uDoc is a new document format based on MicroXML. It is topic-oriented like Darwin Information Typing Architecture (DITA), but has a critical difference in design. Where DITA requires you to fit your data into one of several standard formats, uDoc helps you create a format that exactly fits your own use case, but will still work with a standard toolset and provide good interchangeabiliy with other users.
Note that while uDoc does not impose a schema on you, you are perfectly free to create your own schemas (possibly using Relax NG) whenever you find they are useful for your own purposes. Nothing in uDoc prevents this or even discourages it. As Uche Ogbuji said in a recent post on [xml-dev]:
When a need for an XML schema comes about, it will show itself apparent in the context of practice, often long practice, and ideally it will be minimal and as flexible as possible given the need. That's the point with regard to markup, folks.
The primary designer of the uDoc format is Jeremy H. Griffith, with the invaluable support of the rest of the Omni Systems team, especially the Publications Manager, Carolyn Stallard. As the CTO of Omni, he has had extensive experience with technical writing issues (over 50 years in the field), as well as with computer language design. When he heard of the MicroXML initiative, he became involved right away. He wanted to see what James Clark and John Cowan, two people for whom he has the highest respect, would come up with as a simplified XML. He was very pleased with the result, and was inspired to use it as the basis for uDoc.
At Omni, he led the work that created Mif2Go, a powerful FrameMaker converter that produces numerous formats (HTML, RTF, XML, and many kinds of online help), and its follow-on DITA2Go, with the same outputs from DITA source. Mif2Go™ is known as the top-ranked converter for going from FrameMaker to DITA, a path being taken by many companies now. Designing both ends of the process, conversion to and from DITA, has given him a very deep understanding of what is valuable about DITA... and what isn't. He also designed the DocBook output converter in Mif2Go, providing another view of how technical documents can be structured.
It slowly became clear that while DITA had many benefits, it was being strangled by its own overcomplexity. This was the natural result of starting with a fixed content model, then adding to it as required to handle use cases never imagined at the beginning. Since DITA is based on constraints, limiting what can go into an element and what attributes it can have, new elements have to be less capable than the ones they are based on, which is a constant source of gotchas. This design is not helpful for writers trying to get a job done.
DITA is also hampered by inadequate tools, which is ironic as better tool availability is claimed to be one of its major advantages. The DITA-OT, a collection of XSLT scripts, requires customization by an XSLT programmer for any use case that isn't as trivial as the sample files... and tech writers with no programming experience are expected to do this because, after all, XSLT is just a “script”. A couple of weeks reading the Yahoo [dita-users] list will convince anyone this is a non-trivial task; the list is high volume, and discusses little else. Unfortunately, this also gives users the idea that the DITA-OT is the only approved way to go, and that becomes a self-fulfilling prophecy, as vendors won't invest major resources in tool development (as Omni Systems did for DITA2Go™) if people won't buy the tool because the OT is both “official” and “free”.
So with uDoc, we took the opposite approach. No content model (MicroXML doesn't use DTDs anyway), and a set of elements as small as possible, so they would remain manageable. uDoc also provides a simple copy/paste method of adding more elements if wanted, as opposed to the complex DTD authoring required for even a simple addition to DITA. Support for uDoc's outputs is provided by uDoc2Go™, a very solid commercial-quality converter from Omni Systems, so XSLT scripting is never needed. This makes uDoc a highly-usable format for producing technical and business documentation of all types. A detailed comparison chart of uDoc vs. DITA is at Appendix A. Comparison of Markup Formats .