Repository standards

Tore Hoel tweeted:

The most successful repository initiatives do not engage with LT standards EDRENE report concludes #icoper

pointing me to what looks like a very interesting report which also concludes

Important needs expressed by content users include:

  • Minimize number of repositories necessary to access

Of these, the first bullet point clearly relates to interoperability of repositories, and indicates the importance of focusing on repository federations, including metadata harvesting and providing central indexes for searching for educational content.

Coincidentally I had just finished an email replying to someone who asked about repository aggregation in the context of Open Educational Resources because she is “Trying to get colleagues here to engage with the culture of sharing learning content. Some of them are aware that there are open educational learning resources out there but they don’t want to visit and search each repository.” My reply covered Google advanced search (with the option to limit by licence type), Google custom search engines for OERs, OER Commons, OpenCourseWare Consortium search, the Creative Commons Search, the Steeple podcast aggregator and the similar-in-concept Ensemble Feed finder.

I concluded: you’ll probably notice that everything I’ve written above relies on resources being on the open web (as full text and summarized in RSS feeds) but not necessarily in repositories. If there are any OER discovery services built on repository standards like OAI-PMH or SRU or the like then they are pretty modest in their success. Of course using a repository is a fine way of putting resources onto the web, but you might want to think about things like search engine optimization, making sure Google has access to the full text resource, making sure you have a site map, encouraging (lots of) links from other domains to resources (rather than metadata records), making sure you have a rich choice of RSS feeds and so on.

I have some scribbled notes on 4 or 5 things that people think are good about repositories by which may also be harmful, a focus on interoperability between repositories and repository-related services (when it is at the expense of being part of the open web) is on there.

Tracking the Use of Open Educational Resources

As part of our support for the HEFCE, HE Academy, JISC UKOER programme CETIS are running a “2nd Tuesday” online seminar to discuss the tracking the use of OERs on Thursday 20 Nov (* Yes, I know, perhaps they should be called alternating 2nd Tuesday and 3rd Thursday seminars). Details about timing and how to join will be sent to UKOER projects through the usual strand mail lists; others who are interested should contact David Kernohan (d.kernohan /at/ JISC.AC.UK) about possible extra spaces.

Here’s the full description:

“As far as is possible projects will need to track the volume and use of the resources they make available”

At least that is what the call for projects for this programme said; the aim of this session is to help projects with this requirement. The rationale for tracking use from the funder’s perspective is clear: they want to know whether the resources being released with their money are useful to anyone apart from those who created them. Of course, as any who has tried to work with access statistics for a web site knows, we have to be cautious in interpreting such data. For example, how do we compare a simple “viewing” of a resource with someone taking the resource and embedding it in their own course site? Is it even possible to measure how often the latter happens? Another, perhaps more interesting, aspect of tracking use is what it tells us about what a resource is useful for. Being able to show how other people have used a resource might help someone considering using it themselves, but is there any way to capture this information.

As well as simple access logs and tools like Google analytics, tools similar to track-back on blog postings and the usage information provided by sites such as Flickr and Slide Share (i.e. counting the number of views on-site and number of embeds in other sites) are worth considering. Perhaps more contentious but also worth considering are the techniques such as re-direct URLs and web bugs.

We shall seek to clarify what tracking is required and pragmatically desirable and how it may be achieved. This session will be led by CETIS but we don’t pretend to know the answers to this problem, in fact we’re trying to learn from projects what is useful and achievable, so we will be relying on participants in the meeting to bring their own experiences and potential solutions. For this to work we would like to know in advance who has anything to say, so any project or individual with experience to share should contact Phil Barker as soon as possible.