S1000D or DITA - Which Should You Use?
It's a common question these days for people new to the structured content development paradigm/tech publications world. Here's the simple answer:
- If you have to do a lot of parts related tracking and also integrate with servicing related information, go S1000D. It's really the only thing out there with the muscle to get the job done. (by the way, you may be required to go S1000D by a government agency – so that makes the decision easy)
- If you need to run a website full of all sorts of different kinds of information – that generally does NOT require keeping track of a lot of parts and what they fit into – then go with DITA.
If you compare the earlier versions of the two specs, DITA about three or four years ago had a spec about 200 pages long (before you DITA geeks get too excited, yeah, I do know about all the other papers written to go along with the spec, but those don't really count since those were generally about how to build a system to compile the data and identify if you're DITA compliant). S1000D has been in the 2,700 to 3,000 page realm for nearly six to eight years. Granted, there are a lot of books on the subject about DITA and basically only the free specification for S1000D on the www.S1000D.org website – there are no available books on S1000D just yet, but I expect there will be soon enough. Each has a wealth of information for you to plow through. Each has its challenges. Each has its faithful followers.
I lean toward S1000D because DITA could learn a lot from S1000D (and has). DITA faithful have even created S1000D like implementations (isn't imitation the most sincere form of flattery?). S1000D implementers dream of the flexibility of DITA, but understand why they're locked into the S1000D "standard". That's another thing that separates DITA from S1000D. Basically, DITA is a free form do what you want kind of standard, a "here's how you can go about doing it" approach (can I even venture to say it's a "theory Y" approach?). Where as S1000D is more "if you're going to use it, you must use it this way – OR ELSE!" (I can definitely state it's a "theory X" approach – but sometimes that's a really nice way to make things happen) That's what really separates the two specifications. So, if you need a prescriptive approach, go S1000D. If you want to make a lot of tailored and unique data models, then go DITA.