Re: [orms] Groups - Reference Model, Use Cases, Requirements v0.23 (orms-uc-v0_23.doc) uploaded

[Nat Sakimura <n-sakimura@nri.co.jp>]

Hi

Sorry that I missed the last call. The date somehow mixed up inside me.

Here are some of the things I noted.

Line 39: objective -> subjective

Other comments inline:

--------------------------------------------------
From: "Mani, Mahalingam (Mani)" <mmani@avaya.com>
Sent: Thursday, January 08, 2009 4:43 AM
To: "Anthony Nadalin" <drsecure@us.ibm.com>
Cc: "Bruce Rich" <brich@us.ibm.com>; <orms@lists.oasis-open.org>
Subject: RE: [orms] Groups - Reference Model, Use Cases, Requirements v0.23 
(orms-uc-v0_23.doc)   uploaded

> Our emails crossed. This and any other comments / feedback before and 
> during call; I will consolidate soon after today’s meeting.
>
> -mani
> ________________________________
> From: Anthony Nadalin [mailto:drsecure@us.ibm.com]
> Sent: Wednesday, January 07, 2009 11:39 AM
> To: Mani, Mahalingam (Mani)
> Cc: Bruce Rich; orms@lists.oasis-open.org
> Subject: RE: [orms] Groups - Reference Model, Use Cases, Requirements 
> v0.23 (orms-uc-v0_23.doc) uploaded
>
>
> Some additional comments:
>
> Section 4.
> (1) I assume we want both input and output formats (input/output to and 
> from the calculation engine)

There are two types of input. Domain specific raw input (it will change from 
a domain to another),
and a portable reputation format. The output of the reputation 
calculation/correlation is
actually an input as well.

For the raw input data, we might want to leverage on the works like RDF 
and/or XDI.
If that's the case, what do we need to define in addition? > Bill, Drummond?

> (2) I assume that XML (XDS) is not the only encoding that we would want, 
> maybe ASN.1

Conceptually, yes. We problem is, in our charter, it is explicitly stated 
that it will be XML Schema and there is no mention of ASN.1 etc. Hope it 
does not matter...

> (3) The output does not say anything about the input source (there seems 
> to be an inputdatainfo element, but no details on that), I do believe 
> there needs to be some correlation of input and output so you understand 
> the context of the output

InputDataInfo is supposed to be an identifier (URI, IRI, XRI) to the 
metadata about the input data.