Attribute domain item from today's call

This is related to the following comment from Jarno Elovirta on our TC
comments list, discussed briefly at today's call:
http://lists.oasis-open.org/archives/dita-comment/201002/msg00000.html

I was given the action item to make the suggested change. When I went in to
make the change, I started to wonder about the assertion from the 1.2 spec
that - by convention - attribute domains include -a for entity values after
the attribute name (such as myprops-a for the myprops attribute). That
convention is given in the topic vocabularymodules.xml, which is entirely
new for DITA 1.2. The convention described there does not match the
conventions given elsewhere in the spec, or the conventions given in DITA
1.1. Here is the DITA 1.1 topic that gives naming patterns for attribute
modules (the same content exists in 1.2):
http://docs.oasis-open.org/dita/v1.1/OS/archspec/dtdmod.html

The 1.1 and other parts of the 1.2 spec treat this as a special case of
domain - it should be named with the attribute name + Att + -d-attribute
rather than attribute name + -a-attribute.

So - my impression is that the topic referred to in the email is actually
the one that is incorrect. It should be updated to match both the DITA 1.1
conventions and the other topics in the 1.2 draft. I will plan to make this
update during the week unless there are any objections, so that it is
included in the fourth draft of the spec.

Thanks -

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit