Digital preservation, access control, and scholarly needs

April 16, 2024 9:48 AM

'So, I hope CLOCKSS does have a complete digital copy of the journal, but the question is: will CLOCKSS make it available? This all depends on CLOCKSS assessment of whether a “trigger event” has occurred here.' Ross Mounce, commenting on the disappearance of recently-concluded chemistry journal Heterocycles from the web more than four months ago, raises questions about nonprofit digital archive CLOCKSS ("Controlled LOCKSS (Lots of Copies Keep Stuff Safe)"), and similar "dark archives" that preserve research journals in a kind of escrow. (CLOCKSS has, so far, released "66 journals comprising 13,000 articles" into Open Access.)
posted by brainwane (14 comments total)

CLOCKSS is presumably trying to be publisher-friendly and respect the copyright and licensing terms of the articles? That's nice and all but also the source of the problem. Right now whatever copies of those articles Sci-Hub happens to have cached are way more useful than the ones CLOCKSS is dithering about whether to release.
posted by Nelson at 10:02 AM on April 16 [2 favorites]

Previously, it was librarians whose duty was to preserve hard copies of manuscripts and make them available, Eve notes, but that duty has transitioned to publishers as journals have switched to a digital format.

‘It’s become the expectation that publishers should be the ones who preserve material,’ Eve says. ‘However, that message has got slightly lost in that transition and not all publishers know that they have a responsibility to ensure the continued accessibility of the material they publish.’

I'm not sure 'innocently unaware publishers' is an accurate or useful take, given that there don't seem to be any consequences for publishers uninterested in upholding this responsibility.

I wonder if libraries paying for access would be able to sue over this; they're in a bind because, assuming this journal was bundled together with many others in a single subscription, they probably can't provide a disincentive for this behavior by ending their subscriptions.
posted by trig at 10:22 AM on April 16 [5 favorites]

i have no-one to say this to in this story but i still must utter it

christ what an ⌬
posted by lalochezia at 10:22 AM on April 16 [8 favorites]

CLOCKSS is presumably trying to be publisher-friendly

Or Elsevier-friendly, as Mounce notes

In my view Elsevier could conceivably be a prime candidate as a ‘business successor’ to take over Heterocycles and extract rent from it, for profit. The current executive director of CLOCKSS is Alicia Wise who is well known for her time in employment at Elsevier as its “Director of Universal Access“. Any extra delay in declaring this a “trigger event” at CLOCKSS could be construed as buying time for a business deal to arrange a successor.

posted by away for regrooving at 11:35 AM on April 16 [4 favorites]

I quit my job as a librarian working in noncommercial open access publishing infrastructure when nobody seemed to have a problem paying $3500 to publish an article when the "cost" of two articles could pay for my top surgery. I couldn't disassociate article processing charges from every individual for whom that is a life-changing amount of money
posted by avocet at 12:24 PM on April 16 [2 favorites]

CLOCKSS is presumably trying to be publisher-friendly

Or Elsevier-friendly, as Mounce notes

Associations with Elsevier aside, it's worth noting that CLOCKSS is a US-based 501c3 organization apparently located in California, unlike Sci-Hub which at least started in Kazakhstan. So I suspect their major aim is just not getting sued into the ground.
posted by learning from frequent failure at 12:31 PM on April 16 [4 favorites]

unlike Sci-Hub

I was confused by all the references to "dark archives" at first and only understood this from the Mounce article: CLOCKSS is not a pirate archiver (which is what scihub is); it's an organization with paying customers, namely libraries paying to support it and publishers paying to support it and paying extra for their content to be archived.

In that last link CLOCKS says "Before CLOCKSS triggers content for access, it must be approved by the Board". Its FAQ says "The CLOCKSS Board of Directors includes an equal number of seats for libraries and for publishers, thereby balancing their influence on CLOCKSS policies, priorities, and the decisions that will impact them the most."

They're not an organization acting separately from publishers; publishers are 50% of the decision makers here.
posted by trig at 12:58 PM on April 16 [4 favorites]

Yes, publishers enter into a contractual agreement with CLOCKSS, and give it a license to make the materials available when the CLOCKSS Board decides a Trigger Event has occurred.

The current CLOCKSS Participating Publisher Agreement says that, absent unconditional publisher permission, the content has to be unavailable for six months before a Trigger Event is declared. This is not well-advertised elsewhere on the website. And, of course, six months is more than four months.
posted by grouse at 1:16 PM on April 16 [6 favorites]

grouse, thank you for that! The specific wording, for those seeking it in that boilerplate Publisher Agreement:

the Archived Content is determined in good faith by the Board to be unavailable from any publisher for at least six consecutive months

Alicia Wise, Executive Director of the CLOCKSS Archive, told Chemistry World that:

the Japan Institute of Heterocyclic Chemistry is weighing up future options for the journal. ‘The content is safe and secure and we will be in a position to make it available should they not find a successor organisation to look after it,’ she notes.

Which means we don't know which of the following situations is happening:

* Wise told Chemistry World more specific details of the duration they're waiting before releasing the content, but the journalist didn't include that in the story
* Wise knows but chose not to reveal those contractual details -- possibly because the version of the contract signed by the publisher did not have the specific unavailability duration criterion locked down, or had a much more unpalatably long duration specified
* At the time of interview Wise did not know that contractual detail
* Something else?
posted by brainwane at 2:21 PM on April 16 [2 favorites]

"dark archives" that preserve research journals in a kind of escrow. (CLOCKSS has, so far, released "66 journals comprising 13,000 articles

Dark archives will be invaluable resources for training the appropriate AIs.

If a Med-Chem AI isn’t already feeding on Heterocycles, it will be soon.

Why would anyone who controlled such an archive ever want to make it publicly available and invite competition?
posted by jamjam at 2:38 PM on April 16

Fediverse conversation includes:

"This 2019 document claimed that all triggered content thus far has been through publisher request. Which would explain why so many other cases were faster. http://documents.clockss.org/index.php/CLOCKSS:_Extracting_Triggered_Content" - Michael Hoffman

"It would be a lot of fun to create a chart demonstrating out of the 17,000+ vanished articles at Heterocycles, which ones were funded by e.g. NSF, BBSRC, Estonian Science Foundation, Wellcome Trust, et cetera.

Then email each of those funders asking what they think about the situation - that online access to this research they have funded has gone." - Ross Mounce
posted by brainwane at 5:11 AM on April 17 [1 favorite]

Michael Hoffman

MetaFilter's Own!
posted by grouse at 6:11 AM on April 17 [2 favorites]

This is very interesting. My library is an odd one and focuses on not only collecting science and technology resources but we make a conscious effort to only collect print. Heterocycles is a title we've been getting since it began and we have a full run in print. One of the main reasons we chose to do this is because our previous president was very concerned about things like this happening. When I first got here, I scoffed at the idea that you had to have print to be able to reliably serve your patrons, but this isn't the only journal that has disappeared off of the internet in the years I've been working here. I checked our system and we have had an uptick in requests for Heterocycles via interlibrary loan and I expect that trend will continue until CLOCKSS releases it.
posted by teleri025 at 6:35 AM on April 17 [5 favorites]

I was actually interested in the problems of e-journal preservation -- especially of tiny shoestring journals -- when I was a babybrarian running an institutional repository. I knew my way around digital preservation (well enough for the purpose, anyway) and I'd worked in scholarly publishing. I'd read my David Rosenthal; I knew that the shoestrings were the least-tackled preservation issue.

What I ran into was repeated Giant Walls of No.

Publishers: No. You are only allowed to reposit materials from the faculty at your own institution; if you do anything else, we will sue your institution's azz off.
Library admin: No. Don't you dare antagonize a publisher in any way ever. Also your responsibility ends at the boundaries of campus; we're not paying you to help anybody else preserve their stuff.
Faculty: ... who the hell are you? Oh, a librarian, how quaint, why do we even have librarians these days when nobody reads books? What's a repository? What do you mean you want to copy my shoestring journal into your repository thing? You'll rob me of traffic! No.

I was totally willing to do my bit helping to solve this problem. I had the skills. Nobody would fucking let me use them.

I don't work in scholcomm any more. Don't miss it. Will never do it again. This level of absolute bullshit preventing me from doing something that would have benefited all parties broke me.

tl;dr the problems are systemic and no one intervention can solve them.
posted by humbug at 8:32 AM on April 18 [4 favorites]