RE: [dita] Status of Nested Sections Issue

["Esrig, Bruce \(Bruce\)" <esrig@lucent.com>]

The status of nested sections is that they got lumped in with the
revisions that would make it possible to put something other than steps
(a diagram, substeps) into a task topic. When generalized task got
deferred to DITA 1.2, the compromise version of nesting got deferred
too. They're both desirable features. Notes are available if there is an
advocate ready to write them up and if the release re-opens to
accommodate them. The generalized task stuff also implies a generalized
topic, though, so it's not a trivial change.

The agreement on nested sections, as I recall it, was that each topic
can contain one-user-defined heading (the title of the section). Aside
from that, a separate intermediate heading level would be defined both
above and below section to permit auto-generated headings.
Specialization of the language and processing would be use to obtain the
auto-generated headings.

The real reason for the compromise is a clash over whether and how to
enforce a notion of topic as something that doesn't have arbitrary
amounts of sub-structure.

One might attempt to construct a justification for the compromise: if
user-generated headings are needed, then the information is not
structured in a sufficiently-predictable manner, and the content needs
to be put into context, as topics are. On the other hand, a disadvantage
of the compromise is that it assumes that DITA adopters who need
sub-structure are sophisticated enough to do their own specialization of
language and processing or prosperous enough to hire someone to do it
for them.

Best wishes,

Bruce

-----Original Message-----
From: W. Eliot Kimber [mailto:ekimber@innodata-isogen.com] 
Sent: Monday, October 09, 2006 12:28 PM
To: dita@lists.oasis-open.org
Subject: [dita] Status of Nested Sections Issue

In some work we (Innodata Isogen) is doing, we're finding the lack of
nestable sections to be a very serious impediment to using DITA
effectively for marking up legacy documents as-is (that is, without
re-authoring the content to account for poor writing practice in the
original).

I reviewed the list to see what the disposition of this issue is. I saw
several proposals but no disposition.

To mind the obvious and simplest thing to do is to allow sections to
nest in 1.1.

What do I need to do to formally submit this as a proposal for a vote?

The particular use case in this instance is a semiconductor data sheet,
where the entire sheet is clearly one topic but no subsection of the
sheet can be considered a topic in any rhetorical sense because, by
definition, the information in each subsection only applies to the
specific component. It would be prima-facie nonsense to make, for
example, the "Min/Max" section of the data sheet a separate topic
because that is information that would never be useful either for re-use
or as a topic included in some other package not associated with the
overall component. Even more so for any subdivisions within the major
sections of the data sheet.

And of course the use case for fairly typical technical manuals seems
obvious to me as well but that certainly wasn't in evidence in the
message history.

In essence the notion that information that is not rhetorically a topic
should be marked as a topic just seems so fundamentally wrong to me, as
a technical writer and as a XML practitioner, that I marvel that anyone
involved with DITA would even suggest it. As Paul Prescode pointed out,
to mark things up as topics that are not topics is to erode the whole
value of topics as a concept.

Cheers,

E.
--
W. Eliot Kimber
Professional Services
Innodata Isogen
9390 Research Blvd, #410
Austin, TX 78759
(214) 954-5198

ekimber@innodata-isogen.com
www.innodata-isogen.com