There are lots of websites, blogs, and forums that provide information on DITA. Though I got introduced to DITA in an STC conference way back in 2007, I started exploring it only a few months ago. As it happens to any new learner of a trade, I was awestruck by the paradigm shift. Over a period of time, I came across a lot of wonderful things and a few not-so-wonderful things about DITA and the DITA OT, the real-world implementation of the DITA specifications. I am still learning new tricks every day.
In my first blog post, I sum up my thoughts about DITA. If it helps a new DITA enthusiast, I will be glad.
- DITA helps you focus on a single subject at one time. If you are going to document a window, you will put on the concept mask and write about the whats and whys; the task mask and write about the hows; and the reference mask to include the tidbits and additional information.
- You can generate just about any type of deliverable quickly—such as full-fledged user manuals, quick reference guides, installation guides, and administrator guides—all from the same doc set. You will only need to use the appropriate conditional processing attributes and create the maps based on the deliverables.
- This feature of DITA has been repeated ad nauseam, but I have found it very useful—content reuse. Now, I can reuse the explanation for the basic window elements such as text boxes, drop-down lists, lookup lists, menus, dialog boxes, and record management buttons.
- You can completely automate the publishing process. When all your topics and maps are ready, you will only need to click the mouse to generate PDF, HTML, CHM, RTF, and other types of output.
- In theory, the developers can make use of the DITA topics, which are XML files, for their own needs. For example, they can display grammatically correct error messages and notes pulled from your DITA topics.
A few shortcomings
- Most of the DITA aficionados would agree that it requires a techie tech writer to produce great-looking PDF output using the DITA OT and FOP. Otherwise, you can’t have greater control over how FOP generates PDF. For example, you can’t force FOP to avoid orphan headings if you don’t know a little bit of XSLT and XSL-FO. The flip side: you can accept this as an opportunity and become a techie tech writer.
- You can’t generate index using FOP. It is a fact. You may need to go for XEP to generate index.
- The CHM output generated using WinANT is very good, but the plain HTML output is a far cry from the tri-pane WebHelp. You can, however, import the project file generated along with the CHM output into RoboHelp and publish WebHelp.
So, what is your take? If you asked me “Is DITA for me?” I would say “Go for DITA if you want to reap the benefits of structured authoring, content reuse, and automated publishing”. But remember that implementing a DITA-based ecosystem is a time-consuming process. You will need a different mindset. You will need to immerse yourself deeply in DITA (and in turn XML) philosophy to understand its power and get the most out of it. Your estimation techniques, your content specifications, your templates—all may need to be modified.
Now, if you are hosting the user manuals and online Help on your organization’s website, there is also the possibility of using wikis. I shall discuss this in an upcoming post.