Re: Proposal to modify Classifications in ebRIM

[Farrukh Najmi <Farrukh.Najmi@Sun.COM>]

Martin,

You have distilled Len's proposal into the two fundemental needs or use cases.
Both of which I support. Whats seems to be misunderstood is that both of these
use cases are already currently supported in V1.0.

>a) to recover all the nodes and relationships within a classification scheme
>in a single operation

The above has always been supported since the earliest versions of the spec. I
have a very large XML file for a SubmitObjectsRequest that submits:

-Entire NAICS scheme (2000+ nodes)

-Entire ISO 3166 Scheme (including extra levels added by me manually for
continents and regions as you have suggested in teh past)

-Several Organizations

-Several Classifications using NAICS and ISO 3166 (e.g. Toyota classified by
Japan and NAICS code for auto makers)

What is unique about the RIM 1.0 approach is that one can get a partial scheme
if they want (a single node or a sub-tree upto a speified level from a specified
parent node).

>b) to be able to identify any set of nodes in a classification scheme using
>a web processable address based on the XPath specification.

Again this feature was added is already suported. It was added in Vancouver
based on suggestions from Len, Kathyryn and others. Please refer to line 709 in
OASIS ebRIM v1.0 and look at the getPath method on ClassificationNode. It says
and I quote:

"getPath()           Gets the path from the root ancestor of this
ClassificationNode. The path conforms to the [XPATH] expression syntax (e.g
“/Geography/Asia/Japan”). Maps to attribute named path."

In summary I believe in the use cases identified and thankfully they are already
addressed.

Please let me know if I misunderstood your statements.

I think it is good that we are focusing on use cases and not at speific
solutions in this discussion at this stage.

--
Regards,
Farrukh


Martin Bryan wrote:

> Kathyrn
>
> You stated that
> >However, invocation of that URL should not necessarily return the entire
> classification scheme as an XML document.  Not all users will want to
> retrieve an XML document if all they want to do is look at a specific
> classification scheme.
>
> What do you mean by "look" here? What format will the classification scheme
> be in if it is not in XML? Its one thing to be able to query a
> classification database node by node and another to be able to see all the
> relationships that exist within a single classification scheme.
>
> I think there are two real needs in Len's proposal:
>
> a) to recover all the nodes and relationships within a classification scheme
> in a single operation
>
> b) to be able to identify any set of nodes in a classification scheme using
> a web processable address based on the XPath specification.
>
> Martin Bryan
begin:vcard 
n:Najmi;Farrukh
tel;work:781-442-0703
x-mozilla-html:FALSE
url:www.sun.com
org:Sun Microsystems;Java Software
adr:;;1 Network Dr. MS BUR02-302;Burlington;MA;01803-0902;USA
version:2.1
email;internet:najmi@east.sun.com
fn:Farrukh Najmi
end:vcard