The murky issue of licensing

Over on the excellent new-look Cloudworks, there’s a debate going on about what to do about licensing of content on the site.  There’s two questions: one is what license to choose, and the other is what to do about the stuff that’s already on the site (!).  The latter I’m going to discuss over on that article, since it really only applies to Cloudworks, but what license is the Right One is a big and more general question.

This is far from a settled issue in the educational community. There’s reasonable consensus that Creative Commons licences are Good, rather than more restrictive ones, but as you probably already know, there are multiple versions of Creative Commons licenses.   The details are set out nicely on the Creative Commons Licenses page.  As someone releasing material, there are basically four conditions you can choose to apply:

  • Attribution (BY – must give credit)
  • Share Alike (SA – can make derivative works but only if licensed under a similar licence)
  • Non-Commercial (NC – only if not for commercial purposes)
  • No Derivatives (ND – can only distribute verbatim copies, no derivative works)

You can combine (some of) these to create six distinct licences:

  • Attribution (CC:BY)
  • Attribution Share Alike (CC:BY-SA),
  • Attribution No Derivatives  (CC:BY-ND),
  • Attribution Non-Commercial (CC:BY-NC),
  • Attribution Non-Commercial Share Alike (CC:BY-NC-SA)
  • Attribution Non-Commercial No-Derivatives (CC:BY-NC-ND)

There’s also a newish licence called CC0, which is intended as a way of unambiguously releasing a work into the public domain, free of any restraint.

So – assuming your aim is to promote the widest possible access to the material – what license should you choose?

There is a big, ongoing and fundamental argument going on in the Open Educational Resources (OER) and wider online educational community about this, with Stephen Downes and David Wiley perhaps the most articulate/notable exponents of two different positions.  To caricature horribly (and my apologies to both), Stephen Downes’ position is that the most effective licence is CC:BY-NC-SA, and David Wiley’s is that simple CC:BY is better (or even CC0).  This is overlaid (or perhaps underpinned) by a difference of approach, between a strong principled approach, and a pragmatic one.  (If you’re at all interested, I really do recommend digging in to their ongoing conversation about this.  A good starting place might be this post by Stephen Downes, and this one by David Wiley, mostly on NC – or these lists of the responses of one to the other.  If you’re going to Open Ed in Vancouver, they’re planning to debate each other face-to-face, which should be very illuminating.  A recent contribution to the ongoing debate is that Wikipedia has recently moved to CC:BY-SA.)

The argument for minimal licensing (CC:BY or less) is in essence that the other conditions create un-necessary barriers to reuse and distribution.  So, for instance, insisting on Non-Commercial would stop a company distributing printed copies of the work for profit, which might make it more available than it would otherwise be.  The arguments for more restrictive licensing include a fear that commercial interests will crowd out the free sources, using their greater marketing leverage, and that requiring Share-Alike keeps the ‘open-ness’ attached to the work.

There are obvious parallels with the Free/Open Source Software debate: there, the ideologically-pure line (what you might call the Richard Stallman view) has not been anything like as widely-adopted as the more flexible one (what you might call the Linux view).  Being widely-used, of course, does not mean that the approach is right.

For educational resources, my own current personal view is that CC:BY is the licence of choice, where possible.

It’s the least restrictive license and is the lowest barrier to sharing. All the other CC licenses create speedbumps (or worse) to people who want to use or remix  material.

We know that re-use is not widespread default practice in the educational community, and adding in extra barriers seems entirely the wrong tack to me.  If you’re wanting to use something, the extra conditions create headaches that make it  – for most practical purposes – at least look like it’s easier and quicker to just re-create your own stuff.  It’s hard enough persuading colleagues that it’s a good idea to re-use material where possible rather than re-creating it, never mind if they also need a long course in Intellectual Property Rights to understand what they can and can’t do with it.  Each of the qualifications to a simple CC:BY adds extra questions that the potential reuser needs to think through.

We can dismiss ‘No-derivatives’ fairly easily: it’s an explicit barrier to remixing or editing.   As a potential user, you have to think about things like how much you’re allowed to quote/reuse as fair use/comment.  And if you are prepared to simply copy it verbatim, what constitutes verbatim?  What if you change the font?  Or print it out from an online version?  Put a page number, heading or links to other parts of your course at the top or bottom?  Can you fix a glaring and misleading typo?

‘Non-commercial’ is also full of tricky questions.  Most universities are not commercial for these purposes … except not all university activities are covered.  What about using it on a website with ads on it?  Like, say, your personal academic blog that’s hosted for free in exchange for unobtrusive text adverts?   What about a little ‘hosted for free by Company X’ at the bottom?  A credit-bearing course where all the students are funded by the State is clearly not commercial in this sense … but what about one where (in the UK context) they’re all full fee-paying foreign students?  Or a CPD-type course where there’s no degree-level credit and the learners all pay fees?

‘Share-alike’ means you have to worry about whether the system you’re wanting to use the material on allows you to use a CC licence or not.  Does, say, your institutional VLE have a blanket licence that isn’t CC-SA compatible?  And what if you want to, say, produce a print version with a publisher who (as most do) demands a fairly draconian licence?

For any given set of circumstances, there are ‘correct’ answers to most of these questions.  (And they’re certainly not all ‘No you can’t use it’ in many situations that obtain in universities.)  But you need to be pretty savvy about IP law to know what they are.  And even then, a lot of it hasn’t been tested in the UK courts yet, so you can’t be certain. Worse, what you want to do with the stuff when you’re reusing it may change in future – you might start off making a free online course, but then it might take off and you want to produce a book … but you can’t because some of the stuff you used had NC attached.  Or you might want to transfer your free non-assessed online course to a more formal for-credit version in your University on the institutional VLE … but you can’t because some of the material had SA attached.

You can be a lot more confident about future flexibility if you stick to CC:BY material, and there’s a lot less to worry about whether you’re doing it right.  So my view is that if you want to release material to be re-used as widely as possible, CC:BY makes your potential audience’s life much easier.

Complete public domain release would – on this argument – be even better, except that as an academic, I see attribution as crucial and fundamental, so I can’t let go of that!

I’m not overwhelmingly ideologically committed to this position: it’s very much a pragmatic view of what is most likely to get the best outcome.  I certainly don’t dismiss the counter-arguments about the dangers of commercial, closed pressures: they are real.  But I think on the balance of probabilities that the ease-of-reuse argument outweighs those, and CC:BY is usually the licence of choice.

OLnet: next steps with OER research

I’m involved in the very exciting OLnet project, which is building research infrastructure to understand the Open Educational Resource (OER) world.

The elevator pitch: there’s shedloads of high-quality learning materials available, but not a lot of research in to the entire area, even at a baseline level of project-level evaluation.  The OU’s Open Learn was unusual in having a strong Research and Evaluation strand, led by Patrick McAndrew, and the Hewlett Foundation has now funded OLnet to build research capacity across the OER world.  It’s a partnership between the Open University and Carnegie Mellon, but we’ll be bringing in Fellows from far beyond.

The OLnet project proposal sets out our vision and aims, and the nascent OLnet site itself is the place to go for updates and definitive information.  We had a really successful presence at the OER conference in Monterey last month where we launched the project and captured activity and contacts there using Cloudworks.

We’ve just had another brainstorming activity about what we should be doing (as you do when you’ve been running a project for a couple of months), and here are some desperately rough-and-ready notes I made from my post-its.

NB These are very much drafts: some are silly ideas, and some are in fact things we promised to do in the proposal and are very much still signed up to do and I have merely failed to articulate that here due to cognitive impairment.

So, my notes of Things Wot We Could Do:

Enumerate/map out (dumb and smart – blog post/XML/Cohere map) all known OER projects
Bread-and-butter survey and interviews evaluating all known OER projects. (Have good baseline but not exhaustive; warm contacts for Hewlett-funded projects thanks to Monterey conference last month).

Enumerate/map out (dumb and smart again) all OER bloggers / blog aggregator. Or in fact point to existing authoritative source(s) (Again have good baseline but need to do a little more stuff to span out.)

More active OLnet blog – specified role for OLnet researchers/fellows? Or employ someone for it

Lab/research toolbox – HOWTO – how to research/evaluate your OER project, open-by-default, makes results available in open format instantly, can see how your project compares to others; OR just do the coordinating activity based around microformats/shared practice. Aim to show benefit from researching in open/OER way (see comparison with others)

Tracking package/toolkit – easy drop-in stuff fro tracking learners, tracking re-use. Link to existing Hewlett project trying to integrate Google Analytics across all OER projects; we have close link to it but aren’t part of it at the moment.

Big tech effort to provide automated OER remix/reuse/repost-tracking stuff – like plagiarism detection software. Heavy on the tech stuff but could be very exciting to do in open mode – most plagiarism detection stuff is terribly proprietary.

Working space for researchers – on OLnet site.

Baseline literature review – useful output, good to build on.

Research agenda-setting activity, or research framework setting – meetings, mappings, real, online stuff.   Like Roadmap activities in previous projects but reinvented in OER mode.

Explore impact of licenses on reuse/remix/repost/etc – the David Wiley/Stephen Downes debate about CC-BY (or less!) vs NC-SA – admits of an empirical answer to illuminate the philosophical one?

Explore models of sustainable OER learning. Yet further.

Investigate technologies – scope, pilot new ones – for supporting OER research. E.g. reputation management system, FOAF, SIOC. Cross-site identity management (OpenID vs Facebook Connect vs throwaway logins).

Single sign-on for all ‘our’ stuff – Drupal integration with SAMS (very OU centric but possibly allows us to have single sign-on from Open Learn through to OLnet …)