I think this is a really good move. About bloody time to. I shouldn't have to pay twice to read research papers I helped fund and I certainly shouldn't be paying a private company for it.
EU is saying that when and if you choose to publish papers based on publicly funded research you must ensure those papers will be publicly available. This means you must budget for any cost associated with it but since your budget are public funds it's basically going to be paid for by the EU in most cases.
The "when and if" means that you can choose not to publish or you can choose to publish after you have sought protection (patent). You can also, as a publisher, have a wait period before research papers become "open access" to, I don't know, be an asshole I guess.
The devil is in the details: one possible outcome is that researchers must pay journals for open access publication, and this could change "'small science' fields" into 'big science' ones. I.e., a good but not grant funded researcher will need to find a 'rich' colleague to pay for the publication fee. Whether the 'rich' person should be a coauthor is then of course a matter of negotiation.
> I shouldn't have to pay twice to read research papers I helped fund
Most researchers will publish in journals that publish open access for a fee and the money for the fee has to come from somewhere. So in the end the public will spend as much money as they are doing so far.
Maybe, but currently people outside of universities have a really hard time getting research papers (legally, anyway). So even if the costs stay about the same, a lot more people will have access to the results.
True, but this at least sets the incentives right to create a system outside of rent-seeking publishers. Journal editors are unpaid for the most part anyway, might as well make that a community effort.
Maybe, since publishing an open access paper is (currently) more expensive for the authors, researchers will end up publishing less often (and possibly more significant results), which is not necessarily a bad thing given the bloating state of research literature nowadays.
On a related note, a colleague of mine had this idea of an annual publication cap (3 papers) for researchers. Less burnout, less "superstar" researchers that "contribute" to 20+ papers a year (except for the fact that they have an army of Ph.D. students slaving off for them).
Some of the work I've done that I like the best have been side musings that wouldn't make the "3 paper" priority cap, but have never the less been useful to me and others.
Also, you'd probably get a surge of shit submissions in November from everyone who didn't get their three in.
As long as universities and funding bodies - i.e. your potential employers and the people who give you money - judge you by the quantity of your publications and the impact factor of journals you publish in I'm afraid this won't happen.
We'd need a different way to judge someone's scientific output which is definitely hard - do we start to check Twitter or blogs for scientific outreach? Do we judge scientific "novelty" (of which maybe only the author herself/himself is the expert in, so only s/he can properly evaluate)? Do we start to evaluate preprints (biology doesn't do this much)? Many open questions, and not much is happening in terms of change.
(Edit: This is, by the way, one of the reasons why there is so much more Arabidopsis than Wheat research - Arabidopsis is easier to manipulate and grows faster, so you can get more publications out for the same amount of work)
>As long as universities and funding bodies - i.e. your potential employers and the people who give you money - judge you by the quantity of your publications and the impact factor of journals you publish in I'm afraid this won't happen.
Having seen this evaluation process from the inside, it would still work if everyone was limited to ~3 papers per year. In my experience sheer quantity doesn't land you a job, because quantity usually is a detriment to quality. More important factors are the prestige of the journals you published in, whether those journals are relevant to the position, how consistent your publishing record is (is there a five year break?, have you even stopped publishing?). In CS "best paper" awards from conferences also have significant weight.
Add to that that fewer papers make it easier to asses their quality, and being limited to three high-quality papers a year isn't a detriment to hiring.
I know many may disagree but I'm curious about this movement of 'publicly funded should be publicly available' wouldn't this infringe on national competitiveness?
Like India nor the USA didn't funded these studies but...those people will have access to these research papers? Talk about a free lunch. And potentially demotivating to other countries, as they could be like 'well, EU did it...let's lower our investments in research'.
It's my opinion that there should be a gateway and one must us their state ID or something. And if you are at a public library, easy access. But not public to the whole world.
Edited: I'm heading to sleep and I'd like to discuss this to those that want to. I'll return to the thread in about 24 hours as I have a busy day tomorrow.
What would be the reasoning behind dropping funding? That you can read the output of another country instead of duplicating work?
I doubt heavily that would happen but even if it did that'd be a good thing.
> It's my opinion that there should be a gateway and one must us their state ID or something.
A full, international login system for every person in the EU just in case someone else in the world might want to read a paper? That sounds expensive and complicated compared to just using the web as it exists.
Those other countries already have universities that have buy access to the journals that the papers appear in. Research isn't secret. It's just expensive to access for normal people.
This makes no sense. A "tide" just happens, it doesn't get "created" or "caused" by anyone, which is the case for public scientific research. A crucial difference.
What's really going on here is that some individuals aren't exactly happy/eager/willing about giving away, for free, things that they have funded personally funded (through taxes). This is a perfectly reasonable opinion to hold.
It's the same reason we have things such as patents. So as to reward individuals and groups that create something new. They then get exclusive rights to the profits/benefits that such an invention creates. Yet somehow if it applies to nation states, that no longer holds?
This makes no sense. A "tide" just happens, it doesn't get "created" or "caused" by anyone, which is the case for public scientific research.
It's an analogy, and all analogies are somewhat flawed. The point is, increasing the overall pool of scientific knowledge throughout the world makes life better for everybody in the long run. I mean, if somebody in Botswana invents the first viable fusion reactor, and publishes a paper for free, maybe Botswana as a country won't get "oil country rich" off of it. But as cheap, plentiful electricity spreads around the world, everybody, everywhere, will have a better life as a result. That's what I mean by "a rising tide lifts all boats".
It's the same reason we have things such as patents. So as to reward individuals and groups that create something new. They then get exclusive rights to the profits/benefits that such an invention creates. Yet somehow if it applies to nation states, that no longer holds?
You're asking the wrong person, because I am opposed to patents in general. shrug
That doesn't make any sense... How does having one screwed up system justify another? Maybe we should start paying twice for groceries too, because we pay twice for toll roads and research papers...
My point is, whether or not you pay a fair amount doesn't depend on how many times you pay. Let's say your fair share is $10; $3 goes via taxes and $7 via the other mechanism. Here's a couple of scenarios where it's a good choice:
1) The new road is 60% in the public interest and 40% in the interest of its users, so 60% is paid with taxes and 40% via tolls.
2) Society feels it's reasonable that all children, regardless of family income, should have educational resources. 90% of school is funded by taxes and 10% funded by fees paid by students' families (lunch, field trips, books, etc.).
Both of these are reasonable, as long as the payment is actually covering costs - but in this case the first payment covers some of the costs, and the second payment increases the cost. That is, as a taxpayer I'm paying 1) for the research and also 2) for the extortionate access fees for universities, and as an independent researcher I'm ALSO paying 3) for my own private access to the results. The problem is that the funds from 2 and 3 do not fund the research, they fund the parasites that make 1 more expensive. I have no problem with a two-part payment for things where private benefit and public interest have the relation you describe, but when the secondary payment is a) entirely unnecessary, because the entirety of the work done is funded by the first, and b) feeds disgusting evil parasites that do nothing but harm to society, it's clearly immoral.
> The problem is that the funds from 2 and 3 do not fund the research, they fund the parasites that make 1 more expensive
Speaking generally, paying for someone else's profit is essential to the free market. Without profit, much less would get done.
Speaking about this case in particular, knowledge, at least much of scientific knowledge, should not be sold at whatever prices the market will bear, which is what these publications seem to be moving toward. Imagine the prices you could demand for knowledge of immunizations, relativity, DNA, law, history, etc. (if they weren't already publicly available). Imagine the disruption to science, to the advance of knowledge, and the incredible cost to humanity if only a few with enough money could access these ideas.
What does this mean in practice? Will the EU force the publishers to make them freely accessible? Will the EU only allow scientists to publish in open journals? Europe indirectly funds a lot of research. This will bleed into the workflow of scientists worldwide.
How this happens right now with NIH funding is yes, the government basically forces publishers to make things freely accessible - the paper is going in a public repository. Journals that publish lots of NIH funded stuff handle this for the author, while journals that don't tend to make it the author's problem.
And yes, if an author doesn't comply, it can be a problem for future funding.
>> Will the EU force the publishers to make them freely accessible? Will the EU only allow scientists to publish in open journals?
> How this happens right now with NIH funding is yes, the government basically forces publishers to make things freely accessible - the paper is going in a public repository.
I think you're either mistaken, or you didn't understand the difference between the two original questions. The NIH doesn't force the journals to do anything. The NIH makes it a requirement for researchers they fund to make all publications open access (after some time period), effectively prohibiting the authors from publishing in closed-access journals
For biomedical research supported by the US NIH they require it to be posted to PubMed central (http://www.ncbi.nlm.nih.gov/pmc/) within 6 months of publication. Since so many biomedical papers have at least some NIH funding the journals all went along with it.
I just went through this discussion in an EU funded project, and the consensus was that complying is up to the individual partners. In our case all academic partners decided to comply by posting preprints on their websites (contentwise identical to the accepted version), which is allowed by both ACM and IEEE.
I really dislike this approach. Finding the free version is harder and the links are much less likely to be permanent. I've found a huge amount of data, code and extra material is simply missing now. In 20 years are you confident someone will maintain the same url and content?
Can you afford the open access option with the journals you want to submit to?
I don't know, but speculating on what seems most reasonable, I wouldn't assume that publishers will have any burdens forced upon them apart from losing exclusive control. I'd guess that it means publishers simply don't get exclusive copyrights.
Who has the burden of publishing the free access could be the researchers themselves, or the universities, or the government, I think there are a lot of options besides the current private for-profit publishers.
I also wouldn't assume that much will change in practice immediately -- while the research and data might be available for free, some publishers do currently provide some value above and beyond access. Curation, editing, indexing, quality control, prestige, print versions, conferences, etc. - all things that free access doesn't necessarily come with.
I can't speak for the EU, but since my work last year was NIH-funded, we made it freely available via arXiv (http://arxiv.org/). This seems to be common practice.
The work was done in a professional capacity, so I can't answer your question. Both academia and industry are fraught with pretense, so I wouldn't be surprised if what you ask is true. I can only recommend that you do your best work; if it stands on its own, no amount of pretense will overshadow its value.
I bet, in practice, this will mean some researchers will avoid public funds as it will come with overhead they do not want (publishing rules). But for most this will be a big win.
First of all: Researchers won't avoid public funds, because there's never enough money.
Second of all: Most researchers HATE academic publishers and being forced to sign over their copyright to a paywall in exchange for nothing. But institutional evaluation criteria force you to do so.
EU banning non-open publications will force publishers to allow open access OR force universities to revise their criteria to accommodate the inability to publish publicly funded work in non-open journals.
I think all of my colleagues think this is a great thing. As do I. Elsevier can go blow a goat with their copyright assignment BS.
I'm into the history of logic at the moment. As in not the filtered oh-so-neat version that is presented at third-level but the nitty-gritty twists and turns from Ancient Greece (go team Aristotle!), Indian, Arabic, Scholastic, Medieval, Pre-modern, the rise of symbolic/algebraic logic, Kant/Hegel/Trendelenburg, on to Frege and then the multiplication of logics (modal/tense/deontic, …) in the 20th century ending with type theory.
Recently Elsevier contracted top logicians to produce a Handbook in the History of Logic. It runs to 11 volumes, each volume north of $300. So folks, if you want to get acquainted with the complete history of what is supposed to be a formalisation of rational thought it can be yours for $3,300 or thereabouts. There is no reason this should not be in the public domain, it's too fundamental and central to history of ideas. I could buy two cheap second-hand cars for that price. (Personally I have used SciHub to have a look and I don't mind saying so.)
Now I would like to think that top logicians are not complicit in this scam, but I fear they are. I have a burning desire to crowd-source a website that provides the same information. >95% of the papers/texts are out of copyright. These academics already have very lucrative salaries and grants. I don't think we can point the finger exclusively at the Elseviers of this world, they have had a little bit of help along the way. Here in HN we have a bit of an open-access echo chamber. Some academics are very happy with the status quo, and who wouldn't be with those prices? I think it's _insane_.
I'll post names and links of the details if people are interested.
> Recently Elsevier contracted top logicians to produce a Handbook in the History of Logic. It runs to 11 volumes, each volume north of $300. So folks, if you want to get acquainted with the complete history of what is supposed to be a formalisation of rational thought it can be yours for $3,300 or thereabouts.
Unless you're leaving out important details, that strikes me as a completely reasonable and uncontroversial thing for Elsevier to do. If they're going privately fund the creation of an original specialist work, I think they can charge whatever they want for it. That's worlds away from how they they've subverted the primary scientific publishing process for private profit.
>Ahahahahaha.
>First of all: Researchers won't avoid public funds, because there's never enough money.
What if though, they actually get their money from private funding, like a ton of research projects do. You cast this of as a thing even weird to even consider, but this just implies you don't seem to have found yourself in academia ever (or if so, without any actual colleagues).
I certainly agree that no researcher would prefer a non-open journal vs an open journal, but if the former means a stand-still of their career and the latter means an advancement, then the latter is far more valuable for them.
>What if though, they actually get their money from private funding, like a ton of research projects do.
You'd be surprised how few those are. Even in the biggest private/public companies there are tons of research and tech subsidies, tax-cut deals and other schemes involved which end up in tons of government (our) cash. Those pay for lots of things, including part of their R&D.
I have never, in my entire life, avoided funding that would also make my work more accessible to a broader audience. I don't know any other academics who would do so either.
Responses like yours aren't an unrepresentative sample of academics - the ones who read HN and might have more interest in openness of information? A serious question.
Thanks for responding. Why be "actively skeptical of open data requirements"? From the outside, they seem to have very little downside and occasionally valuable benefits. As always, I suspect the outsider is blind to some important issues.
1. For the work I do, what "The data" is often somewhat ambiguous, and depending on how you define it, falls anywhere from "That's not mine to give" to "This is useless for what you actually want, but obeying the letter of the law".
2. It makes work with minority communities harder - they have a very, very poor history with "Consent to this, and researchers will do whatever they please with your information without consultation."
3. Sometimes privacy issues come up as well with medical data.
4. I think it gives us a false sense of security. "I re-ran their code with their data and got the same answer" is markedly inferior to "I attempted to replicate their study in a different population", but we often do the first and then pat ourselves on the back about it.
Sometimes you are using data that shouldn't be released e.g. for privacy reasons. See the recent controversy about research including data from a dating site.
Under the Horizon 2020 programme, open access publishing (green or gold) is already mandatory today for all EU/EC funded projects. Grant proposals must incorporate open access fees in order to be accepted. It's not much money for most project consortiums as the EU tends to fund big projects with budgets of ~10 million € and more as the norm (there are smaller ones, but ~2 million is really as small as it gets as far as I know). So a couple of 10K€ won't make that much of a difference. If it would become mandatory for all publicly research in the EU, the problem would be quite different. There are really small grants (50K€ and smaller) and if you have 2 publications out of that grant, you now pay 10-20% of your grant money for publishing fees.
Gold open access is where the authors pay the journal a fee for open access publishing. I don't see what's wrong with that and actually I think it's the best approach. The journal offers a service (editing, hosting, ...) and charge for that. Very simple.
The main problem is that the open access fees are so insane. Publishers currently have ridiculous profit margins, and of course they're going to try and maintain them.
Unless some US/EU regulations kick in and set a reasonable max limit to open access fees, the situation is just as bad (perhaps even worse) than the status quo:
Now, researchers from "poor" universities, small research institutuons, third world countries etc. don't get full access to the published literatute. But at least they can publish freely. If gold open access becomes the standard, they won't afford publishing anymore.
> The main problem is that the open access fees are so insane.
I'm confident that the market will take care of this. If there are people who can't or don't want to pay $3000 for an article and the profit margins are huge, there is room for journals who offer a better deal.
I don't think so. The problem is that the market is completely dominated by a few big publishers, and that it's a really hard market to disrupt because everything is about prestige.
If it was so simple as service -> $$$, they should be paying up the referees, which is the only task that requires rarely found scientific skills.
I seriously can't see how editing, typesetting and hosting can host that much, or can make them a respectable journal. No-name journals (for which top scientists don't referee) have that too.
BTW, Americal Physical Society journals don't even do that and outsource typesetting and editing.
This is one of those things thats just so common sense, its hard to imagine a solid argument against it.
Hopefully it will encourage the US to follow, although I'm sure publishers will dig their feet in and make up some bullsh*t reason why it would be a bad move.
"From 2020, all scientific publications on the results of publicly funded research must be freely available." is not "All European scientific articles".
I've got at least three papers that wouldn't fall under that heading.
It's a matter of perspective. I'd argue most European research falls under this. If you work at a public university and are paid for it and do research during that time and publish it that should be covered.
It would be pretty lame if only stuff that was explicitly funded in a research project would be covered.
Research done on unfunded projects as part of a hard money position is super-tricky, which is why for a lot of things the restriction is funded research projects.
For example, if your position is at a public university, but funded by an endowment from a private foundation, is it covered?
If your position is at a public university, you probably work at a publicly-funded building, using publicly-funded electricity and heating, and publicly-funded office material. So I'd say it should be covered.
The building I work in was built using money from two foundations. The electricity and heating, as well as office material, theoretically comes from grant indirects - which at the moment for me come entirely from private industry.
Seems excellent but I'm afraid the EU will just pay a sweet bundle of money to the publishers for the privilege instead of making it a law.
Either way I hope it also extends to the member states level eventually. The circus of avoiding conflict between EU funding and member state funding is often quite amusing so this will be interesting.
It feels almost ethically self evident that any research funded by citizens in any way should be made available to said citizens (and that's more or less 100% of research in Europe). Lets step in that direction.
But they are already cashing in for publishing anything, with additional small charges like few hundred bucks for going over the page limit and big charges like a couple thousand for making the article openly accessible.
" big charges like a couple thousand for making the article openly accessible."
Yes, if you don't want them let them make money off your work, they won't subsidize publication and you'll need to pay. No law can change basic economics.
Fuck them. Publication costs nothing. All they're selling is prestige, and the cost of that will be dramatically adjusted downwards if there is political will to do so.
Yet somehow there are so many platforms that will happily subsidise open publication without even taking the initial thousand bucks for letting me in, like Medium, Wordpress, Blogger...
Seriously, the only reason why academic publishers can charge anything is their established position, where they are a proxy for gauging researcher's output by administration. Peer reviews are done by volunteers, discovery is provided by Google Scholar and everybody uses PDF anyways (usually printing it out, still orders of magnitude faster than fetching from a library).
Those have nowhere near the costs of academic publishers, which are usually in the thousands per paper. There's a reason papers aren't published on medium.
When? Open access after 12 months has different cost than immediate open access, and, at the pace of today's research, is not open access at all. And ERC funded research already requires open access publishing anyway.
Why not demand publishing to open access/nonprofit journals instead? Why does EU have to pay elsevier $2000-5000 per article?
To contextualize the discussion on "open access" (the practice of editors that ask for a sum of money in order to make the paper available). It costs a lot: about 1k for ACM venues. Currently I have to pay for this out of my own budget, and I don't have that kind of money (I'm a PhD student, and in addition to my scholarship (1.8k€/month net of tax, in Belgium), I get a 5k€ budget for two years that must cover material and travel -- I can also get about one additional travel grant per year).
Publisher have about zero added value except fixing the odd latex issue. Actual editorial work is assumed free of charge by professors. Other venues such as Arxiv and CiteSeerX do a better job of distribution for free. Science publishers are leeches in the system and should be removed.
I think the EU a measure will precipitate the current movement against such predatorial publishing practice. Already, publishers are walking on eggs. They move against sci-hub but they'd never dare to go against the common practice to just host everything one publishes publicly, license notwithstanding, something almost everyone does in the CS field -- maybe they say it's a "draft"... (meaning there isn't a copyright notice mostly).
I'm not as familiar with European journals, but if this happened in the US (more or less forced open access) it'd possibly mean that every time we want to publish it would cost researchers ~$3000.
Hopefully this (and similar) legislation comes along with a statement regarding who's supposed to cover that cost. Because some of us avoid that price tag by publishing in non-open access journals.
I haven't read the specifics of this, but typically these kind of mandates come from funding bodies to cover the work they fund. So they're expecting that publication fees will be a budget item in grant applications.
To top it off, I would love that open access is tied to open data and strong councils for standard setting of such formats in a particular domain. The current swath of approaches to open research data is babylonian.
Wait, are you saying that politicians should decide which data formats we're supposed to use when we make research data public? Sounds like a recipe for disaster. I think it's great that we have many competing approaches to open science at this stage. Over time, each research field will identify an optimal approach for their purposes and that approach will win. I just hope that the winner is not going to be GitHub which is becoming more and more popular at least in my field. GitHub is a terrible platform for open science because what happens if they shut down or change their pricing or ...? GitHub is just not made for open science and I think an open science platform should not be run for profit. Also, more technically, git histories can be completely rewritten which means that GitHub would not be a trustworthy historic record.
Bigchaindb is tackling generic open data repositories from a technical pov. As has been said below, the larger part of the sustenance prob is economical of course.
Many people leave academia after a post-doc or two, most actually. These people will not have any incentive to manage their research data for the rest of their lives and the data will eventually be lost, unless GitHub continues to exist in its current form for the next half century, with is rather unlikely. For this reason we need dedicated research repositories that are managed by publicly funded non-profits. Zenodo.org is one example but I'm not sure if they have long-term funding.
The whole idea of git is that it is decentralized. So, even if GitHub instantaneously shut downs and erases all data (unlikely, they'd probably give a few months for people to move their repos), people's copies of the repo can be uploaded to servers.
Not in any legal sense - it applies to new, publicly funded research from 2020, not existing papers. Whether it legitimises Sci-hub in popular awareness is debatable.
If this actually pans out, it could be incredible. It always saddens me to get a link to a scientific paper and to hit my head on the paywall.
There's of course the "(partially) publicly funded" qualifier, but I don't remember too much privately funded research in the journals that I read as a CS undergrad, so hopefully it won't matter too much. (I guess other areas such as Chemistry, Biology might be more prone to falling under this qualifier?!?)
What's even more frustrating is when you're a student at a university and you still can't get access to a paper because your library doesn't have a subscription to that specific journal. arXiv was the best thing to happen to physics papers, and SciHub is doing the same (but less legal).
Interestingly, papers by US government employees are already in the public domain. Journals have no problem publishing those. I think one direction this could go is that public universities could insist that their researchers put their work in the public domain.
Journals don't need copyright to function. There's still value in editing, peer review, and distribution. They're just being greedy by insisting on it.
Slightly related, I just took part in a workshop of experts who are tasked with advising the commissioner on policy decisions for making open science happen in the EU, I've written up my notes: https://medium.com/p/7691802b543a
Does anyone have any insights regarding the "foreign startup visas". Does that legislation have a name I can use to track progress?
It's quite annoying to have to apply for week-long business visas every time I have to travel to Europe. Especially in the summer, when it takes 2 months just to get an appointment with the consulate..
Next step: As they have an exception for "IP" (intellectual property rights like patents) we must promote the logical consequence. All publicly funded research must mean all "IP" must be made available to anyone for free.
Nope. You can seek protection (patent) before publishing your results. See page 3 in the linked pdf. The paragraph begins with "misconceptions about open access.." in bold. :)
Has it been stated anywhere that they won't allow a temporary embargo after publishing? Like how some funding agencies allow papers to be behind a paywall for 12 months and only then require that it be open access?
I can see the evil people behind the TTIP getting in the way of this. Before this goes into effect, the US will execute order 66 and take all the articles down.
Or they can start pulling arguments from TTIP provisions and get away with blocking parts of the initiative. Now it's good this comes from the Netherlands (Elsevier is headquartered in the Netherlands).
But wait, scientific articles are hoaxes made for tax money, so now the lies are freely available to anybody!!! We need to cut more taxes on billionaires and publicly fund more oil projects so we can get these inconvenient science articles out of our way. Where's my next iPhone?!
Some public funding of copyrighted art work purchases would be also useful, if that's the only viable alternative to decreasing copyright age to some reasonable 20-30 years.
My thoughts are for those that are not with us anymore that helped make this slowly happen... I say thank you, but if I could choose I'd rather have them back and get to this results a bit later.
It's nothing new that in this field sometimes people have a tendency to put their ideals of progress before their own life.
No matter how strong is your love for your world changing ideas, YOU as a person come first.
Unless (but you could also argue about that) without your sacrifice the whole world population would be at immediate risk of extinction..
Of related interest, I"m just back from a workshop on Open Science from a group that advises the commissioner. I was an invited expert in to the workshop. This is part of the route through which announcements like the one linked here get made. I've written up my noted https://medium.com/p/7691802b543a
I struggle to see how or why a candidate who's proven to not necessarily have a stance on _any_ issues would want to do something like this to appeal "to the science crowd".
In fact, that's exactly the crowd I'd consider the hardest to win by a candidate such as Trump. You know, critical thinking being a thing and all.
Fair enough, I should have known that putting the words "Trump" in the first 100 words ofa post would be auto downvotes. I meant it to say that this is one way that candidates, particularly those who have most estranged the scientific communities, could shore up or make amends.
EU is saying that when and if you choose to publish papers based on publicly funded research you must ensure those papers will be publicly available. This means you must budget for any cost associated with it but since your budget are public funds it's basically going to be paid for by the EU in most cases.
The "when and if" means that you can choose not to publish or you can choose to publish after you have sought protection (patent). You can also, as a publisher, have a wait period before research papers become "open access" to, I don't know, be an asshole I guess.
Reference (pdf): http://ec.europa.eu/research/participants/data/ref/h2020/gra...
Edit: The little pdf warning.