[sc34wg3] Added scope on <mergeMap/>

Lars Marius Garshol larsga at ontopia.net
Fri Apr 21 11:19:07 EDT 2006

I'm sorry to reopen this can of worms, but I wasn't really very  
satisfied with the discussion of this issue. Let me see if I can  
summarize it in a way that helps us move towards some kind of  
consensus on this (as far as I can tell, nobody in either camp has  
moved in any direction yet).

Below is a summary of all the major arguments in either direction  
that I'm aware of. If I've missed anything, or you disagree with my  
summary of the list "consensus" on any of these, please let me know.

--- #1: Added scope also adds complexity

I think there is general agreement that added scopes do complicate  
both the specification and the processing significantly. (The thread  
starting at


gives a good overview of the issues.)

I think there is also general agreement that it's possible to come up  
with watertight processing semantics, and that it is possible to  
implement this.

Summary: a minus for added scope.

--- #2: Added scope allows unmerging

I think the pro-added scope camp still believes this, while the  
contra-camp never believed it. This is probably the only important  
point that hasn't been settled yet, so I'll start a separate thread  
in the hope that we can reach agreement on this.

Summary: no consensus yet.

--- #3: XTM 1.0 had added scope

I think there's general agreement that this implies that if both #1  
and #2 above are true, then that's not enough of an argument to  
warrant removing added scope. I'm not sure what the consensus is if  
it turns out that only #1 is true.

Summary: a plus for added scope.

--- #4: <mergeMap/> is authoring

I think the consensus here is that whether the statement is true or  
not, this feature was in XTM 1.0 and is so widely used (and useful!)  
that removing it in XTM 2.0 is not really an option.

Summary: makes no difference.

Lars Marius Garshol, Ontopian               http://www.ontopia.net
+47 98 21 55 50                             http://www.garshol.priv.no

More information about the sc34wg3 mailing list