Re: DOCBOOK: Re: any reason why a "procedure" is not a child of "para"?

[Jeff Biss <jeff at marco-inc dot com>]

So what are you proposing? That an inline procedure element be a child of
the paragraph element? For example:

<para>Perform this procedure: <procedure><step>Turn it
on, </step><Turn it off.</step></procedure></para>

or something like:

<procedure><para>Perform this procedure:</para>
<step>Turn it on</step
<step>Turn it off.</step>
</procedure>

or:

<procedure>
<table>
<row><step> Turn it on.</step></row>
<row><step>Turn it off.</step></row>
</table>
</procedure>

I suppose the proposed hierarchy could be defined in the DTD, but I just
haven't run across a good example where the referencing paragraph is not
also part of a coherent whole such as a section or chapter too. The procedure
must be clear to be of use to a reader therefore you should provide a good
example of where this would be beneficial.

Sincerely, Jeff Biss

Robert P. J. Day wrote:
> On Wed, 12 Mar 2003, Jeff Biss wrote:
>
> > Here's my 2 cents:
> >
> > I see no reason that a procedure should be a child of a paragraph. The 
> > paragraph immediately preceding a procedure would reference that 
> > procedure. Why would one need to put a procdure inside a paragraph?
> >     
>
> for exactly the same reason that someone would put a list inside
> a paragraph -- because it inherently forms some of the content of
> that paragraph.
>
> it's not unusual for an XML editor to allow you to expand/collapse
> elements, and to move thoselelements to another part of the document.
> currently, when i make lists in emacs in outline mode, i can do that.
>
> if the procedure is a child element of a <para>, it means that i 
> can logically move a paragraph, and its contents (including the
> procedure to which it refers) comes along for the ride.
> if that's not true, i then must move *both* elements, even when
> they're clearly related.
>
> keep in mind that this is the way it was explained to *me*
> originally when i asked about the efficacy of those two
> approaches.  it made sense then, and it still makes sense now.
>
> conversely, if you don't like this for procedures, by the same
> logic, you would have to
>  argue against it for all of the
> list-type elements as well.
>
> i'm just trying to be consistent.
>
> rday
>
> p.s.  yes, i really have other things i could be doing. :-)
>
>
>
>
>