atom feed74 messages in org.oasis-open.lists.docbookRe: [docbook] Add topic element to Do...
FromSent OnAttachments
Norman WalshOct 26, 2006 5:59 am.pgp
Michael(tm) SmithOct 26, 2006 6:46 am.pgp
Norman WalshOct 26, 2006 7:33 am.pgp
Michael(tm) SmithOct 26, 2006 8:25 am.pgp
Steve CogornoOct 26, 2006 8:51 am 
Norman WalshOct 26, 2006 9:32 am.pgp
Michael(tm) SmithOct 26, 2006 10:17 am.pgp
Yang Tj-ATY010Oct 26, 2006 10:27 am 
Chris ChiassonOct 26, 2006 11:25 am 
Johnson, EricOct 26, 2006 11:40 am 
Sean WhellerOct 26, 2006 12:06 pm 
Melanie KendellOct 26, 2006 4:51 pm 
Johnson, EricOct 26, 2006 5:05 pm 
Michael(tm) SmithOct 26, 2006 10:02 pm.bin
Michael(tm) SmithOct 26, 2006 10:15 pm.bin
Michael(tm) SmithOct 26, 2006 10:17 pm.bin
Michael(tm) SmithOct 26, 2006 10:25 pm.bin
Chris ChiassonOct 26, 2006 10:28 pm 
Michael(tm) SmithOct 26, 2006 10:38 pm.pgp
dougOct 26, 2006 11:32 pm 
Michael(tm) SmithOct 26, 2006 11:51 pm.bin
dougOct 27, 2006 12:19 am 
Camille BégnisOct 27, 2006 1:22 am 
Elliotte HaroldOct 27, 2006 4:47 am 
Michael(tm) SmithOct 27, 2006 5:07 am.bin
Sean WhellerOct 27, 2006 5:24 am 
Michael(tm) SmithOct 27, 2006 5:26 am.bin
Norman WalshOct 27, 2006 5:26 am.pgp
Norman WalshOct 27, 2006 5:39 am.pgp
Norman WalshOct 27, 2006 5:48 am.pgp
Michael(tm) SmithOct 27, 2006 6:53 am.bin
Jirka KosekOct 27, 2006 6:59 am.bin
Michael(tm) SmithOct 27, 2006 7:28 am.bin
Johnson, EricOct 27, 2006 8:15 am 
Rajal ShahOct 27, 2006 8:32 am 
Johnson, EricOct 27, 2006 8:45 am 
Chris ChiassonOct 27, 2006 8:50 am 
Rajal ShahOct 27, 2006 9:02 am 
Chris ChiassonOct 27, 2006 9:12 am 
Rowland, LarryOct 27, 2006 9:35 am 
Rajal ShahOct 27, 2006 9:35 am 
Dan SandersonOct 27, 2006 9:37 am 
Chris ChiassonOct 27, 2006 9:42 am 
Norman WalshOct 27, 2006 9:58 am.pgp
Elliotte HaroldOct 27, 2006 10:06 am 
Dave PawsonOct 27, 2006 10:13 am 
Chris ChiassonOct 27, 2006 10:13 am 
Steven CogornoOct 27, 2006 10:21 am 
Steven CogornoOct 27, 2006 10:36 am 
Eve L. MalerOct 27, 2006 10:47 am 
Bob StaytonOct 27, 2006 10:54 am 
Bob StaytonOct 27, 2006 11:02 am 
Bob StaytonOct 27, 2006 11:28 am 
Steven CogornoOct 27, 2006 11:29 am 
Steven CogornoOct 27, 2006 11:45 am 
Chris ChiassonOct 28, 2006 12:09 pm 
Michael(tm) SmithOct 28, 2006 12:09 pm.bin
Rowland, LarryOct 28, 2006 12:09 pm 
Steve WhitlatchOct 28, 2006 12:09 pm 
dougOct 28, 2006 12:10 pm 
Elliotte HaroldOct 28, 2006 12:11 pm 
Chris ChiassonOct 28, 2006 1:18 pm 
Jirka KosekOct 28, 2006 1:48 pm.bin
Michael(tm) SmithOct 28, 2006 5:24 pm.bin
Elliotte HaroldOct 28, 2006 5:48 pm 
Jirka KosekOct 29, 2006 3:01 am.pgp
Elliotte HaroldOct 29, 2006 3:17 am 
Jirka KosekOct 29, 2006 3:47 am.pgp
Chris ChiassonOct 29, 2006 9:07 am 
Elliotte HaroldOct 29, 2006 10:55 am 
Bob StaytonOct 29, 2006 11:16 am 
Steven CogornoOct 29, 2006 2:13 pm 
Sean WhellerOct 29, 2006 11:30 pm 
Jirka KosekOct 30, 2006 12:07 am.pgp
Subject:Re: [docbook] Add topic element to DocBook?
From:Steven Cogorno (Stev@Sun.COM)
Date:Oct 29, 2006 2:13:54 pm
List:org.oasis-open.lists.docbook

On Oct 27, 2006, at 12:36 PM, Rowland, Larry wrote:

Aren't all of the DocBook elements you identified that have semantic structures block elements?

They are, but I don't see what difference that makes...

I really like the task element that was added to DocBook, since it does exactly what you described -- provides a richer semantic environment for grouping the information related to a procedure -- I have had to do the same thing repeatedly in documenting products. In terms of processing prerequisites and such differently, couldn't the role or remap attributes carry that type of instruction?

Sure, role or remaps could convey processing information. But it would be a massive kludge.

Suppose an author has a two paras of prereq information marked with the prereq role and then decides to split the first para into two. The writer, expecting the the split to work like every other para, simply presses return where the para should split. Now we end up with:

<para role="taskprereq"></para> <para></para> <para role="taskprereq"></para>

That would add tons of headaches into the processing stream.

Also, how would we make sure that the role="taskprereq" is only used in tasks? How can we make sure that the prereqs are only before a procedure and not after?

Role attributes in this context are completely antithetical to structured content.

I guess I am concerned about processing expectations. I want a task to be visually distinct, which is easy when it is a block element. Does a sect2 equivalent task look like a sect3 equivalent task, or do the headings take on the appearance of the equivalent of a section at the same level in the hierarchy? If so, haven't I lost the visual cue that this is a task.

Does it matter?

In our environment, a task title looks just like section title, except that we generate a glyph (a triangle -- &dtrif;). So, if a task occurs in a sect1, then it's formatted as a sect2 would be. If a task occurs in sect2, it would be formatted as a sect3.

But if you wanted a task to format differently, couldn't you do that regardless of where it appears in the hierarchy? That's a processing expectation that I think should be left to the implementors.