"Overwhelmingly positive."

August 16, 2024 6:35 AM

"Since the cancellation, MIT Libraries estimates annual savings at more than 80% of its original spend. This move saves MIT approximately $2 million each year, and the Libraries provide alternative means of access that fulfills most article requests in minutes."

In 2020 MIT ended their contract with Elsevier. This is what happened next.

"As the savings from remaining out of contract with Elsevier continue to accumulate, MIT is interested in collaborating with other libraries to reinvest these funds in community-controlled open publishing initiatives that better serve both their own campus and the academic community more broadly."

posted by mhoye (42 comments total)

Now do Taylor & Francis

https://www.thebookseller.com/news/academic-authors-shocked-after-taylor--francis-sells-access-to-their-research-to-microsoft-ai
posted by lalochezia at 6:51 AM on August 16 [16 favorites]

The academic article publishing industry is disgraceful and I'm glad to see MIT leading the way. Elsevier sucks. as does T&F.
posted by bluesky43 at 6:59 AM on August 16 [23 favorites]

This is one of four unbundling case studies from SPARC. By all means check out the other three.
posted by humbug at 7:15 AM on August 16 [11 favorites]

I sadly just realized I had full digital access to my grad schools library for more than a decade after graduation . There have been a handful of times where I have asked an intern to help pull an article. I am sure if it was open access I could’ve used more of the academic literature in my job.
posted by CostcoCultist at 7:17 AM on August 16 [4 favorites]

I cannot read this without thinking about the death of Aaron Schwartz
posted by CMcG at 7:44 AM on August 16 [68 favorites]

The top tier institutions have enough capital and connections they could quite easily build their own. Tim Gowers & Advances in Combinatorics is a great example of academics taking back their own field.
posted by constraint at 8:16 AM on August 16 [6 favorites]

Big cheers for inter-library loan!
posted by doctornemo at 8:21 AM on August 16 [7 favorites]

There's a big difference between price and value. If you're not providing value, at some point people will stop paying the price.
posted by LegallyBread at 8:26 AM on August 16 [2 favorites]

I do wonder how much of the difference has been made up by researchers just using sci-hub, but of course MIT can't report "oh, we stopped paying, because all our researchers will just pirate it anyway, so no biggie."
posted by BungaDunga at 8:29 AM on August 16 [10 favorites]

I suspect it's made up by the ArXiv in many subjects.
posted by hoyland at 8:45 AM on August 16 [5 favorites]

I am not a librarian or a user of academic research so maybe I don't understand. If Inter-library loan is filling the gap, who is paying for that library's subscription. Someone or some organization is paying. So MIT is just glomming off of some other tax base or some other donation base? A library that has no books (papers) of its own and simply borrows through the inter-library system, is not really acting in the spirit of the process.

Whose money or whose infrastructure is MIT using if not their own or if they are not willing to pay? Seems to me that 80% savings should be shared among the groups helping MIT to achieve such savings.
posted by JohnnyGunn at 8:46 AM on August 16 [2 favorites]

If Inter-library loan is filling the gap, who is paying for that library's subscription. Someone or some organization is paying.
Yeah it's very strange. If other libraries started doing the same thing, they'll have no one to borrow from.

Sci-hub has stopped putting up new articles; since 2021, I think.
posted by dhruva at 9:10 AM on August 16 [1 favorite]

Good on them for this. Yes, there's lots of shameful stuff in the past as well, but in general the Institute does the right thing more often than not for an org of their size, reputation and history.
posted by drewbage1847 at 9:16 AM on August 16

MIT is probably a net lender in interlibrary loan. Most libraries tune their ILL so that it's close-ish to even-steven; they lend as much as they borrow.

But in actual fact a lot of Big Deal replacement happens not through ILL, but through paid document-delivery services.
posted by humbug at 9:19 AM on August 16 [6 favorites]

Big cheers for inter-library loan!

If more libraries follow suit (as I hope they will) expect academic publishers to attack ILL, just as trade publishers have attacked the right of public libraries to purchase and circulate e- and audiobooks.
posted by Horace Rumpole at 9:46 AM on August 16 [3 favorites]

As much as I loathe Elsevier, I feel the need to point out that Academic publishing is just as vulnerable to the same issue as the rest of the media landscape we inhabit:

If you aren't the customer you are the product.

If an academic paper isn't reader-pays, it's author-pays or sponsor-pays. And that comes with issues of its own.
posted by ocschwar at 9:47 AM on August 16 [3 favorites]

From the article, they are using a combination of ILL and on-demand reprint which costs them about $300,000/year currently, and is expected to increase in cost as prior articles are no longer as relevant. Life Sciences is the most impacted field and they anticipate that as other open publication sites proliferate, this may be alleviated.

I believe as far as free-riding on other library's subscriptions to Elsevier, they point to a future reinvestment in savings collectively from many libraries forgoing their subscriptions being applied to appropriate resources, so, I suspect they are considering funding one library's access so all can share. Who knows if Elsevier would support such a move, but if ILL is one of the terms of contract, I don't see how they can block it.
posted by drossdragon at 9:53 AM on August 16 [2 favorites]

If you aren't the customer you are the product.

I think the arXiv model shows that you can in fact have a system, at least for pre-prints, where you are neither customer nor product. Some things can exist outside a capitalist framework.
posted by ssg at 9:56 AM on August 16 [22 favorites]

Open Access definitely comes with its own set of problems - and basically, I'm forced to wonder about why we are in a place where the only two available models are "libraries pay huge, almost extortionate sums of money" and "authors pay huge, almost extortionate sums of money."

Does it actually cost that much money to put an article up on a web site, or is it just a pricing model of "we can charge whatever we feel like because you need us"? If it's not a pricing model of "we can charge whatever we feel like because you need us," then Elsevier (etc.) need to start making a better case for what value they're adding - otherwise a lot more libraries are going to start pushing back. (And they won't have a choice, because higher education budgets are so deeply in trouble.)
posted by Jeanne at 10:12 AM on August 16 [3 favorites]

Long after my last grad program ended (several decades), I find myself looking for academic material to read more than I ever did. I continue to be horrified at the sticker prices.

Anna's Archive / archive.org for the win, of course, but not everything is available. And given how little the authors of these papers are paid by the publishing behemoths, I don't feel like I'm ripping anyone off who matters.
posted by rabia.elizabeth at 10:15 AM on August 16 [3 favorites]

We've been publishing exclusively Open Access since about 2017, sporadically so before that. I work for an employer that has made that basic policy and so budgets for it. I recognize that that is not possible for everyone, but for those of us who are in that position, I think OA has to be a big part of the way forward.
posted by bonehead at 10:19 AM on August 16 [5 favorites]

By the way, the One Weird Trick for accessing articles for free, especially for people who aren't affiliated with a university:

Email the author(s)!

Not always reliable - because academics are almost all terrible at email - but most authors are flattered when there's anybody interested in their research and it's not like they see any money from you paying $40 to read that article via the Elsevier web site.
posted by Jeanne at 10:21 AM on August 16 [21 favorites]

If an academic paper isn't reader-pays, it's author-pays or sponsor-pays.

It was already sponsor pays author to generate knowledge, then sponsor pays again, either OA or subscription to access the knowledge they or others generated. It should be noted that Elsevier have a profit margin generally in the range 30-40% annually. So there is an additional option for your list, publishers charging less. Are they doing anything innovative to justify the level of profit they make? No, they are not.
posted by biffa at 11:05 AM on August 16 [6 favorites]

ocschwar: "If an academic paper isn't reader-pays, it's author-pays or sponsor-pays. And that comes with issues of its own."

In academic publishing, the researchers compete for the grants, pay for the work, write up the results, do peer review of write-ups, and pay to both publish the reviewed write-ups and to access the write-ups. For many funded studies (US federal agencies or EU) researchers are required to deposit an author copy in an open-access database even if they have already paid the publisher to make the paper open access (publishers LOVE to confuse researchers here, so many of my peers don't understand that "open access" is not "NIH open access")

In some cases (MDPI journals for example) the researchers literally have to do the typesetting for the final published version.

At the end of the day, I am left wondering: What the F are we paying the publishers to do? They don't actually publish in many cases (it's all digital) and the author copy is freely available on PubMed or EU PMC, public repositories that are supported by Federal funding to store published work for public access. Can you even argue that we pay publishers for figuring out who should do the peer review? In a whole lot of cases the journal makes us suggest reviewers! We're still stuck with archaic relics of the paper era - page charges, or extra fees for color images, as if black and white pixels cost less than color ones. We still end up with highbrow journals publishing microscopic images (dear Nature and Cell, if the paper is focused on figures or images, MAKE THEM LARGER THAN A POSTAGE STAMP). We still end up with universities plugging millions of dollars into subscriptions to journals that the university researchers provided all the content for in the first place.

I mean, my wife is a university library employee who handles digital licensing, so I guess the academic press justifies her job. But really, aside from that, what are we paying the publishers for? Why do we need them?
posted by caution live frogs at 11:41 AM on August 16 [12 favorites]

I think the arXiv model shows that you can in fact have a system, at least for pre-prints, where you are neither customer nor product. Some things can exist outside a capitalist framework.

Sorry, no.

ArXiv has a good scalable system for collecting and disseminating PDFs. But the people uploading to the ArXiv still have bills to pay.
posted by ocschwar at 11:41 AM on August 16 [1 favorite]

"At the time of his death, Swartz, who had informal connections to the MIT community, faced 13 federal felony charges relating to his downloading of more than 4 million academic journal papers from the online archive JSTOR, or about 80 percent of the JSTOR library. The downloading was carried out surreptitiously using a laptop computer that was left in a basement wiring closet in an MIT building, physically connected to the MIT computer network."

"ndividuals including Swartz’s father, Robert — who did undergraduate work at MIT and is a consultant to the MIT Media Lab — met with MIT officials during the course of the prosecution and asked MIT to support efforts to have the charges dropped or to secure a plea bargain that would not include jail time. Two MIT faculty members also urged the administration to make such an appeal to the prosecution. MIT considered these views as well as those of faculty who expressed opposing views in deciding neither to press for prosecution nor to advocate on Swartz’s behalf."

link
posted by mecran01 at 11:42 AM on August 16 [5 favorites]

ArXiv has a good scalable system for collecting and disseminating PDFs. But the people uploading to the ArXiv still have bills to pay.

As caution live frogs describes, the present commercial academic publishing model does not help those people pay their bills at all. If anything, preparing a papers to meet the requirements of the various commercial publishing houses and interacting with their systems is a nontrivial cost for the people who do the actual heavy lifting of publication.

That's not to say there aren't costs borne by the commercial publishers, but literally none of those costs include supporting "the people uploading" the papers.
posted by multics at 11:47 AM on August 16 [9 favorites]

Jeanne email the authors I've done that occainonally for twenty years, sometimes they've sent me other things as well as what I asked for like 'I never got this published, it might be useful'.

Interloan is the best thing. My librarian told me I'm the highest interloan requesor in their system.

The one thing not borrowable through interloan is Building & Engineering Standards. This is another rort similar to academic publishing. All the work in the standard is paid for from tax, and somehow the product gets owned privately - and prices are steep. Many build contracts stipulate use of the Standard and the content is unknown without buying, another barrier for small contractors.
posted by unearthed at 11:56 AM on August 16 [4 favorites]

expect academic publishers to attack ILL

Honestly, I've been expecting this for a while. ILL is a library superpower, the basic function of libraries on steroids.

Now, years ago - oh, 2006? - I asked the North American head of Oxford University Press what she thought of ILL. This was in a big academic forum. "It's evil," was her immediate reply.
posted by doctornemo at 12:48 PM on August 16 [11 favorites]

Does it actually cost that much money to put an article up on a web site

No, but that's not all they do, long term archiving 'in perpetuity' is considered a core function of journal publishers. How much does it cost to run a website for 10 years? how much for 1000 years? A good way to think about this is to imagine the publication fees as a pot of money that will be invested with the profit made over the years being used to maintain the archive.

I still think they are charging WAY too much for what they provide, particularly given that the Internet Archive seem to be left picking up the pieces for many abandoned journals.
But it is not entirely fair to compare their service with throwing a bunch of PDFs onto megaupload and calling it a day.
posted by Lanark at 1:17 PM on August 16 [1 favorite]

But really, aside from that, what are we paying the publishers for? Why do we need them?
posted by caution live frogs at 11:41 AM on August 16 [

This. Editors get paid very little, occasionally a part time secretary is paid for, reviewers get paid nothing, authors get paid nothing, universities get paid nothing (at least in my world). It's a very stupid system where doing reviewing or editor work for somebody else (a publisher) without pay is supposed to be prestigious.
posted by bluesky43 at 2:04 PM on August 16 [4 favorites]

How much would it cost to create a non-profit version of Elsevier? I'd think Harvard could afford to do it all by themselves, but if the top 20-30 research universities in the world went in together, I'd think the cost would be no problem.

Heck, how much would it cost to buy Elsevier and turn it into a non-profit?
posted by straight at 3:35 PM on August 16 [1 favorite]

Hmph. I guess Elsevier's market cap is close to $48 billion.
posted by straight at 3:37 PM on August 16

SCOAP3, the Sponsoring Consortium for Open Access Publishing in Particle Physics, has played a major role in opening up access to research and should be mentioned in this context.
posted by heatherlogan at 4:57 PM on August 16 [4 favorites]

Academic publishers like Elsevier provide zero real services, just evil parasites. And nobody "uploading to the ArXiv" benefits from them, ocschwar. Your own link proves they do not provide long term archiving, Lanark. If you want long term archiving then write your paper in latex in a github repository.

I'm confident Elsevier invests largely in "hookers n blow", given the folks I know who work there, but even if not then climate driven inflation would destroy any attempts at long-term financial savings.

Academic societies like IEEE have high prices too sometimes, but frequently they're paying for real editing work. We should explore alternative models for "auditing" academic work, like maybe the author pays for the auditors' time, after which the audit becomes public.
posted by jeffburdges at 5:56 PM on August 16 [1 favorite]

If an academic paper isn't reader-pays, it's author-pays or sponsor-pays.

Side note but the last time I published a paper, I had to pay a fee because it was technically an advertisement (and it wasn't open access so then readers also paid (https://scholarlykitchen.sspnet.org/2014/01/09/history-as-caution-when-paid-scientific-articles-were-legally-considered-advertisements/). It has been a minute so is this still a thing?
posted by Tandem Affinity at 10:07 PM on August 16

The one thing not borrowable through interloan is Building & Engineering Standards.

This is a massive problem in many industries, particularly design and construction. Every design office I've ever worked for has had a huge library of bootlegged standards of varying up-to-dateness, which kind of works, but I've also seen architectural docs produced that use a version of a standard that was obsolete five years previously and have to be redone.
posted by deadwax at 12:38 AM on August 17 [1 favorite]

Tandem Affinity, I’m not sure if this is what you’re asking, but these days when I publish a first author paper, the grant I’m on has to cough up around $3500. We’ve moved from a system where it’s free to publish but expensive to read to the converse situation. I don’t honestly know what’s worse, for people from lower-income nations, for new faculty without federal funding, etc.
posted by eirias at 4:17 AM on August 17

This is a massive problem in many industries, particularly design and construction

I realize you're not in the US, but just in case it helps, note that the US structural steel standards are free.

It's more than a little irksome that ACI never followed suit -- IMO if a standard is incorporated by reference in legal requirements (eg: building codes) then that standard should be freely accessible. Of course, that doesn't stop people from incorporating dumbass requirements in their contracts, I suppose (I cannot tell you how many times I've seen contracts requiring adherence to a document that was never written as a standard, which was only ever intended to be an informational discussion, and which you cannot therefore "adhere" to in any meaningful sense, but I digress...)
posted by aramaic at 11:58 AM on August 17 [1 favorite]

Mod note: [btw, this post has been added to the sidebar and Best Of blog!]
posted by taz (staff) at 1:54 AM on August 18

Elsevier is a division of RELX, a huge money-grubbing information company that also owns LexisNexis and other subsidiaries.

As it happens, I work for one of the few RELX subsidiaries that is unionized. We are a very profitable business, but company management doesn't want to share any of those profits with the workers who generate them.

We've been bargaining with them for almost 2 years over our next union contract, but they've been dragging their feet, and making proposals that would actually leave us worse off than we were under our previous contract.

So today, we have publicly announced that if management doesn't get serious about contract bargaining, we will go on strike next month.
posted by Artifice_Eternity at 10:29 AM on August 19 [7 favorites]

what are we paying the publishers for? Why do we need them?
I would have thought we needed them once upon a time, when papers were published on dead trees and sent around on carts dragged by ponies or something after the publisher received a hand-written order accompanied by a cheque or a bunch of notes and coins tightly wrapped in the envelope. I don't know what they do now, when the dissemination of papers and collection of fees is largely automated.

The one thing not borrowable through interloan is Building & Engineering Standards. This is another rort similar to academic publishing. All the work in the standard is paid for from tax, and somehow the product gets owned privately - and prices are steep.
It hadn't even occurred to me that ILL use of standards would be blocked, but the whole private ownership of standards everywhere annoys the fuck out of me. After my taxes and various contributions from industry pay for development of the standards, they should be publically and freely available. It annoys me even more that, often, students have to purchase their own expensive copies because libraries can't afford to own enough copies to satisfy demand.
posted by dg at 3:37 PM on August 19 [1 favorite]