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] indexing question


From: Erik Hennum [mailto:ehennum@us.ibm.com]
Sent: Thursday, 2006 June 29 18:13
To: JoAnn Hackos
Cc: dita@lists.oasis-open.org; Grosso, Paul
Subject: RE: [dita] indexing question

Hi, Paul and JoAnn:

The prolog is a container for metadata about the topic as a whole and thus doesn't occupy a position of the flow. That is, an index on the prolog applies to the entire topic.  

I assume meant "indexterm in" instead of "index on" above, right?

That said, we still need a way to generate a range over the whole topic. 

Huh?  I would have thought what you just said in the first paragraph means that an indexterm within the prolog generates a range over the whole topic.  Now I'm really confused.

 Last year, we talked about whether an index on a container could generate a range if the container spans more than two pages and rejected that approach.

I don't recall whether we've addressed the question of whether an index range could start in the title or abstract and end at the end of the topic's body or even at the end of the body of the last nested topic.

As a minor point, I wouldn't think it would make sense to include the related links in an index range, given that they don't provide any of the topic content and thus don't cover an index subject.  

But if they are part of the topic and you want an index range over the whole topic, it does make sense to include them. 

If we allow index to appear either inside the prolog or before the abstract, do we provide a precedent for for moving other metadata out of the prolog? Do we create confusion by having two places to assert an index on the topic? 

Perhaps, but I think indexing is different from other metadata, so this doesn't concern me as much. 


"JoAnn Hackos" <joann.hackos@comtech-serv.com> wrote on 06/29/2006 03:51:27 PM:

> I'd like to see Indexterm as a child of topic. It feels awkward to wait
> until prolog.
> JoAnn
> JoAnn T. Hackos, PhD
> President
> Comtech Services, Inc.
> 710 Kipling Street, Suite 400
> Denver, CO 80215
> 303-232-7586
> joann.hackos@comtech-serv.com
> joannhackos Skype
> www.comtech-serv.com
> -----Original Message-----
> From: Grosso, Paul [mailto:pgrosso@ptc.com]
> Sent: Thursday, June 29, 2006 4:26 PM
> To: dita@lists.oasis-open.org
> Subject: [dita] indexing question
> Looking at the 1.1 DTDs, as far as I can tell,
> indexterm isn't allowed as a child of topic.
> In fact, the earliest indexterm is allowed
> in a topic seems to be within keywords within
> metadata within prolog.
> Suppose one wants to use index-range-start and
> index-range-end to index an entire topic.  It
> seems the soonest one can put the index-range-start
> in a topic is within the prolog.  But suppose one
> has an abstract longer than a page.  Then the
> index-range-start within the prolog (which follows
> the abstract) will not pick up the first page of
> the topic.
> This seems to be a problem.  Do we need to allow
> indexterm in title, perhaps?  Is there a better
> solution?
> I think we are also going to have a similar problem
> at the bottom of a topic.  Suppose the last thing
> in your topic is related-links.  Where can you put
> the index-range-end so that you can be sure it is
> on the last page of the topic?
> paul

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