CC-BY and licences; we must not get it wrong. I offer some clarification

There has been a week or two of discourse about licences for scholarly publications – much of it attempting to show that the widely used CC-BY licence (from Creative Commons) has problems. The standard of discourse (from academics) has been awful – a first year student would fail courses in logic, law, rhetoric, philosophy if they served up this stuff.

The problem is that we are in great danger of using licences which restrict our action and seriously devalue academic output, while granting more rights to publishers. Given that #scholarypub is of the order of 10-15 Billion Dollars (most of which is not paid by academics, but by taxpayers, funders and students) we deserve at least an informed debate. If we don’t we are wasting (probably) > 1 billion dollars per year.

I posted the following to the “GOAL” Open-Access list. It is not easy to have a fruitful discussion on that list but I have had private mail which supports my posting. I do not guarantee the correctness of everything as licences are difficult and I’d be very happy to be corrected:

I am disappointed by the standard of much of the discourse on this list – I had tried to refrain from posting, but the much of the material on licences is absurdly wrong. I am a member of the Science Advisory Board of Creative Commons and have some acquaintance with licences though I am not an expert.

I’d like to make the following points:
* licences are legal documents. They operate within various jurisdictions – and the law is very different in different jurisdictions. There is a great deal of work required to make a CC licence. The wording is very carefully crafted.
* licences allow two parties to agree a contract. If one party feels they have been wronged they may take the other party to court in an appropriate jurisdiction. A licence (paraphrasing Lessig) is the right to defend yourself in court. There are many other laws in force which may have greater power than a licence and which may be used to control parties’ actions and freedoms.
* CC .licences are valuable in considerable part because they are portable over many jurisdictions and fields of endeavour. Many companies create ad-hoc terms and conditions which are (to my eyes) self-contradictory or ambiguous, while CC licences have been written to be clear.
* the combination of material with different CC licences is very difficult. Generally the more restrictive licence “infects” the less. So that if I take CC-BY-NC material and combine it with CC-BY the result must generally be CC-BY-NC, otherwise the terms of the CC-NC licence will be violated. It is not allowed to change a restrictive licence to a less restrictive one (i.e. the licence holder can sue you). Wikipedia (who has a much clearer idea of licences than many people here) went through major restructuring of material when they changed their licence and it required many individuals to change the licence on their contributions.
* many of the possible effects of licences are unclear.One major problem of CC-NC is that it is extremely difficult to define “commercial”. Many (including me) believe – with strong expert opinion – that CC-NC cannot be used for teaching. Students pay money, so universities are running a commercial activity. Motivation – e.g. non-profit – is irrelevant.
* another unresolved problem is the extent of virality.  If I have a CC-BY or CC0 database and include a CC-BY-SA artefact does that require the rest of the items to carry the more restrictive licence? The answer is that no-one *knows*. We have been discussing this on the OKF open-science list [http://lists.okfn.org/pipermail/open-science/2013-February/002242.html and below] and the opinion of Puneet Kishoor (CC) is that it probably depends on whether the item can be withdrawn without changing to nature of the collection. But that is simply one expert’s *opinion* – probably the only way to find out is to sue or be sued

In terms of simplicity of use CC-BY is far simpler than many of the other licences. I know I can retrieve and re-use CC-BY material without asking permission. And so can my robot (#AMI2). Note also that in 10 years of CC-BY BMC, PLoS, etc none of the suggested horror stories of CC-BY-induced plagiarism, libel, patenting, theft of ideas, etc has happened. It’s a non-problem.

I suspect that many publishers are using CC-NC because they still wish to control the scholarly literature. I am particular upset by NPG who will offer CC-NC licences for one APC and charge more for CC-BY. There is no benefit to the reader or author from CC-NC, the only beneficiary is NPG. (And it doesn’t cost more to use CC-NC because it is free and anyway has 3 fewer characters).

Note that simply because an artefact is visible on the web does not confer any rights. This is the problem with many IRs which do not have clear terms and conditions or have a blanket “copyright the university” or “all rights reserved”.

Now I am off to finish our “Liberation Software” (AMI2) which can technically read a scholarly paper and create semantic XML from it.

If I am allowed to. I can only use CC-BY or CC0.