We tackled this question and then some at the January Central Texas DITA User Group meeting. I’m a little tardy in writing up my notes and thoughts about the presentation but it went really well and I appreciate all the attendee’s participation as well. We had a high school teacher in the audience and I applaud him for wanting to learn more about DITA to pass that knowledge on to high school students.
One of Ben’s answers to the question “What does DITA have to do with wiki?” is “Maybe nothing.” Love it!
Ben introduced another the triangle of choices – you have likely heard of “cheap/good/fast, pick two.” How about “knowledge/reuse/structure, pick one.”
I have to do some thinking about that one and his perception of the limitations and tradeoffs offered by those choices or priorities. Reuse and structure are particularly difficult to pair but also give you the most payoff. Structure and knowledge are another likely pair, but it could be difficult to find subject matter experts who are also able to organize their writing in a very structured manner, and finding writers who know DITA really well and also have specific content knowledge may also be difficult to obtain. His workaround for the difficulty you’d face while trying to come up with a structured wiki is a sluice box – where raw, unstructured data is the top input, some sort of raw wiki is the next filter, and the final tightest filter of all is a topic-oriented wiki.
Original photo of a sluice box by t-dawg.
My take on the question is that there are three potential hybrid DITA wiki combinations, and Chris Almond at this presentation introduced the fourth that I have seen, using DITA as an intermediate storage device, interestingly.
The three DITA-wiki combination concepts I’ve seen are:
- Wikislices – using a DITA map to keep up with wiki “topic” (article) changes. Michael Priestly is working on this for the One Laptop Per Child Project (OLPC.)
- DITA Storm – web-enabled DITA editor, but not very wiki-like. However, with just the addition of a History/Revision and Discussion tab, and an RSS feed, you could get some nice wiki features going with that product. Don Day had an interesting observation that sometimes when you add in too many wiki features on a web page you can hardly tell what’s content and where to edit it. I’d agree with that assessment.
- DITA to wikitext XSLT transform- but no round trip, have writers determine what content goes back to DITA source. Lisa Dyer will describe this content flow in the February session.
The slides are available on slideshare.net. Here are the slides that Ben Allums, Ragan Haggard, and I used.
Here are Chris Almond’s slides and his blog entry about the presentation. I described Chris’s project to Stewart Mader of wikipatterns.com and he blogged about our presentation as well at his blog ikiw.org.