Re: [dita-comment] linktext content model

["Jeremy H. Griffith" <jeremy@omsys.com>]

On Tue, 31 Aug 2010 11:49:15 -0500, Eliot Kimber <ekimber@reallysi.com> wrote:

>The TC has analyzed all the standard topic domains and have determined that
>they are all compatible with the base map module, which means that it is
>valid to integrate them into map document types.

Good...

>The TC is still discussing whether the TC-provided map shell should or
>should not include at least the highlight domain as you suggest.
>
>The initial discussion was that it certainly seemed odd that the map and
>topic document types in the technical content area were not consistent.
>However, the TC needs to discuss it more thoroughly and decide whether or
>not this is a change that can or should be made in 1.2.

Please do!  It is especially bad for keyrefs, which is where my
test case hit a wall.  IIRC, the spec example of using linktext
to provide content shows the highlight domain being used, and I
do not see a workaround for it.  Forcing *everyone* to use a
specialized shell is a powerful barrier to adoption by those
less expert at that than you.  ;-)  Which is almost everybody.

>Note, however, that because all the topic domains are compatible with maps,
>it means that any DITA user can construct their own map document types that
>include any or all of the domains. So to that degree the standard allows it
>even if the TC-provided map document type does not integrate those domains.

However, Oxygen 12 does not... with the draft 1.2 DTDs.

>It is probably useful to provide a statement to this effect somewhere in the
>spec. I'll think about what the best place to do that is.

Better to make them consistent, and avoid the need for a 
statement that will baffle many of your would-be users.

>Thank you for your comment and for making a close read of the specification.

Quite welcome!  I'll try to get more comments in soon.


-- Jeremy H. Griffith <jeremy@omsys.com>
   DITA2Go site:  http://www.dita2go.com/