RE: [dita] Nested Sections

["Paul Prescod" <paul.prescod@blastradius.com>]

This is a long email but a concrete suggestion follows in the next
message.

________________________________

	From: Erik Hennum [mailto:ehennum@us.ibm.com] 
	Sent: Thursday, November 03, 2005 8:50 PM
	To: dita@lists.oasis-open.org
	Subject: RE: [dita] Nested Sections
	
	
> If DITA adds support for extended narrative with division, 
> statements about those benefits would have to be qualified 
> with caveats and provisos. Wouldn't that hurt the appeal of DITA?


DITA does have support for extended narrative, through topic nesting and
through maps. We cannot prevent people from using it incorrectly. I find
it strange that people think that the nesting of sections will somehow
encourage people to forget their goals with respect to moving to DITA.
It's like adopting Java for OOP and then forgetting to use objects. It
isn't going to happen much and if it happens it isn't our fault, it is
the fault of the people who did not do their information architecting
well. If we are to be honest, we already have to make statements about
DITA with "caveats and provisos." There are many, many ways to misuse
DITA already: 

 * using generic topics when you should use task,
 * using "ol" when you should use "steps".
 * Using bold when you should use a semantic element. 
 * using inline links when you should use relationship tables
 * using out-of-the-box topic-types when you should specialize
 * using conref when you should use maps
 * using subtopics when you should use independent topics
 * etc.

Proper DITA use has a learning curve and IMHO, this small change will
barely effect that.

> Java offers an analogy. Undoubtedly, the Java cultivators considered 
> whether they should add support for procedural programming. They
decided 
> instead to keep a strong object orientation, and, ultimately, Java 
> was more successful because of it. 

I disagree on so many levels that it would double the size of this email
to enumerate them all. Nevertheless, I'll point to the stunning failure
of other B&D languages[0]: Pascal, Ada, Eiffel versus C and C++. Java
married twenty year old programming language ideas with a hundred
million dollar marketing budget[1][2]. In my opinion, that's its
innovation. Nevertheless, it gives you plenty of room to hang yourself
if you wish to. The static function feature _is_ explicit support for
procedural programming. "Almost everybody" knows that you shouldn't
write Fortran-like code in Java because they are intelligent,
knowledgable adults. The same sorts of trustworthy people will use DITA.
;)

(I could go on forever, citing other cases where Sun and others have
decided that Java programmers need to be given enough rope to hang
themselves: [3], [4],)

[0]
http://computing-dictionary.thefreedictionary.com/bondage-and-discipline
+language

[1] http://news.zdnet.com/2100-3513_22-1013860.html

[2] http://wired-vig.wired.com/news/business/0,1367,8974,00.html

[3] http://java.sun.com/products/javabeans/faq/faq.bridge.html

[4] http://www.eclipseplugincentral.com/displayarticle228.html

> While I'm delighted and encouraged by the positive comments on 
> constraints, my thought is that contraints offer fine tuning for 
> a shared understanding of content expressed by the specialization. 
> The reason for the hesitation about the recursive <section> 
> element isn't because of structural issues but, instead, because 
> recursive sections seem to belong to a fundamentally different 
> model of content (divided narrative).

I disagree that there is typically a correspondance between
narrativeness and nesting. Novels tend not to nest much and yet are our
canonical form of narrative. Airplane manuals nest a lot and yet are
seldom considered narrative. Narrativeness-is driven by dependence of
each information object on its predecessors and successors. This can
only be reduced by actually rewriting the content. Structural
limitations cannot force that rewrite. Nested sections or not, DITA is
_technically_ quite appropriate for writing a novel, whether that is our
intended usage or not. (especially with subtopics!) I don't see how the
proposal of letting sections nest changes anything substantial
whatsoever.

 Paul Prescod