[Standards-JIG] Re: JEP-0004 - <desc/> replacement

Peter Saint-Andre stpeter at jabber.org
Fri Jul 30 19:15:05 UTC 2004

In article <200407190157.04231.sneakin at semanticgap.com>,
 Nolan Eakins <sneakin at semanticgap.com> wrote:
> Section 4.2 presented an additional question. The last line in the
> description of <desc/> elements: "If a <desc/> element is not provided, a
> user agent SHOULD show the value of the 'var' attribute instead." Could the
> 'var' attribute be completely useless to the user? 

Sure. That's why it's better to provide a <desc/>.

> If there is no <desc/>,
> the client won't show a description or anything else. But if there is no
> 'label' attribute, the client SHOULD show the 'var' attribute in the
> label's place?

Well, since <desc/> is supposed to be a longer description such as a 
tool tip or popup help (think of the difference in HTML between the 
'alt' and 'longdesc' attributes of <img/>), probably it is best to show 
nothing at all if no <desc/> is provided.
> I thought there was going to be a second question about <option/> having
> more than one <value/>, but it makes sense to me after giving it
> some thought. An example showing that may be nice though.

The XML schema clearly states that <option/> is allowed to contain at 
most one <value/> child (the default maxOccurs in XML schema is 1).


More information about the Standards mailing list