RE: [xri-editors] Clarification

["Drummond Reed" <drummond.reed@onename.com>]

Nat,
 
Funny you should ask. We discussed this at length on the last editor's
call and decided to remove all examples of this type of "dots used as
relative cross-references" because it was too confusing for the main
spec. We know we have to deal with this issue in the Primer.
 
The short answer to your question is that you are correct that 
 
            xri:=JohnDoe/(+phone.number)/(+work)
 
(note the first set of parens added) could also have been expressed as
 
                xri:=JohnDoe/(+phoneNumber)/(+work)
 
because the dot in this context is only a "separator" as you call it. In
fact, the first example should probably be expressed as
 
                xri:=JohnDoe/(+(phone.number))/(+work)
 
In this way it's totally syntactically clear that the purpose of the dot
is simply to link the identifiers "phone" and "number". And of course
there are many other ways of expressing that, including camel case
phoneNumber.
 
See my next message regarding the RC1b draft.
 
=Drummond 
 
-----Original Message-----
From: Sakimura, Nat [mailto:n-sakimura@nri.co.jp]
Sent: Sunday, November 09, 2003 3:55 AM
To: xri-editors@lists.oasis-open.org
Subject: [xri-editors] Clarification
 
Dear Editors, 
 
While I was reviewing the spec, I got confused a bit. 
 
In the following example: 
 
xri:=JohnDoe/+phone.number/(+work)
 
the dot "." between +phone and number does not seem to be a delegation
character, but just a dot for the readability purpose. In other words,
it may have been +phoneNumber. Is that right? 
 
Also, how dows one resolve the cross-references in local part? 
 
Nat