Full Article, Immediate, Permanent, Discoverable, and Accessible to Anyone Free of charge
Background: We have an article that is peer-reviewed and accepted by an academic community, however we do not agree with the publisher's copyright form (essentially giving exclusive rights). Here is what we are proposing instead.
Your feedback and contribution to this proposed agreement (in CC0) is much welcome. You are invited to use it as a template.
- CC0 1.0
- Full Article, Immediate, Permanent, Discoverable, and Accessible to Anyone Free of charge
The authors provide the contribution to Acme Corporation with the Creative Commons Attribution 4.0 International license (https://creativecommons.org/licenses/by/4.0/) under the following conditions:
- the full article; text and media
- immediately and permanently
- discoverable and accessible from their services
- to anyone (humans and machines)
- free of charge
Springer's copyright form: http://www.ida.liu.se/research/semanticweb/CopyrightFormESWC2017.pdf
Charles Oppenheim replied on
Looks fine to me. Whether Springer would agree to it is quite another matter!
Egon replied on
I am not sure why you need to impose conditions on top of the CC-BY license... that makes it a non-standard license, which makes it much harder to deal with.
Marc Couture replied on
In fact, you can’t impose conditions on top of a CC license. More precisely, you can’t use the CC trademark (name, logo) if you make any modification to one of the predefined licenses. The only thing you can personalize is the way the work must be attributed.
What you’re trying to do here is mixing two types of licenses: an end-user license (CC BY) and a private publishing license (between the publisher and you) describing what the publisher may or may not do. This entails a basic contradiction: you provide a manuscript with a CC BY license; this allows Springer (and anyone who gets a copy of it) to publish it, under the sole condition of attribution. According to the license, Springer can modify it, sell it, distribute it under a different license, with any access restriction, as long as it is attributed. But at the same time, you tell Springer that they must allow free, immediate, permanent access, machine readability, and so on.
In my opinion, the way to obtain what you want would be to state that you authorize Springer to publish you work under the following conditions (1) that they agree to meet your access requirements (FAIPDAAF + machine readability); (2) that, upon publication (meaning your requirements will be met), a CC BY license will be applied to the work, with the following attribution information ... [the main issue here is the URL of the "source” of the original work (see https://wiki.creativecommons.org/wiki/Best_practices_for_attribution); this could well be the URL of the version published by Springer, as it’s supposed to be freely available forever; the URL with the DOI would be the safest].
This way, the (private) publishing license, between you and Springer, and the end-user CC license (applied only upon publication, with no supplementary conditions/restrictions) are not mixed-up.
Note that I’m not a lawyer, so that doesn’t constitute legal advice. Also, I agree with Charles that it’s far from certain that Springer would accept such a proposal. But I used a similar reasoning when I recently revised the publishing contract for a small OA journal using CC BY.
Thank you Marc. That was a very insightful breakdown. I now see better how what I have said about our article lead to this conflict. Your proposal makes perfect sense, and that was what I was hoping as well. But that didn't come out as planned.
To be clear on the original "source" of the work: the article at http://csarven.ca/linked-data-notifications (uses CC BY) is for all intents and purposes our canonical published version. The body of the content is versioned at https://github.com/csarven/linked-data-notifications (master branch) and uses HTML. For the conference, what we did was created a branch from the master (eswc2017 branch) and then complied with the formatting requirements (of Springer) by creating a LaTeX version of the article. This is all after peer-review. We submitted the LaTeX as the camera-ready to the conference and eventually Springer if things go through.
The master branch receives updates (editorial, extended text, structural etc), and is a live document. The updates will eventually make their way to this site. One thing that I'd like to be certain of is that, our master on this site is not considered to be a derivation of what Springer gets. [Aside: a couple of years ago I explained this branching for another article of ours, as well as the publication on my website to Springer, and they considered it to be a "preprint" (or something along those lines) since it wasn't really published by another publisher/commercial entity, or had a DOI/ISBN etc. That was all fine by me.]
Coming back to licensing: We would like our LaTeX (eventually PDF) version that Springer publishes on their site to be in CC BY as well as the FAIPDAAF. It doesn't prevent Springer (or anyone) from making derivations from that or be compensated on the new version that they've created. That is also okay with me.
As far as I'm concerned, I don't particularly care about the version that Springer has (even if it comes with a DOI). It is a dead-end document (eg., not machine-friendly, not social..) in contrast to our rich "canonical" on this site (as far as I'm concerned).
I hope I understood you correctly and this information perhaps fills in the gaps. I'm happy to change the wording to minimise confusion. PS: None of us are lawyers :)
Marc Couture replied on
I hadn't had a look at the article, so I wasn’t aware that a CC BY license had already been applied to it.
This situation raises a very complex issue regarding CC licenses, namely the possibility of equivalent versions (in terms of copyrighted content, thus with only different formats and/or other non-substantial modifications) of the same work being disseminated under different licenses. For instance, you could send a copy (what you call the Latex version) without a CC license (you can do that as the copyright owner), so that the strategy I propose could still make sense. I assume the eswc2017 branch is not available online with a CC license; if it’s not the case, Springer could claim that they can publish it the way they want, because they just have to use the same exact version available under CC BY.
The terms of Springer copyright agreement illustrate well the complexity of the situation, and the difficulty in reconciling two opposite dissemination modelsThe agreement asks for an exclusive license that amounts in practice to a full copyright transfer. But a non-exclusive, non-revocable license (the CC license) has already been granted, and this license allows anyone to do all what Springer wants to get the exclusive right to do through the copyright agreement.
A mention of the existence of the CC license could be added to the agreement, meaning that the author could not enter into any further agreement except in the situations and under the conditions mentioned in the “Rights Retained by Author” section. But that would mean the author (who is bound by the copyright agreement) would have less reuse rights than what CC BY license grants to every ‘end user’.
But the bottom line, from a practical point of view, will be the decision of Springer to agree or not with your access requirements. I went to their site though, and the 2016 proceedings papers of the same conference are all paywalled (USD 30 per paper). [By the way, I don’t understand why people publish in paywalled conference proceedings or, more to the point, why conference organizers choose to help publishers make money by restricting access to the papers. (To be honest, I know perfectly well why they do it, but that’s a discussion for another day.]
Finally, I’d like to point out that this discussion is not purely hypothetical. The same issues can be raised with Elsevier copyright policy. According to this policy, authors, who grant an exclusive license to Elsevier, can “share” their manuscript in a limited way. At the same, they must apply a CC (BY-NC-ND) license to the manuscript; this allows others to do things that are forbidden to them, for instance freely disseminating it on their institutional repository (see https://www.elsevier.com/about/company-information/policies/sharing). It’s no surprise that authors are at loss to understand what they are allowed or not to do (and that they often quite purposefully ignore to try and untangle this mess).
Based on internal discussion, it doesn't appear to be that our proposed agreement actually made it out to Springer. We responded with:
"Publisher is entitled to have the Contribution peer‐reviewed by external reviewers of its choice."
"Publisher is entitled to have the Contribution peer‐reviewed by external reviewers of its choice, and if the review leads to any changes in the Contribution, Publisher agrees to release all of the reviews in full content with CC BY 4.0 license, with reviewers attributed, to the Authors."
or alternatively, that the text on external reviews can be removed.
We would like to request clarification on the following:
Will the Authors remain to be the original copyright holders, and then waving some rights so that through CC BY, Springer is allowed to publish the Contribution within their services?
It is not clear from this agreement if Springer could not sell access to the original "agreed to publish OA content" to readers, i.e., a "no paywalls" clause. If it is in there somewhere, can you please point at the sentence(s)? If not, we would like the agreement to have the following statement:
"Springer agrees to publish the Contribution in full, immediately and permanently accessible for anyone to read and/or download a copy of for free of charge, without requiring any intermediary steps (e.g., creating an account at Springer's website)."
Marc Couture replied on
I’ve examined the new form (“Consent to Publish”). First, let’s say that it’s always amazing to see the efforts publishers make to ensure that nobody (except lawyers and legally inclined scholars like me) will try to understand these agreements, or even keep on reading past the first sentences.
Though this agreement seems at first glance appropriate for the current situation, the entire second paragraph of section #1, in which the author grants Springer a non-exclusive license, is highly questionable.
On the one hand, the CC license already allows Springer to do all that is mentioned in the endless enumeration in the paragraph (except for two specific rights; see below). For instance, Springer wants the right to “permit others to use individual illustrations, tables or text quotations”, and “may use the Contribution for advertising purposes”, but the CC license give “others” and Springer all these permissions.
The only explanation I can imagine for requiring the authors to grant these redundant rights is that to be able to “take steps” to protect the work again infringement, Springer must have acquired these rights through a license or an assignment. But here we delve into complex legal matters I’m not expert in.
On the other hand, two of the enumerated rights are not redundant with the CC license.
(1) The right to sub-license, which seems also pointless to me, as others who want to use or disseminate the work need nothing more than the CC license. The only situation I can think of would be to engage others, who don’t realize that a CC license is used or don’t understand what it means, into an superfluous licensing agreement, for instance to reproduce a figure (requiring a fee?).
(2) The right “to be the first to communicate and make the Contribution available to the public in any form or format. “ (This was at the very end of the paragraph, and even yours truly, legally inclined and all, had missed it on first reading).
This puts the author in the very same type of contradiction I mentioned in my previous comment. As far as a CC BY license is already applied to the work, anybody can legally distribute it. The only way for this to work is if the author don’t distribute copies of the work, or don’t include the CC licence in them until Springer has published the work. I don't know for you, but for me it would be a no-go.
Note that it took me a lot of pondering to reach this far from obvious conclusion. But, as I said, few people read that far (or with enough attention).
A final thought. For all the superfluous content, something is missing: the designation of the “source” of the work for attribution purposes. It’s normally the copyright owner who makes the decision. I don’t know if, when the author doesn’t specify the source, someone using the work make that decision, and indicate their version as source? In the present case, could Springer indicate its published version as the source? That would be coherent with having obtained the “right” to be the first to distribute the work. In any event, I think this should be part of the agreement.
I am afraid that we cannot change the CC-BY-Consent to Publish form. This is used for a number of different types of publications and we are not permitted to change the wording for our proceedings.
In our case, we do not have the Contribution peer-reviewed by external reviewers. This is the task of the Programme Committee.
Authors retain the copyright. Springer is given consent to publish the work as detailed in the form.
There is no paywall for any of our Open Access contributions. Please take a look at some examples:
So, although we cannot change the CC-BY-Consent to Publish form, I believe we fulfill your requirements.
We can sign the agreement for now, however we would like to get in touch with those who can work to make the agreement more specific for future proceedings. Would you mind directing us?
The fact that Springer has not paywalled those example articles does not mean that Springer can not in the future, or any other article from the start. An explicit clause will clear this up.
We appreciate your cooperation on this, and hope that you understand where we are coming from.