SV: [xri] updated proposed XRI 1.1 ABNF

[Bryan Rasmussen <brs@itst.dk>]

Well I've just joined this group as an observer, although I have read
sporadically on the technology for a while, my reason for joining is I think
it might help me with a bunch of things I need to do, and because I am also
interested in making an implementation (probably wouldn't start for at least
a month from now, I'm hoping this list together with evening reading of
specs will help sketch out my path). That said I'm not sure I understand why
XRI is at 1.1, when I hear 1.1 after something I automatically think there
must have been a 1.0 with implementations, and that those implementations
were fairly wide spread and canonical, have there been? 

>1.  Concern about allowing the "//" at the beginning of an XRI to be
optional.  Resolution: "//" is never optional.  
>Absolute XRIs begin with "xri://" or with either an xref or gcs character.
Anyway, since the implementation I was thinking of would be for IE if xri:
without // is encountered an error must be returned, are there any
requirements on errors that would answer this, a malformed xri?