SV: [codelist] Fwd: Re: Impact of changing a specification from URI to IRI and XRI

["Bryan Rasmussen" <BRS@itst.dk>]

Hi, Quick response just back from vacation and need to catch up on emails:


Fra: Anthony B. Coates (Miley Watts) [mailto:abcoates@mileywatts.com]
Sendt: 19. december 2006 12:53
Til: Code List Representation TC
Emne: Re: [codelist] Fwd: Re: Impact of changing a specification from
URI to IRI and XRI


>If we allow XRIs for identifying code lists, code list versions,  
>column/keys, column/key versions, are their any special semantics that the  
>XRI spec would apply to the construction of these?
No, XRI doesn't have a concept of these things but what XRI does have (that
might be pertinent in the context of codelists)are special syntactical
features for describing the organizations, people etc. that are the
authorizing entities for the XRI.

in the typical case
"the XRI authority segment contains a dot- or colon- separated list of
authorities. This delegated authority identifier is resolved left to right,
with each authority supplying the
location of the next. The last authority knows how to resolve the local
access portion of the XRI." xri faq, page 4
http://xml.coverpages.org/XRI-FAQ01-20040120b.pdf 

so some examples from the XRI faq which took them from the XRI 1.0 spec:

xri:@ExampleCorp  - this authority xri is ExampleCorp, the @ sign indicates
an organization
xri:=JohnDoe - this authority is JohnDoe, the = sign indicates a person,
people are identified via i-names http://www.xdi.org/i-names-explained.html
xri:+flowers - this authority is the 'concept' of flowers, the + is a
concept. 

Concepts can be described in dictionaries -
http://www.equalsdrummond.name/?p=39

For example an XRI asserting the dictionary authority for Codelist could be:
xri://=coates/(+CodeList)


Any way, what I'm getting at there is nothing that the XRI spec would require
one to do for describing Codelists. But I suppose there are lots of ways
these things can be integrated. For example an xri could be the organization
iso and the concept domain which the codelist enumerates. 


 
>If so, what would that be, and what would be need to put in the spec?
>Would the XRI semantics change the nature of the way URIs currently work  
>as identifiers in the spec?

I would say no, however what might make sense is a whitepaper on use of XRIs
with Codelists. 

>Has anyone had a chance to look at this?  Thanks,

Well I've followed XRI lightly over the years, basically keeping an eye on it
to see if it ever got that important momentum that causes a standard to
really lift off.  

Cheers,
Bryan Rasmussen