New US Federal guidance on open access publishing

[Parasaurolophus]

Why not just requiring authors to upload accepted versions of papers on arXiv?

Problem solved (and for free...)...

In the Great White North, that's what SSHRC requires of us (along with acknowledging its funding, of course). As long as there's a version freely accessible in a repository somewhere, we're good.

Not every discipline has that sort of repository in place, of course. But there are always other places you could upload the document--academia, your website, your faculty page, etc.

[research_prof]

Why not just requiring authors to upload accepted versions of papers on arXiv?

Problem solved (and for free...)...

In the Great White North, that's what SSHRC requires of us (along with acknowledging its funding, of course). As long as there's a version freely accessible in a repository somewhere, we're good.

Not every discipline has that sort of repository in place, of course. But there are always other places you could upload the document--academia, your website, your faculty page, etc.

That's my point. As long as you can upload a (pretty close to) final version of your paper on a public website, why do we need to pay thousands of dollars to publish papers as open access? I can definitely see NSF grants not being enough to even cover a single PhD student per year in the future if all of us have to budget several thousand dollars every year for open access publications.

[Puget]

This is a step in the right direction, but APCs need to be limited to the actual cost of processing and posting articles, otherwise we are still wasting taxpayer money (just on grant budgets instead of access fees). I've seen estimates that this is more like $500 at most, not several K as some journals charge. Make publishers negotiate a rate with federal funders like universities do for indirect rates.

This is something that funders can fix immediately, by setting a limit on allowable costs per article. The language the OSTP is using seems to allude to that (mentions of reasonable costs etc).

Bigger issues for me:
1. What happens when societies complain they can't keep their doors open with lost revenue (almost definitely not true, but they have and will lobby for this).

2. How to deal with the review-for-free system that doesn't seem to be working well now, and with less money in the system (for editors, prestige games, etc) not likely to get better. APC vouchers could be part of a solution, but suspect we need more (not sure what).

Yep, I'm very much hoping that the federal agencies will cap the APC rate and force the publishers to play ball. A good analogy would be health insurance -- providers can nominally charge whatever they like, but they are still only getting what the insurance company is willing to actually pay.

The other issues are indeed pressing, but will require other solutions. For (2) they can raise membership fees, and the membership can decide if they are getting value for money. (3) is obviously a lot more complicated with no easy solutions. I'm not sure paying reviewers is a good solution (e.g., can have a bad incentive structure unless there is some sort of quality evaluation). Something like getting credits for reviews and needing 2-3 of them for each paper submitted as senior author could do the trick-- to really work these would have to be fungible across journals I think, but since most journals now use Publons  to document reviewing, that's doable.

At institutions that are not wealthy and give mediocre start-ups (or for unfunded middle-career researchers), those without grant funding (many, since the success rate for NIH funding, anyway, is abysmal--some institutes are ~8-10%), who had difficulty competing before will have even more difficulty when all papers are open access. I have seen many journals charging $2,3 even $4k per article. This doesn't effect big well funded labs--just ask for more money in your next grant. However, when you can barely scrape together enough to pay for experiments, how do you turn around and then pay $9k to get 3 articles published? Publish or perish is real (and not just for tenure). Catch22-->Without publishing you can't get a grant, and you need a grant to pay to publish? Is the answer close down the mediocre programs and let the strong survive? Or do the publishers get strong-armed (by who?) to lower their open-access fees to something more reasonable?

Thoughts?

I understand the concern, but I don't think journals would necessarily go to all OA, at least not right away. More likely most would still be a a "gold" OA model with APCs to make an article OA, and paywalls if not. So those without grants would still be able to publish at most of the same journals without APCs.

In the longer term, if most research is OA, libraries are going to be increasingly unwilling to subscribe to journals to access paywalled content though. However, if everything eventually does go OA, this will save institutions a ton in subscription fees, which they could repurpose to pay OA fees in behalf of their faculty. Also, as noted above it seems likely that federal funders will cap what they will pay APCs, which will hopefully bring down the cost for everyone.  Even if not, for  the type of institution you are describing, they would likely still save money overall even covering all APCs for everyone without a grant (ask your library what the yearly subscription costs are--you will likely be shocked). (My institution already has a program where if you don't have budget for APCs you can get a few papers a year covered.)

Why not just requiring authors to upload accepted versions of papers on arXiv?

Problem solved (and for free...)...

This is already what happens, but with PMC (required for federal funding)-- the change is journals can no longer impose a 12 month embargo. That is no doubt going to change the publishing model, such that journals will want to start charging APCs to recoup their costs. Nothing is actually free. . .

[dismalist]

Why not just requiring authors to upload accepted versions of papers on arXiv?

Problem solved (and for free...)...

Absolutely!

That's an example of a competing publishing outlet that's instrumental in breaking the publishing oligopoly. If a paper went through a review process, it doesn't matter how it's disseminated, just name the review process.

[fizzycist]

Why not just requiring authors to upload accepted versions of papers on arXiv?

Problem solved (and for free...)...

Absolutely!

That's an example of a competing publishing outlet that's instrumental in breaking the publishing oligopoly. If a paper went through a review process, it doesn't matter how it's disseminated, just name the review process.

That's how folks handled Plan S, but with the US joining the mandate game, it is not a sustainable endpoint. If all articles are on arXiv, then no need for subscriptions. If there are no APCs or subscriptions, there is no revenue; journals will close and there will be nobody left to facilitate peer review.

As someone said upthread, there is a true cost to publishing pre-publication peer-reviewed science, which has been estimated to be at the level of $1k/article (assuming free reviews). If we move to curation and post-publication review, the costs likely go down, but we need incentive structures to ensure ppl keep reviewing. And anyway this is now a pretty big transformation beyond just post to arXiv.

So ti is not simple, but it will be fascinating to see how this will play out, and I'm glad to see the (absolutely awful) system get shaken up.

[Hibush]

The OA repositories appear to be egalitarian, but in fact deepen inequality.
Sure, anyone can post but what matter is whether it gets read.
The peer reviewed literature is filled with enough dross that I only read ~20% of the papers I see that pertain to my research. I can usually tell halfway throught the abstract that the article doesn't actually contribute anything meaningful. And that is not only reviewed and edited, but then curated specfically for me.
The unreviewed literature additionallly has all the rejected stuff, the never-to-be submittted stuff and mostly challenging to read because it hasn't seen an editor. There is no way I'm going to wade through that.
The result is that the only thing I read on the repository are new uploads from the well established labs, since I know those are well done and will see print soon. Most scientists will have to do the same as a simple survival tactic.

[Puget]

That's how folks handled Plan S, but with the US joining the mandate game, it is not a sustainable endpoint. If all articles are on arXiv, then no need for subscriptions. If there are no APCs or subscriptions, there is no revenue; journals will close and there will be nobody left to facilitate peer review.

Yes, this is what I was trying to explain in my earlier post-- with no more embargo period (which I fully support), you do need some way for journals to cover their costs. I don't even begrudge them a small profit. But the the major for profit publishers are currently raking in huge profits, between downright extortionary library subscription costs and APCs, and that isn't sustainable either. Caping APCs would go a long way toward fixing this.

The OA repositories appear to be egalitarian, but in fact deepen inequality.
Sure, anyone can post but what matter is whether it gets read.
The peer reviewed literature is filled with enough dross that I only read ~20% of the papers I see that pertain to my research. I can usually tell halfway throught the abstract that the article doesn't actually contribute anything meaningful. And that is not only reviewed and edited, but then curated specfically for me.
The unreviewed literature additionallly has all the rejected stuff, the never-to-be submittted stuff and mostly challenging to read because it hasn't seen an editor. There is no way I'm going to wade through that.
The result is that the only thing I read on the repository are new uploads from the well established labs, since I know those are well done and will see print soon. Most scientists will have to do the same as a simple survival tactic.

I am also not optimistic about the "burn it all down and replace with post-publication review" approach for these and other reasons. For all its imperfections pre-publication review is still the best system we've got for screening and improving science. I don't think we are anywhere near ready to replace it with something else.

[research_prof]

Why not just requiring authors to upload accepted versions of papers on arXiv?

Problem solved (and for free...)...

Absolutely!

That's an example of a competing publishing outlet that's instrumental in breaking the publishing oligopoly. If a paper went through a review process, it doesn't matter how it's disseminated, just name the review process.

That's how folks handled Plan S, but with the US joining the mandate game, it is not a sustainable endpoint. If all articles are on arXiv, then no need for subscriptions. If there are no APCs or subscriptions, there is no revenue; journals will close and there will be nobody left to facilitate peer review.

As someone said upthread, there is a true cost to publishing pre-publication peer-reviewed science, which has been estimated to be at the level of $1k/article (assuming free reviews). If we move to curation and post-publication review, the costs likely go down, but we need incentive structures to ensure ppl keep reviewing. And anyway this is now a pretty big transformation beyond just post to arXiv.

So ti is not simple, but it will be fascinating to see how this will play out, and I'm glad to see the (absolutely awful) system get shaken up.

I am afraid the way this will play out is forcing each PI to budget tens of thousands of dollars for publication costs per year. If you add indirect costs to that, then we will end up not being able to budget for our own salary or students. So, yes we will have money to publish stuff, but we will not have money for anyone to work on the stuff that we are supposed to publish.

The problem I see with this "breakthrough" that comes from the US government is that it seems it will be implemented (again) by non-scientists. It is not a coincidence that barely any of the US senators or representatives over the years has been a scientist. Scientists are simply too busy to regularly participate in policy making (to some extent this is understandable), thus leaving the implementation of policies to people that have no clue about how the economics of science work. Yes, scientists show up from time to time being part of a taskforce, but still in the long run they do not participate in policy making. That's my 2 cents.

[Kron3007]

The academic publishing system is broken.  This will not help.

Some open access journals are rigorous, others are not.  I have experienced both.  However, the more rigorous ones tend to also be very expensive.  The real issue is that these are for-profit companies syphoning funds from public research.  The grants are unlikely to get bigger, but will require larger amounts to support publishers.  I also never know how many papers a project will produce, science is not that predictable....

It would be better for the government would subsidize non profit publications (ie society journals) to allow them to publish open access articles for free, or cheap, instead of propping up the many for profit publishers who rely on donated time from reviewers and editors to drink from the public teat.

[Kron3007]

I am particularly bitter as I just received the $3000 USD invoice from Frontiers for an article that was recently accepted.  This should be good news, but that grant expired last year and now I need to cannibalize other grants to pay this stupid fee....

And for some reason I keep donating my time to them as well?!?!

[Puget]

I am particularly bitter as I just received the $3000 USD invoice from Frontiers for an article that was recently accepted.  This should be good news, but that grant expired last year and now I need to cannibalize other grants to pay this stupid fee....

And for some reason I keep donating my time to them as well?!?!

I will no longer submit to or review for Frontiers, since their system constantly spams with with requests that are increasingly indistinguishable from scam journals and clearly do not have a human editor involved. Send them straight to spam where they belong.

The policy statement makes it pretty clear they are planning to impose some sort of cap on APCs (reference to "reasonable fees", which 3k is not given actual costs). .

[AJ_Katz]

Digital Commons paid and maintained by our libraries is the best way to make science more open and accessible, without needing to pay for open access.  In fact, I liked publishing in jounals that were free if it was closed and then sent my post-publication article to my library and told them to post it on Digital Commons -- making it completely open to anyone, worldwide.  I wish more people knew about Digital Commons and that more libraries / universities invested in it.

[quasihumanist]

In math there are a few completely free journals - no subscriptions and no APCs.

In addition to the usual editors, they have some mathematicians who volunteer to do some of the less glamorous work of managing peer review.  Usually they have negotiated with their chairs to get credit (and maybe even a course release) for this work.  These journals don't do any typesetting or copy editing; papers look a little uglier and everyone knows if you can't spell.  Some university donates web hosting.  There are a few small annual costs running in the several thousands (total for the journal) and folks just put up their own money.

The problem is that they have collectively exhausted the volunteer pool for peer review management while publishing at most a quarter of the papers in their subfields.

[research_prof]

I am particularly bitter as I just received the $3000 USD invoice from Frontiers for an article that was recently accepted.  This should be good news, but that grant expired last year and now I need to cannibalize other grants to pay this stupid fee....

And for some reason I keep donating my time to them as well?!?!

I will no longer submit to or review for Frontiers, since their system constantly spams with with requests that are increasingly indistinguishable from scam journals and clearly do not have a human editor involved. Send them straight to spam where they belong.

The policy statement makes it pretty clear they are planning to impose some sort of cap on APCs (reference to "reasonable fees", which 3k is not given actual costs). .

I will never submit a paper or contribute to Frontiers as an editor or reviewer for the same reason. If you charge mandatory fees, then you need to pay me if you want me to review a paper or be the editor that handles papers.

They reach out to me almost every week with fee waivers or they invite me to review papers or manage special issues. I responded to them once asking if I will be paid since they charge mandatory fees. They said "no, you will not be paid", so I never responded again.

[Kron3007]

I am particularly bitter as I just received the $3000 USD invoice from Frontiers for an article that was recently accepted.  This should be good news, but that grant expired last year and now I need to cannibalize other grants to pay this stupid fee....

And for some reason I keep donating my time to them as well?!?!

I will no longer submit to or review for Frontiers, since their system constantly spams with with requests that are increasingly indistinguishable from scam journals and clearly do not have a human editor involved. Send them straight to spam where they belong.

The policy statement makes it pretty clear they are planning to impose some sort of cap on APCs (reference to "reasonable fees", which 3k is not given actual costs). .

I will never submit a paper or contribute to Frontiers as an editor or reviewer for the same reason. If you charge mandatory fees, then you need to pay me if you want me to review a paper or be the editor that handles papers.

They reach out to me almost every week with fee waivers or they invite me to review papers or manage special issues. I responded to them once asking if I will be paid since they charge mandatory fees. They said "no, you will not be paid", so I never responded again.

I also generally dont publish with them, but in a couple collaborative projects I have been a part of this is where they went.  Unfortunately, I am on the hook for the fees due to a previous commitment I made.

They also asked me to join one of their editorial boards recently.  I asked if there was any incentives or benefits of doing so.  They said they are "considering it in the future", but currently no.  I declined and gave them some constructive criticism.  Instead, am accepted to be an editor for my society journal, which I feel better about even though they publish through a for-profit entity as well.  Frontiers is one of the worst offenders IMO.