RE: [dita] bookmap > bookid

[Robert D Anderson <robander@us.ibm.com>]

If there are no other comments, then I think we can make the bookpartno
element repeatable. I know that traditionally each of our books has a
single part number, so we will probably warn our users if more than one
appears (or specialize to only allow one). Your use case seems perfectly
valid, though.

With regards to specializing off of <data> -- the bookmap prototypes, which
are what I started with in defining the new version, were created before
data was available. Starting over, that would have been a better base for
most of the new bookmeta elements - as it is, I've changed many to data,
but not as many as I should. I think as part of the cleanup for these
prototype DTDs, we should change the base for any that make sense. If
anybody sees one they think should change, please let me know on or off the
list. The TC has not had any votes about the specialization base for any of
these new elements, so as long as it does not add or remove anything, I
think this can be done as cleanup without a TC vote.

Thanks-

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

"Esrig, Bruce (Bruce)" <esrig@lucent.com> wrote on 06/20/2006 10:46:00 AM:

> What would be an appropriate way to generalize bookid to cover the
> desired book numbering techniques?
>
> 1. bookpartno
>
> Lucent has an unlabeled number and a labeled number on many documents.
>
>         1234567
> comcode 9876543
>
> If bookpartno were made optional repeatable as in (%bookpartno;)*
> then Lucent could specialize it twice: ordering-number, comcode?
>
> On the call, Erik Hennum suggested that bookpartno could be
> specialized from <data> rather than <ph>, but I didn't capture the
> mechanism or advantage of doing so.
>
> 2. There has also been discussion of standardized numbers, which can
> be continued on this thread.
>
> http://lists.oasis-open.org/archives/dita/200606/msg00059.html
>
> 3. To start a separate debate ... if Erik Hennum's suggestion in
> item 1 is good ... now that the <data> element has been accepted,
> perhaps a lot of bookmap ingredients could be retargeted so that
> they are specialized from <data>.
>
> Best wishes,
>
> Bruce
>
> ========
>
> From bookmap.mod as of 30 May 2006:
>
> <!ELEMENT bookid           ((%bookpartno;)?, (%edition;)?, (%
> isbn;)?, (%booknumber;)?, (%volume;)*, (%maintainer;)?)>
> <!ATTLIST bookid           %data-element-atts;>
>
> <!-- DocBook:   bookinfo / invpartnumber     IBMIDDoc:  prolog /
> ibmbibentry / ibmpartnum     -->
> <!ELEMENT bookpartno        (#PCDATA)*>
> <!ATTLIST bookpartno        %univ-atts;
>                           outputclass CDATA #IMPLIED
> >
>
> -----Original Message-----
> From: Robert D Anderson [mailto:robander@us.ibm.com]
> Sent: Monday, June 19, 2006 3:21 PM
> To: Esrig, Bruce (Bruce)
> Cc: dita@lists.oasis-open.org
> Subject: RE: [dita] Complexity of bookmap - sequence of volumes
>
>
> Hi Bruce - there is a volume element in the bookmeta. It's part of the
> <bookid> container, which also contains isbn, edition, and some other
> information. I'm not sure if that would work - if all of your books in
the
> sequence have the same title, then this could include the volume number.
>
> Another thing that occurs to me is that you might want to use the
booktitle
> element, which includes both a library title and a book title. The
> booklibrary gives the title for the sequence of books, while the
> mainbooktitle element gives the title for the current volume.
>
> Robert D Anderson
> IBM Authoring Tools Development
> Chief Architect, DITA Open Toolkit
> (507) 253-8787, T/L 553-8787
>
> "Esrig, Bruce (Bruce)" <esrig@lucent.com> wrote on 06/19/2006 02:11:45
PM:
>
> > Hi too,
> >
> > Is there a notation that defines a sequence of volumes?
> >
> > According to the Lucent DTD, to define multiple volumes, you define
> > multiple books.
> >
> > Bruce Esrig
> >