[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [dita] ISBN Element Trapped in Bookmap Specialization
Based on today's TC call, I said I would follow up with some comments on
the technical side of this.
Currently, ISBN is used as follows:
<bookmeta>
<bookid>
<isbn>1234567890</isbn>
</bookid>
</bookmeta>
bookid and isbn are both specialized from <data> today, and bookmeta is
specialized from topicmeta. So, this generalizes back to:
<topicmeta>
<data>
<data>1234567890</data>
</data>>
</topicmeta>
If we add the isbn element to the topic prolog, I think we would do so as
topic/isbn rather than as "topic/data topic/isbn" - the core topic model
does not have any specialized elements, and I do not think that isbn would
be an exception.
If we then want to define <isbn> in a way that bookmap can pick it up
unchanged (aside from a temporary bookmap/isbn token in the class
attribute), we would need to define it in the same location that bookmap
already uses it. This means it would have to be inside an existing data
element, and I'm not sure how we can do that unless we add the whole bookid
structure to topic.
One alternative that I can see is to add <isbn> into topicmeta in whatever
spot seems appropriate. The bookmeta specialization drops the core <isbn>
and redefines its own version as a specialization of data. In DITA 2.0,
bookmeta would drop its re-definition of isbn and switch over to the new
definition. Not pretty ... have any better ideas?
Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit
(507) 253-8787, T/L 553-8787 (Good Monday & Thursday)
Eliot Kimber
<ekimber@reallysi
.com> To
dita <dita@lists.oasis-open.org>
05/29/2008 10:44 cc
AM
Subject
[dita] ISBN Element Trapped in
Bookmap Specialization
The bookmap specialization is in general not usable for publications that
do
not or cannot conform to the constraints imposed by the bookmap
specialization. This includes, many, if not most, publications produced by
publishers. These constraints include requirements/limitations in bookmap
metadata as well as constraints and requirements imposed by the bookmap
content structure model.
However, there is one element defined only in the bookmap specialization
that is potentially applicable to *any* publication, the <isbn> element.
ISBN is the only bookmap metadata element that is truly universal and not
really a reflection of (potentially) enterprise-specific practice.
Any publication, even product-specific technical manuals, may have ISBN
numbers associated with them.
Thus it seems to me that <isbn> should not be bookmap specific but should
be
part of the base set of DITA-defined metadata elements.
As it is, in order to create a publication map that includes ISBN numbers
in
its metadata, I either have to copy the declaration from bookmap into my
specialization (and maintain the bookmap class value), define a new element
type that has the same meaning but a different name, or simply use "isbn"
as
the local element type name and put it in my own module space (since
there's
no possibility of future derivation from bookmap). I'm not sure which of
these options is the best, but definitely declaring an element with a name
other than "isbn" seems like the worst choice.
I think this could be fairly classified as a bug in the 1.1 design: isbn
should never have been specific to bookmap.
Is there any possibility that this could be addressed as a bug and isbn
moved to a more appropriate module?
I think the appropriate fix would be to declare "isbn" in topic.mod with
the
class "- topic/data topic/isbn " and then declared in bookmap.mod with the
class "- topic/data topic/isbn bookmap/isbn "
This would allow existing bookmap processors that expect "bookmap/isbn" to
still work but would make topic/isbn available to all topics and maps.
The type "bookmap/isbn" could be flagged as deprecated so that we can
remove
it in a future release.
Cheers,
Eliot
--
Eliot Kimber
Senior Solutions Architect
"Bringing Strategy, Content, and Technology Together"
Main: 610.631.6770
www.reallysi.com
www.rsuitecms.com
---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail. You may a link to this group and all your TCs in
OASIS
at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]