EdReNe – Current state of educational repositories – national overview

• Länkskafferiet (The Swedish Link Library) was established already in 1995. The repository contains 4000 links and the mark-up is based on Dublin Core. The repository contains metadata about quality-assessed web resources and is maintained by a number of part-time subject editors. The editors operate according to a set of quality criteria. The service is very popular in schools.
  http://lankskafferiet.skolverket.se/. Information in English can be found at http://lankskafferiet.skolverket.se/information/eng2.html.
  
  Screenshot from Länkskafferiet

• NoT-navet (The Science Hub) from The Swedish National Agency for Education contains 2000 Internet resources about natural science and mathematics. NoT-navet uses semantic web technology. The repository will be rebuilt in 2008.

• Multimediabyrån (The Multimedia Bureau) from The Swedish National Agency for Education is a website that supports the use of ICT in schools. It also contains a repository.with 2000 images and sounds.

• Museifönstret (The Window to Swedish Museums) contains appr. 500 digital resources from Swedish museums. 90 museums cooperate in providing access to their digital resources for schools.
  
  Screenshot from Museifönstret

• Kursnavet (The Course Hub) from The Swedish Agency for flexible learning. 10,000 free digital learning resources and courses in 60 subjects can be exchanged. Kursnavet supports IMS content packaging. CfL will close on 30 September and the service will be available from another organisation. Information on the transfer of responsibility will be announced in due course.

• UR Mediebiblioteket – UR Media archive (UR is the Swedish Educational Broadcasting Company). The repository has descriptions of UR’s learning resources, both analogue and digital.

• Lektion.se is a private initiative founded by teachers. It is the most popular service currently with more than 160.000 users, and is a rapidly expanding base of shared lesson plans. There is only a rather unsophisticated quality control done by the web editors.
  
  Screenshot from Lektion.se

The Spider (Spindeln) is a new search facility developed by MSU and IML for the Swedish schools. The purpose with the search facility is to give schools access to high quality digital learning resources directly from the school’s own website (see an example below). It has been very important to ensure that the Spider is easy to use and implement for the schools.

The spider search through the following repositories:

• 4000 Internet resources – all curriculum subjects, The Link Library (MSU).
• 2000 Internet resources – nature, science and mathematics, The science Hub (MSU).
• 2000 images and sounds, Image and sound archives from Multimediabyrån (MSU).
• Web, radio and video resources, Media archive from the Swedish Educational Broadcasting Company (UR).

Based on the ongoing work the general impression is that the repository owners are very interested in giving access to their repository.

This is an example of the Media Centre in Umeå has integrated the use of the spider on their web site
http://www.regionvasterbotten.se/mediacenter/

First of all, The Spider is only dealing with the metadata residing in the various repositories, not the learning object itself. There is support for several application profiles. Semantic web metadata is managed but not used actively yet.

The technical solution includes both harvesting and federation. Federated searches were used in the beginning. However, this solution gave the repositories a bigger implementation job. For that reason the harvesting model has been introduced.

Relevance ranking between the two models is a problem, which IML still has to solve. With “federation alone” it is also necessary to sort and rank the results; otherwise the results from the repository that answers first, will simply be shown as the best results.

The front-end:
The spider can be regarded as a service. The service can be accessed in different ways: JavaScript, portlet and webservice. The JavaScript model is the easiest to use. The result is shown on the school’s own website, which is quite important.

Example of script:

The future of the spider:
New archives (repositories) will be added. Some work must be done on the metadata quality (some repositories have good quality, others don’t; when they are combined it gives different problems). License management is an issue too, the Swedish National Agency for Education and IML wish to support searches for resources with a specific license, etc.

The platform used for the spider is open source. Fredrik Paulsson can be contacted for further information.

It might be added that the MELT project (another brokerage service lead by EUN) actually obtain access to the Swedish metadata through the spider.