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

 


Help: OASIS Mailing Lists Help | MarkMail Help

docbook-tc message

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


Subject: Re: [docbook-tc] Corrected DocBook Accessibility Issues


"Rowland, Larry" <larry.rowland@hp.com> writes:

> DocBook Section 508 HTML Issues
> 
> 
> DocBook Grammar Issues/Questions
> --------------------------------

[...]

> 4) Title attribute for ulink, link, xref and any other elements
>    producing an <a href=""> result.
> 
> 5) abbrv and acronym in HTML have title attributes

About these two points in particular: The issue of providing markup in
the DTD to generate HTML title='Some note here.' attributes has come up
in the past, and here are some of the issues with implementing it:

  - What do we call it?
    We already have a Title element, which is for marking up, well,
    titles. I would argue that the html title attribute is, actually,
    a note or annotation, and that appropriate corresponding markup in
    DocBook should probably be named Annotation (we already have Note).

  - Element or attribute?
    It seemed, when we last discussed this, that we had a general
    consensus that if we were to provide such markup, it should be
    through use of an element and not an attribute; that is, like:

      <ulink url='http://foo.net/bar'>
        <annotation>A note with info about this link.</annotation>
      </ulink>

    Instead of, say, putting a title='A note with info about this link.'
    attribute/value on the Ulink element.

  - What content model?
    We discussed the problem of how to model the content for the
    something-to-map-to-HTML-title element, given that the
    content of the HTML title attribute is limited to PCDATA.

    If we limit the content model of the something-to-map-to-HTML-title
    element in DocBook to PCDATA, then we are a making a decision based
    on how we think it should best be modeled, but instead on the
    limitations of one particular output format. We won't be able to do
    something as simple as:

      <ulink url='http://foo.net/bar'>
        <annotation>A <emphasis>very</emphasis> good site.</annotation>
      </ulink>

    If instead we permit markup within the something-to-map-to-HTML-title
    element, then processing applications will have the problem of
    trying to figure out what to do with the contents when
    transforming to HTML. I'd suggest that they should just strip it
    out such that the above would just render as something like
    <a href='http://foo.net/bar' title='A very good site.'>

  --Mike

PGP signature



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