Over at Auricle, Derek Morrison has gathered together some information about "learning object repositories". It's an interesting article, but it some ways it left me even more puzzled than before I read it. I'll try and explain.

Derek writes:

We want easy authoring, updating, archiving, retrieval and overall management of such material. We also want it to be finding and retrievable by software clients and even software agents/robots as well as humans.

This seems a reasonable set of requirements for a repository. What puzzles me is the proprietary and complex nature of the solutions reviewed.

Let's look again at the list of requirements. I'm not entirely sure what "authoring" means in this context, so I'll skip that for the moment. "Updating" and "retrieval" are commonplace services offered by every web/FTP server on the planet. To me "archiving" usually refers to keeping copies of obsolete material for potential but unlikely later reference. This also seems pretty easy using ubiquitous technology like FTP and zip. I may not understand everything that "Management" entails, but certainly things like associating objects in directories, moving and renaming them are also "business as usual" on the web.

So, it seems that storing, updating, archiving, retrieval and some basic "management" are available simply by placing learning materials on a garden-variety web server. Now, however, we get to "finding". This is arguably more tricky, but stop and ask yourself how you find stuff on the internet at the moment. How did you find this article?

I'm still unclear what anyone actually means by the term "learning object", but if the stored "learning objects" are primarily textual in nature, then they should be relatively easy to find using regular search engines such as Google to look for known content phrases. There will obviously be lots of cases, though, where searching by textual content is inappropriate. In this case it seems more reasonable to search by some form of metadata. So write this metadata in a text file or web page, include a link to the URL of the real resource, and you're back able to use regular search engines. Find a useful resource? make a link on one of your own pages so you can find it later ...

Some sorts of web sites have an even smarter method of finding things. A typical example is that of weblogs. Many weblog authors use software to create the entries which sends a "ping" to one or more "aggregation" services. Exactly the same methods could be used to track additions and updates to arbitrary collections of learning objects. They could even use exactly the same RSS, RDF or Atom message formats that the weblog systems use to distribute "feeds" of updates.

Now back to "authoring". I really don't know what is meant by this term here, but I'll hold out my hands for clarification, and try a few guesses while I'm waiting. At it's simplest level, I guess "authoring" is the same as writing. If I prepare a handout, or some lecture notes, or a graph, or whetever, I just need to get it into a form that can be stored on a web server. Stuff that only exists on paper, or a hand-drawn OHP slide, or a whiteboard, will need to be scanned or photographed. Most other forms of "learning object" will probably be in electronic form already.

It may be that what distinguishes "authoring" from mere "writing" is the production of accompanying metadata. If so, then it might make sense to produce some simple applications to walk "authors" through the choices and ensure things make sense. Such applications can also take care of naming and uploading the content and metadata too, if they want.

To summarise. I have a strange feeling that I must be missing something significant about repositories, management and authoring. If I'm not missing anything important, what's stopping us from just using the world wide web as it was intended? Why would we need specialist learning object repositories? What are the significant benefits such systems offer, that make it worth paying for, configuring, learning to use, and living with the bugs in these systems?