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


Help: OASIS Mailing Lists Help | MarkMail Help

dita message

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

Subject: Re: [dita] Missing <address> elements in DITA 1.1 bookmap <organization> and<person>

Hi Chris,

I found the proposal for this change [1] and the minutes from the meeting
where we discussed it [2], but the minutes do not record any of the actual
discussion. I'm pretty sure this side effect was raised at the time, but I
can't find a record of it.

The authorinformation element is based off of <author>. This line in
bookmap says that any place we allow author, we should also allow XNAL
specializations of author:
<!ENTITY % author "author | %xnal-d-author;" >

The personinfo and organizationinfo elements are specializations of <data>.
So, the only way to add them directly to bookmap outside of the
authorinformation container would be as a domain replacement for data. The
personinfo and organizationinfo tags would then show up virtually
everywhere in bookmap, re-creating the problem that forced our original

Given that both person and organization are defined in bookmap, it is not
possible to specialize them directly in the XNAL domain. There are still
proposals for DITA 1.2 that would make it easier to integrate bookmap and a
domain, but that won't help with 1.1.

The person and organization elements are allowed in the following contexts:
publisherinformation, published, reviewed, edited, tested, approved,
maintainer, bookowner

All of these have an href attribute. Would it be possible for you to define
a full set of information in the authorinformation tag, and then reference
it with an href? I know many of my users are using this method to provide
more information about the maintainer:
<authorinformation id="maint"> ... </authorinformation>
<maintainer href="#maint"/>

I think that the only other options are 1) to allow personinfo and
organizationinfo anywhere we allow data, which is a big usability hit, or
2) remove publisherinformation from bookmap and move it entirely to the
XNAL domain. Both of these seem like very large changes at this point in
the review cycle. What do you think?

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

[1] http://lists.oasis-open.org/archives/dita/200606/msg00084.html
[2] http://lists.oasis-open.org/archives/dita/200606/msg00119.html

dita@seicodyne.ch wrote on 04/27/2007 04:32:54 AM:

> For DITA 1.1 we invented xNAL to describe addresses and names.
> While upgrading a customers environment from DITA 1.0 to DITA 1.1 we
> recognized that the address definitions e.g. in publisherinformation
> of bookmap have not been replaced with xNAL, they have been
> completely removed.
> I guess that removal happend by accident. Indeed, it is a shame that
> I have not noticed that earlier, and needed a cutomer notified us,
> but at least it can just in time.
> Correcting it might be a minor issue, as we may have to change
> bookmap.dtd
> <!ENTITY % author "author | %xnal-d-author;" >
> into
> <!ENTITY % author "author | %xnal-d-author;" >
> <!ENTITY % person "person | %xnal-d-personinfo;" >
> <!ENTITY % organization "organization | %xnal-d-organizationinfo;" >
> and in xNAL.ent
> <!ENTITY % xnal-d-author "authorinformation" >
> into
> <!ENTITY % xnal-d-author "authorinformation" >
> <!ENTITY % xnal-d- person  " personinfo " >
> <!ENTITY % xnal-d- organization  " organizationinfo " >
> Best regards
> Chris

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