RE: returning nodes which have a specific child

> From: Michael Kay [mailto:mike@xxxxxxxxxxxx] hild
> 
> > But that's no reason
> > to try to provide post-hoc justifications, either, IMO.
> 
> Except that the post-hoc justification is exactly the same as 
> the pre-hoc one, namely that a very high proportion of XSLT 
> usage is rendition (often to HTML), in which case skipping 
> unknown tags makes excellent sense.
> 

Well I do see the logic of that.  In my experience though, when rendering to
HTML, we do mostly end up following a pattern more akin to the one David
Carlisle posted, in which we keep going until all known tags are explicitly
handled, so in the end the default behavior is largely masked.