from CS. In my area practically all top journals reject more than 95% of all submissions, and if you get accepted, it takes about a year until the article is published. Many journals get so many submissions that they do not provide any reviews or feedback - not even a paragraph or line - if they reject a paper.

Although I have been able to stay a fully paid postdoc for more than 10 years without problems and get decent amount of publications out, I'd agree that the peer review system is broken and seems to be strongly biased in favor of mediocrity over originality. It's particularly bad in fields like philosophy in the humanities, where you could argue that there are no clear-cut criteria for evaluating the content other than merely formal issues. (Philosophy is not a science.) There are lots of people who game the system, I'd say many of my colleagues are doing it, though not (yet) the majority. By gaming the system I mean writing articles they barely endorse themselves, tailored to get past reviews. I've literally had a very successful colleague say "Who cares? It's one more publication!" when I raised doubts about a joint paper, because it involved too much hand-waving and didn't seem to work.

Long story short, I have the impression that in the humanities whether you're accepted or not mainly depends on the writing style and on whether you win the "reviewer lottery." One and the same paper can get accepted in one journal and rejected in another, for journals with similar reputation. I also believe that there are certain niche areas in the humanities that are (almost) pseudo-sciences and based on a small group of researchers colluding with each other and passing opportunities for publication to each other. I won't mention the disciplines, though.

I'm a bit dismayed that you describe CS in a similar way, I've always thought its much more rigid there and easier to evaluate CS papers.

Having said all that, I don't think there is any alternative to the peer review system. Any alternative I've seen so far or come up with myself seems to be even more prone to manipulation and gaming the system. I believe that it's not the peer review system itself, but over-reliance on objective indicators by university administrators that has led to most of the problems. But the humanities have certainly changed to the negative, at least in philosophy I'm certain that many seminal and famous articles from the 70s of last century and earlier would not make it past the peer review today.

currymj2• 6 years ago

sadly, CS is also about the reviewer lottery. however because it is conferences rather than journals, turnaround is much faster at least. and there is a culture of preprints, so having a paper rejected once doesn’t delay getting it out into the world.

papeda7• 6 years ago

Your points are mostly fair, though I'll push back on one. For context, I'm a late-year PhD student in ML, but I primarily do theory.

Reviewers might not have relevant expertise to judge whether a paper is good or not...

Finding enough quality reviewers is hard. There are so many papers (heck, you put out 30 papers in 3 years!) and there is not much incentive to review. I review maybe 5x more papers than I write each year. It's doubly hard to get good reviews out of senior people. For this reason, I think grad student reviews are not a bad solution. In my experience, grad students puts in more effort and write decent, careful reviews. The very worst, low-effort reviews seem to come from established people who can't be bothered. A lower reviewer bar at workshops is fairly common and reasonable. Workshops are not the main publication venue (at least in my areas), so they mostly just want to filter out papers that are bad rather than only accept papers that are good.

Politics...

I can't say too much about this since I primarily work in theory, but my perception from the outside is that the more "lab"-dominated a field is (i.e., accepted papers tend to come from some group with 20 PhD students and 10 postdocs) the more likely there are to be "mafias". Maybe this is a function of the kind of people who want to head up a lab, maybe it's a function of the kind of work that gets produced in factory environments, or maybe it's baseless prejudice from me.

higeorge132• 6 years ago

You are probably right about postgrads. Although i would not consider myself the best reviewer i was really investing time and effort to produce a fair and honest review with the "limited" knowledge of the field i had, and i was probably a better reviewer than "experts" who would produce 1 line reviews or generic comments "add more references". However i would beg for highly-technical reviews, providing harsh theoretical and technical feedback, which i guess would come from more experienced or expert reviewers than postgrads or myself. And by expert i mean people who have knowledge and publications on that specific paper topic, e.g. you might not be able to provide expert feedback on P2P or distributed networks just because you are an expert in QoS (network buzzwords here, please ignore :) ). Which is probably an effect of what you mentioned; the amount of conferences/workshops/journals and papers submitted and lack of incentive from people to review.

The ~30 papers were an outcome of collaborative R&D projects with universities and companies, most of those just having my name in the authors and requiring me to quickly read and comment them, instead of spending effort to produce the results presented. From a third person perspective this somehow justifies the 3rd point of my original post on how these mafias are created and maintained. The thing is that i didn't care for citations or more paper invitations because my position or wage didn't depend on them (i was just doing my research for the implementation and the trips in the conferences my papers were accepted) and eventually i am not doing R&D anymore, however i can imagine how unfair this whole setup is for people who want to follow an academic path.

jseliger2• 6 years ago

From the original piece: I don’t have a solution, but I hope my observations here might be interesting to some.

I have a partial solution: researchers "publish" papers to arXiv or similar, then "submit" them to the journal, which conducts peer review. The "journal" is a list of links to papers that it has accepted or verified.

That way, the paper is available to those who find it useful. If a researcher really thinks the peer reviewers are wrong, he/she can state why, and why they're leaving it up, despite the critiques. Peer-review reports can be kept anonymous but can also be appended to the paper, so that readers can decide for themselves whether the peer reviewers' comments are useful or accurate.

laretluval• 6 years ago

If the paper is on the arXiv then it will not be anonymous to the reviewers.

currymj2• 6 years ago

This is how many CS conferences already work (even with public anonymous peer review, go to openreview.net to see it in action). It does seem to be better, although the frustration level for researchers is not really less.

Edit: also, in CS there is a norm that reviewers will not go out of their way to figure out if a submitted paper has been previously posted on arXiv. I don't think it's possible to have perfect anonymity without forbidding researchers from publicly talking about work in progress, so this seems like a reasonable compromise.

yqx• 6 years ago

While peer review has its problems, I have the feeling that the problems that we're talking about here are not primarily caused by peer review per se, but by a set of perverse incentives that surround academia these days. I'm talking about the "publish or perish" culture that is results from the (sometimes) automated metrics based on which promotions are decided, grants are handed out and academic staff is assessed. Isn't it these incentives that can create attitudes like "Who cares? It's one more publication!" (as quoted by another commenter in this thread). And isn't an overwhelming amount of mediocre journal submissions that result from this one of the reasons peer review is under pressure, qualified reviewers are rare and politics are so common?

albertzeyer2• 6 years ago

This topic comes up again and again. (Here some references from a quick search: https://news.ycombinator.com/item?id=10531374 https://news.ycombinator.com/item?id=18523847 https://news.ycombinator.com/item?id=8731271 https://news.ycombinator.com/item?id=18595074 https://news.ycombinator.com/item?id=22251079)

But there are not many good solutions.

Most seem to agree that we should move over to a more open system, like OpenReview, and also that all publications should be open for everyone afterwards, and not behind some paywall. But these are only two aspects of the system, and this is less about the peer reviewing itself.

The community already has the trend to publish more and more directly on ArXiv. Just in addition to submitting it to a conference. If the conference peer review filters it out for some unlucky reason, it's still open for everyone to see. That might be already an improvement of what we had before. But probably it's not really the optimum yet. But what is the optimum?

jseliger2• 6 years ago

There are good solutions, but existing publishers like Elsevier like the current situation because it makes them rich (https://www.universityofcalifornia.edu/press-room/uc-termina...). Any existing individual academic is strongly incentivized to work within the rotten system, even though most know its pathologies and want to defect—but one person defecting alone will screw up their own career while likely leaving the system intact. The equilibrium is bad even though the solutions aren't complicated. https://news.ycombinator.com/item?id=23291869

buboard2• 6 years ago

That's peer review from 3+ decades ago. Things have changed substantially. PIs dont have enough time to review all that stuff, lot of it is delegated to students, reviews can be hasty, with reviewers who may latch on to a specific detail and let the rest pass through. It's not even that they are not trying hard enough - Peer review is not doable in 2020. If you honestly want to scrutinize and improve every detail of a manuscript, you re better off posting it in reddit or twitter.

clamor for double blind anonymous review

That's not even possible in most fields. you can easily tell which lab it comes from (you can even often guess who the reviewer is)

specialist• 6 years ago

I love this topic of peer review. Feels like arguing about methodologies, software QA & test, learning organizations back in the 90s.

Peer review and the reproducibility crisis are the absolute bleeding edge of the human experience. Science, governance, transparency, accountability, progress, group decision making. All of it.

I encourage all the critics, thought leaders to also consider how to apply these lessons elsewhere.

Every thing said about peer review also applies to policy making, product development, news reporting, investigative journalism.

Peter Drucker, way back when, stated that management is a technology. Same applies to peer review and reproducibility.

For whatever reason, this post prompted me to think "sociology of scientific progress". I encourage critics to also factor in the collective human psychology. Covered by so many books, like the classics Structure of Scientific Revolutions and The Diffusion of Innovation. (Sorry, none of the more contemporary titles, a la Predictably Irrational, are jumping into my head at the moment.)

papeda7• 6 years ago

Peer review is flawed, but I don't think it's broken. For context, I'm a late-year PhD student in machine learning theory, and I submit almost entirely to conferences, which is the norm in my field.

To me, the big problem is that there is little incentive for experts to review papers. I write a few papers every year, and I review 5x that number. Slightly more senior people -- postdocs and junior faculty -- may review more like 30-40 papers. Most papers I review are not good papers. This is especially true at the big machine learning conferences like NeurIPS, ICML, and ICLR, where a disconcertingly large fraction (1/3?) of submissions are outright bad. Most of them are at best "meh". So there's not much benefit to the reviewer beyond some notion of community service.

I think my ~colleagues (PhD students, postdocs, young faculty) have similar opinions. We've certainly discussed these issues in the past. One positive step NeurIPS has taken is giving "best reviewer" awards that come from with a free registration. That's a nice benefit, even if it won't on its own incentivize good reviews. Some people have suggested outright paying people for reviews at some quasi-competitive rate, maybe $100 per review (note that this is not great -- a good review takes at least several hours if the paper is not terrible). NeurIPS also added a requirement this year that all submitting authors must review, which is an interesting experiment.

My impractical wish is that we also have some kind of "worst reviewer" sticker for all the reviewers who just paraphrase the abstract, add some milquetoast feelings on novelty, and weak reject. Less facetiously, some kind of StackOverflow reputation system might be useful. As is there's so little external signal about being a good or bad reviewer that the incentives are pretty wrong. Some kind of points system might help.

But my overall point might be: reviewers by and large aren't malicious, jealous people who are only rejecting your work because they want to protect their own turf. They are more likely good faith actors who are trying to balance time spent on a pretty thankless but societally useful job (reviewing) with time spent on much more personally enjoyable and rewarding activities (research, family, hobbies, etc). There aren't really obvious solutions to tilt the balance toward better reviews that can still handle the volume of papers.

JoeAltmaier2• 6 years ago

Dunno about paying. Remember Amazon Turk? Folks just spammed it with zero-effort submissions and 'closed' thousands of contracts with bots. Instantly gamed, so some jerk could walk off with thousands, $1 at a time.

The shaming idea, that could work.

papeda7• 6 years ago

My understanding of the payment strategy is that it would be targeted at people who already qualify as experts but are on the fence. For example, I know postdocs and young faculty who would like to review more as a kind of community service but can't quite justify it in the face of other demands on their time. The money might be enough to tip the scales.

I don't think it will work on the other more senior pool of expert reviewers. On the other hand, I think most of the really bad reviews come from this senior pool, so maybe that's fine.

boomboomsubban• 6 years ago

Amazon Turk is still around, and if it had that problem they fixed it pretty quick. The poster reviews the submission before you get paid, and the same system would be in place in this scenario.

canjobear3• 6 years ago

Mechanical Turk is still useful, though. And there’s a big difference between paying a bored rando $1 to identify pictures of frogs and asking an academic who is highly invested in the relevant ideas to review a (hopefully interesting) paper.

canjobear3• 6 years ago

You’re fortunate to be in a field of theory that is (1) relatively close to a source of feedback from objective reality and (2) formal. Then theoretical discussions can center around what will work and whether the math makes sense. The horrible politics and vitriol happens when you have neither (1) or (2).

guptaneil• 6 years ago

Peer review is something I've been thinking about a lot lately as an outsider. I've never gone through a formal peer review process or worked in academia, but it's pretty obvious that the process has significant problems.

However, the problem that peer reviews are trying to solve (conferring legitimacy on a paper by the scientific community) is so important that we can't give up on it. Sure, pre-print servers mean science can move faster, but it also means anything can be "science." If we normalize publishing papers as preprints, there's very little stopping conspiracy theorists from publishing a paper and having it look plausible.

The sudden rush of media coverage for COVID research has really exasperated the problem. Journalists should responsibly cover preprint studies, but frankly most don't, and who can blame them when even the scientific community is struggling to review the papers itself.

I volunteer with a group[0] that's trying to address this problem in a small way by creating visual summaries/reviews of popular preprints for mainstream readers. This is a side project, so the review process is just a shared Google doc for each paper, but what I'd like to see:

As a reviewer:

  * View diffs between paper revisions
  * Comment on papers inline, like a pull request
  * +1/-1 a submission, or just leave questions without a decision
  * Gain "karma" for good reviews to build my reputation in the community (like Stack OverFlow)

As a paper author:

  * Reply to comments to understand and explain context
  * Post revisions

As a reader:

  * See visual summaries of papers, with inlined reviews (like Research Explained)
  * Vote on both studies and reviews to surface the good ones
  * Browse profiles of authors and reviewers to understand their qualifications/contributions

I do think https://distill.pub is on the right track. Their papers are significantly more visual/readable than most, and they require contributors to open a pull request against the distill repo. However, their process isn't approachable for non-technical fields.

[0] https://www.researchexplained.org

jkhdigital4• 6 years ago

I have only the slightest introduction to actual peer review (pre-qual computer science PhD student) but I have to say that my views have shifted dramatically after moving from “outsider” to “insider”.

The problems you mention seem more about how the public consumes and interprets scientific research, rather than how scientific communities share knowledge and keep their fields moving forward. These are two different challenges and it seems unreasonable to expect the peer review process to solve both of them.

marcus_holmes• 6 years ago

The problem also works in reverse. Probably the biggest example is Plate Tectonics - this idea took decades longer to be accepted than it needed to because the generation of expert geologists who were the gatekeepers for the journals refused to consider it seriously, and the guy who came up with the idea wasn't a geologist [0].

Scientists are just humans, after all. Subject to human flaws. And the scientific method works "in the end", not immediately.

[0] https://en.wikipedia.org/wiki/Continental_drift

haltingproblem3• 6 years ago

Brooks initial papers like "Intelligence without representation", "Elephants don't place Chess" and robots like Ghengis and Attila changed the robotics landscape. It was a leap akin to going from a Mainframes to an Apple II.

Brooks was extraordinarily well placed to lead this switch in Robotics. He was at MIT, a place with very talented engineers - mechanical, electrical - etc. who could work on the robots. MIT also had limitless funding and a culture for disruptive research with long payoffs. This might not have been possible at other places like CMU where robotics has more industrial applications. I remember watching the movies of Atilla and Genghis over and over again, downloaded over an FTP connection!

https://www.youtube.com/watch?v=-6piNZBFdfY

LockAndLol• 6 years ago

How is it that peer review hasn't changed in 30 years? With the amount of people afflicted has nobody had the time and/or influence to improve the process?

There are no doubt lots of smart people concerned by this...

Peer Review

Related Stories