I'm trying to establish whether anybody out there has successfully developed and engaging learning content in XML (DITA) from within a CMS. We are under increasing pressure to reuse content objects, and our Technical Authors are convinced that our CMS will enable us to create and redeploy elearning content which is fit for purpose. The examples I have seen to date have been flat and uninspiring...
What you say about DITA is exactly why we prefer DocBook - it is much more expressive, so issues like video handling are easily handled. It is also extensible so we extended it for our own quiz/question mark-up. We chose to design our own element set for this, to make mark-up much more intutive, but it allows us to map readily to platfrom specifics such as Moodle XML or even to QTI if need be. QTI is a bit of a dog to use.
Chosing DocBook is one thing; using it well is another. The goal that we are chasing is to use semantic mark-up properly, i.e. mark up to keep the semantics correct as opposed to what many people do, which is to mark up for delivery. There is a very big distinction and one that significantly affects the re-use potential. With a semantically rich repository you can then perform a range of 'semantic interpretations' that allow you to deliver in a very rich, highly functional and highly integrated manner to the output target. The example I pointed you at above I think illustrates this point.
What's more, what you see there can be 100% engineered. It takes our systems <10 minutes to interpret the XML and deliver into Moodle, PDF and an HTML 5 for tablets/iPad with no intervention. All the Moodle structure is also described generically in XML and interpreted accordingly. For what it's worth, I think this is the future as it puts control right back to the content and hence the teacher. The engineer's job is to understand the semantic interpretation requirements and deliver accordingly. Once that job is done then the content creator has control via a button.
Interesting area this.
Many thanks for your reply. You're right, we are in a very similar position; the Technical Authors are producing documentation in XML which is stored in a CMS and they are keen to repurpose/reuse content and topics where possible.
There is a fundamental difference between a document or user guide and effective training materials. However, in recent weeks I have seen some good examples where more interactive content has been embedded in XML. Our challenge will be upskilling our small team to maximise opportunities for reuse.
I think for now that our approach will be to use XML content, where feasible, but this will probably be limited to reference materials that are associated with our elearning activities (to your point, stand alone objects).
I'd be very interested to hear your thoughts on the future of elearning development - I have a very talented and innovative developer working for me who's always looking to raise the bar....
Overall I agree with you, you are clearly ahead of the game. How may people do you have engineering and running the system?
Another problem we have found using DITA (yes there are advantages too), is authoring, the learning curve is long and we end up spending time which should be put into developing content, actually fixing the DITA itself. How do you deal with this?
What do you use as a repository, it's not clear from your site? I think where you are is where everybody else is heading, however we need to have systems that make this easier and cheaper to deliver lower value courses.
I believe that the future is exactly what Ken's organization is doing, we just have to figure out how to create and deliver this content in a cost effective way. I see on-demand courses overtaking eLearning, so in future we see a learning module being made up of a number of components, with the bulk of the information being delivered as a video material.
At the moment I'm trying to justify some changes to our LMS to provide a delivery mechanism for this new approach. I'm happy to share out thoughts on this.
You don't say what your organization does, do you make money on training, is it provided as part of the product, how large is the training audience. This makes a difference to the amount of investment one can put in. Ken's organization is built around doing this as a core business for large customers.
Nothing comes for free, and authoring is a problem with DITA, DocBook, ... etc ...
Perhaps our main task at CAPDM is to capture what the authors write into our XML, i.e. we absolutely allow the authors learning freedom and we have to translate what they want into smart markup. It's not a trivial learning curve, but we do this pretty well. The name of our company is taken from the five main stages in information management and delivery - Capture, Author, Publish, Deliver and Manage.
However I think the end result proves the value of putting the effort here, as all our clients end up with a semantically rich, standards-based information repository - with no proprietary hooks! The OU do this now, all the major publishers are making this move, and there are many smart industries who have been following this for years.
We use any repository tool (within reason). Sometimes our clients already have such a tool so we are forced that way, but internally we have used RCS and more recently SVN. All we really need is revision control and group awareness. Our own tools take care of repository 'gatekeeping' (e.g. correctly formed, parsable XML, usage rights, etc). We are big fans of XML databases and XQuery - for content agility - though we don't use these for basic repository management.
We are a small, Edinburgh-based company of 10 people, but we have been very focussed in the area of semantic markup for over 20 years.