The Challenge of OER Sustainability

Sustaining the outputs of projects and programmes beyond their initial phase of funding is a weel kent problem but it is one that we still struggle to solve. Back in 2009 when Cetis were working with Jisc to scope the technical guidelines for the forthcoming UKOER Programme we attempted to address this issue by recommending that projects deliver their content through multiple platforms. One of the few actual requirements among the programme guidelines was that projects must also deposit their content in JorumOpen, in order to act as a safeguard against resources being lost:

Delivery Platforms

Projects are free to use any system or application as long as it is capable of delivering content freely on the open web. However all projects must also deposit their content in JorumOpen. In addition projects should use platforms that are capable of generating RSS/Atom feeds, particularly for collections of resources e.g. YouTube channels. Although this programme is not about technical development projects are encouraged to make the most of the functionality provided by their chosen delivery platforms.

OER Programme Technical Requirements

Six years down the line and attrition is taking the inevitable toll. Several of the sites and repositories that hosted UKOER content have disappeared and the sustainability of the content hosted by the national Jorum repository remains uncertain following Jisc’s announcement in June that it intended to retire Jorum and “refresh its open educational resources offer”.

These problems were brought into sharp focus by Viv Rolfe (@VivienRolfe) of the University of West England this week when she tweeted

The shear volume of our publicly-funded OER that are no longer available at all is quite staggering. #ukoer #oer pic.twitter.com/Shtef5NQOa

— Viv Rolfe (@VivienRolfe) October 10, 2015

Viv’s tweet sparked a lengthy discussion on twitter that drew in several of the community’s most incisive critical thinkers on open education including Simon Thomson (@digisim), Pat Lockley (@Solvonauts, @patlockley), David Kernohan (@dkernohan), Leo Havemann (@leohavemann) and Theresa MacKinnon (@WarwickLanguage).  

The wide ranging discussion touched on a number of thorny issues relating to OER preservation and sustainability.  I’ve created a Storify of the entire discussion here: The Challenge of OER Sustainability

Self-hosting was seen as one alternative to using institutional or national repositories to host OER, with WordPress being a popular platform in some quarters.  David Kernohan took this one step further, asking if individuals who want to self-host OER should run their own repositories. While this is an interesting idea it was regarded as a rather heavy weight solution to the problem and Pat argued that repositories are the wrong tool for the job as they sit outside standard academic digital literacies.

The discussion then turned to cross-publishing. The LOCKSS (Lots of Copies Keep Stuff Safe) approach to digital preservation was regarded as one good way to ensure that content does not disappear.  However if content is deposited in multiple places and scattered across the web then other issues arise relating to how to find and curate content. Pat commented that multiple deposit may solve “lost hosting” but complicates “find”. Aggregators and the dark arts of search engine optimisation clearly play a role here, however search engines’ ability to accurately interpret licence information is still problematic.

The Solvonauts aggregator and OER search engine represents a good example of one sustainable approach to locating OER content.  Solvonauts has aggregated 141867 OERs, it costs around $50 a year to run and the code and database are shared on Github. If Pat falls under a bus tomorrow, it’s business as usual for Solvonauts. (Pat’s phrase, not mine.  Please don’t fall under a bus Pat!)  Of course Solvonauts can only find content that it is there; it can not solve the problem of how to sustain content if servers are switched off or repositories shut down with little or no warning, which brings us right back to the issue of repository sustainability.

Leo Havemann commented that the main problem is lack of funding rather than the failure of repositories per se and Simon Thomson suggested MERLOT as a good example of a sustainable OER repository.  This resulted in a rather heated discussion about whether MERLOT can be regarded as an OER repository as not all the content is CC licensed and there is a cost associated with deposit.  Simon has already blogged an excellent summary of this discussion and the points he made regarding MERLOT which you can find here: The challenges of maintaining OER repositories, but why we must never stop trying.

Ultimately there is no simple answer to the question posed by David.

 Where should I put my OER so people can find and use it?

Pat’s answer may suggest a way forward in the short term.

I would place content into any platform which supported some licensing, or was free hosting, caveated with a bulk download option should the platform close.

Even if there is no easy answer, sustainability of OER is a pressing issue that requires immediate attention and a collective response from the community.  Digital curation and sustainability of OERs may represent a challenge, but as Simon pointed out in his own blog post, we must never stop trying.

Advertisement

LRMI Implementation Case Study: Jorum

Project lead – Ben Ryan, Jorum Technical Coordinator.

“Jorum is a Jisc funded Service for UK Further and Higher Education, to collect and share open educational resources, allowing their reuse and repurposing. Jorum’s free online repository service forms a key part of Jisc’s Learning and Teaching digital content offering. It is the first port of call for thousands of resources, all shared and created by those who teach or have been inspired in the FE and HE and professional skills community.”

http://www.jorum.ac.uk/

The Jorum LRMI implementation project began in September 2013 and work is ongoing as the platform develops. Jorum is a DSpace repository containing a wide range of educational resources in many different formats, including a considerable volume of IMS Content Packages. The repository contains approximately 16,000+ resources, most of which are openly licensed. Jorum uses a combination of LOM, DC, LRMI, Jorum metadata, plus custom fields for different collection ‘windows’. LRMI properties are crosswalked to LOM education fields.

Jorum search returns page

Implementing LRMI in Jorum presented something of a challenge to the project team as DSpace only allows metadata to be added to ‘items’ which are collections of individual files.  Metadata can not be applied at the file level, e.g. an html page within a content package, and there is no way to add microdata to an individual page. If Jorum ingests a resource that already includes LRMI it can be cross walked to map to the LOM fields, however if a content package contains multiple pages that are marked up with LRMI it is difficult to extract this data as there is nowhere to store it in DSpace.  In addition, in order to enable a search engine to access a full set of LRMI metadata it is necessary to generate a page containing the LRMI which the search engine can hit. The default page generated by DSpace only includes title, author and description. To generate a page with a richer set of metadata it is necessary to create a custom theme, which is a non trivial task.

Jorum resource page

Despite the technical difficulty of implementing LRMI in DSpace, the Jorum project has made good progress and the team have plans for further development in this area including implementing the new Schema.org License property, implementing the alignmentObject and aligning it to the UK higher education Joint Academic Subject Coding Scheme.  The project team also plan to build a Google Custom Search engine to demonstrate how Jorum LRMI metadata can be surfaced.

Jorum has implemented the following LRMI properties and types, though this is dependent on what metadata has been supplied for each resource.

• educationalAlignment: plan to implement in the future
• educationalUse: yes,
• timeRequired: yes
• typicalAgeRange: no
• interactivityType: yes
• learningResourceType: yes
• useRightsUrl: no, but plan to implement License property

The Jorum project team have not yet shared their LRMI metadata with the Learning Registry as they are waiting to upgrade their platform to DSpace 4.0 before taking this implementation forward as DSpace 4.0 includes an upgraded OAI PMH interface providing a good flexible way to generate metadata from different schema.  It should be noted that Jorum have already undertaken a successful Learning Registry implementation project, the Jisc funded JLeRN Experiment ( which successfully built a Learning Registry node in the UK and developed a number of prototype services for LR ingest and querying. The same developer responsible for the JLeRN implementation has been commissioned to undertake the Jorum LRMI implementation and this work will be taken forward once the upgrade to DSpace 4.0 is complete.

Links

Jorum
Jorum Case Study Questionnaire
Jorum Blog: Implementing the Learning Resource Metadata Initiative (LRMI) in Jorum

Encouraging news from the Wellcome Library and Europeana

I’m a bit pressed for time for blogging at the moment, but there have already been two news items this week that are worth highlighting.

First of all, the Wellcome Library have followed the lead of the National Portrait Gallery, the J. P. Getty Museum and many other institutions worldwide, and announced that they have made over 100,000 high resolution historical images available free of charge.  All the images, which include of manuscripts, paintings, etchings, early photography and advertisements, carry a a CC-BY licence and can be downloaded from the Wellcome Images website.

Among many fascinating collections, Wellcome Images includes works by my favourite Georgian satirical cartoonist Thomas Rowlandson, along with his contemporaries James Gillray and George Cruikshank.

“Side way or any way” by Thomas Rowlandson

In a press release accompanying the launch, Simon Chaplin, Head of the Wellcome Library, said

“Together the collection amounts to a dizzying visual record of centuries of human culture, and our attempts to understand our bodies, minds and health through art and observation. As a strong supporter of open access, we want to make sure these images can be used and enjoyed by anyone without restriction.”

The BBC also published a rather entertaining article about the collection here:  Grin and bare it: buttock cupping & other health ‘cures’.

The other announcement that caught my eye was the launch of the second release of the Europeana Open Culture app.  I haven’t had a chance to try the new app, but I haven’t had too much success searching Europreana in the past, so I’m hoping that it will be an improvement.   The new app promises to bring “enhanced functionality, new content,  a more user-friendly layout” and is available in seven languages (English, German, French, Spanish, Dutch, Bulgarian, and Swedish).  The press release states:

“you can now download the hi-res image for free and use it however you want to – for your work, study or leisure projects”

However, as I haven’t had a chance to load up the app, I don’t know what licence or licences these images carry.   However the app code for the Muse (Museum in your pocket) Open Source iPad App used by Europeana is available from Github.

It’s really encouraging to see more and more museums, libraries and galleries making their content freely available under open licence, these are invaluable resources for teachers, learners and researchers worldwide.  I just hope we will see more education institutions joining them!

Spotlight on the Digital: Technical Focus Group

Last Friday I attended the Spotlight on the Digital technical focus group led by David Kay, of Sero Consulting and Owen Stephens, of Owen Stephens Consulting. The workshop was attended by institutional repository and archive managers, together with representatives of a wide range of bodies including the Welcome Trust, EDINA, the Archives Hub, Knowledge Integration, the Wikimedia Foundation and the British Library.

Spotlight on the Digital is run by Jisc in collaboration with RLUK and SCONUL.    The aim of the programme is to

“identify practical solutions that will support Higher Education (HE) institutions to enhance the discoverability of their digitised collections for the benefit of their key audiences, in particular researchers, teachers and students.”

The project is focusing on three user groups; researchers, students and university teachers. Two hundred publicly­ funded digital resources have been examined to assess their discoverability and a literature study of research on search behaviour has been undertaken.  In addition to the workshop I attended, the project team will also consult with a group of library directors.

I’ve created a storify of tweets from the event summarising the main points raised, which you can find here storify.com/LornaMCampbell/spotlight-on-digital

It was a very interesting workshop that raised many important issues and highlighted some really innovative approaches to discoverability including; British Library Labs‘  Mechanical Curator project, which selects a small image taken at random from the pages of digitised works held at the library and posts it to tumblr ever hour on the hour; the Imperial War Museum‘s use of History Pin to crowd source information about their collections; and Cymu1900, a crowd sourcing project to transcribe Welsh place names.

However what struck me at the end of the workshop was that we had barely scratched the surface of issues relating to the discoverability of educational content. Educational materials are frequently scattered, messy, poorly catalogued, lack persistent identifiers, and rarely reside in well managed archives or repositories.  This raises a whole host of issues and problems in relation to discoverability.

While context is an important aspect of discoverability for all kinds of digital artefacts, it is critical for educational content.  One inspiring project highlighted by Owen Stephens which addresses both contextualisation and educational resources is the National Maritime Museum‘s contextualisation of the digitised Board of Longitude papers by creating stories around the artefacts and developing Key Stage resource packs for schools.

Board of Longitude Key Stage 3 resources from the National Maritime Museum

My Cetis colleague Brian Kelly, who also attended the workshop, has written up this thoughts on the role of Wikipedia in this context: Spotlight on Wikipedia: the opportunities and the risks.