Wow…

Time has just… gone!

It seems like last week I was delivering my opening address, somewhat exhausted, at Congility 2011.

The event wasn’t even over and we were letting people know about the DITA Training courses on offer at Congility South. Not to mention my having to prepare the course itself.

A heartbeat after I was whisked away to begin work on what is one of the largest DITA Content Management projects of my career, and the biggest Content Technology Audit (CTA, similar to a Content Strategy Audit but systems-focussed) I’ve ever done. They have 1000 (one THOUSAND) dedicated technical authors worldwide with 10,000 (you heard me) engineers looking to contribute natively into DITA topics. They said full roll-out numbers might be 30,000 users globally. If we opt to go for direct Engineer contributions, they will have more DITA authors than most companies have staff!

Oh, and they’re working off the file system today – no CMS support.

DITA Gone Wild

It’s been a fascinating example of DITA pushed to the limits. They products are set up in a ‘platform>product’ way where specific products are built off the platforms, so there is extensive reuse. But:

  • They various builds of the software run in parallel, meaning there’s reuse from platforms but also any number of parallel builds of the software
  • The need to keep maintaining all the various versions – current, plus 2 versions back
  • They have to reuse extensively into training materials
  • There’s hardware too…
  • Oh… and just for fun, it’s full of client-specific stuff

They’re handling this all with extensive branching and merging, and more conditions than I’ve ever seen. They’re conditionalising content by:

  • Version – because of all the branching and parallel development, different versions are considered different products
  • Audience – Including the employees, the public, and customers
  • Outputclass: e.g., chm and pdf
  • Platform – what base platform is the product derived from

Single lines like ‘The XYZ product can boil a chicken in 5 minutes if you press the “boil” button’ might be conditionalised to the point where it can say:

  • The XYZ23 product can boil, or broil a chicken in 5-10 minutes if you press the “cook” button
  • The XYZ27 product can cook a chicken in 15 minutes automatically
  • The ABC123 product can boil a turkey in 20 minutes if you press the “boil” button
  • The 321CBA product can nuke a turkey, squab or chicken if you press the “Angry birds” button

And so on, and so on and so on aaand so on…

Imagine working on the sentences while looking at a document module (DITA Topic) where all that is being expressed simultaneously. and then sharing modules like that that were created by other people for use in your documents. Without a database…

What is amazing is how well they’re managing. I have amazing respect for what they’ve already accomplished. Now they’ve asked for help to go from managing to optimising and really tuning for the customer’s interests.

I’ll let you know how we get on.

Overall I’m really enthused by seeing such a huge DITA project be a success, and now, having the opportunity to help take it to the next level. Lame, but that gives me the warm and squishies inside.

Geek corner

For the nerds out there, I’m seeing now just how interesting DITA Keys and Conkeyref are to the scalability of reuse, especially combined with Conditions. I’m upping the focus on them significantly at Congility South.

Also, I’ve finally got a little workshop together on Context-sensitive help in DITA. I had a real developer build me an actual application that I can use to demonstrate the concepts and make new context sensitive help on the fly for various formats from DITA source

Now I just need to work out how to use it to teach conditions as well… and make it call a website instead of a local help file…