@minimammoth I think that is one underestimated (or understudied) value of preprints: we can finally ask, with data, the question: what value do journals add to the scientific process?

@minimammoth I just think that this needs to be asked at the individual journal level - in the interest of full disclosure, I've been (but am not anymore) an editor at J Exp Med. In my opinion some editors and journal do add a great deal. Many do not, and just drain cash from a system that requires "peer reviewed" publications without carefully monitoring the quality of the process at different journals

@cyrilpedia yes, exactly. I think journals (but not individually, as a corpus), have a... sort of... administrative role? They are an efficient way to connect papers with good peer reviewers, and to provide editing, copy-editing, etc. Personally (partly because I am a professional editor working for eLife), I think these are valuable services, but not everyone has to agree. Some will be happy simply reading preprints, and that's OK, and should affect journals' bottomline, especially if most ->

@cyrilpedia scientists prefer to read preprints and either make their own judgements or discuss amongst themselves or provide public peer review.

@minimammoth In my (obviously biased) opinion, people underestimate how much works goes into curating proper peer review. Making sure you have the right expertise, screening for conflicts of interest, arbitrating the result and chasing reviewers

@cyrilpedia having had to manage peer review a few times, absolutely. I think it's the sort of thing you don't realise until you do it.

@minimammoth It's going to be difficult to work out - because Science and Nature for example have great science journalism in the front matter (again with the disclosures, I also write front matter for Nature Medicine) the work of people like Kai Kupferschmidt (hope I spelled that right) and Amy Maxmen was amazing during the pandemic (or in her case, with the great writing on Ebola before, pandemics). But that certainly does not apply to the endless trickle down of journals in the Ikea now you can't get out of here model

@minimammoth sorry - one pandemic, one epidemic.

@cyrilpedia agreed. I think the science journalism aspect is incredibly important (though I wonder if that doesn't mean that manybe we need more science magazines – places where journalism about science is published – and fewer journals?).

@minimammoth The niche they fill in this respect is science journalism for a scientifically literate readership - so it's not like SciAm or the (again great) work of people like Carl Zimmer or Ed Yong. The economics of all this is a mess, but it was less complicated before commercial publishers began the brand driven expansion phase

@cyrilpedia completely agreed. I don't know if this type of science journalism could be funded (or realistically, would exist) any other way. Because it's funded by the subscription charges and APCs that scientists pay for the research. I'm not sure how many scientists would buy a science-journalism-only version of Nature or Science (unfortunately). (And of course this goes back to funding and also how funding for journalism is... well...)

@minimammoth Definitely is the case now - but I do remember a time when individual subscriptions (I had one) the Science and Nature were not uncommon, and they were read as magazines for the weekly news, book reviews, interviews etc and then you read the papers that caught your eye or were in your field. That could still work, I think - but with OA and preprints, that means enough subscribers for the news and front matter. I don't know if the numbers could work, but they might

@cyrilpedia I suspect it would mean these sort of publishers losing out on a lot of profits, which I suspect they're unwilling to do in the end, since they're private enterprises.

@minimammoth Yes, and now they often belong to ever larger conglomerates (and I suppose ultimately to the Scheinhardt wig company)

@cyrilpedia Exactly. Thanks for the convo by the way. It's one of my main interests, as an ex-scientist and currently an editor who is invested in keeping science open and accessible.

@minimammoth Thank you - and I hope to have more convos here. Cheers

@minimammoth The only other scientific journal where for the last almost 30 years I've consistently read the front matter is Current Biology. (and back matter, the last page with Sidney Brenner was a reason to wait for the next issue at my university library)

@cyrilpedia yes! Always a pleasure to read.

@minimammoth @cyrilpedia IMO, as a collections librarian, a lot of the value I see from Science and Nature is their front matter and it makes it easier for me to somewhat justify the crazy prices...

I know our researchers would sacrifice limbs to get their research published in either place but I remain skeptical as to how much value they are adding on the research side...

I also recognize that my views are definitely not representative :)

@cyrilpedia in itself: peer review could simply happen in the public forum after posting of a preprint (this is what a lot of public peer review initiatives are about), and revisions wouldn't necessarily need to happen (because someone could be aware of the limitations of a study by reading the public peer reviews). But then, we'd need to be really careful about how preprints are being used, and make sure that peer reviews are available and easily accessible, which again, isn't always simple.

@minimammoth In the middle ground you have refereed preprints like Review Commons at EMBO (again, to be clear, I have worked for EMBO and I still do freelance for them, including the podcast) https://www.reviewcommons.org

@cyrilpedia There's of course the question of what the actual point of peer review is in itself. Peer review can only do so much: in essence, it can check whether the conclusions in a paper follow from the data presented. Occasionally it might detect fraud, but for the most part, peer reviewers have time to read a paper and find obvious flaws, but can't be expected to e.g. replicate the experiments to see if the reporting is accurate (does that make sense?). So one important thing is for ->

@cyrilpedia everyone reading papers (with and without peer review) to be aware of what role peer review actually has, what it can do and what it's limitations are.

@minimammoth I think most editors would agree that the function of peer review is not to detect fraud (though if it does, wonderful). But good publishers have image, data and text screens for manipulation and plagiarism that run parallel to and are independent of the peer review process

@cyrilpedia yes, of course. The reason I'm saying that about peer review is that there seems to be this idea that a preprint can't be trusted and a 'published' article automatically can, and this is also not accurate.

@minimammoth We also have to be careful not to just by extreme cases - as my courtroom drama TV and film education taught me "hard cases make bad law". But what journals are for, in the age of preprints is a question we've been systematically asking authors on the EMBO Podcast and on the RC blog https://www.reviewcommons.org/blog/

@minimammoth A consistent pair that turn up is "journal help me find content and my readers/community find my own work" and "I know the bar/value the QC of certain journals". (note that this last does not at all imply valuing glamour mags)

@cyrilpedia actually this is a point I talk about a lot: most scientists (at least the ones I interact with) want to publish good science they believe in. We should be infrastructure for those scientists, not for the exceptions or rare cases that break the rules (even if we still try to detect those cases).