RE: [xri] A different perspective on * and / (long)

["Wachob, Gabe" <gwachob@visa.com>]

Drummond-
	Comments inline.

> Gabe, you're correct that, for the XRI resolution, we still 
> need to "add
> semantics" to the two segment types, i.e. to define that within an XRI
> authority segment, one of the two delimiters (slash) is the 
> delimiter that
> marks the end of the authority segment, and the other (star) 
> is used for
> subdelimiting authorities within the XRI authority segment.

But its more than "adding semantics" - we have to treat the whole set of stuff before the first slash as a unit, and thats why / and * would be different. * is not required to have any such chunking behavior, whereas / does. This is irrespective of "semantics" - its a syntax issue. 

Now, I'm having a hard time parsing the BNF (because I'd have to spend more time with it) - maybe in fact, you are, for the authority, defining the syntax differently so that, from a BNF point of view, / is treated differently for the identifier authority part. Fine.

But I *still* want the ability to chunk things together beyond the authority - if we say / and * are peers, then the only namespace-independent chunking syntax we have is () -- ie cross references... is that acceptable? 

For example, i'd no longer be able to reliably break up the following identifier:

@GabeWachob/address*home/city

I'd have to use cross references to ensure its "broken up" the right way:
@GabeWachob/(+address*home)/(+city)   --or--
@GabeWachob/(+address*home)*(+city)

At this point, if we are going to make * and / semantically equivalent (absent some specific application), whats the point of having * at all? Why not just say '/' is the only XRI-defined delimiter outside the identifier authority? What does * add outside the identifier authority? 

	-Gabe