The story so far

As we all know by now, barrier-based publishers like Elsevier and Springer sometimes offer authors a choice to upgrade their papers to open access by payment of a fee: Elsevier calls this a “sponsored article“, Springer calls it “Open Choice“, and other publishers have other names for similar facilities.

Springer’s page is very up front and explicit about the license that Open Choice articles are provided under — it’s the same Creative Commons Attribution (CC BY) license that PLoS uses, which is an excellent choice: it’s true open access, conforming to the Budapest definition, and freely allowing them to be extensively quoted, modified into teaching materials, be used in text-mining, analysed statistically, etc.

By contrast, Elsevier’s page on sponsored articles says nothing about the licence, and is very vague about what you’re allowed to actually do with such articles.  All it really says is that authors can “sponsor their article to make it available to non-subscribers”.  So maybe you can do all that good stuff; and maybe you can’t.  The sponsored article order form is no more informative.

So three weeks ago on this blog I asked what I thought was a very simple question: what actually is Elsevier’s open-access licence?  I publicised that article as widely as I could in the hope of bringing in an answer, and mentioned it in comments that I left on other blogs.

Early responses

For a long time, the closest thing I saw to a response from Elsevier was this comment from Alicia Wise (Director of Universal Access), a reply to a comment that I made on an Occam’s Typewriter article:

Hiya Mike,

If memory serves I tweeted this info to you a few days ago. Elsevier is experimenting with various licenses for our OA content. There are some bespoke licenses which permit non-commercial reuse, and some CC options including BY and NC-ND. This information also appears in different places on different articles/screens. We’re in a test-and-learn phase.

With kind wishes,

Alicia

(For what it’s worth, I don’t remember such a tweet, and if it existed I’m not able to find any trace of it.  But then Twitter’s search facilities are pretty lame, so maybe it was sent and I somehow forgot.)

I then replied to Alicia as follows:

So it varies on a per-article basis? But the fee is the same $3000 irrespective? Are there pages somewhere on the Elsevier site that explain this? It would be very helpful.

That was sixteen days ago, but there’s been no reply.

I tweeted the query directly to Alicia and Tom Reller (Elsevier’s Head of Corporate Relations) asking them to comment:

No response to that one, though.

Episode IV: A New Hope

A couple of days ago, an informative tweet from a new player: @elsevierscience.  (I don’t know who operates that account.)

The link is to http://ow.ly/i/tmPC, but since I’m not sure how persistent that service is either, here is a copy of the image:

It seems a bit strange to convey such information as a JPEG, but it’s better than not conveying it at all.  What disturbs me more is the implication that only authors published by Elsevier get to see this information.

More important still is what it actually says.  So I replied:

Which was followed up by:

But there was no more from @wisealic on this subject.

However — I just typed in a characteristic phrase from the JPEG (“any translations, for which a prior translation agreement with elsevier”) and gave it to Google.  And I immediately found that these rules are on Elsevier’s web-site after all — on the page Supplemental Terms and Conditions for sponsored documents published in Elsevier journals (1.0).

(The same language also appears in the FEBS Open Bio User Rights page.)

So what have we learned?

Nothing very satisfactory, I’m afraid.

• Elsevier’s sponsored article page doesn’t say what a sponsored article is.
• Neither does the sponsored article order form.
• Elsevier are very slow to respond to queries, when they respond at all.
• It may be the case that the terms of a sponsored article are revealed only to authors published by Elsevier.
• The terms are in fact on Elsevier’s web-site, but not linked from the main sponsored article page or anywhere else obvious.
• No-one at Elsevier appears to know this.

All of this is bewildering.  But much more importantly:

• Sponsored article are not in fact open access.
• Articles published in FEBS Open Bio are also not open access, but we already knew that.

Oh dear oh dear.

The way forward

I know that this may seem hard to believe, but I am really am trying to be as nice to Elsevier as I can in recent articles.  But holy poop, they can make it difficult.  Here’s what they need to do in organisational terms:

• Clearly state on the Sponsored Article page what a sponsored article is.
• Restate it on the order form.
• Make sure that relevant pages are linked.
• Respond to queries much more promptly and systematically.

And the big one is:

• Adopt a true open access licence such as the CC BY that Springer uses.

I imagine that, while Elsevier would agree with the previous four bullet points, they will push back on this one, imagining that they would be giving up valuable rights.  But they simply will not be able to compete in the open access space if they don’t actually provide open access.  It’s not just that PLoS are doing it right, as you’d expect them to.  It’s that other barrier-based publishers, who are competing directly with Elsevier on the same terms, are doing it right.

But, really.  The way things are now, it’s no wonder that PLoS ONE publishes more open-access articles in a month than all of Elsevier’s 2637 journals put together publish in a year.