RELOAD releases beta of open source Metadata editor
One issue that keeps popping up in debates about learning objects is the metadata question; how are educators supposed to make an interoperable description of a learning object that will allow it to be found by others? The Reload elarning tool development project decided to attack that problem before any others, with the first results already appearing.
The problem of learning object metadata authoring is roughly twofold; you have to know which category elements there are and where they go, and how to write the lot in XML. In professionally authored material for a large market, that problem is often solved by hiring a librarian or information specialist for the classification, and a techy to write it up. Great if you can afford it, but not the sort of thing that is likely to help establish the much vaunted learning object economy.
So something is needed to tame the 60 odd fields of the IEEE LOM- the standard for learning object metadata. A part of the solution is the use of application profiles to narrow down a lot of the options built into the LOM, and focus the application to a specific context or use. This is so necessary that virtually all stored learning objects now have metadata that comply with one of many profiles of the LOM: CanCore, UKCMF, SCORM etc.
Which is nice, but doesn't make the job of part solution no. 2 --a program that streamlines metadata authoring-- any easier.
The solution that the RELOAD people came up with is twofold. The app itself has no built in awareness of any of the metadata standards or profiles. It simply reads an XSD file (the ultimate authority on how a compliant XML document of a particular kind should be structured) of the relevant spec, and a couple of optional helper files to show or hide the fields for a specific profile and to provide labels intelligible to mere mortals.
Armed with this information, the editor then presents the user with a choice of a simple form, where only information that can't be predicted needs to be filled in, or, for the diehards, a full view of the entire document tree. With the application profile information, it even presents the right vocabulary to be used for a specific field in the full view.
Before you starting eagerly pointing your browser to a download fix, Phil Beauvoir, RELOAD's senior developer, would like to remind you that this is an beta release. I.e. a lot works, but not everything. Specifically, multiple vocab values per one field needs to work in the form view as well as full view. It only works in full view now. The default language interface widget is there, but doesn't do much yet, which is also true for a widget that allows you to define your own specific profile of the LOM. The editor has some remaining differences of opinion with the draft IEEE LOM XML binding, and the IMS Meta-Data binding and therefore doesn't do validation yet.
That's the bad news. The good news is that this tool has been built in pretty short order, and continues to be improved. The whole content packaging code, for example, has been put in over the last two days. Even better; since this is open source, anyone with a bit of Java knowledge can muck in, and start improving it themselves.
Besides, even in its present state it will allow non-technical users on PCs or Macs (OS X only) to edit instances of many of the most popular LOM profiles.
But Reload won't stop there. At the time of writing, the first bits needed to unpack and pack an IMS Content Package (the standard Learning Object exchange fromat) were already put in place. It reads everything and allows some editing, but doesn't do all of the automagic wrapping up yet. To follow after that are tools to author SCORM behaviours, IMS Simple Sequencing and IMS Learning Design.
More information on the whole of the Reload project is available from the RELOAD website. The editor can be downloaded from there as well. For Java developers, the project's source code can be hacked on at Sourceforge.