Ray Denenberg writes:
 > > I think the mere fact that we are not understanding each other
 > > (or purposefully disagreeing) and the dearth of folks who
 > > actually need the functionality is reason enough to not
 > > complicate the spec by adding it.  There is nothing to prevent
 > > the one person who wants this ability to add it to their server
 > > is there?
 > 
 > (If I seem to belabor this, it's not because I think this extension
 > itself is that important, but there is a larger, philosophical
 > issue about how we develop specifications for the at large SRU
 > community.)
 > [...]
 > I believe that the SRU implementors group is the appropriate body
 > to develop commonly used and extensible specifications and
 > extensions for the universe of SRU applications.

I strongly agree.  As painful as this particular extension design is
proving to be, I think it's much better to push through with it, if
only to gain a better understanding of how extensions fit into the SRU
world.

 _/|_	 ___________________________________________________________________
/o ) \/  Mike Taylor    <[log in to unmask]>    http://www.miketaylor.org.uk
)_v__/\  "Man, if only taxonomic ranks were real! We could look at some
	 pretty cool questions in a really rigorous fashion" -- Matt Wedel.