I agree with the aim of getting FpML into the ISO 20022 Repository.
I agree we should give it its own FpML name, and "Master
Schema" is a good one, but our Master Schema is a type of Conceptual Model and we should use a well defined and
well understood term to describe it. ISO 20022 does the same in that its master
model is named "Business Layer" and described as a type of conceptual
model (in the 1.5 iteration). Please don't change the name, but please do
change the description to refer to existing concepts using their well known
names.
It is interesting to compare the FpML 5.x Architecture to ISO 20022 for
the view generation syntax. IMHO the adoption of the Master Schema makes FpML
much closer to ISO 20022 because:
·
Both FpML and ISO20022 have two
layers of models, with the master layer being annotated to generate the lower
layer. FpML and ISO 20022 are the same in this regard.
·
The upper layer in ISO 20022 that
acts as the master is named the "Business Layer". This fulfils the
same role and is equivalent to the "Master Schema" in FpML. It was
described as a "conceptual model" by Dave Frankel in his editing of
the ISO 20022 documents. The FpML Master Schema and the 20022 Business Layer
are roughly equivalent.
·
The annotations in ISO 20022 were
given the name of the UML feature they use which is "trace". They
annotate the master model/schema and are used to generate the lower layer. They
are described as Decorators because this is the pattern used to annotate the
master model/schema. The FpML Master Schema Annotations and the 20022
Traces/Decorators are roughly equivalent.
·
The lower layer in ISO 20022 that
acts as the message specific definition is named the "Message Layer".
This fulfils the same role and is equivalent to the "View Specific Message
Schema" in FpML. The FpML View Specific Message Schema and the ISO 20022
Message Layer are roughly equivalent.
·
The sole significant difference is
that ISO 20022 allow far fewer features in the annotations:
1.
Change cardinalities from defaults
2.
Rename Elements
3.
Add and remove new Elements
Tony – you know ISO 20022 well. How close is this to what you
were thinking of? Which 20022 layer you were referring to as the model? How
would your proposal work with the Master Schema? Whether the Master Schema goes
into the 20022 Repository or whether it is replaced by something in the 20022
Repository amounts to the same outcome and start point, so is it a significant
difference in path? Do you think that if we described the Master Schema as a
Conceptual Model it would preclude creating a conceptual model in a different
technology later?
I propose we take the feature-set of FpML Master Schema Annotations and
use them as a submission to ISO 20022, because Brian’s work is richer in
this area than 20022’s. This would ensure better support for FpML in
20022. Are we happy to do this? Tony, Marc, Andrew, Andrew, or I could do this
as we are participants in both groups.
Matthew Rawlings
+44 791 539 7824
-----Original Message-----
From:
Sent: 07 January 2008 13:42
To:
Subject: Re: FW: FpML-AWG View generation syntax
Per
model. Put it this way - I like to think that at some time in the
future,
we will have a model for FpML in the ISO 20022 repository. That
to me
would be the conceptual model. The existence of that model
wouldn't
preclude the existence and use of a master Schema as it is planned to
use
it now.
So, I guess that I don't like the idea of identifying the master Schema
as
the conceptual model, because I don't want any kind of
"overloading of
nomenclature" being a bar to the future creation of a conceptual
model
that *isn't* written in an XML schema language.
Cheers, Tony.
On Mon, 07 Jan 2008 12:18:28 -0000,
<matthew.d.rawlings@xxxxxxxxxxxx>
wrote:
> The point I was making is that "If it walks like a duck and
quacks like a
> duck, I would call it a duck." We use the Master Schema as a
conceptual
> model to generate view-specific schemas, and it has the
characteristics
> of
> a conceptual model. Ducks come in many shapes, sizes, and colours
and in
> the same way conceptual models differ from each other, but they
are all
> still conceptual models.
>
> I would call it a duck.
--
Anthony B. Coates
Senior Partner
Miley
Experts In Data
Mobile/Cell: +44 (79) 0543 9026
Data standards participant: genericode, ISO 20022 (ISO 15022
XML),
UN/CEFACT, MDDL, FpML, UBL.
http://www.mileywatts.com/
-------------------------------------------------------------------------------
To unsubscribe: Email majordomo@xxxxxxxx with a blank subject line
In the body include the line: unsubscribe awg youremail@address
To view archives: http://www.fpml.org/_wgmail/_awgmail/threads.html