OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

dita-machine-industry message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: WG: Structural specialization vs. domain


 
fyi

-----Ursprüngliche Nachricht-----
Von: Robert D Anderson [mailto:robander@us.ibm.com] 
Gesendet: Dienstag, 1. Juli 2008 21:01
An: SeicoDyne DITA
Betreff: Structural specialization vs. domain


Hi Chris,

Here are the pros and cons of your two options (as I understand them). You
may have additional pros or cons that I have not thought of.

Structural specialization
Pro: Much easier to define and maintain
Pro: If somebody wants to specialize machineTask, it is easier than using
the domain, because they do not need to know how to implement DITA
constraints
Con: Requires 2 new elements (machinetask and machinetaskbody)
Con: The elements are only available to machinetask and specializations of
machinetask

I do not know how useful these elements are to people who are not using
machinery, so I do not know how big the last con is. If the domain is called
"machinery task" then - as an outside observer - I expect it is only to be
used in machinery related tasks, so this might not be a problem.

Domain specialization
Pro: Any task specialization that still includes prereq and postreq may use
the domain
Pro: 2 fewer new elements
Con: DTD and schema are more complex to implement and maintain, because they
require an extra constraint module
Con: Specializers must understand DITA Constraint Modules in order to use
the domain in new task types

I am also not sure about the last con here. DITA Constraints are harder to
understand than basic specializations. However, they are part of the 1.2
standard, so we should not avoid them just because they are complex.

My personal choice would be to use a structural specialization - but that is
somewhat selfish, because it would make the maintenance much easier for me.
The Machine Industry subcommittee should consider each option and decide
which is best for the group.

Thanks -

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit
(507) 253-8787, T/L 553-8787 (Good Monday & Thursday)



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]