Split domains from DITA 1.1 core

[dita@seicodyne.ch]

Based on some discussions I had at the DITA Europe Conference and within the DITA Machine Industries SC I would like to forward the following request.

As already mentioned the three domains: software, user interface and programming, have caused some missunderstanding about the purpose of DITA. And there are these two issues about it:
1. it looks like DITA is for software documentation only
2. it extends the set of elements and the volume of descriptions (e.g. the language guide) and makes it looking as being something complex.

This point 2 (higher complexity) becomes even more significant with DITA 1.1, because of the added bookmap specialization and supplementary domains.

Proposals:
To support easier understanding of the DITA standard and to bring in more simplicity I would like to add the following three proposals:

Proposal 1: Priority "high"
Splitting the language reference guide for DITA 1.1 into 3 volumes:
Volume 1: The DITA core (topic, reference, task, hi-domain, utilities-domain, indexing-domain, prolog, map)
Volume 2: Special Domains (software, user interface, programming and further domains in the future)
Volume 3: Bookmap (bookmap, xnal domain)

The html version would not be affected.

Proposal 2: Priority "high"
Removing software-, ui- and programming-domain from the DITA core DTDs.
To give a user a simplified start, these three domains should not appear when started. The code may stay in the DTDs but just get commented out. If someone needs these domains, he/she has then just to remove the <!--   -->.

Proposal 3: Priority "low"
Physical split of the dtd files.
To keep it simple I would like to recommend of having in the dtd folder just the files required for the DITA core, as mentioned in proposal 2.
The special domains may be moved into a domains folder and the bookmap related files into a bookmap folder.

I can not estimate the impact of these proposals to the tool vendors, but I am convinced these proposals would help gaining more acceptance and easier assimilation of the DITA standard by the user community, especially when looking to the extended level of complexity with the upcomming DITA 1.1 standard.

Chris Kravogel
SeicoDyne GmbH