Comparison
S1000D vs DITA: Two Architectures, Two Worlds
The Short Answer
Contracted vs Chosen
S1000D is the aerospace and defense industry's specification for technical publications: content authored as data modules, held in a Common Source Database, filtered by a formal applicability model, and governed by business rules agreed per program. You rarely pick it — a contract does, and the specification carries the lifecycle machinery that world requires. DITA is an open OASIS architecture for topic-based authoring: typed topics assembled by maps, reused across publications, and — its defining trait — extensible through specialization. You choose it, and you supply the operational machinery around it.
That difference in origin explains most of the practical differences below. S1000D assumes a program: configurations, suppliers, decades of sustainment. DITA assumes an enterprise: products, audiences, release cycles. Both reward information architecture done early, and both punish reuse-by-copy — the failure mode they were each invented to prevent.
Side by Side
The Structural Differences
| Dimension | S1000D | DITA |
|---|---|---|
| Who governs it | The aerospace and defense industry bodies (ASD, AIA, A4A), jointly | OASIS, the open standards consortium |
| How you end up using it | A contract cites it — adoption is usually an obligation | You choose it — adoption is an architecture decision |
| Unit of content | The data module, identified by a structured Data Module Code (DMC) | The topic, typed by what it is (task, concept, reference) |
| How publications are assembled | Publication modules drawing modules from the CSDB | Maps and bookmaps assembling topics |
| Source model | The Common Source Database is part of the specification's world — a managed system, not a folder | Left to you: files in a repository, or a CCMS when reuse demands one |
| Configuration handling | A formal applicability model — content filtered by product configuration | Conditional processing — attribute-based filtering at publish time |
| How it's extended | It isn't, really — business rules layers (BREX) constrain the specification per program | Specialization — the vocabulary is designed to be extended without leaving the standard |
| Toolchain | Predominantly commercial CSDB and publishing suites | The open-source DITA Open Toolkit plus a broad commercial ecosystem |
| Where it dominates | Aerospace and defense programs, where it is contracted | Software, technology, and enterprise product documentation |
The extension row is the one that surprises people. DITA's superpower is specialization — extending the vocabulary while staying in the standard. S1000D points the same energy the opposite way: programs don't extend the specification, they constrain it, through business rules layers like ATA Spec 1000BR, MIL-STD-3031, and MIL-STD-3048. One architecture trusts you to grow it; the other exists so that a thousand organizations produce data that combines.
FAQ
Questions We Hear
Should we choose S1000D or DITA?
If a contract specifies one, that is the answer and the rest is scheduling. Where the choice is genuinely open, it turns on whether you need what S1000D carries for you — a managed source database, lifecycle handling, and configuration-driven applicability across tail numbers or serials — or whether that machinery is overhead for content that does not vary by configuration. DITA is materially quicker to adopt and easier to staff; S1000D solves harder problems you may not have.
Can DITA content be converted to S1000D, or the other way?
Structurally yes — both are modular XML, and transformation between them is well-trodden. The real cost is semantic, not syntactic: assigning Data Module Codes, mapping topic types onto information codes, and expressing conditions as applicability are decisions, not transformations. Budget for the mapping design and the exception content, not for the transform.
Is DITA free to use?
The standard is, and so is the DITA Open Toolkit that publishes from it — a genuinely open stack. Costs arrive with scale: a CCMS when reuse outgrows files in a repository, commercial editors, and the information architecture work no tool does for you.
Do organizations run both?
Yes, and it is less awkward than it sounds. A manufacturer with commercial product lines and defense contracts will often hold enterprise documentation in DITA and program deliverables in S1000D, because each body of content answers to a different master. The mistake is not running both — it is running both accidentally, without deciding which content belongs where.
Get In Touch
Weighing the Choice?
Tell us what your content has to do and who it answers to. We work in both architectures daily, and we will tell you honestly which one your problem needs — including when it's the cheaper one.