When building software that is to give language education, the issues associated with the training material itself is only one aspect of bringing it to a public. You have to have the material in order for the secondary issues to become relevant. This article is about some of the parts that are infrastructural to the delivery of language training as developed by the University of Bamberg and reflects discussions and thinking that grew over time.

When we discussed about how to do provide access to the training material, we discussed several scenario's; individual people using the material, schools using the material as exercise material and even having a full language course in "Wikiversity". The consequence of all this is that people from many organisations will have a need to gain access to the educational process data that is needed. To me it is obvious that this is certainly as complicated as the development of the language training software itself, it is equally obvious that it is best to link into existing programs. When these programs are part of an international effort to make educational resources available on the Internet we are onto a winner.

There are several components that are required. Security for access. Security for the educational process data. Availability to the data and the software. And it is at least as important that the software can be found by the educators and by students.

The last few days I have been looking into this and I found that there are many projects that are Open/Free and that can fit the requirements that exist. In this it was quite fortunate that I was at the I&I conference in the Netherlands where I learned many of the necessary buzz words.

For authentication the a-select project is really great. It allows for the federation of trust and it does it in such a way that even commercial banks cooperate in this project. Storing the educational process data is operational information that may fit in an e-portfolio, this in turn is part of the IMS framework. For finding the existence of the material, there is good old advertising but having standards like IEEE LOM on our side makes abundant sense.

In a way it may be premature to consider all this for the language training. But given the amount of time that it will take to implement all this I am afraid it is not really. The point that is driven home by making use of these standards is that indeed we care about the sustainability of the effort that is put into the development of the language learning software.

As the Wikimedia Foundation is at this time considering the implementation of single login for within its resources, it is an opportune moment to see if the WMF can be persuaded to look into the a-select authorisation system. Because of its ability to federate security, it has the necessary hooks to give teachers access to their students and not more than that. I spoke with Brion, the main developer of the Mediawiki software and he can be persuaded to look into this. I will need assistance to make a good pitch to him. It must be clear that by implementing A-select within the WMF environment we are talking about serious scalability; 80.000 people doing 10 or more edits a month is just a tip of an iceberg. The iceberg is Alexa.com http://Alexa.com stating that we are the 36^th website of the world with all that it implies..

The operational education data is part of a students "portfolio", what I have not understood yet is if the level where we need this data is to low level but when skills are acquired it does make sense when it finds its way into a students portfolio. The current approaches to portfolios are not going to the detailed level of the individual training modules. This is however relevant as it has its impact on the amount of time / training necessary to acquire a skill.

By having the data and software on WMF servers, I expect we will host it with an organisation that is neutral to many organisations and is there for the long haul.

We can create the greatest software, have it in the greatest infrastructure but people will still have to find it. IEEE LOM is a standard that is really there for finding educational material. At the I&I conference I learned how much the Dutch education is getting into this. Edustandaard is a Dutch implementation of the IEEE LOM standard. Many companies that supply schools with software are interested in having Mediawiki support IEEE LOM natively. Certainly when there is cooperation to make this work this will become a reality.

Resources: http://entree.kennisnet.nl/ http://www.a-select.org/ http://e-portfolio.kennisnet.nl/ http://www.imsglobal.org/ep/index.html http://trefpunttalen.kennisnet.nl/ http://contentketen.kennisnet.nl/programma/edustandaard http://www.edustandaard.nl/ http://ltsc.ieee.org/wg12/

Thanks, GerardM