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


Help: OASIS Mailing Lists Help | MarkMail Help

docbook message

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

Subject: Re: DOCBOOK: re step container for procedure

Norman Walsh <ndw@nwalsh.com> writes:

> [...]
> If we had instead changed procedure for consistency, then all these
> lists would have been:
>   optional title
>   one or more items
> I don't think this would satisfy Micheal Smith's use case, since it
> would not easily allow reusing the steps by reference to their
> container without simultaneously reusing the title.

Actually, that is what I meant, though I've done a very poor job of
saying it. I wrote "group of steps" when to be precise I should have
said "group of steps with an optional title and titleabbrev".

To clarify, what I'm suggesting is that we:

  * return the *lists-- Orderedlist etc.-- to their pre-RFE 144
    content model: ((title, titleabbrev?)?, listitem+)

  * create a coordinate content model: ((title, titleabbrev?)?, step+)
    and call it Stepset or whatever.

> I'm opposed to adding a 'stepset' wrapper around the steps (and
> by extension a 'listitems' wrapper around the list items):

I didn't mean to suggest creating a new "listitems" wrapper. What I
meant and should have been more clear about is what I've said above:
*lists in their pre-RFE 144 state, something coordinate for steps.

> 1. If required, it would introduce a new level of wrapper in every
>    procedure, even when there was no need for most authors in most cases.
> 2. If optional, it would allow procedures to contain either a set
>    of steps or a stepset and having two ways to do the same thing
>    is bad.
> 3. If stepset is allowed to have a title, then you'd have titles
>    at two different levels in the procedure and that strikes me as
>    awkward.

I can see that Stepset will create additional complexity, but the only
alternatives seem to be:

  X. keep Procedure as is-- realizing that although "Step" is a
     semantically explicit name for the equivalent of a Listitem in an
     Orderedlist, the content model for containing Step is not
     coordinate with the *list content models for containing Listitem

  Y. forget about creating Stepset and restrict the content model of
     Procedure to ((title, titleabbrev?)?, step+)

  Z. make the content models for all *lists coordinate with the
     Procedure content model: e.g. with 

      <!ELEMENT orderedlist ((%formalobject.title.content;)?,
           (%component.mix;)*, listitem+)>

X preserves an inconsistency and doesn't allow for programmatic
storage/reuse of groups of steps (with optional title/titleabbrev).

Y I like fine, but I suspect many other people don't agree.

Z I pray no one thinks is a good idea.

    --Mike Smith

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

Powered by eList eXpress LLC