US: Immediate free access to government-funded publications (2026+)

[mfb]

Source memo
News report - the link goes to an archived version in order to have free access (...)

Apart from minor exceptions: If the research is at least partially funded by the US government, then from 2026 on publications have to be freely accessible without embargo period.

This won't matter much in particle physics - CERN already has a similar policy and everything is uploaded to arXiv.org anyway - but it will help in places where this is not standard. Medical results and COVID in particular are explicitly mentioned in the memo.

It looks like the US wants the individual organizations to fund this as needed:

In consultation with OMB, federal agencies should allow researchers to include reasonable publication costs and costs associated with submission, curation, management of data, and special handling instructions as allowable expenses in all research budgets.

That point could get interesting in international collaborations. Authors A, B, C from some other country want to publish in a journal with an embargo, author D from the US is not allowed to. Will they switch to a different journal? Will the institute of author D pay the whole amount to remove the embargo?

 

[Vanadium 50]

One of the great things about the US political system is that whoever is in charge, OSTP can make a mess of things.

I love how the provide a mandate and at the very end say that they haven't actually developed a process to meet it.

Comments:
1. The US federal government is a major reason that journals are expensive. It used to be that every research group got its own copy of the journals. Then the feds stepped it with "we're not going to pay for N copies - you should just get one and share it". And lo and behold, the cost per copy by a factor of N. I am sure there will be unintended consequences here. I just don't know what they will be.

2. Some of the grants at the time of the deadline have already been issued. There are some 4 and even 5 year grants out there. The three year awards are likely too late to change.

3. They hold up Covid as a shining light of success, when in fact a pile of crap got published.

4. "Scientific data" is vague and likely impractical. Many years back there was a FOIA request for the Tevatron data. All of it. That's tens of thousands of tapes. It was handled by asking the requestor to provide the cash for the blank tapes in advance.

5. The requirement that unpublished data also be made available will make a mess.

6. One can read this as requiring that a history of sources of funding be stored by researcher. That is not "this was funded by X" but "in the past, this researcher took funding from X". I can see this causing trouble, particularly in the health and social sciences. I also foresee some nasty politically-minded data mining in the future, "Professor X's work can't be taken seriously because in the past he collaborated with Prof. Y, and before that Prof. Y got support from Out-Of-Favor Group Z"

 

[Office_Shredder]

They hold up Covid as a shining light of success, when in fact a pile of crap got published.

As opposed to normal science, where a pile of crap would never have a chance of getting past peer review.

Remember the goal here is to find and socialize the truth. If a bunch of crap is published but mostly ignored, that doesn't actually matter.

 

[Tom.G]

If a bunch of crap is published but mostly ignored, that doesn't actually matter.

But it raises havoc with the Signal-to-Noise-Ratio.