Open Education Handbook/Creating & developing OER
Both educators and institutions need to understand the landscape of open education. As an educator you need to familarise yourself with your institution’s licenses and policies. You can start to find alternatives to questionable resources in one of the OER repositories and talk to OER practitioners, or join a group like OER-Discuss. Look at what is out there and see if there is anything that you could use or re-purpose, and talk to colleagues to get their perspectives.
Once you have made the decision to develop OER you need to think about a strategy for moving forward. Successful approaches have used the following ideas:
- Develop incrementally, making generic versions available too
- Each part of an OER, such as a picture, or text, can also be an OER and can be shared as well
- You probably already have potential OER - any resource you use which does not use other people’s copyrighted work could be an OER
- You don’t need to be an elearning genius to make an OER: a Powerpoint file can be an OER
- Once developed - all you need to do is choose an open licence
- Construct the resource with the intention of releasing it as an OER from the start to avoid 3rd party copyrighted material rather than fix it retrospectively.
When creating OER you will need to:
- Check the license
- Attribute the author, and include a disclaimer and takedown
- If stuck, people working openly tend to like helping
- Share what you’ve made
- Share what you’ve learned
OERs in a completed state can sometimes be difficult to reuse. It is often the case that the separate components or elements of an OER have more reuse potential.
There has been some exploration around this idea. Megan Beckett from Siyavula proposes that “when creating/authoring/aggregating OER/open textbooks for reuse, the final step should then be to disaggregate it into its component parts to allow for easy and accessible remixing (ie. make up then break up!).”
Other ideas are the use of http://coursefork.org/ described as a sort of a OER github.
The Open Educational Ideas project is looking at this area. It argues that one of the main challenges of OER reuse is their complete state: "What is clearly lacking is a feeling that learning opportunities have to be created by educators themselves. We call this concept emotional ownership which describes what kind of emotional / affective relation an educator has towards certain resources. Thus, the not-invented-here syndrome seems to be even more relevant in the educational domain".
If you have made the decision to develop OER at your institution it is worth being aware of some of the biggest challenges. The three most significant challenges that you will need to address are copyright issues, quality control and sustainability of any OER developed. Researching in advance and planning for the future should help mitigate issues around copyright and sustainability. Starting out with a well-defined specification and testing content on both teachers and learners should help with quality assurance.
Challenges
editThe main challenge in OER creation is striking the balance between simplicity, as a requirement for educators, and complexity, as a requirement for developers. A number of other issues that might arise are listed below:
- Be sure to have appropriate permissions before you assign an open licence
- Think about what kind of metadata will be relevant and include only this
- Make sure that consent has been attained from relevant parties
- Don't ignore rights other than copyright (such as performance rights and data protection)
- Include a disclaimer and takedown policy, and act on it if necessary
- Think carefully about attribution
- Practise what you preach - use stuff that’s already out there where possible rather than making more
- Encourage re-purposing and re-use
Educators need an editor that is simple and easy to use, otherwise they will not use it, developers need a level of complexity to achieve the functionality required,
Dirk Uys from P2PU explains: “I think the problem has two parts to it. The first part is to provide a tool that is easy to use. The second part is to teach the user more about the medium they are using. Without the second part I feel that we are just providing tools and not empowering educators. The questions for me becomes - in what way do you teach more about the medium without distracting from the short term goal of creating/remixing some content and without intimidating the user too much?”
Another question posed by Raniere Silva is "what way do you teach more about the medium without distracting from the short term goal of creating/remixing some content and without intimidating the user too much?"
Some of the proposed solutions pose barriers for most educators self-publishing materials due to their technical nature. Many argue that tools for remixing ultimately need to make it easier to ‘copy and paste from one place to the other’.
Other challenges include licences, which as well as technical formats, effect the practicalities of moving course materials; version control which can also cause problems; and the benefit of using a distributed version control system versus a centralised system.
There are also issues with tools being editor dependent resulting the user will be limited by the editor features. It should be possible to use git and allow the user choose the HTML convert tool to allow he/she use the editor (or markup language) that best suit his/her needs. E.g. for a K-8 teacher a WYSIWYG editor is better but for a math high education teacher LaTeX can be preferable and for a engineering high education teacher IPython Notebook. MathML is also an issue, to some degree addressed by using the Aloha editor.
Michael Chesterman of FLOSS Manuals makes an argument for using ePubs. He explains that most OER are shared by the person / team that writes them and that online courses, Moocs, OER repositories are increasingly the place where OER are collaboratively written using blog type, WYSIWYG tools which output HTML pages. Format specs like Scorm and metadata standards like LOM are too hard for self publishers to use, however ePubs are the most suitable candidate to allow importing and exporting of OER into these platform allowing us the freedom to exit and remix between repositories. EDUPUB is in danger of bringing a lot of complexity to the equation and hindering uptake. He argues that we shouldn’t get hung up on interactivity but should get the workflow working with simple ePubs first and use the web coding principle of "progressive enhancement" to bring more interactivity to OER. He points out that exported ePubs work well on mobile devices. However math on the Web and ePub has many issues. MathML is a W3C recommendation for how to insert math in Web pages and was adopted by EPUB3. The first issue is that only a few browsers and ePub readers support it. Firefox is the web browser with best support to MathML and it doesn’t support many important features. iBooks support part of MathML specification there are currently no ereaders that do it. Right know, almost all web pages that need maths is using some polyfill solution, e.g. MathJax, and this approach has some problems, specially for ePubs. Another issue is that type math isn't easy. Some times you request that the user knows LaTeX or other markup and others time the user need to spend time selection "anchors" at a WYSIWYG editor.
Pat Lockley argues that most reuse of OERs is through linking. With this practice comes the massive problem of link-rot. Especially given the influence of venture capital in the sector where OER are online for as long at the funding is coming in. A case in point is coursefork.org - which stopped before it even really started. Linking out to resources on a platform that invites user contributions with no real commitment to keeping them there is a real problem. Data portability should mitigate the problem so users of the platform can at a minimum archive their own data and upload it somewhere else. And ideally it encourages reuse /remix. Also, as a hack, where the licence permits it, users should grab HTML pages and import them into longer lasting community driven OER repositories which are in it for the long run. It is possible handle URLs, or mirror sites (some form of LOCKSS) would help here.