Hi, The good news - as mentioned on the call this morning, I've been working on consolidating the tables based on John Hunt's suggestion from last year. The general idea is that if an element in the base topic has 20 different parents, and in concept it has the same 20 plus 3 more, the "contained by" row for the concept document type should read "Everything from base topic, plus [link to the extra 3]". I've got that change working, along with a similar change I was able to make for the "Contains" section on some elements. For the full package with every element, that actually drops almost 40,000 duplicate links from the tables - saving almost exactly 10 MB from the generated PDF. The unexpected news - I couldn't figure out why some rows were not collapsing, until I realized that the Ditabase document shell did not get any of the new tech-comm domains (or the new deliveryTarget domain). In 1.2, this document type was in sync with all of the tech-comm shells, so they clearly should be added. (I've done this locally, and adding these actually further reduces the link count in the spec, by allowing more consolidation.) Similar to that - the bookmap shell does not have any of the new tech-comm domains. Oddly, it also has two of the older tech-comm domains (UI domain and Software domain) that are missing from the tech-comm maps. I think it's pretty clear that for anything we added to the tech-comm topics (task, concept, etc), we also want to add them to ditabase. It's less clear for the maps, but I'm reasonably sure that 1) For most of the tech-comm domains that are in bookmap but missing from map, we want them in map. Only uncertainty is the glossref domain, which is defined in the tech-comm section but we may only want it in bookmap. 2) For the new domains that were added to map, we also want to add them to bookmap Thanks - Robert D Anderson IBM Authoring Tools Development Chief Architect, DITA Open Toolkit (
http://dita-ot.sourceforge.net/ )