Why doesn't Springer use a CC licence?

Posted on July 11, 2007 by pm286

In a reply to a post of mine, Jan Velterop of Springer writes, and I comment, I hope, constructively.

  1. Jan Velterop Says:
    July 11th, 2007 at 10:29 am eDear Peter (and Bill), [Bill Hooker]
    JV: The wish to improve findability is a fair one and we are working on that. But I’m afraid I cannot see how a document carrying the publisher’s copyright is by definition restricted in terms of open access. If that were so, carrying the author’s copyright would be just as restrictive.

PMR: we clearly have a major problem in communicating our views here. Copyright is, by default, restrictive. If I have a film copyrighted by Disney I am restricted in what I can do with it. If I have a book by J K Rowling I am restricted as to what I can do with it. If this blog simply said “Copyright P. Murray-Rust” you would be restricted as to what you could do with it. By default the world assumes that IN THE ABSENCE OF ANY OTHER INFORMATION copyright on a scientific article restricts its use. There are recent cases where publishers have taken action against scientists who they believe have broken their copyright.
Scientists and other scholars are trained to observe and honour copyright. There are limits to what we can do with photocopying, multiple copies, etc. We can be audited. We normally have to ask permission.
So I cannot see how you can assert that a document carrying a copyright is not restricted.
Please also note that most scientists do not have the time to spend working their way through complex arguments on copyright. They simply note “this is copyright – I have no other information – therefore I cannot re-use it”.
It is, in fact, precisely because copyright is so ineffective at asserting the rights of the user that publishers have resorted to using licenses (such as CC or GPDL).

  1. JV: The Bethesda Statement says this (and the Berlin Declaration, too, without the ‘copy-’):“The author(s) and copyright holder(s) grant(s) to all users a free, irrevocable, worldwide, perpetual right of access to, and a license to copy, use, distribute, transmit and display the work publicly and to make and distribute derivative works, in any digital medium for any responsible purpose, subject to proper attribution of authorship[2], as well as the right to make small numbers of printed copies for their personal use.”JV: It says ‘author AND copyright holder’, implicitly recognising that they can be different. Otherwise it would have said ‘author AS copyright holder’. Any copyright holder can attach an open access licence. Indeed, ONLY a copyright holder can do that. Besides, ‘condemning’ any open access articles to a non-OA status because the copyrights are held by someone other than the author is doing great injustice to all those articles that have been made fully open after they were published. That would be a mistake. It seems to me that doing so is indeed declaring nail varnish more important than food, to extend my earlier analogy.

PMR: “Any copyright holder can attach an open access licence”. Exactly. What could be simpler? Yet Springer have failed to do so, and by failing to do so have implied that they do not regard the document as having the same level of open access that other publishers assert by attaching licences.
By default your articles are Open Access (rather than freely viewable) by reason of the complex structure of your web pages. The only argument I can make for the document being open access runs something like this (forgive me if I have missed some part of the trail):

  • Springer have asserted on their publisher site that SOME (not all) of their articles have “full open access”. It goes on to list the conditions which, though not a licence, might be reasonably read as such. So far, so good. (I disagree with the NC phrase – below – but that is a separate argument at present). So with that caveat I would accept (reinforced by your statement on this blog) that Springer intends certain articles are “full open access”.
  • Any particle article – divorced from the web site environment – carries NO INDICATION of the fact that it is “full open access”. It carries the standard copyright applied to all other articles. So if I were shown two Springer articles – one closed access, one full open access – on someone’s web page or in a repository I would have NO idea what rights I had. By default I would assume none.
  • A scientist visiting your site through a DOI (one of the most common ways) is presented with an abstract page which carries the following:
  1. A page about rights (all pertaining to the publisher)
  2. A option to pay for the article (which would presumably be enforceded)
  3. A small black icon called “Open Choice”. This icon gives no indication of its purpose and has no links. While Springer staff and regular vistors might know what “Open Choice” means, the average scientist has no clue. Given we already have “Online Open”, “Free Access”, “Author Choice” from other publishers, some of which are severaly restricted, the author cannot be blamed for regarding the article as closed or partially restricted access. There is no indication that the author has to refer publishers home page to find out what it means
    1. JV: This phrase in the Bethesda Statement: “the right to make small numbers of printed copies for their personal use” clearly limits commercial use. That never seemed to be problematic. Why is it problematic now? Non-commercial is in no way a limiting factor for open access to the on line version. … Jan Velterop

    PMR: I may come back to this later – it seems to be the only reason for preferring NC over BY. It only occurs in one statement – I am sure there was a good reason at the time but it  hasn’t been seen as important by many other OA publishers.
    PMR: The real point – and to me it seems obvious – is that copyright per se is inadequate to indiacte anything other than the denial of an unspecified set of rights. The simple way to tackle this is to add a licence. I do this for my software, PLoS/BMC and many others do it for their papers. If the licence is located IN THE ARTICLE there is no confusion. I see the paper, I know what the rules are. It takes 10 seconds.
    I would have no problem with a “Springer Open Access Licence” – indeed I would encourage it. There are many organization-based licences in Open Source – Mozilla, Apache, Eclipse as well as  the more widely used BSD, GPL, Artistic (which I use). SOAL would solve all the problems (assuming it read as your home page). I’d prefer CC-BY and I suspect the authors would as well. So the simple question is:
    “Why doesn’t Springer simply adopt a CC license and add it to the text of its OA publications?”  I cannot think of any reason.

    This entry was posted in open issues. Bookmark the permalink.

    6 Responses to Why doesn't Springer use a CC licence?

    1. Andrew Walkingshaw says:
      July 11, 2007 at 7:07 pm

      Peter, on the copyright thing, isn’t this a bit of a non-sequitur/red herring? This isn’t actually as complex as it looks, and I think you’re in danger of conflating a couple of related, but separate, issues.
      (I know we all know much of what’s going to follow, but this is one of those areas where we have to run through the entire series of arguments, I think; otherwise we’re all gonna get confused…)
      As I understand it, and this being the Internet, everyone knows I’m not a lawyer; copyright exists in everything as soon as it’s created, regardless of whether it’s asserted or not (copyright is unlike trademarks in this regard; you don’t need to defend copyright, whereas you do need to defend a trademark from genericisation). It follows that any article written within the last 50 years, under UK law, is under some form of copyright. It’s certainly possible to refuse to enforce your copyright, but under UK law it’s very, very unclear whether there’s any way to permanently revoke your rights under copyright; there really isn’t the American concept of the “public domain”, which in any case only really applies to the output of the US Government. So every article or blog-post we write is under some legal entity’s copyright from the instant we write it. Refusing to enforce your copyright doesn’t have any legal force; it’s just a gentleman’s agreement. You could change your mind tomorrow.
      However, if you own the copyright, what you can do is grant general – or specific – licenses to the work thus copyrighted. This ability depends absolutely on copyright; it is only copyright law which gives the copyright-owner the right to issue licenses. If you obtain a copy of a document under a given license, and that license is irrevocable, then regardless of what the copyright holder chooses to do in the future, the copy you have remains under whatever license that is. The upshot is that if a document in your possession is under CC-BY, for argument’s sake, it’s irrelevant who the copyright holder is; you have exactly the rights that the license grants you, in perpetuity, plus any fair-use rights in your jurisdiction. The article remains OA for ever, whether you hold the copyright, or whether Springer does, or the ACS, or a spinster in Droitwich.
      So the real issue here is that much of the licensing of so-called-Open-Access documents out there is at best ambiguous and at worst nonexistent, and without those licenses you don’t have the legal right to do anything at all. That’s independent of who owns the copyright; the problem is the lack of issuance of licenses.
      However, that doesn’t make transferring copyright to the publisher per se a non-issue, because the author is signing away a lot of rights they may not realise they’re giving up in so doing. Just a few possibilities;
      The copyright holder (the publisher, in this case) could relicense documents under different terms – say, selling copies of the article into a database, which then is under some more restrictive database copyright. (Imagine a hypothetical CAS or Thomson-ISI like abstracting service) – which you may not agree with.
      The copyright holder might change the license on the document full-stop; this doesn’t affect the status of any copies already floating around (they remain under the previous license), but would be mighty inconvenient – it might break website links, for instance, or in general make technically-OA papers harder to obtain than they need be.
      You lose the ability to relicense your own work; the classic case here is that you have a series of papers published CC-BY, which you then wish to aggregate and turn into (say) a textbook. If you don’t own the copyright on your papers, you can’t relicense them, forcing your entire book to be CC-BY – which you might well not want
      and so forth and so on. Some of these actions, at least in the case of a publisher who you’ve paid a page fee to, would probably give you a very decent case to sue; they’d be a breach of the contract you entered into with the publisher, if they’d undertook to make the article available in perpetuity under Open Access terms – but that’s contract law, not copyright law. The problem here isn’t transferring copyright per se, it’s signing bad, one-sided contracts with unexpected consequences.
      The free-software world’s learnt this lesson. Licenses really matter. Actually, the original Artistic’s a pretty bad license – it’s ambiguously phrased in some places; check the FSF’s website out, there’s an analysis of the problems on there. That’s why Perl’s available under two licenses, the Artistic and the GPL, simultaneously…

      Reply
    2. Bill says:
      July 12, 2007 at 6:16 am

      I think Peter has gone right to the heart of the matter: why doesn’t Springer use CC-BY [1], and label papers accordingly?
      [1] Or, if they must, CC-BY-NC; though it is beyond me why they should care if some enterprising soul finds a way to make money by re-using something from which they’ve already had their profit.

      Reply
    3. Egon says:
      July 12, 2007 at 10:20 am

      Bill, regarding [1]… I think the following plays a role here. Say the make it CC-BY, and someone extracts the data (the most important bit for chemoinformaticians), set up a database with that, competing with the databases the publisher already provides… so, I would say, it is not just the profit from the article, but also the profit from the data that is at stake. The NC clause would prohibit at least commercial competition.

      Reply
    4. Bill says:
      July 13, 2007 at 12:00 am

      Egon,
      I take your point but I see no sense in us, as researchers, supporting publishers who hold data for ransom as part of their business plan. Databases should compete, if they are going to, on the basis of value added (search tools, visualization tools, and so on); the underlying data should be Open.

      Reply
    5. pm286 says:
      July 13, 2007 at 4:23 am

      (3) Bill – you and I both misread Egon – he was simply putting the publisher’s point of view – not his, which is the same as ours. See my later post.

      Reply
    6. Bill says:
      July 13, 2007 at 6:17 am

      I should have phrased more carefully — I saw the update before I responded, and did not mean to aim at Egon (with whose blog I am familiar, so it is no surprise to me that he is on the side of the angels here). I meant only to reply to the publisher argument — Egon’s remarks brought it home to me that I frankly think it wrong — immoral — to profit by restricting access to scientific data.

      Reply

    Leave a Reply

    Your email address will not be published. Required fields are marked *